_ _ _ _ __ ___ __(_) |_(_)_ __ __ _ (_)_ __ \ \ /\ / / '__| | __| | '_ \ / _` | | | '_ \ \ V V /| | | | |_| | | | | (_| | | | | | | \_/\_/ |_| |_|\__|_|_| |_|\__, | |_|_| |_| |___/ _ _ _ ___ _ __ ___ __ _| | | _ __ (_) ___ ___ ___ ___ / __| '_ ` _ \ / _` | | | | '_ \| |/ _ \/ __/ _ \/ __| \__ \ | | | | | (_| | | | | |_) | | __/ (_| __/\__ \ |___/_| |_| |_|\__,_|_|_| | .__/|_|\___|\___\___||___/ |_| ---------------------------------------------------------------- 2024.04.26 I'm slightly tired this morning. And glucose is up (from ~150 to ~200), having started rising at about 03:30. I'm eating a light breakfast, and hoping it'll drop back down over the next couple of hours. ---------------------------------------------------------------- 2024.04.25 I'm working on the Konilo packaging today. Let's see if I can get this completed. Evening: - the konilo packaging is basically done. I'll finish the docs tomorrow - second endo visit went well. Changing insulin to Lispro for meals & Tresiba for basal. - initial dose of Monjauro is wearing off? I feel slightly hungry, and found myself thinking about food a bit Tried to go to bed early, but Tammie stayed up late reading, so this didn't work out well. ---------------------------------------------------------------- 2024.04.24 Last night was not good for sleep. I started off ok, but woke up at 3:30 and couldn't get back into sleep, so I'm tired :( Work was busy, but not as bad as expected given that we still had several people out. Hoping the next couple of days go as well. I didn't get to work on the Konilo packaging today. Hopefully tomorrow will let me have some time to get the scripts to build the release tree finished. Starting to prepare for sleep (early again). Hoping I'll be able to get a few solid hours tonight. ---------------------------------------------------------------- 2024.04.23 Last night was restful. I feel ok for starting off, but expect I will still end up tired in a few hours. Since starting the Mounjaro, my morning glucose spikes have not been as severe. Hopefully with the U500 (finally being filled!) I'll get even more control. Hunger is down, and side effects are minor so far. I've committed a patch to milo/x which tweaks the screen redraw code. This improves the overall performance a bit. Other than this: I made a bunch of small improvements to the konilo.sh and started preparations for 2024.4 release of Konilo. ---------------------------------------------------------------- 2024.04.22 Sleep was subpar, and we're going to be shorthanded at work once again. Ugh. Good news so far: my morning glucose is only at 238, and hasn't had a major spike yet, so I'm hoping it'll be able to drop a bit over the next few hours. It's evening now (about 20:00), and I'm starting to prepare for sleep. Dustin & I had lunch at On The Border (I skipped on the tortillas, but did eat the chicken & onions, and a few chips & salsa). No big spike from this. The spike from dinner was higher even though dinner had less carbs. A good news: the pharmacy finally got the script for the U500, so I can pick that up tomorrow. And now off to bed. ---------------------------------------------------------------- 2024.04.21 Trying to avoid anything stressful to help manage glucose. It's been a decent weekend, though my numbers are still high. Not yet certain, but some slight effects may be appearing from the shot of Mounjaro. I've done a bit of Swift coding, on Paipera. I'll need to bring it into Xcode to finish (I don't know how to do fonts and such inside Swift Playgrounds), but I think the rewrite will be done in early May. For right now I'm not quite feeling well, so am going to bed early in hopes of being well rested for tomorrow. ---------------------------------------------------------------- 2024.04.20 The pharmacy was able to fill the Mounjaro, so I've started on it. I'm hoping it helps. I decided to have a very light lunch (avocado & an egg), and to leave anything heavier to dinner. My glucose has still been elevated today, I'm starting to use more of the R to compensate as I wait for the U-500 to be dealt with. No programming today. I probably will do something tomorrow. There are several unfinished projects in the blocks, so maybe I can get a bit further on one of them. (One coding thing: there's an oddity with ABE's text insertion on characters other than the first in a line. I need to look at this to figure out the reason for the misplacement). ---------------------------------------------------------------- 2024.04.19 I'm feeling modestly rested this morning. My glucose is up, but that's due in part to being told to stop taking the R insulin. I haven't received the script for the U500 yet though, so will be calling my doctor sometime today. I'll take some R to try to get back below 250. Annoyance: I called their office during lunch, but had to leave a voicemail. Now several hours later, the office is closed until Monday, and nothing changed at the pharmacy. I have enough R to get through the weekend, but would have preferred to start the new stuff on the days I'm off work. A bit of good news: the preauth for the Mounjaro was approved, so I might get that tomorrow. ---------------------------------------------------------------- 2024.04.18 Sleep was, while not as long as I'd have liked, pretty good. I will try to go to bed early again today. I'm still trying to work out the remaining bugs in ABE. There is an issue with text entry that's not at the start of the line. Busy day for health stuff. Finally was able to see an Endo for my diabetes. She wants me to make a lot of changes to my meds, so I'm waiting for scripts to be filled. (I may have to call her office tomorrow regarding one of them). Also had a dental visit, with teeth being cleaned & four cavities being filled. First visit in ~ 30 years. Going to bed early again I think. Goodnight! ---------------------------------------------------------------- 2024.04.17 I've switched one of my keyboards (a Logi K380) to a newer one (a Logi K380S). It feels basically the same, but has one change that's proving annoying: the Fn and CTRL keys have been swapped, which breaks a bunch of finger memory. Work was busy. And a bit stressful. But the evening has been nice. I went to the chiropractor. I'll need a few more sessions I think. For the evening, I've done a bit of writing and testing some adjustments to milo/x. I might be able to get the screen draws working more quickly soon. Going to bed earlier than normal. Goodnight. ---------------------------------------------------------------- 2024.04.16 Last night's glucose spike started earlier and wasn't quite as high, so I was able to get some sleep during the night. Saw a CVE on PuTTY. I'm now attempting to use SSH under the Windows Terminal instead (on my work machine). I'll try this for a few days to see if it works ok. I was able to finish updating the Konilo image & blocks for Napia. It's now in sync with the main Konilo system, and I'll soon have an automated patching script. ---------------------------------------------------------------- 2024.04.15 I thought Saturday night's glucose spike was bad. Well last night was worse: it jumped to over 400, and stayed up there for several hours. Getting it back to a safe level required 143 units of insulin and six hours. So I'm exhausted from being up dealing with this. I wrote a document that explains Termina in more detail. It may be helpful for those seeking to make use of it. ---------------------------------------------------------------- 2024.04.14 High glucose last night. But ok during the morning. Arland finished the Konilo parser/structure printer. I've run it against some of the blocks, and it looks good. (There will be a patreon post on this sometime soon). ##forth has been more active than normal, with actual Forth discussion. I'm happy to see this. ---------------------------------------------------------------- 2024.04.13 We had Ben's birthday party today. It was a quiet gathering of just our family and my Mom & Dad. I think it was a pretty nice time, though both Dad & I ended up sleeping a bit during it. I discussed the idea for the parser/structure printer wih Arland and he has a mostly functional implementation written. With a few display fixes it should be good. ---------------------------------------------------------------- 2024.04.12 The sleep project went well. I was able to get some sleep from 17:00 - 19:30, and then spent a couple hours reading before going back to bed at 00:15. Sleep from there went well, with a few short awakenings before getting up at 04:45 this morning. It was busy at work again. I've done some X11 programming this evening. I'm trying to make ilo/x & mult/ilo/x have faster redraws of the display and better latency on input. Arland did some work on a syntax parser for Konilo. I'll be looking at this soon. ---------------------------------------------------------------- 2024.04.11 Had an issue with the air conditioner last night, so it was hot. I've managed to fix it this morning, but it'll be another day in which I'll be exhausted. It was a busy day at work. We had half of our people making del- iveries, and a "lunch & learn" with a vendor as well. This left us shorthanded all day. I'm going to bed early tonight in hopes of some sleep. ---------------------------------------------------------------- 2024.04.10 I got a bit of sleep last night. Not enough, but I'm a little better this morning than the last couple. Plus my blood sugar was at 144 when I got up, so starting below 200 is nice. For Konilo, I rearranged some of the system blocks a little. Dentist appointment. Initial examination was done, I go back in a week for a cleaning and repair of four cavities. ---------------------------------------------------------------- 2024.04.09 I'm very sleepy today. The last several days have not been good for blood sugar levels. Maybe I'm a bit sick or stressed more than normal? No programming today. ---------------------------------------------------------------- 2024.04.08 Starting another week. This afternoon is a solar eclipse, which may be interesting to see. Otherwise, it should be a normal day. I've worked on the manual a bit. ---------------------------------------------------------------- 2024.04.07 On Android, I have begun using Unexpected Keyboard, which is pretty nice when paired with Termux. I'm still getting used to some of the layout, but it works well in landscape, letting me run Konilo in a large enough size to be readable. I'll still prefer a physical keyboard, but this is enough to let me do at least a little writing with just the phone. ---------------------------------------------------------------- 2024.04.06 I've written an ilo in Retro, It's slow (taking over 3 minutes to load (std) on my phone). I'll put in some effort to try to improve this. On my Android phone I started testing non-Samsung software keyboards. So far, nothing seems good in landscape. gboard is decent in portrait, flesky feels buggy and has some weird flashing when I hit shift. I used to use a Thumb Keyboard and a Hacker's keyboard, but both aren't working well for me. I kind of would like something like the Pinephone keyboard that would work with this S22 Ultra. It'd make using termux much nicer. ---------------------------------------------------------------- 2024.04.05 Too tired to do any significant programming. I fixed more bugs in Retro's GNUmakefile and set up a local environment for doing a heavily commented ilo-unix (using unu). This will take a few days to go through and add commentary to, but it'll hopefully be helpful. ---------------------------------------------------------------- 2024.04.04 I fixed some issues in the GNUmakefile for Retro. For Konilo, some size optimizations in the (std) blocks. I've shaved 14 cells. (I'm hoping to get the overall size back down to under 9k cells) ---------------------------------------------------------------- 2024.04.03 I've made a change to the shell script that publishes the WISP export. It'll now update my entry on plan.cat with the latest block. The konilo.sh has been rewritten again. It's under 100 lines now (and still shrinking a bit). I'm getting happier with it, so it might be time to start publishing it again. First doctor visit in several years. I have a bunch of tests scheduled, and then need to schedule visits with a number of specialists. ---------------------------------------------------------------- 2024.04.02 I probably should have taken today off; I really didn't get much (any?) sleep last night, so as of 0549 I'm very, very tired. I built a current Python on the ancient server we are using for convey. In a few small tests it seems to be working, so now I am going to look at installing it alongside the existing copy. ---------------------------------------------------------------- 2024.04.01 Last full vacation day. We had french toast for breakfast, then drove home. Alex got sick & vomited during the drive, so we had to stop to clean that up. I dealt with a few work issues during the drive, as well as a banking issue. We arrived at home about 1700. Dinner was a few pepperoni & cheese sandwiches. I did a bit of work on redoing the konilo.sh script; it's mostly working on OpenBSD. I'll need to do some testing on other hosts, but I like this one better. Not really looking forward to returning to work tomorrow, but I suspect if I don't go in, I'll still be working remotely again. ---------------------------------------------------------------- 2024.03.31 With gzexe(1) it looks like the konilo-$ARCH-$HOST single file thing will be doable. I have binaries for aarch64-linux, amd64-openbsd, and amd64-freebsd now. With a prompt from Arland, I added `a:new` which takes a length and a name, and creates a new array. ---- Vacation Notes Breakfast was a nice potato & egg bake w/bacon and green onion. Had another bad glucose spike, but got it under control after a few hours. We visited w/Tammie's family for most of the day. ---------------------------------------------------------------- 2024.03.30 So the ilo w/embedded rom & default blocks was easier to put together than anticipated, so I have a working test under my termux system. It's about 335k when compressed with upx. Not terrible. I'll need to look at storing the blocks in compressed form for other OSes w/o upx though. Big glucose spikes with breakfast & dinner. Ugh. Hoping to get it back down during the night. For now (23:31) it's still over 370 though. ---------------------------------------------------------------- 2024.03.29 Breakfast was quiche w/tomato & green onion, sausage, coffee cakes, and a backed apple dish. (I skipped the coffee cake, and only at a little of the baked apple). After lunch, some programming. I brought in the `quad` words from Retro as a loadable block, did some chatting in ##forth, and a little work on the convey backend. Tonight (probably) I will configure one or more of the domains we have to start the actual testing parts. After a discussion with Arland, I have begun an ilo branch with an embedded copy of the rom and initial blocks. This is very much experimental at this point, but might be useful in the future... ---------------------------------------------------------------- 2024.03.28 Vacation, day 1. Sleep was ok. Most of the day was spent on the drive up to the B&B we are staying at. Things at my employer were apparently hectic. I spent a lot of the drive reading & replying to emails and texts to help resolve problems. Dinner (the only meal today) was pizza. Not bad, but also not good for my glucose. Arland bought me a black & white mocha. It was good, but overly sweet. I might try a cappucino or americano next time. I've done a little work on the convey backend code. Going to try to resolve the login issues during the vacation, and maybe start on the data update code. ---------------------------------------------------------------- 2024.03.27 It was a a bad night for sleep, but vacation starts tomorrow. I am looking forward to the break. abe still has a few little issues to look into, and needs more testing, but it's functional. I'll move it to the public blocks this weekend, after I get to do a little documentation on it. ---------------------------------------------------------------- 2024.03.26 Some good sleep last evening, though the during-the-night part was interrupted by a glucose crash. Still feel ok overall today. abe (the arks block editor) is now in my blocks, so it's much more comfortable to work on. I've been able to resolve the bug in the partial line entry. The windowed view is also working, so it's now functional enough to use. I'll need to write a fallback mode for editing normal blocks. ---------------------------------------------------------------- 2024.03.25 Another QOL change: tuhi & rem now display the current block number in their title tags. Sleep last night was not as good as Saturday night, but was better than most days last week. I'm hoping the upcoming time off work helps me reset more. ---------------------------------------------------------------- 2024.03.24 I was able to get some solid sleep today, so am feeling better thank normal. Programming stuff: a new Konilo/iOS is waiting for review from Apple. I moved (rem) into the system blocks area and merged it. The `s:get/token` was causing a crash if I attempted to back- space past the start of the token. This is now corrected. A similar bug was found and fixed in `s:get/line`. The updated Konilo/iOS fixes a bug in the DEC escape sequence filter code. This resolves a crash. ---------------------------------------------------------------- 2024.03.23 I merged in the work-in-progress REM rewrite. It's functional, though I still need to add a few small things like "leap" to it. Today has been a day for rest. I was able to sleep/rest in, and spent some time catching up on reading, doing a bit of chat in ##forth, and a little time tweaking my X11 configuration. I want to write a small X program to let me quickly (re)launch ilo, and maybe a couple of other programs. I might start on this before bedtime today. ---------------------------------------------------------------- 2024.03.22 Almost four hours of sleep last night, was quite nice. The REM rewrite is progressing nicely. I'm doing a keyboard layout that will work with both qwerty & dvorak, and making use of color as part of the editing process. It'll be pretty nice when done. Arland noticed a bug in my modification to `p`, which I have corrected. Nearly midnight, so I'll stop for now. I have planned out the remaining keymap for the updated REM, so should be able to have it done tomorrow or Sunday. Then I'll need to document it in the manual. ---------------------------------------------------------------- 2024.03.21 Back to work on the hypertext manual; I'm expanding the sections on the basic editor and termina currently. It's nearly at a 1/4 of the reserved space (248 of 1024). I extended (termina) to add support for displaying a title as part of the hints area. I was not sure if I would actually like this, but it ends up being a very helpful touch. ---------------------------------------------------------------- 2024.03.20 Not enough sleep once again, but seeing some improvements to my blood sugar (hopefully this stays consistent...) in the morning. I've finished the first part of the updated glossary. All words are now described and the blocks are updated. Still error checks to do, and adding more examples, but this is a solid jump. An annoyance at work: Windows 10 is pretty useless. I'm at a point where trying to open a spreadsheet causes the file manager to hang until I kill the process. Opening via the File>Open dialog works fine though. ---------------------------------------------------------------- 2024.03.19 Still tired, but was able to sleep a bit last night, so today was better. I continued on my work to expand the glossary, ending with about 120 words remaining. So there's been a lot of progress on this over the last few days. I'm hoping to wrap up this tomorrow or Thursday. ---------------------------------------------------------------- 2024.03.18 No sleep last night... I'm now a week and half away from a short vacation. I hope it's restful; the lack of sleep lately is becoming a problem. I have added blank pages for all the words in the glossary and begun filling in the details. My pre-lunch work ended with 83 complete. There's still a lot to go to finish the 432 words. ---------------------------------------------------------------- 2024.03.17 Moderate sleep. I was up until around 4, but was able to get a little sleep on and off between then and 9. Blood glucose was on the high side all day. My dinner was just chicken, cheese, and a bit of enchilada sauce. This spiked my glucose by about 110 points, but I was able to get it back down to around 140 by midnight. ---------------------------------------------------------------- 2024.03.16 I didn't plan to do any programming, but I did add a small thing to the manual browser last night. [ and ] are now bound to prev and next, so I can quickly flip through the manual blocks (a very helpful bit for things like the *WordsByName pages). I seem to be getting sick again. Hoping it stays mild and clears up soon. For a non-programming thing: I've started looking at the option of moving to a recumbent pedicycle (probably a trike). I miss cycling, and with ongoing back and hand issues, this might make it possible again. ---------------------------------------------------------------- 2024.03.15 Lots of work on the manual today. Only a small part is visible at present (adding in the Teensy related bits), but I have made solid progress towards updating the glossary. All told there are now links to every (non-internal) word, though there's no docs yet for the ~180 or so from the blocks. When I code next (maybe Monday?) I'll be working on getting the stub pages for all the words in place, and trying to get the pages filled in over the next couple of weeks. Since I'm starting to have some hand issues, I've pushed off the week-in-review for a couple of days; it'll probably be done on Sunday. ---------------------------------------------------------------- 2024.03.14 The watchlist code is functional and has been committed. So I'm a little closer to having takawiri ready to actually use. Next up is the string pool preview... I dropped the number of sessions in my personal m/ilo to 3. This more closely matches my actual usage as I seldom need more than three sessions at a time. Only a few days left until the first anniversary of this little log. It's been fun so far. ---------------------------------------------------------------- 2024.03.13 I posted a couple more documents to konilo.org/articles. Over the next few weeks several more are coming. The manual is now at 197 pages. I've added some material on the variants of ilo (ilo/x86, mult/ilo). ---------------------------------------------------------------- 2024.03.12 Another sleepless night... ---------------------------------------------------------------- 2024.03.11 I didn't mention this yesterday, but I added a (short) vocab. This provides (simulated) 16-bit integer math, which may be of use for some host hardware interfacing or in implementing some compatibility things. Getting back into my attempts to find a way to diff the blocks, I started a new C implementation of a block diff tool. It's now able to id the differing blocks. The next step, I think, will be writing the altered blocks to disk. Then a review tool. More work on the manual; adding 14 new blocks and editing some existing ones. It'll pass 200 blocks in another day or two. ---------------------------------------------------------------- 2024.03.10 Daylight savings threw off my sleep schedule a bit. And I had a rough night with my glucose (rose 170 points after dinner, even after I had factored in insulin to cover everthing I ate). Sigh. The ESP32 ilo port is being worked on. I'm still using the Arduino IDE (for now), and have started trying to get the SD card and serial functional. I'll leave the VGA part for a bit later. The ESP32 compilation is *much* slower than the Teensy4 which is quite annoying. I found that Sharp has a ~4" memory LCD at 320x240. I might see if I can order one or two; this, a PS/2 (or other) keyboard, and the Teensy4 would make a nice little Konilo machine. ---------------------------------------------------------------- 2024.03.09 My blood sugar was running high all day again. I have a few ideas as to why it seems worse on the weekends, but need to try to gather a bit more data. I've revived my old formula-based art thing. It's now rewritten and working better, with the "x" graphics block. (I need to give that an actual name soon). Groceries are done, I picked up more insulin & needles, and ordered more Libre 3 sensors. That's enough for today. ---------------------------------------------------------------- 2024.03.08 More internal expansions and cleanups to the Konilo hypertext manual today. I've added a few new pages to cover various points that I felt needed some discussion. I have added the stack guards and some extra optimizations into a m/ilo vm. This is somewhat opposite of my preferred direction (reducing host dependencies), but is much less likely to crash so the tradeoff may be worthwhile overall. I'm not sure if I'll merge this, or if I'll just post it as an extra for patrons. I also put together an ilo.c with the termios stuff from m/ilo; this might be worth merging after I get more testing done. ---------------------------------------------------------------- 2024.03.07 For today, I worked on improving some pages in the manual, and then turned some attention to Arland's blocktools.py. I have done some tests and am starting to add additional comments as I work through it. ---------------------------------------------------------------- 2024.03.06 I finished a short article on doing external interfacing to a Konilo system. I've been able to control a system using a pair of fifo's, which is pretty neat. Soon I'll have a demo of this. No work on the manual today. My hands hurt a bit, so I'll stop for now, and resume tomorrow. ---------------------------------------------------------------- 2024.03.05 The weather has shifted; it's now warmer and rainy. I'm still hoping for one decent snowstorm, but it doesn't seem likely at this point. A number of years ago I started an implementation of an old game, Hammurabi, in RetroForth. This was never completed, but I found my old sources and decided to move it into Konilo. The existing code has been brought into the blocks, and updated, so I'll be working occasionally on it until it's done. The Hackernews thread on Konilo has brought up some areas of lacking documentation that I'm going to try to address as my general writing continues. Enough for now though, I'm off to bed. ---------------------------------------------------------------- 2024.03.04 I had trouble sleeping again. Not quite unexpected - it's been warmer than I like, and today will be a little stressful at work (we have a new person starting, and I get to train them). While I want to try to get the new listener finished, I can't quite get my head into the right state, so I'm just continuing to write in the konilo manual. ---------------------------------------------------------------- 2024.03.03 Tammie didn't feel well today. I took another rest day, just reading all day. Arland sent me the latest things he's been working on, so I'll have a few new words to add to Konilo blocks tomorrow. I did check the Konilo entry on hackernews and reply to a few more comments. Matthew Keeter has sent me an update ilo.rs implementation. It looks like a solid set of improvements, so I'm looking forward to testing & merging it in the next day or so. ---------------------------------------------------------------- 2024.03.02 Saturday is a rest day. It rained (which wasn't in the forecast), and was a bit windy. This prevented us from doing the BBQ we had planned for lunch, though Arland did roast the meat in the oven instead. I made a peanut butter & soy ramen for dinner. This turned out pretty good, but I'll tweak the amount of peanut butter next time. Konilo hit hackernews today. The server handled the load ok, which was good. I replied to a few comments. ---------------------------------------------------------------- 2024.03.01 Arland started working on a (math) vocabulary last night. This will eventually cover a lot of useful functionality that's not in the base system. I fixed backspace handling in mult/ilo (terminal). Much, much nicer to work with now. ---------------------------------------------------------------- 2024.02.29 Leap day. No sleep (or near enough to it). I'm very tired today. I finally deleted the old ansuli server today, leaving just one VPS (arneson) on Linode. For now I'm leaving that alone, but am going to start evaluating options. Glossary lookups in the manual are working, and I have made a variety of improvements to help with performance. This is merged into the main repository. ---------------------------------------------------------------- 2024.02.28 Spent my time working on the manual expansions. Its now live on konilo.org and has 95 pages (plus approx 250 for the initial glossary). I'll probably be around 110-120 pages by the end of the night. The growth after this will be slower, but should be steady as it's more organic than the rigid manual structure I was trying to do before. A little more programming is needed; I still haven't written the + handler yet or finished the HTML generation for that. Even so I like it. ---------------------------------------------------------------- 2024.02.27 I continued work on the new manual/wiki. The basic HTML export is working, and a number of pages have been converted. I think I'll be able to get it ready to merge into the system blocks in a couple of days, and will have the old content fully moved over next week. Arland is working on a Python class for block operations. It's already looking good to me. Tomorrow I will start writing some tests for it. We'll use it to help improve some external tools like the wiki & manual export code soon. Last bit for tonight: doing some reading on floating point. We'd like to get a soft float model working eventually. More on this eventually... ---------------------------------------------------------------- 2024.02.26 Arland & I discussed the manual project and decided to make some changes. The embedded manual is being replaced with a topical wiki (which will tie into the Glossary data as well), and the full manual will be formatted to be printed as a physical book. My initial code is written, drawn mostly from the (wiki) blocks. This will form a superset of the wiki code (due to adding in the Glossary blocks), and should (due to a smaller, more constrained space) be faster when done. After I finish moving the existing stuff into the new framework I'll write an HTML export (using Retro) and begin publishing it to konilo.org nightly or weekly. ---------------------------------------------------------------- 2024.02.25 Arland is home for a week, so we'll get a few things done. He's already provided me an updated Procreate file for the Konilo logo, and is working on his Graphica code. I ordered a chest freezer to allow me room to store frozen food in larger quantity. Tomorrow I resume programming. I'll be working on a variety of small things this week. It'll be interesting to see how much can be completed. My hands are not in good condition at present. ---------------------------------------------------------------- 2024.02.24 Still not a great night of sleep, and my glucose was high most of the day. I decided not to do any programming today, and to try to rest until Monday. ---------------------------------------------------------------- 2024.02.23 I couldn't sleep last night. Not a good start to the day. The code for the watchlist in takawiri is nearly done. Another day or two for testing and I should be ready to commit it. Then I'll spend some time to finish the interface scaling code (for the width). I am hoping to wrap up the programming for this by early April. An aside from programming: I'm drained. I need to get some sleep soon, so might try to take a break from computing for a few days (or maybe a bit longer). ---------------------------------------------------------------- 2024.02.22 Busy day at work, with a lot of quotes and followup on orders. Today is the start of a transition back to a lowcarb (but not keto) diet. I'm hoping to get my blood glucose stable over the next month. I worked on (tuhi) a bit today. The help screen was redone, and the "S" command was cleaned up a bit. ---------------------------------------------------------------- 2024.02.21 I've had to use Windows for much of my life. It's grown much, much worse over the years. Both it and Office decide to do updates at random times, with no option to delay. And in the case of Windows, it'll just inform you that it's going to reboot with no alternative. Very frustraiting. In Retro, I merged in a Zig implementation of Nga and did some work on Takawiri. The TOB code now scales based on the terminal height. I have not finished with the watchlist code, so will continue working on that tomorrow. Evening project: updating the 68k mac to use the latest ilo rom & blocks, and updating the ilo in swift playgrounds. ---------------------------------------------------------------- 2024.02.20 Today's adventures in programming were: adding a new "R" key to (tuhi) to run the current set of blocks, and doing more work on the manual and glossary. I'm a little less than a month from the first anniversary of WISP. I'll need to decide how to handle the publication side of this soon as I don't really want an ever growing single file since that's wasteful in terms of transfer & time. Maybe I will split it by year? Or perhaps by year, and have the main text be the last month of entries? Tomorrow I will be working on the RetroForth listener again. In hopes of getting the watchlist code mostly complete. ---------------------------------------------------------------- 2024.02.19 Arland reminded me that I haven't uploaded WISP data in a few days, so I've done this this morning. I have a good variety of implementations of ilo. But one thing is lacking: a properly safe version that checks for errors and responds gracefully. (There are some that do partial checks for a few cases, but none that are exhaustive). Today I start on this. Other than the above, today I'm also doing a little more for the manual. I'll hope to have something to share on this soon. ---------------------------------------------------------------- 2024.02.18 It was a typical Sunday morning. Get up, go to church with my family. My glucose remained high, but I was able to get it to an acceptable level by half noon. My sister is visiting from Texas, so we went to my Mom's for the afternoon. It was a nice time. Arland & I spent a little time testing things in ilo on Swift Playgrounds and his Python implementation. I'll start making a few changes to help with performance soon. Going to bed early again. I'll have some programming time in the next few days, and I have a small list of projects to work on. ---------------------------------------------------------------- 2024.02.17 After yesterday's glucose issues I went with less insulin last night, so my numbers this morning are higher than ideal. But at least I was able to get some sleep. No coding today. I spent time listening to music and playing Stardew Valley. ---------------------------------------------------------------- 2024.02.16 It's Friday. Last night I had a glucose crash, so I'm now very tired. Work was slow, so it was at least possible to rest a little. I am going to bed early tonight though as I still feel off. ---------------------------------------------------------------- 2024.02.15 I'm doing some work with ilo & x today. Following Arland's desire, I'm using a 720x720 X display, but with a 640x640 ilo display. I like the border. Running with no window manager; just X + ilo. With some changes to the block locator code the * loader broke on missing block sets. I've fixed this in my blocks and will push this to the main repo tomorrow. A little progress on the ilo-only computer project. I have the keyboard plates designed and ordered. Next will be sourcing the displays, then working on the rest of the enclosure. More on this soon... ---------------------------------------------------------------- 2024.02.14 Bob retired today. He had been working at Denney for 47 years. I am now the longest running employee at our branch, with a bit over 20 years. It'll be odd without him. Not doing programming today. ---------------------------------------------------------------- 2024.02.13 It's supposed to snow today. Mixed with rain. Joy. I'm dabbling a bit with my ilo-on-x setup. Currently using just mult/ilo w/320x240 & the small 8x4 font. My desktop is just this with a small ~10 pixel border. I'm running with a zoomed in view via VNC. It's pretty usable, and after a few minutes of using it the font remains readable. A little future direction: I should do a UTF32 ilo (prob. just mult/ilo and mul/ilo/x). This would make strings more logical as in recent Retro releases. Other bits: PS/2 keyboard and mouse should arrive late this week and I'm ordering an SD card so I can work on the ESP32 stuff. ---------------------------------------------------------------- 2024.02.12 I've been using OpenBSD.Amsterdam for a month now. Overall I am happy with it. My only (small) complaint is the i/o speed for disk access. It's a bit sluggish in this regard. But it's been stable and is fast enough to be usable, so I see no reason to seek alternatives. ---------------------------------------------------------------- 2024.02.11 I have an initial import of konilo.pali into the blcocks (at 45056 in my set). I'll need to do a lot more work on updating the comments to use the `c` directive instead of being separate and will want to reaarange things to break between functions more cleanly. But this is another step towards eventually being self-hosted. Healthwise: still sick and no real improvements over the last few days. This is very annoying now; I really just want to feel better again. ---------------------------------------------------------------- 2024.02.10 My early programming days were using BASIC. I still have some fond memories of the QBASIC environment. Today I considered the language a bit. Specifically READ/DATA. This is something I did not use often, but which was useful at times. So I decided to write an implementation for Konilo. There are a few parts. A DataSource variable holds a pointer to the stored data in memory. A parsing word, data:add, reads and stores numbers, ending when `end` is encountered. And data:read reads from the stored data. I have a couple of thoughts playing around in my head regarding future expansions of this. But that will be a topic for a later time. ---------------------------------------------------------------- 2024.02.09 RSV likes to linger. I'm much healthier than I was a few days ago, but still have a cough and feel somewhat unwell. Work on the (nonix) replacement has moved forward a little with the new "dir"ectory words being nearly complete. (I want to write a "dir" combinator to reduce some repetitive bits). I think this part will be done over the weekend, then I'll spend some time on the move/copy/rename stuff. I have done a little on the new Retro listener, starting on a bit to use the d:source data and tags data to help locate definitions. ---------------------------------------------------------------- 2024.02.08 I'm down to a lingering cough and some chest tightness. Hoping this improves over the next couple of days. The first changes relating to my (nonix) replacement have been added. This slightly shrinks the initial memory footprint (though it will grow larger as the work continues). ---------------------------------------------------------------- 2024.02.07 Health is slowly improving. I did a little more website work (now live), and sent Arland a copy of the ilo + graphica in javascript since he's planning to do more work on it. Early work on a replacement for (nonix) is progressing now. This will add more `block:` words and a new set of command line tools for working with the blocks. ---------------------------------------------------------------- 2024.02.06 Still sick, but starting to feel better. I did some work on the HTML for konilo.org. It's not live yet, but should be in a day or two. With a bit of luck, tomorrow I might be able to start actually working on projects again. ---------------------------------------------------------------- 2024.02.05 It's been 9 days and I'm still feeling sick. Back at work today, so it's a little more normal. As of lunch time I feel a little better. The tacos are driving my glucose up, but I think it'll be manageable. Tonight I'll try to work on Konilo/iOS a bit, and maybe on Retro or iOS as well. I'm still stuck a bit on the SwiftUI changes in this though. ---------------------------------------------------------------- 2024.02.04 8 days in, and still sick. I'm just continuing to rest, in hopes that next week will be better. ---------------------------------------------------------------- 2024.02.03 I'm pretty sure at this point that I had the flu, and am now dealing with RSV. Quite annoying to still be sick enough to prefer being in bed to being active. Given the ongoing illness, any programming is still on the back burner until my health improves. I am doing a bit more writing in the new manual, as I feel up to it. This is only for very short periods though; I'm getting quite tired very easily at present. ---------------------------------------------------------------- 2024.02.02 I did not end up needing to report in for jury duty, so took a day to rest. I feel a little better now, but still far from healthy. ---------------------------------------------------------------- 2024.02.01 I may be starting to get better. Or, at least my throat is less raw, so I'll consider that a win. I'm trying to spend a little time on the (nonix) blocks, to see if I can expand them into something more usable. I now have a nice little SSH + VNC setup on my Android phone. I can work comfortable from it via my iPad, in either a terminal or graphical mode. I'm using 2bwm, and a 1080x540 display size. It's a little cramped, but is readable on the iPad display. ---------------------------------------------------------------- 2024.01.31 At the end of January and this illness just keeps lingering. My throat is still raw, and I'm seriously lookig forward to just being able to rest in bed this weekend. ---------------------------------------------------------------- 2024.01.30 Still sick. My throat is raw, and it hurts to swallow or speak which will make work today difficult. Dizziness is reduced so far. I made it through the work day, and did a little reading in the evening. That's enough for today. ---------------------------------------------------------------- 2024.01.29 Still sick, but have uploaded my latest blocks to the dev box. I'm trying to make it through a work day, but it's questionable so far... It's now 20:41. I made it through work, but still don't feel well. I took a covid test which was negative. So that's good at least. In mult/ilo, I am testing binding CTRL+C to restart the current session. If this goes well, I'll probably commit the change tomorrow. I now have X11 working with Termux, so I can do ilo+x work more easily now. ---------------------------------------------------------------- 2024.01.28 Still sick. Taking a full rest day. ---------------------------------------------------------------- 2024.01.27 I am sick. I spent most of the day in bed, dizzy and with an increasingly sore throat. I was able to do a little work on Konilo, getting the size down to under 8k cells (excluding buffers). ---------------------------------------------------------------- 2024.01.26 My back & right arm/hand are both getting worse. But I'm doing a little programming today. Specifically, recoding much of extend.konilo in assembly. This will make the core smaller, faster, and get me a step closer to being able to self-host in the future. While I was not expecting to finish today, the entire recode is now done (though some bgs may exist). 8,116 cells now. I'm happy with this. ---------------------------------------------------------------- 2024.01.25 Still having a lot of issues with my back and right arm. Won't be writing today. ---------------------------------------------------------------- 2024.01.24 I woke up this morning with a massive headache. After several hours, it is still lingering. Additionally, I am having numbness in my right arm & hand. ---------------------------------------------------------------- 2024.01.23 For the last several days I've been having back pain. It's not improving. This is making sleep difficult, and it is also making mundane tasks like walking agonizing. I don't know what to do for it. While working on the new listener for Retro, I found and fixed a bug in `s:get-word`. This was followed by noticing that `n:put` was using the temporary string buffers, which was one source of my evaluation issues. These have been corrected. ---------------------------------------------------------------- 2024.01.22 I continued writing code for Takawiri today. The Watchlist is now functional (though addresses are hard coded still); and the interface is starting to make use of the terminal configuration for sizing purposes. Other stuff: adding a -p for plain source files, an ioctl() device, and some experiments with colors. I'm happy with my progress. Tomorrow I will try to finish the layout sizing code, then it will be time to finish the new string evaluator. ---------------------------------------------------------------- 2024.01.21 I woke up with a major headache today. Not an ideal start. But otherwise not too bad. The normal complaints: my back hurts, blood glucose is a little higher than ideal, and my hands are hurting a bit. Given the wrist/hand pain I'll probably not be doing much on programming or writing today. I ended up deciding to build a 32-bit DOS binary for ilo, using DJGPP. This is much faster than the 16-bit binary, but the need for a CWSDPMI binary is annoying. I'll see if I can find an alternative path here. ---------------------------------------------------------------- 2024.01.20 Glucose numbers are still high, peaking at about 297. I'm back below 180 as of 1300, but have been fasting since dinnertime yesterday. I've continued working on takawiri. This includes some interface updates and the start of a new Watchlist section. I think it will be usable by the end of next week. ---------------------------------------------------------------- 2024.01.19 Today was interesting for blood glucose. Until evening I was trending low (even dipping below 70 at a few points), but in the evening it jumped to 227 from 93, then stayed over 180 since. Not fun. I added the new listener code to the RetroForth repository and Arland's latest ilo.py to the ilo repo. The new listener is getting closer to being usable. I started on a new `s:evaluate` for use with it; this will solve the string buffer consumption issues. ---------------------------------------------------------------- 2024.01.18 After yesterday's site change for the CGM, I ended up going high (347!), and now (0541) am trending down (at 93). No morning rise yet, which may be an issue as I dosed expecting one. I have been doing work on the new listener today. The TOB code has received some bug fixes, and is now split out into a library (there wil be more of this in the future). Some use of color in the interface is now done. I found a new bug in the TOB code which I will try to fix late tonight or sometime tommorow. Evening update: other than lunch (which went horribly wrong), my blood glucose has been good today. Tomorrow I'll try to stay in range all day, though I'm still struggling to control the dawn rise. ---------------------------------------------------------------- 2024.01.17 I'm having a hard time getting adaquate sleep once again. This has been getting worse over the last two weeks, which has me a little concerned. Today was my first switch of the CGM. Removal of the old one was a little painful. The combination of skintac, libre 3, and cover patch held up well. It took several minutes of work to remove the old adhesive. ---------------------------------------------------------------- 2024.01.16 For Konilo, I've removed most of the remaining uses of DTC from the kernel and made a few optimizations via use of inline code instead of calls. This, coupled with use of the reserved space in the system buffers, has reduced the initial memory footprint to under 9k cells and made things slightly faster. A second Konilo thing: I have written a prototype for a self hosted image build. This is not something I'm going to do more work on now, but I'm confident I can fully implement it if/when the need arises. ---------------------------------------------------------------- 2024.01.15 I have started work on a faster ilo again. I'm trying to see if caching the blocks in RAM will lead to a faster overall system. (This will increase the initial load time, but may be of benefit on hosts with slow i/o, but lots of RAM). Other than this, I did a bit to make fossil pulls/sync faster, and to improve my ability to easily push changes from my perso- nal set to the default ones. For health, I started with a prototype of the IOB tool. I don't quite like how it came out, so I'll rethink it a bit and try again. ---------------------------------------------------------------- 2024.01.14 I've managed to keep my blood glucose in range all day. I'm hap- py with this, though it took a toll: I had to sacrifice most of my little sleep, and use more insulin than I'd prefer. I'll pro- bably got out of range tomorrow. Speaking of blood glucose, I decided that I need to write a pro- gram to track my insulin on board (IOB). I'll need to think on the details, but I have some rough ideas on it. The ilo repository now has a larger block set, which will be much easier for me to keep in sync with my set. I posted some details on this to Patreon on the 3rd, but will write more on it soon. ---------------------------------------------------------------- 2024.01.13 I received a jury summons. Joy /s. I'll find out on Feb 1 if I have to report in for this. Today will begin a lengthy project: writing proper documentation for Konilo. I'm intending to have a very complete thing once this is completed, including a tutorial, which I've failed to actually complete in the past. I also had some thoughts last night on how to make Konilo self hosting (apart from the ilo computer). I'm not sure I want to actually do this, but it'd be interesting to have the option to do so. ---------------------------------------------------------------- 2023.01.12 I was able to get the block writing code working in ilo.ha, so the first Hare implementation is done. The migration of the forth.works server to OpenBSD.amsterdam is completed without issues. I'll turn off the Linode on Monday, and probably decomission it at the end of the month. And I think that's it for today. ---------------------------------------------------------------- 2024.01.11 I was able to get a little sleep last night. This is good, but caused me to miss my early AM insulin, so my morning glucose is higher than I'd like. One of my goals for 2024 is to finish my server transitions. For this the forth.works server (ansuli) is going to be moved away from Linode. I've decided to switch to a system on OpenBSD Amsterdam. Today I paid for the first year; tomorrow I will begin migrating my stuff to it. The migration will take a day or so to copy over the files, and I'll need another day or two to ensure all the automations are working as expected. So maybe Monday for the migration to be complete? ---------------------------------------------------------------- 2024.01.10 After a few hours work last evening I was too tired to write. I didn't get a good night's rest, so am still quite tired today. It's the end of the day now. Work was busy, I feel a little sick and need to try to sleep now. I was able to finish the 2024.1 release of Retro. Hopefully tomorrow will be better... ---------------------------------------------------------------- 2024.01.09 I setup Hare (harelang.org) on my dev box today and began an ilo in it. This is still very incomplete (no keyboard or blocks yet) but the rest appears to be working as expected. ---------------------------------------------------------------- 2024.01.08 I forgot to take a dose of insulin befrore the morning rise from the dawn phenomenon. But thankfully the spike was only 53 points (maybe the metformin is helping?). Today I have a small email-related program to write for work, & normal quotes / orders to process. I'll try to keep stress low and hope it is more relaxed than last week. Since I was taking a break from active computing, I didn't push my blocks to the server for a couple of days. The missing WISP entries will be published tonight. ---------------------------------------------------------------- 2024.01.07 Much better glucose control today. I was able to remain around 220 all day, finally dipping below 180 near bedtime. I kept carb intake low, which probably helped with this. The me- tformin is causing some issues, which I'll use immodium to help. Hopefully I'll adapt quickly to it so this can be stopped. It was another day spent reading instead of coding. I finished the New Management trilogy by Charles Stross. Not as enjoyable as the main Laundry novels, but I did like stepping back into this fictional world for a few days. ---------------------------------------------------------------- 2024.01.06 My morning glucose spiked to over 300, which made the day start off poorly. But otherwise it went ok. I took a day off computer (mostly), went to lunch with Tammie, and did some reading. The Metformin arrived today, so I'm starting to use it today. I hope the side effects aren't too bad. ---------------------------------------------------------------- 2024.01.05 I woke up at 0300 today and could not get back to sleep. Not a good start. 1112: less busy at work, which is nice. I'm getting caught up with both my backlog & the delivery schedules. Still taking the weekend off. ---------------------------------------------------------------- 2024.01.04 Now a bit over 12 hours into using the CGM, and I like it. My arm is a little stiff, but not painful. And the numbers have come through without issue. I removed the old `blocks` tutorial. It's a bit smaller in terms of memory usage, and frees an output line in the startup screen. Two very stressful days at work completed. I am very much hop- ing that tomorrow is quiet. I have a lot to do, and no pati- ence left. Only one programming thing today: I fixed a typo in the life blocks, so it's not crashing on load anymore. ---------------------------------------------------------------- 2024.01.03 I've finished a dry run, then merged in the VOA Special English dictionary. So I have Toki Pona and VOA dictionaries now. This pushes my total block count to 3074. With this I have also finished organizing my blocks (for now), and have update the ilo repository with the latest set. A tool to merge future updates is not yet written. My CGM (Freestyele Libre 3) has arrived, so I'll be starting to use it tonight. Here's to less finger sticks and a more steady stream of measurement data. ---------------------------------------------------------------- 2024.01.02 I attempted to post a reply to a reddit thread, but reddit remo- ved it. No explaination was sent to me. I think I'm done with trying to participate with reddit. In the dictionary project, I merged in a Toki Pona dictionary. This went smoothly, but I'll probably need to write a dictionary lookup vocabulary/tool now. Tomorrow I'll do a dry run of the English dictionary data. This probably won't be merged quite yet, but I will try to have it ready this week. ---------------------------------------------------------------- 2024.01.01 A new year has begun. I've managed to keep WISP up through 288 days of writing last year. It's been fun so far, so I'll try for a full 365 posts this year. I woke up early, as normal, but went back to bed for a few hours to get some more rest. This was helpful as I feel better now. My plan for today is to start writing some things for the Konilo documentation. (Getting this complete is a small priority; I'd like to have Konilo & its core blocks be a finished thing by the September release.) I sent Arland a possible instruction set for ARKS last night, if he likes it we'll start working on implementing it this week. ---------------------------------------------------------------- 2023.12.31 The end of the year, the start of the next. I'm thankful to have made it through 2023 without any serious issues. The last DecAdv log entry is done and published, closing out an enjoyable series of little adventures in programming. I am ready to start 2024. It'll bring some big changes for me. I have a CGM coming sometime this week, which will give me much more insight into my diabetes, some things are changing at work that will help me reduce some stress and get a better work/life balance. And I'm taking some steps to avoid things that make me unhappy. ---------------------------------------------------------------- 2023.12.30 I had a glucose crash last night, and then ended up running high all day. Hoping tonight goes more smoothly. Arland cooked hot dogs & some bratwurst on the grill. These were good. I will enjoy using the grill when the weather warms up a bit. Otherwise, it's been quiet torday. Tomorrow is Sunday, so it should be peaceful, then I have a day off work on Monday... I look forward to this. ---------------------------------------------------------------- 2023.12.29 Good news on the health front: I was able to get a prescription for a CGM (Freestyle Libre 3), so I'll be able to start getting more reliable/frequent data on my glucose state. Next up is to get a new script for metformin... My earbud ... pads? Finally need replacing. The left earbud pad has shredded, and now I need to decide on how to replace it. I did get two years out of the current set, so hopefully replacing the pads will get me another 2-3 years. Heading to bed now; will write again tomorrow... ---------------------------------------------------------------- 2023.12.28 2023 is almost over. As I look forward to 2024, I have a few things I'd like to do next year: - finish the RetroForth new listener - complete a manual for Konilo - migrate ansuli to OpenBSD Amsterdam - learn to use ZFS - get better with ed(1) - setup a local, low power system for development work - work w/arland to complete a portable forth machine - finish the arks spec & get an initial vm written - continue to make changes to improve my work/life balance - find a way to actually get my diabetes under control - take a real vacation from work ---------------------------------------------------------------- 2023.12.27 I did some cleanups to the new RetroForth listener and then spent the evening completing some prototype stuff for Convey. Convey is now partially functional, though I need to finish the backend updates before I can go too much further. (If reading this later than 12/27; I forgot to upload the blocks last night, so this update will appear late - sorry!) ---------------------------------------------------------------- 2023.12.26 My block usage is now at 1542. This is up from a little over 1300 a few weeks back. (There will be a large jump when I add in the dictionary data soon). I'm doing some work with Xcode today. I really hate this. Xcode burns through disk space and RAM. It's really annoying as I don't use it often, so every time I start it I need to get mor updates. . ---------------------------------------------------------------- 2023.12.25 Merry Christmas! I didn't do much on nauhea, but I did finish & commit an update to Konilo. This changes the dictionary structure and handling of immediate words. The net result is a savings of over 600 cells. Other than this, I made a batch of chili, burned my left hand, and did a small BBQ in the dark. It was a nice day overall, but my left hand will probably be hurting for a few days. ---------------------------------------------------------------- 2023.12.24 I took a detour to explore getting a dictionary into the blocks. For this I took the Oxford 3000 word list, stripped out a few, and pulled from some free dictionary resources. Now I need to clean up the output, setting the blocks up. It'll be somewhere over 3000 blocks when done, but will be useful when I'm writing. Otherwise, I spent a half hour on (nauhea), then spent the day offline. It was nice to take a slower, relaxing day. ---------------------------------------------------------------- 2023.12.23 I think the basic entity code in (nauhea) is done. I still need to do a few things for initialization, and flesh out the docs more, but will stop on this for today. With ilo, I have refreshed my experimental vm that uses a single file for the blocks & rom. This actually works pretty well, and might be something to look at actually doing for the native sys- tem. (It would open a possibility of eventually self-hosting the rom build process). Other than that, I'll probably do a little work on Convey later today, if my hands stay ok. I should try to get the HTTP setup done so I can test the latest login process. ---------------------------------------------------------------- 2023.12.22 For a Friday before a major holiday we've been quite busy at work. Work died down after lunch, I was able to finish the month end preparations, including clearing out over 300 old orders from our data tracking system. I might make some changes to my backup scripts soon. I'd like to only run a backup if the blocks have actually changed. Maybe tomorrow night... ---------------------------------------------------------------- 2023.12.21 I'm really tired this morning. I'll start the day with a tea (a Lapsang Souchong from Twinings), and see if that helps. (Update: it did not). Other than my morning work on nauhae, I did a little prototype for a Konilo using subroutine threading. This isn't complete, but the part I converted went smoothly enough that I should be able to do a full STC image in the future. I expect it to be a little faster, but somewhat larger overall. ---------------------------------------------------------------- 2023.12.20 Despite dosing carefully for the carbs I ate last night, eating much less than normal, and stopping earlier, my morning glucose is still much too high (345). For now I'll be returning to taking a greater number of smaller (3-5U) doses of insulin hourly and supplemental doses for any food I eat. Diabetes is terrible, and a huge drain on me mentally and physically. ---------------------------------------------------------------- 2023.12.19 I was tired most of the day. But that's ok. With Christmas comi- ng up, I'll have a long weekend to rest. For this week in DecAdv, I'm focusing on the nauhea game code. This is something I've wanted to do for a while now. As of today, the map editor is almost done. I should be able to wrap that up tomorrow, then I'll start the code to move a player aro- und the map. A small hickup in my plans might be the need to get Convey to a testable state. Tomorrow night I'll start to tie the pieces into a functional prototype. ---------------------------------------------------------------- 2023.12.18 I did not sleep well last night, so the day started off rough. At work, a delivery driver crashed into the air curtain. It, the support framing, and some drywall was destroyed. This was irritating and dealing with it ate much of the morning. Onto happier thingss: I've finally worked out the needed bits to write the ilo + graphics extension for Windows (C#, under .NET) and Arland should be able to port his graphica to this. So no more need for SDL2. And I started on (neuhae), which is a textual interface , tools, and vocabulary for writing small games. ---------------------------------------------------------------- 2023.12.17 Today was a day spent with family, celebrating Christmas. I'll resume programming tomorrow with something new to start for DecAdv. ---------------------------------------------------------------- 2023.12.16 I spent the morning resting, as I now do most Saturdays. This is becoming a habit, but a beneficial one, as I don't get enough sleep during the week. In January I'll be starting a new project with Dustin. (Or maybe I should say resuming; we did some prototyping, but want to try to take it further now). This will be a web & iOS application, with some social aspects (but not intended to be a social netw- ork like Twitter, or Mastodon, or Facebook, etc). I'm still working through some parts of the design, which I want to get right before real programming begins. ---------------------------------------------------------------- 2023.12.15 Optimizations can be fun. Today I rewrote several Retro words in assembly, and moved the full listener into the base image. And began work on the first part of a 2024.1 release. Christmas shopping is complete, and the last items should be arriving tomorrow, just in time for our early celebration on Sunday. ---------------------------------------------------------------- 2023.12.14 A few weeks back, Arland & I discussed borrowing an idea from Unix to create a location database so we could scan the blocks more quickly. Today I took the first few steps towards this, allocating 69 blocks for the data and several for the words to generate and use it. This is likely to be a hybrid project: I think that I'll be writing a RetroForth program to be run under cron. This will run nightly and update the index data. The words to search this will be in Konilo. (I'll have words to do the updates from within Konilo as well, but would prefer to have this be automated). ---------------------------------------------------------------- 2023.12.13 I'm still very tired as sleep was not happening last night. And, I forgot to upload my ilo.blocks, so the WISP for last night won't show up until tonight. For DecAdv, I just worked on the RetroForth argument parser. I'll try to do something more substantial tomorrow. ---------------------------------------------------------------- 2023.12.12 Good evening. Today was spent at work, continuing preparation for the upcoming year end inventory. (I do not look forward to this). At home, I just tried to rest. It didn't work well, so I'm still very tired ash I head to bed. ---------------------------------------------------------------- 2023.12.11 I ported (termina) to RetroForth. It needs more testing, and I still need to write the help system, but some limited tests went well. This, the Konilo block editor, and the x11 words can now be loaded via `library:load`, which will find and load them from either ./library or ~/.config/retroforth/library I've also added another half dozen blocks to my wiki. ---------------------------------------------------------------- 2023.12.10 Blood glucose is high again. Ugh. Just doing a little cleanup on RetroForth's device code again. My hands are less painful today, but I'm just going to relax and not push things. I'm hoping to avoid a full RSI recurrance. ---------------------------------------------------------------- 2023.12.09 Switching my decadv focus to RetroForth & nga for at least a few days. I'll be working on cleaning up some parts of the source. I read https://marc.info/?l=openbsd-tech&m=170205367232026 which has some interesting details on a change coming to OpenBSD. I may need to update the assembly to add a new section for the pinned syscalls. ---------------------------------------------------------------- 2023.12.08 It's interesting to work with a system where I don't feel a need to constantly change the core of the sytstem. With RetroForth, I was frequently making little changes, adding words, renaming, replacing stuff. I don't feel the same compulsion with Konilo. (Since starting on this journey, I've also become happier with just using Retro, not constantly rewriting it). I do like using my languages and tools. With Konilo, it's all in one system. I can just use the same rom and blocks everywhere, with all the data and tools present. RetroForth's close ties to its ost make it very flexible in interacting with the world, but it's a much larger & more complex system as a result. ---------------------------------------------------------------- 2023.12.07 Cleanups to ilo/native: I reorganized parts of the C and fixed a number of formatting issues. Also updated the system blocks in the main ilo repository. Other than this, I now have an ilo/native with working interrupt support and began doing source cleanups to nga-c. ---------------------------------------------------------------- 2023.12.06 Last night was less restful than the previous two. But there is a small silver lining. I read a lot on VGA, and found the register information I needed to get the mode switch working. So now ilo-native-x86 has a new variant with 320x200x256 graphics, (termina) support, and a 4x8 font. Tomorrow I'll do some work with RetroForth. ---------------------------------------------------------------- 2023.12.05 It was another good evening for sleep. I'm hoping today isn't too busy at work. End of dat update: work was busy, but manageable. I'll be happy when the week is over though. More work on mult/ilo and I spent time on graphics in the native system. The graphics aren't working yet though. ---------------------------------------------------------------- 2023.12.04 I was able to get several hours of sleep last night, so today is off to a good start. For #DecemberAdventure today I'm just going to work on a few easy items. I fixed a small annoyance in mult/ilo on unix, and moved a few words out of (nonix). I also added a key binding to see sys:info from inside tuhi. ---------------------------------------------------------------- 2023.12.03 Bad glucose spike last night (peaked at 529). This lead me to be up very late (7AM) before it was comfortably down enough to get some sleep. I spent most of the day with a bad headache. For Konilo/native, I worked on the start of an "fdisk"-like tool. Specifically, I read up on the x86 MBR and wrote words to parse and access specific fields. Due to exhaustion, I'll not be writing more today. Here's to hoping tomorrow is better. ---------------------------------------------------------------- 2023.12.02 I wrote a mult/ilo for the native Konilo system. This is pretty nice. I'll likely be using this as my main system for native use going forward. (I will make sure to keep the basic ilo/native system in sync). I was able to record a short video of it in action and make an mp4 (and upload a copy to YouTube). I may do more of these in the future. ---------------------------------------------------------------- 2023.12.01 #DecemberAdventure has begun. I'm starting with work on the native ilo system. My work today was focused on getting the port i/o and memory exposed via ilo's io device system. With that done, I wrote a few words to use them, then added CMOS RTC and serial port drivers. I plan to start adding in mult/ilo support tomorrow. Maybe as a second kernel option? ---------------------------------------------------------------- 2023.11.30 Today was bad at work; half of the people were outfor various reasons, so we were quite shorthanded. This is very tiring. I worked on updating the keyboard driver for ilo/native. It can now report -1 to -10 for F1 to F10, and -29 or -30 for F11 and F12. With this I should be set to start work on adding mult/ilo support to the native system. A nice update on the native system: I'm now able to boot it on one of the systems we have at the house, and it has a working keyboard, display, and can read data from disk. I've not tested writes yet. With this I should be able to start some tests on framebuffer support on real hardware soon. ---------------------------------------------------------------- 2023.11.29 It's getting cold outside. Down to -8C this morning. I notice the cold more each year. A few years back this would have been comfortable. Now it's harder to deal with. Over the last year I have noticed that I use RetroForth and Konilo differently. I write a lot of code in Retro, and run on my Unix systems. And it's great. But I use Konilo interactively which ends up being different. I always have a Konilo open, and I write little vocabularies to interact with it. There are some full applications (like wisp or the wiki browser), but I'm spending more time creating ways to converse with the system. It's an interesting difference. ---------------------------------------------------------------- 2023.11.28 My multiple ilo (milo) system is working now. I can easily move between copies of Konilo. It's pretty nice, and should be adapt- able to the native ilo system. After uploading a copy to patreon, I've added CTRL+/ to shut down, and shortened some variable names in the milo.c so all lines are now under 65 characters again. I need to decide how many ilo systems milo needs to handle. I set it at 32 initially, but this might be too many? There will be a balance point between the resource use, functionality, and mental effort to keep track of things. ---------------------------------------------------------------- 2023.11.27 I'll start #DecemberAdventure a little early: starting today I will write some code every day through the end of December. I'm probably not going to stick to a single project though. Today I wrote a small RetroForth & shell script to export text copies of blocks to a file in my ~/forth.works/share dir. This makes it easier to share sets with others. Blocks 4992 - 5023 are set aside for the December Adventure. ---------------------------------------------------------------- 2023.11.26 Not much to report today. For Konilo on iOS, I added enough of (termina) to allow a little use of catalogue, tuhi, wisp, wiki, the manual browser, etc. It doesn't do color or cursor positioning yet. In December, I'll be partaking in #DecemberAdventure, where I'll write some code each day. This should be fun. ---------------------------------------------------------------- 2023.11.25 I have now written a multiple ilo implementation. This allows a single process to switch between several (currently 32) ilo instances, each with separate RAM and stacks, but all sharing a common block set. The ilo systems are not running concurrently; only one is active at a time. This is intended to be more useful in the future, with plans to allow switching systems via hot keys on a future hardware device. ilo/native (x86) has better backspace handling now, and I can update the floppy image from OpenBSD now, so future updates will be easier. Now to tackle graphics... ---------------------------------------------------------------- 2023.11.24 I rested in for a couple of hours, then kept an eye on emails. Nothing significant came, so I was able to avoid working. It was nice. For dinner I made a split pea soup. Arland roasted the veg & cubed pork, and I did the broth & split peas. 8 cups of water, eight cubes of bouillon, green onion, herb de provenance, salt, pepper, and garlic. It turned out very nice; I'll make it again. I am going to start reading the multiboot documentation more to see if I can get a VGA or VESA display mode working. I also want to setup testing under GRUB2, but have not done this yet. ---------------------------------------------------------------- 2023.11.23 Thanksgiving today, so I'm off work until Monday. I rested for several hours, then got up. I prepaired and tested a combined update/install kernel, so I will be able to make this all fit onto one floppy. I'll build that tomorrow or Saturday. ---------------------------------------------------------------- 2023.11.22 I'm really tired today. As I lay unable to sleep, I spent some time considering Konilo on x86. I am going to make some alterations to aid in making it easier to get started. There'll be an "install" kernel that will copy the system blocks to disk, and an "update" kernel that will replace the system blocks (other than configuration ones) with the latest set. This will still require a dedicated machine (or at least drive) and a separate boot disk, but will make it much easier to get started on real hardware. Update: it's 21:46, and I have now created an "install" kernel which copies the system blocks to disk. I'll do the "update" kernel shortly, then make a disk image. ---------------------------------------------------------------- 2023.11.21 I setup some emulated NetBSD and OpenBSD systems so I could get a few more binaries. NetBSD/pmax (mips) and OpenBSD/luna88k (88000) are now covered. For the luna88k build, this needed a big endian adapted ilo, so I've also added this to the source tree. There are a number of targets I'm interested in. I'd like to do RISC-V, Sparc, PowerPC, and Alpha as well. It'll be a while though. I had talked with Arland in the past about doing floppies with Konilo. I've now crafted a floppy disk image with the image, system blocks, binaries, and source. It's pretty full, but it fits, and seems to be easy to load into emulators. ---------------------------------------------------------------- 2023.11.20 When working with Konilo, I have to remind myself that this is not a Unix. I have a few things (like nonix) that are lightly rooted in Unix, but it really is a chance to have a fresh start. This doesn't necessarily mean that I can't leverage Unix though. Retro is a very solid system on Unix, and with block support it can let me do some useful things like generating HTML copies of blocks in my wiki, or backing up changed blocks. I can't lean too heavily into this though, since Konilo does run on non-Unix hosts (e.g., Mac/68k, MS-DOS, or native). I feel like Konilo is slowly progressing into a very useful (for me) system. It's going to be interesting to see what changes await over the next year. ---------------------------------------------------------------- 2023.11.19 I continued to work on huatango (the block tools). The `-titles` and `-locate` are added. Several more of the easier ones will be added tomorrow. Arland has done more work on Braceless C, intending to fix the issue with blank lines in a function. I'll test this tomorrow. With this being Thanksgiving week, I'll have a shorter work week. I'm hoping the week will be calm and quiet. ---------------------------------------------------------------- 2023.11.18 I've been relaxing today. It's a little cool outside, which was nice after the warm day yesterday. I do love fall weather. This evening I started a fresh rewrite of the block diff tools. The original ones I wrote are functional, but I decided to make several changes that I think will make them better for me. A repository for the diff tools now exists, and I'll be pushing commits to it soon. ---------------------------------------------------------------- 2023.11.17 Today I worked in my wiki; adding 32 new blocks. I have some rough ideas on locating links to missing pages, so may start writing code for this soon. I've restarted some work on the block editor for ARKS. This will use byte-packed blocks, so will have longer blocks. I think the title line will maintain use of 32-bit characters though, to keep better compatibility with the existing tooling. Following some convention, these blocks will be identified by a few cells in the title line. Arland will be home for a day or so tomorrow, so we may get a fe more issues resolved in Braceless C. ---------------------------------------------------------------- 2023.11.16 This morning I wrote a decompressor for ulz, a small compression format from @neauoire. This was done in RetroForth. It was a fun exercise. I'm doing some cleanup work on the diff tools and have begun a bit of a writeup on them. I made a small display tweak to (tuhi), so a space is shown both before and after each line, to make it more consistent. Should I write an X11 window manager? I don't /need/ to do this, but since I'm using X11 now, it might be nice to have another piece of /night software/, to borrow a name from Alex Schroeder. ---------------------------------------------------------------- 2023.11.15 bsd.cafe now has a lemmy instance. I'm registered as crc there. It's FreeBSD 14 release day, so I get to update arneson. This is always a little uncomfortable. Since everything on the system has been up to date, this should go smoothly. Update: server upgrade finished w/o any issues. On the block diff/patch tools: I have the initial code for all of the main pieces done. Still lots to do, but I can make binary diffs now, so I'll start to work this into my backup and fossil workflows. ---------------------------------------------------------------- 2023.11.14 I have ported several small programs (retro-unu, pali, part of the new block diff/patch tooling) to Braceless C. Doing this has identified a couple areas needing work; I emailed some details to Arland. Update: Arland fixed most of the reported issues. On the diff/patch tool, creation and basic display of patch sets is done. Tomorrow I'll do a textual comparison view, then start the remaining tools needed. ---------------------------------------------------------------- 2023.11.13 Arland is working on Braceless C; a couple of updates were added today. There's still work to be done, but I'm starting to make use of it in other projects now, so I'll be able to provide feedback over time. I am writing a set of tools for creating and managing a diff & patch set for the block file. After some experimenting today, I have a functional `altered-blocks` which creates a binary file with the modified blocks. Tomorrow I'll do a tool to review the changes, then work on tools to apply the changes. This will get me closer to having a much more space efficient way to store & review changes over time. ---------------------------------------------------------------- 2023.11.12 I added a fossil repository for Arland's Braceless C. There are going to be a number of things to do before this is ready for use outside of testing, but I did a quick adaption of ilo.c into it, and it works for that. ---------------------------------------------------------------- 2023.11.11 Arland has the initial Braceless C working now. I'll do some work on it soon. Just resting today. Tomorrow I'll do some coding. ---------------------------------------------------------------- 2023.11.10 I have a color-capable ilo-x11 now. And it's a lot faster to update and redraw the screen. Arland & I did some work on a "C Without Braces or Semicolons". This is not working yet. Arand thinks he'll be able to get it working soon. Arland has begun a new Graphica implementation atop the faster ilo-x11. ---------------------------------------------------------------- 2023.11.09 Mostly done ilo-x11. I have some ideas for color support, and am still looking at performance improvements. I need to finish the nga-x11 soon. ---------------------------------------------------------------- 2023.11.08 I have now uploaded my RAW fonts (in 8x16) to charles.childe.rs These are all suitable for use with Konilo or RetroForth on X11. In the evening I was able to implement lines and triangles. It's enough for now, but I should probably do filled shapes and set- ting the drawing color. Tomorrow I may write a bit on my early goals for 2024. While still several weeks out, it'll be good to try to organize my thoughts a little. ---------------------------------------------------------------- 2023.11.07 Color selection. Konilo + X11 is a monochrome experience. The only small splash of color is the cursor. I'm using a bright red for this, which is easy to see against the black backdrop. For RetroForth + X11, I'm doing a white on dark blue, with a cyan cursor. This makes it easy to quickly identify them. RetroForth now haas support for the pixel, get_pixel, and mouse devices. For my patrons, I posted some documentation on the drawing words. Tomorrow I will try to finish more drawing words. ---------------------------------------------------------------- 2023.11.06 I spent my programming time doing more work on X11 bits. I now have a partial Nga with X11 support. Tomorrow I'll start the process of adding the i/o interface code so I can port the ilo drawing words over. Otherwise, not much to report. It was cool today, but seems to be warming up outside. My glucose issues are improving a little. ---------------------------------------------------------------- 2023.11.05 I added some initial cursor support to ilo/x. This needs some refinements, but already makes a big difference in usability with the termina programs. The next steps are to get the stability improvements from the ilo-safer.c in place, then to add support for Xresources to set foreground & background color. Health: I was able to get several hours of sleep this morning, and my glucose is still quite high. I'm going to start back on IF and working towards a short 1-2 hour eating window to see if I can get my glucose levels to lower levels overall. ---------------------------------------------------------------- 2023.11.04 I continue to suffer from high blood glucose. This is troubling. I've started on a kqueue approach to uxn11, but it's not working yet. I'll be taking a break until later as my hands are starting to hurt a little. Up to this point, my X11 setups have either been using dwm or cwm with four xterm. I am going to change this, to let me have a layout focused on a Konilo window. As work on this progresses, I may be able to drop the xterm sessions and just run my system on X. It's already tempting to try this... ---------------------------------------------------------------- 2023.11.03 I attempted to build uxn11 on my OpenBSD box, but the lack of a set of Linux-specific system calls kept me from completing it. I was able to build the SDL2 version, and have the "essentials" up on my dev box. For RetroForth, akritid in #retro prompted me to finish up some work on retro-describe(1). This is now considerably faster (as is anything using file:read-line). I have one more word to work on (file:write-line), but am pleased with the results so far. ---------------------------------------------------------------- 2023.11.02 I did end up writing a turtle (t:) vocabulary, along with circle (no filled variants of rect or circle yet, and still no triangle or arbitrary lines). I'm going to take a short break from this to work on RetroForth tomorrow. My blood glucose is high again (running between 300-430, even after over 200u of insulin on the day). Hoping this resolves soon... ---------------------------------------------------------------- 2023.11.01 November is here. And it's finally getting cooler, so I'll be able to spend more time outdoors. In ilo/x, I've added basic lines (hline, vline, and diagonals) as well as rect. I didn't get to do the circles or arbitrary lines yet. I might try for turtle graphics (at least to a limited degree). ---------------------------------------------------------------- 2023.10.31 The ilo/x system has seen more work done. It's a little faster now and I added i/o ports for reading and setting pixels. Now I can write some words to do basic drawing. Doing this in Forth will be slow, but I can live with that for now. ---------------------------------------------------------------- 2023.10.30 I didn't get the Napia code packaged. But I did work on ilo/x. And it's working. I'm using a 1bpp framebuffer, with the DEC bits from Arland. Just terminal emulation so far, but it's all working. I'm writing this entry under it. Tomorrow I will try to do some performance improvements and general source cleanup. ---------------------------------------------------------------- 2023.10.29 Today was Tammie's birthday. We spent it at home, resting. Arland is working on a cake, hopefully for tomorrow. I have to smuggle a few supplies to him. Tomorrow I'll finish Konilo+Napia release, and get it published. It's yet another overdue task. I need to start budgeting more time for these projects. Also tomowrrow: I will be working on the next bits of the X11 ilo. Initial goal is to be able to emulate a basic terminal; Arland will look at expanding to support (graphica) later. ---------------------------------------------------------------- 2023.10.28 I've caught up on some rest (slept from 0600 to 1100), so that's a good start to the day. I wrote & published a weekly Patreon update. Today will be mostly resting further. I'll be doing a little in the wiki, and probably a little more on robot-finds-kitten. It's probable that there wont be a lot of progress today. Benjamin is off at Dorney Park as part of a band trip. He won't be home until about 0100 tomorrow. That's quite likely to be an interruption of sleep, so I expect to still be tired tomorrow. ---------------------------------------------------------------- 2023.10.27 Another sleepess night. I seem to be coming down with something. At work I spent most of my day sorting tools and getting them separated for shipment to our nine branches. This was long and tedious. It'd have been better if the team working on this yes- terday had actually finished, or at least accurately tracked the part they did. I'll write the Week in Review post tomorrow. Going to rest now. ---------------------------------------------------------------- 2023.10.26 Yesterday evening I setup what should be my final VNC configs on ansuli & Arland's box. I'm running 4x terminals w/Konilo. The layout is a 2x2 grid, using cwm, and I have xeyes & xclock at the bottom. I like it. I began working on (tito), a vocabular for writing interactive fiction. This is being based on my earlier work in Retro 11, but will not be a direct translation. I intend to take some advant- age of the block based nature of Konilo. I am also starting on something like cal(1) in Konilo. I have a couple of prototypes, so hopefully can get a full implementation done pretty quickly. ---------------------------------------------------------------- 2023.10.25 Back to not sleeping. Ugh. In #retro on irc.oftc.net, fangchar reported a couple of bugs in RetroForth. I've fixed both now. (I'm likely going to do a new release before the planned 2024.4 one, as a number of issues have been fixed now) The (termina) interface to (wiki) now has a key binding for adding new pages. Arland will be home tonight. We may discuss Graphica a bit. I am considering some of the keyboard/mouse handling stuff and how to make it more responsive. ---------------------------------------------------------------- 2023.10.24 I was able to get a few hours of sleep at last. Feeling a little better this morning. I've merged a copy of Arland's s:dedup code into my personal block set. I made a couple of small changes to naming to keep it in a single block. This depends on (ll) so I added a load block as well. ---------------------------------------------------------------- 2023.10.23 Another sleepless night... I merged the .h files into the .c for ilo-x86-native. There's no functional change with this, but it makes the source more consi- stent with the other implementations in terms of keeping files to a minimum. In Retro, I moved some files to package/extensions. This will be done for several of the Forth-side device vocabularies (for less major ones) to hopefully make it easier to manage some parts of the build process. ---------------------------------------------------------------- 2023.10.22 I woke up with very high glucose and a massive headache. It took about six hours to get my glucose back to a safe-ish level. I'm trying to setup an ESP32 toolchain on OpenBSD. It'd be nice not to need the Arduino software & Windows for building ilo for the dev board I have. For RetroForth, I added a bunch of missing source data, then began making the output from `make` less noisy. Also fixed some casting issues, and may have found an issue with `s:dedup`. ---------------------------------------------------------------- 2023.10.21 Too tired to write today. My glucose issues are getting worse. This is troubling; I will be trying to find a doctor soon. ---------------------------------------------------------------- 2023.10.20 I was able to get a 32-bit Nga in D written. This took a little longer than the ilo in D, but I'm happy that it's functional. i still need to do some optimizations, and add stack class. ---------------------------------------------------------------- 2023.10.19 Going to dive into RetroForth for a bit. I'm still exhausted, so I just added some NOP skipping today, and setup a source tree for a D implementation. ---------------------------------------------------------------- 2023.10.18 In Konilo, I've cleaned up the D implementation a little more. I'll probably start a Nga in D soon. ---------------------------------------------------------------- 2023.10.17 It's the second day in my work week, and I'm still covering for a coworker for another week. I'm quite tired, and not really feeling like doing much. In the D implementation, I've added classes for stack & memory, and some range checks. (This is for Konilo). ---------------------------------------------------------------- 2023.10.16 OpenBSD 7.4 released today, so I've updated my dev box & the one Arland uses. Updates are easy: sysupgrade; pkg_add -Uu; sysmerge then any manual adustments the install guide lists. Stuff today: a little work towards the "robot finds kitten" game (will be for both Retro & Konilo), and I'm doing some tests on SwiftUI, trying to figure out a good way to deal with the connection between the file manager & the editor. I've also setup a fresh ports tree so I can finish the update to the FreeBSD port of RetroForth. ---------------------------------------------------------------- 2023.10.15 This was a lazy day: just focused on resting and a bit of reading. I started setup work to try to get my Mac to backup via rsync, but am getting the stupid "broken pipe" errors that plague SSH on macOS. ---------------------------------------------------------------- 2023.10.14 I wrote an implementation of ilo in D. This was quite easy; and is quite similar to C. I had a few small issues around division, but was able to resolve them quickly. I'll mark this as one to work on more (I'd like to do a version using D's classes for the stacks). Haskell would be an interesting target (suggested on Mastodon), but there's not enough space on my dev server to install GHC. I've attempted to free up space on the VPS for GHC, but there's just not enough space with the partition setup I have. Maybe I can set this up on Arland's system... ---------------------------------------------------------------- 2023.10.13 The small delays in input handling under c-r-t are not bad; I am finding that they actually fit with my slower typing of late. Next up for me is getting Konilo/napia ready for release. It's similar to the main Konilo system, so should be easy? ---------------------------------------------------------------- 2023.10.12 The "cool-retro-term" works much more smoothly when run directly on a physical machine. I quite like it. For VNC, I've turned off all the effects except for burn in and bleed, and I'm pretty happy with the result. On the release front, I'm about half done with the release notes and am mostly finished packing the files up. This will be done today. (14:15 update: it's done!) ---------------------------------------------------------------- 2023.10.11 Good news is that I was able to get some sleep last night. I'm still very tired, but if I can repeat this for a few days, I should be feeling better. I saw an article (https://www.theverge.com/2023/10/10/23911145/) discussing a "personal cloud computer". This is quite opposite of what I think of as a personal computer. A personal computer really shouldn't be cloud-first and heavy on AI. Local storage, and understandable by its user are important here. I've been playing with cool-retro-term on OpenBSD. I like it. Under VNC, it's sluggish unless many of the effects are disabled but it definitely reminds me of my times using an actual termin- al. Maybe I should start watching for a real terminal again... ---------------------------------------------------------------- 2023.10.10 It was a busy day at work. I had an early dinner, then went to bed. During lunch break I did some prep work on the 2023.10 release packaging of Konilo. I might be able to finish up on Wednesday if it's not too busy. ---------------------------------------------------------------- 2023.10.09 Still not sleeping well, but I've been semifunctional today. For ilo, I've done a little work on ilo+graphica. The character display code has been refactored, and I added NOP skipping. Some experiments have been done on autoselecting the opcodes/event tuning, but this is not yet done. I wrote a little overview of the * load blocks. This will be put on Patreon tomorrow. Tomorrow I'll do more on the verbose execution in RetroForth. ---------------------------------------------------------------- 2023.10.08 No programming today, instead I spent the day setting up some emulators on the M1 Mac mini. It seems to be pretty good for this. For the iPad, I now have a USB-C HDMI input dongle, so I can use it as a display for my S22 Ultra or other devices. I'm using an application named Orion for this. It's a little odd to run an Android interface on the iPad, but the extra screen space is nice. ---------------------------------------------------------------- 2023.10.07 I wasn't able to sleep until almost 3AM, but did manage to get some sleep after that, until about 9AM. Not doing much typing today as I'm still exhausted. I placed the order for the TTGO VGA Controller w/PS2 & ESP32. It should arrive in a few weeks; I'm quite looking forward to implementing ilo for it. ---------------------------------------------------------------- 2023.10.06 I did not get any sleep last night. The exhaustion is making it difficult to do anything. Despite my exhaustion, I was able to get some typing done, so the updated Konilo manual is ready to merge. ---------------------------------------------------------------- 2023.10.05 Last night was a little better; I'm a bit less exhausted today. I spent a few minutes last night working on RetroForth/iOS. I'm hoping to have my source editor fully tied to the file manager in the next day. At that point I can release the beta, even if a variety of other things aren't quite done. ---------------------------------------------------------------- 2023.10.04 I was not able to sleep last night, so I am very, very tired today. ---------------------------------------------------------------- 2023.10.03 I didn't sleep well last night, so am very tired today. In the IRC channel, sevan reported a bug in retro-muri.c, and also found issues with the test code for the instruction set. I have fixed these in the repository, and will be issuing a new release merging in the fixes in a few days. For Konilo, I need a little more time to finish some revisions in the manual. I should be done these within another day or two. There's also a few needed updates to the glossry blocks. Tonight I'll start considering how to get started with writing a tool for tracking bugs. And toy with adding color to Arland's (game:pig). ---------------------------------------------------------------- 2023.10.02 I've made more progress on the iOS interface. I should have the editor tied into the file manager in another evening, then I can finally get a TestFlight build issued. Using a much larger font (28pt) is kind o nice. I much prefer a easily readable display. Hard to do this on my phone though... ---------------------------------------------------------------- 2023.10.01 Today was spent on RetroForth/iOS. The new SwiftUI based code has seen significant progress. I have a functional Listener, and most of the editor functionality done. The file manager is also getting close to being usable. TestFlight build *will* be published this week. RetroForth 2023.9 is now out. ---------------------------------------------------------------- 2023.09.30 I worked with Arland last night; we now have his Pig game in the blocks. I refactored the display code and added a (termina) wrapper; he added code to display the dice face. I've had some ideas for a faster way to find things in the blocks. It's not something I will be writing yet, but I'd like to start experiments on this towards the end of the year. ---------------------------------------------------------------- 2023.09.29 It's been raining on and off today. I'm enjoying it. The block reorganization should be done tonight, then I can prepare the subset for the sample blocks in the upcoming Konilo release. ---------------------------------------------------------------- 2023.09.28 Trying to balance multiple projects is difficult. I'm going to try to plan my time more carefully, in hopes that I can divide my efforts in a more balanced fashion. ---------------------------------------------------------------- 2023.09.27 After I get the 2013.9 releases of Retro & Konilo finished, I'd like to take a short diversion to document my two personal comp- uting evironments. At this point there are two, one based on Retro, and one based on Konilo. Last night on Mastodon, eli_oat shared a link to Uiua, a stack based array language written in Rust. It looks like this will be a fun thing to play around with. (Eventually I want to make a vocabulary for working with APL-style arrays, so things like this are helpful in my research towards this.) ---------------------------------------------------------------- 2023.09.26 I've noticed that the native version of Konilo does not seem to have the colors quite right in (termina). I'll look at this. I'm going to write a bug tracker in Konilo & RetroForth. Or at least attempt to do so. (Details on the bugs will be in the blocks, with Retro used to create a web/gopher/gemini display of them). I gave up on fixing RetroForth/iOS. The storyboards are proving annoying to try to fix, so instead I'm rewriting the interface using SwiftUI, but leaving most of the backend code in place. It's my first mixed-language Swift & Objective-C project now. So far I have the basic framework for the editor, documentation, & settings. Tonight I'll work on the file manager. ---------------------------------------------------------------- 2023.09.25 Rain again. At least it's comfortably cool outside now. I decided to setup a directory to start working on an x86 native + graphica system. This will hopefully (eventually) let me have Arland's stuff working without an underlying OS. I will be adding NOP skipping and a couple of superinstructions to the DOS ilo vm. This should make it slightly less painful to use on older hardware. ---------------------------------------------------------------- 2023.09.24 Another rainy day. I've done some work on the block reorg. It'll be a couple of day to finish. I also have started updating the Storyboards for RetroForth/iOS to try to fix the keyboard issue. It's unlikely to be complete today, but maybe on Monday? ---------------------------------------------------------------- 2023.09.23 Still having weird spikes & crashes in my blood glucose levels. It's a cool, rainy day today. Just spending some time reading and resting currently. ---------------------------------------------------------------- 2023.09.22 I made headway in getting a Konilo + ilo using floating point stacks and memory working. The Konilo image isn't yet working, but I have been able to load and run some smaller test images. I moved forthworks.com to the new server today. There'll be a few more things to get up and running (the glossary server and dict servers particularly), but it's working and I think that I should be nearly done now. The only remaining thing is my sisters website, which uses PHP. I'm looking at my options for dealing with this one. ---------------------------------------------------------------- 2023.09.21 I've tracked the issue with RetroForth on iOS to the keyboard display. I'm resizing the views when the keyboard opens and then resetting them when it closes. But iOS 16 or 17 changed some parts of how the keyboard is handled, so I'm working on adding revised constraints to the views so I can hopefully have it fixed soon. Rick posted a picture of an ESP32 board with VGA and PS/2 mouse and keyboard. I'm going to order one to port Konilo to. It'll be interesting. There's mention of a `FabGL` library for the VGA, but this is GPL, so I might want to write my own. Tomorrow starts the next round of block reorganization as I get ready to finalize the 2023.9 release. ---------------------------------------------------------------- 2023.09.20 I've run into a number of issues with RetroForth on iOS after building with the latest Xcode. Most are pretty small, but one is causing the display to be replaced by a blank view after the keyboard is closed. I'm working on fixing this before I push a public TestFlight build. ---------------------------------------------------------------- 2023.09.19 In Konilo I've added a couple of blocks with initial support for unsigned maths and comparisons. These are in `(crc:unsigned)`. I also began working on a vocabulary for double cell numbers. This is new: I've never actually implmented this in my systems, so I'm learning as I go. The double cell code is in the `(crc:double)` blocks. ---------------------------------------------------------------- 2023.09.18 I've added a VGA-based color palette for Arland's Graphica and set it as the default. It's now close to identical to the normal terminal colors, making stuff using Termina display as intended. I now have a remote VNC connection to my OpenBSD box, so I can test Graphica when not at home. This will be helpful. My remote setup is basically the same as at home. X, a lightly tweaked dwm, and three xterm windows. I've dabbled with a few other window managers, but don't really want a non-tiled one and since I only ever have 3-5 windows, dwm works well. ---------------------------------------------------------------- 2023.09.17 Another bad night w/high glucose. I wasn't really functional until late in the day. I need to figure out why this is happening. I may not be able to finish the TestFlight beta today; if I can't, I'll probably push for Monday or Tuesday. Hopefully tonight will be better... ---------------------------------------------------------------- 2023.09.16 Not feeling well. Possibly due to glucose issues from the last few days. I worked a little on the iOS RetroForth build. I need to get a few more hours of testing before I can push the TestFlight beta. ---------------------------------------------------------------- 2032.09.15 Last night was really bad. My blood glucose spiked to 506, then dropped down to 148. (Up from 197 initially). All of this over a five hour span. Ugh. I hate diabetes. ---------------------------------------------------------------- 2023.09.14 The manual is up to 98 blocks, with several more in various stages of completion. I'm hoping to be around 150 blocks by Monday. I also noticed that the ilo-x86-native floppy image used in the demo was running an old, pre-(termina) capable display driver. I've updated this, and it now works properly. ---------------------------------------------------------------- 2023.09.13 It was a stressful day at work. But the evening was calmer and I was able to rest a bit. Arland sent me the first draft of his Pig game. I merged part of it (the random number generator) as `(dusk:rng)`. After a read, I identified a few issues that he is addressing before he moves on to working on a (termina) interface for it. I've expanded my backups to rsync.net. At present I have about 35gb of data backed up via nightly cron jobs. Before bed I'll start writing a new section in the manual. ---------------------------------------------------------------- 2023.09.12 I finished migration of retroforth.org to the new server today. The big remaining one is forthworks.com. I'm still looking at what will be needed for this. It's the only one running Atua (as it's a Gopher server), so I'll need to see if I can run Atua-WWW on a separate port and have httpd redirect port 80 to it for this domain. Other bits: I wrote a bunch of new words for doing indent, unindent, cut, copy, and paste in the block editor. These are in (opt:editor-ext) if you want them. They are pretty handy, but will use ~600-700 cells of space. Arland has been writing a 'pig' game. I'll look at it later. ---------------------------------------------------------------- 2023.09.11 I have a small schedule change at work (now 6:00 to 15:00). So the change is going to take a few days to get used to. I've started reading a little on 'virtio' and doing a closer look at x86-64, in preparation for hopefully starting a second 'native' ilo implementation. (Or perhaps third, since I do have the Teensy one). At work they told me to expect a new laptop next week. So now I get to start making backups of the current one to prepare for migrating to a Windows 11 laptop. Other things: I finished setting up forth.chat yearly archives, and am getting very close to pushing the new file: words. ---------------------------------------------------------------- 2023.09.10 I setup the staging environment for RetroForth 2023.09 and Konilo 2023.09. There'll be two Konilo releases: one with ilo and one with napia. I'm expecting to publish the a release candidate at the end of next week. Things to look forward to over the coming week: finishing file: additions in RetroForth, preparing the final starter block set for Konilo. Finishing the Graphica documentation. I'd like to expand the Konilo manual a bit more as well, but am unsure if I'll have time to do so this week. ---------------------------------------------------------------- 2023.09.09 In Chrono Trigger I'm now done about a third of the game, and am taking a break until at least Monday. I have the rsync.net account created, and have run the first backups. I've crafted cron jobs to do these daily. No programming today. Just resting for the rest of the day. ---------------------------------------------------------------- 2023.09.08 The only programming done today was tests on new file: words in RetroForth (for use with utf8/32 translation). This is going well, so should be committed next week. Tomorrow I'll set up an rsync.net account and begin getting a proper backup of the server in place. For now, I'm going to spend a little time playing through Chrono Trigger. ---------------------------------------------------------------- 2023.09.07 I've moved Arland's server contents to the new server. My personal wiki has picked up several new blocks. I hope to be able to continue adding at least one block a day with various notes and informational bits. I now wish to write some tools to help find wiki blocks more easily. Not sure what this will look like though. Tomorrow I will be trying to get back into Graphica to finish updating my draft of the updated documentation for the core functionality. A side thing: I should have a new beta of RetroForth for iOS next week unless my hand pain starts increasing. ---------------------------------------------------------------- 2023.09.06 I've finished much of the work to get the new server functional. At this point I have OpenBSD httpd configured, working with TLS, using certbot to get the certificates and a cron job to update them. Other than retroforth.org, I have all the content from the main server copied over. Tomorrow I'll be working on setting up the various cron jobs, then starting to look at getting Atua up and running. I'll need to write a new tool to generate index.html files from the goph- ermap files before I can bring over forthworks.com. Forth stuff: I'm looking at some small programs and games for a 'zine. Will be discussing this more with Arland soon. ---------------------------------------------------------------- 2023.09.05 I started setting up a new server. This will replace the two which host retroforth.org and forthworks.com, and also several others that I use for other things. The new one is more capable. It's a Linode with 4 CPU, 8GB RAM, 160GB disk space. I have it running FreeBSD 13.2 with ZFS. I'm running it with OpenBSD's HTTPd. I'll maintain an option to flip to using Casket for testing, but until I can get it working better under load the pragmatic choice is to use a more conven- tional server. Gopher will continue to be under Atua. The full migration will be done in stages, over this month. I have moved forth.chat, retroforth.net, and retroforth.com over. Tomorrow I'll begin the shift of retroforth.org over to it. ---------------------------------------------------------------- 2023.09.04 Today is Labor Day in the United States, so I was off work. It is nice to be off on holidays as I generally don't face any inquiries from customers and can relax more than normal. It's hot again, and looks to remain hot for the rest of the week. I'm not happy about this, but will try to stay cool. I started a few early bits towards the new filesystem code. It isn't going to be a high priority project, and might never get released, but this one might work out better than the last few. Tomorrow I'll be starting on words to read/write uncode charac- ters in data files. This will add several new word to the file: namspace. ---------------------------------------------------------------- 2023.09.03 No programming today. But some reading of old Unix sources was done, and I created a fresh block set for future file system experiments. I had a hard time sleeping last night, so spent much of the day resting. I am starting to experiment with using `rsync`. ---------------------------------------------------------------- 2023.09.02 I was up late, and slept in. Some time was spent on server maintenance. I'll be moving some things to a new server soon (consolidating three of mine into a single one with more resources). I'm gathering usage data from them to prepare for setting up the new one. ---------------------------------------------------------------- 2023.09.01 It's September. With no one bugging me I was able to get the small amount of needed work done early, then I was able to rest for a few hours. On the Mac, I had no luck with OpenBSD or FreeBSD. I did get a Linux install running. I tried out KDE a bit, then Gnome. Some quick observations: KDE is fast under Wayland. Gnome is slow, but notably quicker with X11. (Still slower than KDE). I dislike seeing so much running under systemd in the process list. So what's next? I have a "week in review" to finish writing, then will probably write a "pig" dice game (or at least start this) ---------------------------------------------------------------- 2023.08.31 I'll be (mostly) off work tomorrow. Looking forward to this; a four day "weekend" is a pretty nice break and I can hopefully get some rest in addition to doing some writing & programming. It's warm today, but the humidity is down. This is nice; I hope for a long break from the summer heat. On my Mac mini, I setup Cheetah, which is a front end to the macOS virtualization framework. I'll see if I can figure out how to setup a BSD on it, or at least a Linux. ---------------------------------------------------------------- 2023.08.30 I'm starting to feel better. I added a few new blocks to the wiki. I am hoping to add a block at least every other day until the overall contents get up to a large enough set to help idenitify additional tooling needed to make it more usable. Good news: I'll be off work Friday and Monday, giving me a nice four day weekend to rest and hopefully do some programming and writing. ---------------------------------------------------------------- 2023.08.29 Still not back to normal. Work was stressfull. UTF32 for source files is now working in RetroForth. This was a longer than expected process, but I'm pleased to say that it seems stable and doesn't appear to have broken anything. Space on the OpenBSD server has been freed up. I'll be good for a while now, though I will look at replacing it with a new install after I retire toot.forth.works. Tomorrow I start on UTF32 support for general file i/o and the UTF3 conversion device/vocabuary. ---------------------------------------------------------------- 2023.08.28 I'm still feeling sick. When I got to work I felt ok, but have been feeling worse as the day goes on. I moved the insulin cost & dosage calculators to be under (health/) and added a BMI calculator. I ran out of space on my OpenBSD dev server. This is due to a bad partitioning (~6.5G set aside for /usr/obj and /usr/src). One day I'll fix this with a reinstall & manual partitioning. ---------------------------------------------------------------- 2023.08.27 I ended up sick with a stomach issue last night. Still not quite feeling right yet, but definitely improved. I've now merged in some Graphica fixes from Arland's source tree and will continue on the documentation this afternoon. Evening update: Arland has (paint2) mostly working now. I'll be merging in his latest few changes to the backend code before going to bed tonight. ---------------------------------------------------------------- 2023.08.26 I found & fixed a serious bug in the (nonix) `xrm` word. This was deleting the wrong blocks... ---------------------------------------------------------------- 2023.08.25 Some stuff: I added a version of ilo.c for OpenBSD. This uses pledge() and unveil() for added security. I'll be adding in some error checks soon. For Graphica, I was able to get most of the backend code updated to use the new point structure, write new Forth blocks to call into these, and do a little testing. (I woke up this morning to a little numbness in my hands, but almost no pain; quite odd after yesterday). So I'm basically a week or so behind on what I wanted to get done, but hopefully the next week will be better. ---------------------------------------------------------------- 2023.08.24 I've not uploaded the WISP entries in a few days. I should do this sometime... Today has been particularly bad with RSI pain. Not sure why; I have been trying to keep typing and general hand use down to a minimum. Made tacos for dinner. They turned out well. I'll try to do another batch of them next week. ---------------------------------------------------------------- 2023.08.23 Halfway through the week. I have a list of words I think should be good for Graphica 0. I'll send Arland a copy soon. I have prepared a couple of small source files using UTF8 multi- byte characters for testing RetroForth's UTF8/UTF32 code. I hope to get back to working on this next week. ---------------------------------------------------------------- 2023.08.22 Due to wrist pain I'm probably not goung to write much. It's been getting a little worse, so I don't want to risk any more damage. I'm thinking through the Graphica vocabulary to decide on what words should be in the initial "level 0" vocabulary. ---------------------------------------------------------------- 2023.08.21 Well it worked pretty well. There's still some issues (colors are not quite right, and cursor movement is a little off for the second row of text), but I think I can finish it tomorrow. I also started updating the Graphica drawing backend to use the packed point (x in upper; y in lower bits) approach instead of an in-memory representation. If this keeps going smoothly, (and if the RSI pain remains at a tolerable level), I might (with Arland's help) be able to get it done this week. That's probably wishful thinking though. I already have a little pain back in my left hand... ---------------------------------------------------------------- 2023.08.20 Still tired. But the pain in my hands is not as bad today, so I will try to do some programming tomorrow. I think it's time to start on the Graphica terminal emulation. Hopefully this will be easy; I *should* be able to mostly just drop in the ilo-native-x86 code for it, with a few small tweaks. But there's no guarantees... ---------------------------------------------------------------- 2023.08.19 Sleepy today. I ended up sleeping about half the day, which was much needed. Then watched a movie, took a walk, and spent some time with the kids. It was a pleasant day. ---------------------------------------------------------------- 2023.08.18 At work I've been focusing on cleaning up billing issues. This is quite time consuming, but I think I'm about a third of the through. Tomorrow I'll start some work on the new terminal code for Graphica. Then Arland can try to finish the drawing functions. Otherwise, nothing to report today. Still having some pain in my hands, so I'll probably still be limiting my time typing for a few more days. ---------------------------------------------------------------- 2023.08.17 Video work today. I finished the initial draft & test recording of a video on the evolution of the VM models (Maunga, Ngaro, Nga as well as ilo & Napia). Tomorrow I'll try to do a test record of the audio. Arland is going to look at doing some drawings to add visual interest. ---------------------------------------------------------------- 2023.08.16 ilo-x86-native now has colors working. Taking a break from this now. I'm making a small patch to ilo.c to test out the OpenBSD pledge and unveil system calls. For pledge() it appears we need to use "stdio rpath wpath". I still have a few more tests to run before looking at unveil(). If successful, I might add use of these to the amd64 assembly implementation. Arland has been making 1-2 loaves of bread each night. It's fun to watch. I do need to try to get him started on these earlier; starting them at 2200 or later doesn't give a lot of time to work on them. ---------------------------------------------------------------- 2023.08.15 I fixed the cursor positioning in ilo-x86-native. Next up is colors. At least it's now functional enough to run monochrome (termina) programs. Due to hand pain I'm not planning to do any actual programming tomorrow, just reviewing the existing code and VGA docs to make sure I'm ready for adding the color support later this week. Bob (at work) has me playing Wordle, Connections, and some other puzzle games. I might try writing something like Wordle in Forth at some point. ---------------------------------------------------------------- 2023.08.14 I attempted to merge in code for UTF8 to UTF32 support for source files, but this did not work out and the VM started to hang. I'll investigate more later. Onto Konilo. I started merging in the terminal emulation into ilo-x86-native. It's not quite working properly yet (still seems to have a small issue with cursor positioning, and colors are not in place yet), but I can now run some (termina) programs mostly successfully. My hands are starting to hurt again. I'll probably try to debug the cursor positioning, then take a break for a couple of days. ---------------------------------------------------------------- 2023.08.13 The console code rewrite from Python to C is done. I also did a bit of refactoring, so it's ready for me to start merging into ilo-x86-native and then Graphica. What else? It's hot and humid again, so I'm not going to be on the Mac mini this weekend. I'll try to get to it once conditions improve. I'm working on a little thing that might end up being related to Arland's Namiko project. This is currently in Python, but may be rewritten in C and Forth later if it works out. ---------------------------------------------------------------- 2023.08.12 I've mostly finished rewriting Arland's terminal processing code in C. I should be able to finish debugging later tonight, or tomorrow afternoon at the latest. Then we can start merging this into the ilo-x86-native branch as a test case. ---------------------------------------------------------------- 2023.08.11 I'm less tired today. Stressful at work, but now (1821) I'm off until Monday, so I'll see if I can unwind and do some reading, writing, and maybe listen to some music. Ok, so programming stuff. I will be working on the terminal code this weekend. And the Life (termina) interface. And trying to complete a beta of RetroForth on iOS, which will require some work with the Mac. (This will be temperature dependent; if it's too hot/humid, I'll have to wait until it cools down.) ---------------------------------------------------------------- 2023.08.10 Still tired, but I did write up some notes for the weekly update on patreon. Arland sent me his latest DEC terminal code, so I started review of it. I'm planning to work on this and some of the Graphica code this weekend. ---------------------------------------------------------------- 2023.08.09 The month is rolling by quickly. I decided last night that I should write a (termina) interface for editing & running Life. It'll sit over the existing implementation, much like the JMF Mandelbrot viewer. I was too tired last night to do any programming. But I did some thinking through parts of the Nga code, and I have a plan now for implementing the UTF support in source files. This is probably going to be ready over the next few days. ---------------------------------------------------------------- 2023.08.08 I'm going to rearrange my blocks again. (This seems to be a thing I do every couple of months). I'll be moving source code to 0-8192, and devoting the higher blocks to data. I will also be writing a small utility to patch the differences between my personal block file layout and the default blocks in the ilo repository. ---------------------------------------------------------------- 2023.08.07 I'm attempting to spend less time online (apart from work hours) so updates to WISP might be delayed a bit. What I hope to do is structure my online activities into batches, connecting briefly to pull in updates and push out changes & responses. From past experience, this wil be difficult, but probably better for my mental health. With that being said, here's some quick updates for today: both (kowae) and (taotaka) launchers have been merged into ilo. Retro has UTF8 <-> UTF32 support at the interactive listener (as a build-time option). More UTF stuff is coming. I've begun writing a post introducing both (kowae) and (taotaka) It'll likely be published near the end of the week. ---------------------------------------------------------------- 2023.08.06 Not feeling well today. My blood glucose is high this morning, and I feel somewhat queasy. Will rest as much as possible. ---------------------------------------------------------------- 2023.08.05 I woke up today with a severe headache. Spent much of the day avoiding light, loud noises, and trying to rest. I was able to do a little work on the RetroForth/68k Mac code. This is not yet working, but I am now able to get the image to load and some code to start executing. There's still a lot of issues. I'm going to try to do a little more tonight, but may need to wait until tomorrow if the headache gets worse. It's now 2138 and I'm wrapping up for the night. The Retro/68k is still not working properly. I'll review tomorrow, comparing to the original sources, and then compare modifications to the ilo sources. Hopefully I can find the cause of the issues. ---------------------------------------------------------------- 2023.08.04 It's getting late, so I'm going to enter today's notes before bed. I have a working kowae (the second launcher, with a 3x3 grid) It's nice, but not as flexible as the unnamed termina-based launcher. I'm leaning towards including both in the blocks. We had a BBQ for lunch. I cooked sausages and burgers; they were quite good, though there was a bit more leftover than normal. Dustin says we might do another one next Friday, so I'll prepare for this. I'll spend a few minutes to finish the weekly update before sleeping. ---------------------------------------------------------------- 2023.08.03 It's a tired day. While I got some sleep, I'm still suffering from the severe sleep deficit from the last few weeks. Before bed I'll be doing a little more work on the launcher. I have two models and am not sure which I prefer yet. 2153: I'm going to shut down in a few minutes. Arland helped me identify a bug in kowae, so I'm now very close to having the second launcher working. I should be able to get it to a usable state tomorrow. New word in (std): "invoke": '(block-set) 'word-name invoke Will load the block set if not loaded, then find and run the word. This will be enhanced a bit further soon. ---------------------------------------------------------------- 2023.08.02 I was able to get a few hours of sleep last night, so I'm much less exhausted than yesterday. So what to work on? (tuaki) a single step debugger (kowae) the new launcher program (kohure) adding optional color to the memory browser? (peita) ascii art editor These are a few ideas to look into. I think Arland is coming for a few days, so I'll discuss these with him when he arrives. I also need to get back to work on the DEC escape sequence code; it'd be nice to have this functional soon... ---------------------------------------------------------------- 2023.08.01 Today was a very tired day: I'm going to bed early, so won't be programming tonight. I think I'm going to try writing a single step debugger for Konilo sometime soon. Maybe starting with Retro's Autopsy code? ---------------------------------------------------------------- 2023.07.31 Today was a retrocomputing day: I spent my evening working in System 6 on an emulated Mac. The results? I've made the ilo-68k a little faster (adding in NOP skipping). I was also able to update to both the latest Konilo image & blocks. The good news is that it does work fine. It's slow when run at 1x speed, but that's expected. Before I update the repository I plan to add the limited set of superinstructions (handling of lica...., liju...., etc as single units) to eke out a bit more performance. Other stuff. I updated the Konilo blocks. And started another branch on the launcher interface. ---------------------------------------------------------------- 2023.07.30 Last night Arland & I rewired an old light fixture in our kitchen. It's now using line voltage LED T8 lamps, so is more efficient. I did get a small cut on my left pointer fixture, but that was the only issue we encountered. On the hardware front, Arland is interested in Sharp's memory LCD (400x240). It looks like this can be used with a few i/o pins and a small bitmap framebuffer. I'll look at getting one of these for testing. ---------------------------------------------------------------- 2023.07.29 It's still hot. I spent the morning watching Oppenheimer. It was better than I expected, but long. The theatre was nicely cooled, which was nice. Today I start rewriting Arland's Python code for DEC terminal processing in C. It'll be tied into both ilo-x86-native and the Graphica system. Once done with the native side, the tools using Termina will work, which will make that nicer to use. New hardware: in August I'm going to be picking up a second RP2040 board and something with an ESP32. I'll also order some FRAM chips from Adafruit, in hopes of getting a full ilo system running on these. ---------------------------------------------------------------- 2023.07.28 It's very hot, and I'm not doing much today. I spent a little time with Arland's PlayDate; it'd be amusing to have something like this with a 30% keyboard and Forth. ---------------------------------------------------------------- 2023.07.27 I made a small change to tuhi, which makes it a little nicer for non syntax highlighted editing. I also identified & fixed a bug in `s:get/line`, which now works correctly for both delete and backspace. This has been committed in the repository. An interesting change: with the tweaks to tuhi, the current text entry is slightly brighter, so it's easier to spot what is being edited. ---------------------------------------------------------------- 2023.07.26 It's getting hot again. Hoping it cools down tonight. For the video, I'm going to start with two. One will be on the dictionary structure in RetroForth; the other will be on blocks. In other news, I have an experimental ilo that does conversion of keyboard input/output between UTF8 and UTF32. I need to do a lot more testing on this, but it appears to be something that would open up the world of Unicode to me if I need it. I added a block with `s:pad/left` and `s:pad/right`. The current code has a memory leak; I will fix this before comitting it to the public blocks. ---------------------------------------------------------------- 2023.07.25 It was hot today, but we had a short afternoon thunderstorm that cooled things down significantly. The forecast says it'll get hot again as the week progresses, but I'll enjoy this brief respite until then. I have begun porting some parts of the extended Konilo system to RetroForth. This started with adding `fill`, and is continuing with (termina). The termina application scaffold is not yet working, but the color & cursor positioning code is fine. This process will help me identify differences between the systems. I will try to document these. Arland is coming over this weekend; we'll try to get a test recording of the first video done so he can start editing. ---------------------------------------------------------------- 2023.07.24 I refactored the (wisp) code, making it easier to relocate the data blocks. This was done as a precursor step towards a termina interface. Then I proceeded to write the termina interface. It's two blocks, and provides keys for moving through the blocks, starting tuhi, creating a new post, and jumping to the latest post. I've altered (termina) in a breaking way: the bg: and fg: words now set the colors directly. This will require updates to existing programs, but makes the code cleaner overall. Apart from this, I wrote an " sigil (in the examples) for strings with spaces. And I wrote a bit more into my wiki. ---------------------------------------------------------------- 2023.07.23 I'm still quite tired from yesterday, though I did finally get more than two hours of sleep. I'll probably be resting more this afternoon. On a development front, I now have a working SSH server on my Android phone, so I can use Termux via Blink (mosh) or Prompt (ssh) from my iPad. This is nice, though I still need to see if I can work out doing the connection via USB or an ad-hoc wifi network. (I like the larger screen on the iPad for readability) I'll be making a few small revisions to my launcher code today. It should be ready to merge into the blocks in a couple of days. ---------------------------------------------------------------- 2023.07.22 I was able to fix `restart` in the amd64 assembly ilo. I updated the OpenBSD binary, but am out of time. Today I'll be visiting my my brother and his family. They have been in Grenada for the last five years, so it'll be nice to see them again. (Late update: we got home around 22:00; I'm exhausted, so just going to have a late dinner, then go to bed). ---------------------------------------------------------------- 2023.07.21 Still having pain, but am doing a little work. I'm sending notes to Arland concerning the Graphica updates. And re-reading my launcher code. I have noticed that the amd64 assembly ilo has a problem when running `restart`. I think I can fix this, probably tomorrow. It'll be a little annoying (as I'll also need to update all the binaries), but I'll probably wait on that until my hands are doing better. ---------------------------------------------------------------- 2023.07.20 I have updated the Napia source tree to use the latest Konilo image & blocks. Other than this I'm not doing any programming today; I've been having increased pain in my hands, so will slow down a little to try to rest them. ---------------------------------------------------------------- 2023.07.19 A bit of non-Forth (though still related) today. I wrote the core code for processing the subset of DEC/VT100 escape sequence stuff that (termina) uses. Arland ported this from C to Python and has fleshed out the missing bits. We'll rewrite back into C, then update the Graphica text output to use it. Once this is done, we have plans to simplify the Graphica system in a variety of ways. I'm hoping that we will be able to have it ready (including documentation & some examples) by October. ---------------------------------------------------------------- 2023.07.18 I've not been sleeping well, and it's starting to become notica- ble. I'll try to get to bed early tonight, with hopes of enough sleep to feel better tomorow. The update handling code has been added to (ripanga). This now needs two blocks. I've done some testing on the update code; it appears to work ok. Tomorrow I'll try to get the spreadsheet notation code written. I'm dealing with some numbness in my right hand and a little pain in my left, so I might have to take a short break in the near future. ---------------------------------------------------------------- 2023.07.17 I merged in Kohure. Looking at my old notes, I decided to pick back up the previous work on a headless spreadsheet. I've reworked the vocabulary, and written an implementation. The data storage parts work, so next up is writing support for calculations. I described my planned approach to Glen; he thinks it'll work, so I'm going to proceed with this tomorrow. ---------------------------------------------------------------- 2023.07.16 Other than the help screen, Kohure is done. I'll finish this tonight, and will merge it into Konilo tomorrow. With my weekly patreon postings, I'll be starting to include some of my handwritten notes on occasion. This might be of some interest. I've finished the first PDF of the Konilo dictionary. It's sized for A5 paper, and will be included in the nightly snapshots starting this week. A supplementary dictionary for the (std) blocks is now being worked on. I hope to have this done by the end of the week. ---------------------------------------------------------------- 2023.07.15 I installed iPadOS 17 beta on my iPad last night. I've run into a few small bugs, but nothing too annoying yet. Stage Manager is much improved. Tonight I will begin adding direct entry of values or assembly in Kohure. Then write a help screen and it'll be ready to merge int the repository. I didn't write up the week in review for Patreon yet, I'll do this before bed tonight as well. ---------------------------------------------------------------- 2023.07.14 I now have a functional core for Kohure (the memory browser) in my blocks. I still need to write a help screen, and maybe add a function to edit memory cells directly, but the browsing part is finished. As with most of the recent stuff it has a (termina) interface. It's still hot and humid, but I have one room that's mostly tolerable today. So getting a little work done. Tomorrow is a real-world task: helping Arland set up some fans to improve airflow in the upstairs of his place. Then I might work on finishing the scripts so we can start the audio parts. ---------------------------------------------------------------- 2023.07.13 My blood glucose crashed hard last night, resulting in a night without sleep. Thankfully it's been stable today. Humidity is up again, making my house uncomfortable. I was able to get a few more fans to help circulate the air better, which may help. I've started work on a memory browser. With the small screen sizes I use, this is kind of tight. It displays the address, three cells (as instructions, values, and ASCII [when valid]). I can already browse a portion of memory, but it does crash after several screens. I probably missed a small detail; I'll take a closer look after I sleep for a while. ---------------------------------------------------------------- 2023.07.12 Some fine tuning on Casket: I'm now having inetd run the program without using a shell script. It also no longer gets run under timelimit, so it saves two processes per connection. This should help scalability a little. The reaper script is now run every few minutes and kills any processes older than 180 seconds, so accomplishing the same limits I had with timelimit. I added `clear` to (nonix). Tomorrow evening I have to pick up a ping pong table & a chest. There'll probably be a little programming, but I suspect I won't get a real chance to resume significant work until Friday. ---------------------------------------------------------------- 2023.07.11 Today was a little stressful. I published Casket v6. And it promptly got field tested by both lobste.rs and hackernews. It did an acceptable job under load from lobste.rs, but spent the entire afternoon and evening under a (for me) massive load from hackernews. The bad: lots of timeouts, inability to connect, and very heavy server load. The good: this is the first time the server has remained responsive to SSH connections, so I've been able to keep an active eye on things. I'm pleased overall, but will probably look for more ways to handle scalability in the next update. ---------------------------------------------------------------- 2023.07.10 I'm a bit over halfway done my launcher program. It's not a very fancy system, but it'll serve my purposes. (Arland's launcher will be nicer; with much more flexibility). Weather update: it's still quite hot, but the humidity is down, so it's much more tolerable. This should be helpful in making progress towards video production. Health updates: my RSI issues are mostly staying under control, but I'm going to start reducing some of my typing soon to help ensure that this remains true. Blood glucose levels are down a little, but still too high. I'll be transitioning back to a very low carb diet starting tomorrow. More plants than meats this time. ---------------------------------------------------------------- 2023.07.09 I've started writing my take on a simple launcher program. This will probably take at least a few days to complete, but it'll be interesting to work on. An update on the above as I wrap up for tonight: I have the data structures created, and am able to populate the categories. Next will be words to access the corresponding addresses, then I can write the display code and key handlers. It should be straightforward at this point, but I'm sure there will be some unexpected issues. ---------------------------------------------------------------- 2023.07.08 It's now the evening, so I'm just doing a bit of recap. The humidity inside is down, so while still hot, it is much more bearable. Arland has begun working on his launcher program. I helped debug it a little. I'll also be writing a launcher program; mine will be less fancy thank Arland's. It should be fun to compare these. I have had a few thoughts about words for saving/loading image (and ASCII artwork) data across multiple blocks. I need to spend some time considering this more before starting to implement it. ---------------------------------------------------------------- 2023.07.07 I've refactored more of Casket. I should be done with this on Monday. Arland comes home tonight for a few days. I'll see if he is willing to work on a launcher program. It's still too hot and humid. Not going to write much more code or text today. ---------------------------------------------------------------- 2023.07.06 I forgot to run the upload script yesterday. I'll do that in a few minutes. Casket using the new file words is *much* faster. A transfer that took almost 16s on my dev server now takes 0.3s. It's only about 3x slower than the OpenBSD HTTPd in this test case. I swapped the main retroforth.org server to use this version of Casket as I continue to improve various other areas. I will write a bit on this for patreon. I'm going to try to have Glen record the audio for the first two videos in the next week. This will need a bit more writing, but I really want to get these moving forward. ---------------------------------------------------------------- 2023.07.05 It's still very hot and humid. And it looks like this will not be breaking anytime soon. I'm making headway on various fronts. The Konilo dictionary is now finished draft 3 (waiting for Arland to review again). I'll upload a copy to my patrons tomorrow. RetroForth has gained file:read/bytes and file:write/bytes. I started work on using this in Casket. ---------------------------------------------------------------- 2023.07.04 It's July 4, so I'm off work. But still programming. I fixed the memory leak with the 'hints' part of (termina). This was bothering me, so I'm glad it's no longer an issue. Next up: trying to get nested (termina) programs working. It took longer than expected, but the nested (termina) stuff is now working. I've set (tuhi) as the default `Editor` for the `catalogue`. I've also written a `ti:help` as part of (termina). It'll be used to display a simple help screen for (termina) programs. Last bit for today: I'm working on more (wiki) words. ---------------------------------------------------------------- 2023.07.03 Today was spent on Tuhi and trying to stay cool. I added cut/copy/paste (via XCV) and indent/de-indent (via []). A little work on the video scripts was done, and some steps towards a possible quarterly 'zine. I added a right column to the Tuhi display. This makes it easy to make sure the code being entered doesn't overflow the line. So all that's left for Tuhi is a general help block (which will need some (termina) additions to do the way I want) and the "leap" function. Going to wrap up for now; maybe I'll write the "leap" tomorrow. ---------------------------------------------------------------- 2023.07.02 I've started writing the additions to Tuhi, but didn't get very far. It's too hot to be productive today. I attempted to do a little work on the kitchen sink, but wasn't able to turn off the hot water line. I'll try this again after it cools down. (The single handle control is loose; I can't remove it until I can get the water turned off.) ---------------------------------------------------------------- 2023.07.01 Today was a rest day. Tammie & I stopped at an Olive Garden for lunch, then did a bit of grocery shopping. I didn't do any programming today. I did think over my planned approach for the Tuhi expansions, and started considering how to extend (termina) to support nested programs. I'll likely start writing code for these tomorrow. ---------------------------------------------------------------- 2023.06.30 I'm off by a day; I had thought yesterday was Friday. It's a bit disappointing that it's not quite the weekend yet. I ended up being too tired to work on the scripts last night; hopefully today will be better for this. 1300: It's been somewhat better. I have drafts for three scripts written & sent to Arland for review. Deciding on weekend projects: tuhi improvements (add leap, indent/deindent, cut/copy/paste lines), update the Teensy build, expand the wiki: vocabulary. Finish some reading backlog. Sleep. I probably won't finish all of these. ---------------------------------------------------------------- 2023.06.29 The core functionality of Konilo works on a non-visual-terminal environment. With (termina), I have a vocabulary for writing things that use this, but I still desire to support a simpler output model for most things. As such, I'm starting (with the wiki) to add variables that disable the use of (termina) words for the things that can reasonably support both. ---------------------------------------------------------------- 2023.06.28 I'm about halfway through the first draft of the script for the first video. Arland thinks it's ok, though he said he may wish to alter some of the phrasing. I am going to try to have my part done by end of Thursday so we can do some revising on Friday and a test recording on Saturday. ---- On a Unix host, if performance matters, it may be best to cache the blocks in RAM. In my tests it saves considerable time when searching through the blocks. I'll probably publish an ilo-fast at some point that makes use of this and other techniques. ---------------------------------------------------------------- 2023.06.27 Last night I merged a patch to fix an issue with the GNUmakefile as well as adding `block:for-each` to the (std) blocks in Konilo and also committing the current error handling device code for RetroForth. I have begun working on the scripts for the initial set of videos. I'm hoping we can record & prepare some longer tests in the next week or two. For Arks, some work has been done towards supporting packed blocks. This will not be done quickly. I will be trying to have the core words needed done over the next few weeks, then Arland will need to work on making a usable editor. ---------------------------------------------------------------- 2023.06.26 I renamed the (tools/titles) to (tools/catalogue) [and aliased catalog to catalogue]. It's looking like a strange week; rain is expected every day, but we are seeing long sunny periods as well. It's very humid. Some minor issues in the new RetroForth error handling device are now resolved; I can now proceed to flesh it out further. Work related note: I had thought I was done with purchasing for now, but our purchaser wasn't back, so I'm still covering his tasks. This requires much more typing than normal, so I have to be careful to avoid another RSI flare up. ---------------------------------------------------------------- 2023.06.25 It's going to be a hot (~30C every day), humid week, which does not bode well for working. (Our house is not well insulated, and is basically the same temperature as outside during the summer). A possible list of initial videos about Konilo: Introducing Konilo; An Overview of Blocks; Locating and Browsing the Blocks; The Basic Block Editor; The Tuhi Block Editor; An Overview of Forth; Basics of Programming in Forth; Exploring the Vocabularies. For RetroForth: Introduction; adaptions of the overview/basics/vocabs videos. ---------------------------------------------------------------- 2023.06.24 Tammie and I watched Pixar's Elementals. It was ok; there wasn't anything new or innovative in the story, but it was fun little excursion with my wife. I started some experiments for production of videos relating to RetroForth and Konilo. Arland and I hope to begin the actual videos over the summer, with a goal of having two initial series (one on each system) ready later this year. We expect to publish them to Patreon first, with possible a possible YouTube channel later. Initial production is likely to involve use of Asciinema, screen capture, voice overlay, and some art done in Procreate on iPad. The initial tests look promising; more details to come later. ---------------------------------------------------------------- 2023.06.23 I published a week-in-review to patreon this morning, then had a coffee before starting work at my day job. Hopefully I'll have time to finish the tuhi revisions at some point today. ---- I don't have the tuhi "leap" working yet, but the editor now uses the syntax mode for source blocks and plain mode for normal blocks. It also realigns the keys to be closer to that of the titles tool. ---------------------------------------------------------------- 2023.06.22 I didn't merge in the (list:syntax) blocks; I was too tired after work. I'm doing this now. ---- For tuhi: I'm going to make some changes. I will add a "leap" key, rearrange some of the hotkeys to be in line with `titles`, and am making the display switch into a toggle. The display toggle will switch between (list:syntax) and the default mode. ---- I started some notes for revising Graphica to what should be the final state. Hopefully this can be completed over the summer. ---------------------------------------------------------------- 2023.06.21 I added a `wisp:latest` to (wisp) to make it easier to bring up the most recent entry for edits. This should simplify adding to an entry as the day proceeds. I've made a few small revisions to the (list:syntax) blocks. I still want to add a couple of things (detecting . comments, and maybe the [ ] ; immediate words), but it's working well enough for now. I'll merge it into the ilo repository today. (tuhi) now has support for using (list:syntax) [if the blocks have been loaded]. ---------------------------------------------------------------- 2023.06.20 Covering for multiple people at work is draining. I'm too tired to do much in the evenings. Thankfully there's only three more days of this until we get back to normal. In the ilo repository I've grown the blocks, added in (wisp), and relocated the Glossary blocks to the end. Adding in (wiki) will be done later this week. For the wiki; I've updated the wiki-export.retro (not yet published) to add in some CSS and generate an "all-pages" page. Some minor revisions to the pagename filtering have been added. Arland suggests adding something to display the page name. I'll try to add this tomorrow. ---------------------------------------------------------------- 2023.06.19 I wrote a small program to export the wiki blocks as HTML files. It's written in Retro, and runs pretty quickly. I'll be working on a variation for Gemini as well. ----- Today was hot (31C); I'm hoping it'll cool down some tonight. It's just barely entering summer and I'm already tired of the heat. ----- Tomorrow I'll start work on words to identify missing links in wiki. ---------------------------------------------------------------- 2023.06.18 Happy Father's Day I fixed the basic formatting issue with the new (list:syntax) blocks. It's now at a minimal, but functional state. Arland will probably work on the color selections before we merge this into the main ilo blocks. The most recent updates to iSH seem to have fixed most of the problems I was having. It's still slower than ideal, but I can live with that. It does still segfault with my non-libc binaries though. I'll do some investigations into this over the summer. It's warm, bordering on hot, outside. Tomorrow is suppose to be worse though, so I'll be trying to stay indoors if I can. ---------------------------------------------------------------- 2023.06.17 No coding so far today (19:37). The pain is down to my 'normal' background levels. I'll try do some programming tomorrow. I will be increasing the ilo.blocks in the konilo repository to 4,096 blocks. This is a big increase, but I am expecting to use a fair amount of this with documentation related blocks and more examples. Arland did some image logos for Napia & Arks. I'll be updating the Napia HTTP page soon, and we'll get a placeholder up for Arks. ---------------------------------------------------------------- 2023.06.16 As expected it was a long, draining day at work. Thankfully we will be mostly fully staffed next week. Here's to hoping that I can get some rest over the weekend. Arland helped add a "Leap" function to the new `titles`. This lets you leap to the start of a block set, by providing the first part of the block title. It's really handy, and I'll be merging it into the ilo repository tonight. Other than that no programming today. ---------------------------------------------------------------- 2023.06.15 We didn't get home from Arland's graduation until pretty late yesterday. I'm glad I did a partial day off/work from home; I'd have been exhausted if I had gone into work. Today's work has been a little expansion in the (wiki) blocks. I refactored `wiki:render` to separate out `wiki:locate`, as a step towards some later words. I'm not going to do much tonight. My right wrist is in a lot of pain tonight. Hopefully tomorrow will be better... ---------------------------------------------------------------- 2023.06.14 Arland graduates today. I'm happy for him, though it's been a stressful couple of weeks leading up to this. I took a (mostly) off work day since the graduation is at 18:00, and I didn't want to be exhausted after work (unlike with Glen's last year). Konilo: the initial wiki code is done. I have a few more words to write, but I'm now able to start writing things for it. I expect to grow this in interesting ways over the summer. The rewritten (wisp) is working out nicely, and I have the server side stuff for generating the .txt completed. ---------------------------------------------------------------- 2023.06.13 Maybe I should add a "jump to block starting with ...." key to titles. This would make it easier to quickly locate sets of related blocks. I've long liked the idea of a personal wiki. As part of using Konilo as a personal knowledge base, something like this would be very helpful. I have a few ideas on implementing this; I'll be starting to experiment on thesee tonight. Based on some quick estimates, I expect the backup set (assuming 1 backup per day) to be around 160MB per year. This seems small enough to be easily dealt with on modern systems. ---------------------------------------------------------------- 2023.06.12 I couldn't sleep last night, so I'm quite tired. I need to get some sleep soon. I'm moving my old WISP blocks into this bigger set. I'll have room for more entries than I think I'll ever actually write, but that's fine. I'll rewrite the words for generating the text output and starting new entries. With a much larger space for the blocks, I added a "4 Jump" command to `titles`. With this I can quickly jump around in the block listing. Enough for now. Time to head home for dinner, a shower, and (hopefully) some much overdue sleep. ---------------------------------------------------------------- 2023.06.11 Very tired today. My blood glucose started ok, but spiked and stayed high. I have setup a new project directory for my personal konilo branch. This is not fossil tracked, but does have some limited backup/restoration scripts. There is also a nightly cron for the backups. This system has 65,536 blocks. By my quick calculations, WISP will use 18,250 blocks at a rate of 1 per day for the next 50 years. At that point I'd be 90; I think this will be enough, unless I decide to try somehing really crazy like storing video in the blocks. ---------------------------------------------------------------- 2023.06.10 Took a break from programming today, and from the manual work. Arland asked about the possibility of having konilo.sh handle multiple projects. This could be done, but at the moment I'm not sure about possible issues that this might introduce. I'll be spending a bit of time considering this. We might do Python & Retro alternatives to konilo.sh. I am considering how to improve performance and reliability of HTTP(S) access to retroforth.org (and the other domains). My Casket server remains slower than ideal, and frequently gets tripped up by improper HTTP headers being sent. ---------------------------------------------------------------- 2023.06.09 Working on the manual; mostly documenting the konilo.sh script. I have added backup/restore functionality to konilo.sh. This is something I wanted for a while. It includes an option to remove backups older than 30 days. There is some room here to explore more, but I don't want this script to get too long... ---------------------------------------------------------------- 2023.06.08 The smoke from the wildfires is starting to clear. It looks like it'll linger for a few more days, but we should be past the worst of it. ---- I'm working on the manual. There are going to be a bunch of new things added: sections on strings, characters, buffer:, byte addressing, assembly, termina, tuhi, and more. It'll be a quite significant expansion of coverage for the Konilo system when done. I think I need to rebuild forth.works to improve the partition layout. The auto one I've been using is wasting space I need. ---------------------------------------------------------------- 2023.06.07 I've added documentation (in the form of glossary blocks) for all of (std), (nonix), the extra tools, and most of (termina). This pushes the overall block usage to 64%. I expect to reach at least 70-80% usage as I get back into the manual. I will need to grow the block space soon. ---- The air quality today is horrible. The Canadian wildfires are causing heavy smoke, even here in southern Pennsylvania. This made for a colorful sunset yesterday, but today the sky is just a haze of smoke. Hopefully the weather will change to help return to normal. ---------------------------------------------------------------- 2023.06.06 Working on documentation today. I have basic glossary data now for the (std) words and am about halfway done with (termina). Later this week I'll add the Glossary: blocks for these. It will consume another 135 or so blocks though. I anticipate running short on blocks by the end of the month. Increasing the block store is doable, but it will make the local size considerably larger. I'm planning to move from 1,024 blocks to either 8,192 or 16,536 blocks. Both seem pretty big, but that isn't likely to be a problem on most systems. I'll provide a variety of sizes (all preloaded w/the system blocks) soon for those wanting or needing a smaller block store. These will not include the full documentation, as that ends up consuming much of the space. ---------------------------------------------------------------- 2023.06.05 I was able to wrap up my work project by 9:30 am, so had most of day to rest. This was good as I ended up not going to bed until 4:30 am. Staying up to read several books was probably not the best idea... I watched the WWDC keynote. Nothing really noteworthy apart from the "Reality Pro" announcement. I'm not planning to do any AR/VR stuff (currently), and don't have funds for a headset anyway. It will be something to watch a bit, much like the ongoing AI stuff. We made a birthday cake, but won't serve it until tomorrow as it's too late for me to safely eat tonight. ---------------------------------------------------------------- 2023.06.04 I am now 40 years old. ---------------------------------------------------------------- 2023.06.03 It's Saturday. Going to watch Guardians of the Galaxy, Vol 3. Update: the movie was enjoyable. The CGI & effects weren't great but this didn't prevent me from having a good time. The only actual issue I had was with the air conditioning in the theatre: it was too hot. PECO was supposed to shut of power to do some line upgrades, but didn't do so. This is the third or fourth time now; I'm getting quite irritated with them. ---------------------------------------------------------------- 2023.06.02 I'm going to be spending the weekend resting a bit, reading, and hopefully not doing much computer work. I have Monday partially off work (I have one project to finish, but can do this from home), so this will hopefully be a restful time. I think I may need to increase the size of my default set of blocks. Of the 1,024, I am now using more than half, and, if I keep going towards some of my projects, will likely use most of the remaining ones over the next few months. The question here is how many blocks should I add? ---------------------------------------------------------------- 2023.06.01 I added a new (nonix) vocabulary. This provides a few words (cp, mv, cat, locate, and so on) to help in the block reorganization process. I added `n:get` to (std). This really should be in the image, so I'll add it there in the near future. (mandelbrot) now has a (termina) based interface. This puts it on par with Arland's (graphica) version. (It also means that the set of (termina) programs now increases to four. More will come) Updating (graphica) [and hopefully finishing it!] will be one of the summer projects. ---------------------------------------------------------------- 2023.05.31 The end of May. The start of summer. It's getting notably warmer outside. I'm not a fan of this. With the new `titles`, I've begun reorganizing the blocks a bit. This is somewhat overdue, but once I'm done there'll be a nice set of empty pages between my examples and the documentation. ---------------------------------------------------------------- 2023.05.30 The new `titles` is pretty nice now. It presents a page (16) of blocks, allows flipping through the pages, and jumping into the editor. It also allows changing the title line or display of a block. With this, I can much more easily see what I'm working with and find a space in the blocks for a project. It also has prompted me to think about the blocks a little different: perhaps a page of blocks should be enough for almost all of my programs? ---------------------------------------------------------------- 2023.05.29 I have the miniwi font working in an ilo+graphica build now. It isn't ideal yet (there's some header junk I don't really need), but it does work nicely. I didn't do anything on the error device code. Some time this evening on the new (termina) based `titles` tool. No coding, just planning. ---------------------------------------------------------------- 2023.05.28 I'm working a bit on converting the miniwi BDF into a format I can use. I probably won't finish on this today. For Retro, I've made a little progress on the remaining issues with the new error handling device. If my hands are ok tomorrow I'll try to clear up the remaining problems I'm aware of. With tomorrow being Memorial Day, it's the start of three short (4 day) work weeks for me this year. Looking forward to a bit of a break from the daily grind. One more thing: I'm starting to dabble with writing a compiler for something that resembles a C like language. Not sure if it will pan out though. ---------------------------------------------------------------- 2023.05.27 I slept in today. Haven't rested this well in weeks, so it was nice. Ate some noodles w/veg & sweet chili sauce for lunch. I found a nice little font (miniwi) that we will try to adapt for use w/ilo+graphica; this should allow an 80x30 display on a 320x240 display, which will work quite nicely for my purposes. Or a 180x90 display on Arland's 720x720 targets (!). ---------------------------------------------------------------- 2023.05.26 Konilo does not have `words`, as it's not really doable without saving actual names. But I did write a `words` (loadable from tools/words) that displays the names of anything with a Glossary block. I also moved around some of the (std) blocks, to leave room for modified pali assemblers with larger numbers of instructions (in prep for napia & arks). Arland is planning to make some instruction set changes to ARKS. His changes will replace some of my superinstruction handling I had planned. More on this as it develops. ---------------------------------------------------------------- 2023.05.25 Hands are hurting today. I'm not working on Retro or Konilo, though I did write some stuff for my employer. I'm looking for small fonts. I'd like to get an 80x25 (or more rows) that's actually somewhat readable on a 320x240 display. This would get us closer to building a Konilo-only system around the Teensy 4.1. I may be able to order the FRAM chip I'm hoping to use with the RP2040 soon. ---------------------------------------------------------------- 2023.05.24 In ilo.py, I have separated the big if/elif block into a dict based dispatch, and started adding handlers for some full inst. bundles. ("superinstructions"). With this, the Python startup is now much faster (about 24s, down from 34s). Still a lot of work to go, but as we decide on additional superinstructions, it should get closer to being usable. ---------------------------------------------------------------- 2023.05.23 Keeping this brief as I'm starting to fall asleep. ilo-x86-native; the embedded image now has a consistent format. retro: I did some work on the error handling device. It's still exposing issues with the handling of fatal errors. I'll try to fix these sometime tomorrow, but it seems to be taking longer than expected. Heading to bed now; goodnight. ---------------------------------------------------------------- 2023.05.22 Konilo: I updated the .sh script to make a few improvements: curl now pipes directly into gunzip; machine names are now fixed to a standard set (e.g., x86 instead of x86, i386, i686). Other than this, I also updated the startup code to restrict the blocks being searched during startup. This lets the Python VM start in about 34 seconds on my OpenBSD system. I/O under the Python VM is slow. We'll explore some caching in the near future. I've started gathering data on the specific instruction bundles and their usage frequencies. With this Arland can start making some "superinstructions" in his VM that should allow faster execution. ---------------------------------------------------------------- 2023.05.21 UTM.app for macOS is quite buggy. After three attempts I managed to get a 32-bit x86 OpenBSD installed, so I now have another binary for the collection. This is currently the C VM, but I'll be able to start work on a non-libc backed one now. Arland successfully finished clearing the old garden bed and planted his sprouts and seeds. Now the long process of waiting and tending begins. It should be a good experience for him. Dinner tonight was burgers & some popcorn. This will be repeated over the next several nights. Initial stab at a quick port of the amd64 to use libc backend is not yet working. I'll need to look at the ABI docs a bit. ---------------------------------------------------------------- 2023.05.20 Arland has noticed that I don't publish updates daily. I'm going to try to push the latest export more frequently. For Retro, I'm continuing to work on the error handling bits. It is going slower than I would like. Maybe I can finish this over the next couple of days? Not certain though. For non-Forth stuff, I'm making headway on finishing my schemas for Convey, and have begun work on the code to read and update the data. This evening I'll do some reading on the requirements for push notifications. ---------------------------------------------------------------- 2023.05.19 I'm tired today, so not really doing much programming. I finished reading the Paksenarrion books by Elizabeth Moon. I still enjoy this series, but am looking forward to choosing a new book or series to start. Dinner tonight is tacos. I'm using a bit of meat, a lot of mushrooms and lettuce. For flavoring, a blend of salsa, tomato, and a bit of lime. ---------------------------------------------------------------- 2023.05.18 Beginning to setup a source tree for the next Konilo release. I'm not planning to push a new release yet, but want to have things ready for when the time comes. I started a little work on adapting the amd64 assembly to work with the c standard library, which will make a generic unix target possible. On other fronts, I'll be loading (termina) in the standard init blocks soon. This will facilitate more friendly applications. For Arland's graphica system we will support enough of VT100 to let termina applications work without change. ---------------------------------------------------------------- 2023.05.17 Fixed the arguments under FreeBSD by adjusting the stack pointer Veltas suggests just changing the offsets used instead; I'll make this change later. I was also able to get NetBSD working. Will start on DragonFly. DragonFly has an issue w/lseek; I need to work on this, but am starting to get painful. Will take a break for a bit ---------------------------------------------------------------- 2023.05.16 So on FreeBSD, it has a few issues: command line arguments and system calls aren't working. It does work (other than the args) if I run under truss(1). So it turns out that FreeBSD is clobbering more registers than Linux or OpenBSD. It's now working, after saving/restoring them. I'll take a look at the argument issue later. ---------------------------------------------------------------- 2023.05.15 I'm reading through and testing an amd64 assemby implementation of ilo. This was sent to me by veltas. This is quite nice. It's for Linux, but should be easy to port to OpenBSD. And that was indeed easy; a few adjustments to the system calls and it works fine. I'll give FreeBSD a shot now. Not working on FreeBSD: odd issues. ---------------------------------------------------------------- 2023.05.14 Wishing my wife & my mother a happy Mother's Day. My wife desired steak & potatoes; I made mashed potatoes and a couple of sirloin steaks. The steaks were done in a sous vide, and finished on the stovetop w/avocado oil, butter, and a few herbs. She was pleased with it, so I'm happy on this front. Not working on RetroForth or Konilo tonight; I'll resume on these tomorrow. ---------------------------------------------------------------- 2023.05.13 The iSH app has been updated; Konilo (when built from C) now works. (Specifically; the old release of iSH had bad terminal emulation and would drop random lines from the output; this no longer appears to be the case). The x86 binary using syscalls causes iSH to crash immediately. I'll look into this further. The kids spent the day playing Tears of the Kingdom. They seem to like it. Yesterday and most of today my left wrist & hand have been too painful to use. It seems to be getting better; hopefully it'll be ok tomorrow. ---------------------------------------------------------------- 2023.05.12 My left wrist is very painful today. Not going to be coding or writing other than this block. I started a little thing to use blocks as additional RAM. This should be useful for the time I occasionally need larger data sets. Arland & I had a little discussion on interrupts and using some of the registers to provide information on system state at the time of interrupt. This may require some changes to napia and arks. ---------------------------------------------------------------- 2023.05.11 Lots of work on the canvas stuff. It's getting close to being almost usable. I might be a bit odd in that I'm mostly programming using an Android phone, an iPad mini, and a keyboard. This setup has been working pretty well for me though, and I'm not missing using a desktop as a primary system. I'll write more on this later. ---------------------------------------------------------------- 2023.05.10 I've been playing with both LocalStorage and IndexedDB to see if I can get in-browser storage/restoration of the blocks working. This isn't done, but I've ruled out LocalStorage: the 5MB data limit is too tight to store the full set of blocks. Arland has started working on the canvas support. When I last checked, he has support for drawing pixels and lines. I'll write code to load the font data once he gets the other parts of the graphica code working. Rick sent a TypeScript implementation. This is pretty neat. I setup a Node environment to try it. The TS version has separate the interface and actual ilo instruction processing. I'd like to explore this in more depth, but need to read up on TS first. ---------------------------------------------------------------- 2023.05.09 Arland reminded me that I've not pushed updates to WISP in a few days. I'll do so tonight. WISP is now being mirrored to charles.childe.rs as well. I published a quick update on the ilo-js progress to Patreon. Rick has a variety of ideas that sound interesting. I'll be looking at these soon. I did a little work on the new error handling device for RetroForth. This is lagging a bit behind where I wanted, but I'm preferring to take time to make sure it works as expected. Work on documenting the RetroForth FFI is ongoing. More soon. ---------------------------------------------------------------- 2023.05.08 Work was long, but we're getting caught up after a serious lack of people last week (due to illness) Arland has added color themes to ilo-js. I like the amber and gas plasma themes. The whole thing is progressing nicely thanks to his help. Tomorrow I'll ask him about making the amber be the default and maybe see if he can start work on saving the settings data. I need to start reading on IndexedDB and LocalStorage to see what makes sense for storing the blocks. I need to look for some small containers so we can start the sprouting of the seeds for the garden. ---------------------------------------------------------------- 2023.05.07 Tammie is feeling sick today. I'm just continuing to rest. I've still been congested, though it is clearing up at last. The temperatures are getting warmer. This is not to my liking. Arland is working on some improvements to ilo-js. This has become interesting now. It's usable, and he's made the interface scale with the display size. ---------------------------------------------------------------- 2023.05.06 I'm quite tired today. We did some grocery shopping, then picked up some tea at Panera Bread. After this was selecting seeds (and some soil) for the garden Arland is working on. ---------------------------------------------------------------- 2023.05.05 I was able to get the block stuff working, so a demo copy of Konilo is now running at forth.works/demo . I've published a copy to my patrons. Arland is working on some CSS for this, and I just saw a note from Rick with a patch that will fix the blocking issue. I'll merge this in tomorrow - thanks to both for helping! Kiyoshi has asked me about a Retro or Konilo that uses floating point only. I've not tried this, and it might cause some issues with regards to data in memory, but I'm quite curious to see if I can make something around this idea, so I'll be starting work on a prototype soon. ---------------------------------------------------------------- 2023.05.04 The initial ilo-js is done. I still need to add blocks, and hopefully loading the image & blocks from a server, but it's probably only a day or two away from being publishable. On the health front, I have some lingering congestion, but now feel fine. Modest hand pain today. Blood glucose levels still high, which may indicate that I'm not quite past the illness yet. ---------------------------------------------------------------- 2023.05.03 I'm starting to feel better. It is time to start a javascript implementation of ilo. I'm not good with js; so hopefully someone more experienced will be able to make suggestions on how to improve it once I get it working. ---------------------------------------------------------------- 2023.05.02 I added the `dump` word to the konilo blocks. The main web server has been updated to FreeBSD 13.2. This was straightforward, but more difficult than an OpenBSD update. I also updated the dev server to OpenBSD 7.3. Some chat in #retro and ##forth today. Nice to see activity on the IRC channels. Very tired, so I'm wrapping up for bed and sleep. Maybe I'll be able to get more progress done on projects tomorrow. ---------------------------------------------------------------- 2023.05.01 Still sick. I published the Konilo release. I'll have some block updates to work on. I'm going to add a `dump` word. Enough for today. ---------------------------------------------------------------- 2023.04.30 I'm still sick. It seems likely that it's just a cold, but if it gets worse, I might do a covid test. For now I'll just continue to rest. If I feel better tonight I will try to do a little work on the ilo for RP2040. ---------------------------------------------------------------- 2023.04.29 It as difficult to sleep last night, and when I finally did, I woke up congested and feeling a bit sick. Trying to rest today in hope that I'll feel better tomorrow. For arks, I did some work on Graphica; reducing the number of places where the framebuffer was directly accessed. I think we will be good to go on the 1bpp mode soon, and I've raised doing more abstractions to isolate the SDL bits further. ---------------------------------------------------------------- 2023.04.28 I updated the main retroforth website. I still need to add links to the patreon page and a list of former & current patrons. This should be done tomorrow. I'm experimenting with getting the error handling device working under RetroForth. It'll take a bit longer before it's ready to commit. It's supposed to be a rainy weekend. There's a bit of housework to do, but that'll be pretty quick to accomplish. Arland was going to do some garden work, but that'll probably be pushed to next week. He and I need to go over his plans so I can make sure to get the materials he needs to repair the beds and plant. ---------------------------------------------------------------- 2023.04.27 Another painful day. I did some work on the Konilo (and ilo) web site. This now has Arland's logo (thanks to him for making it; I like the tiny hexaghost as the dot on the "i"). I'll be updating the main RetroForth website tomorrow. It'll be similar to the Konilo one. Websites for the napia & arks systems are coming; probably this weekend or early next week. I'll be publishing the first arks snapshot tomorrow to the project patrons. It's still quite a way off from being ready for daily snapshots, but we'll hopefully get there after Arland graduates in June. ---------------------------------------------------------------- 2023.04.26 I've not done any actual programming today. My hands hurt too much at present. I took my youngest kids to watch the Super Mario Bros movie. They both enjoyed it; Alex found the musical bits to be, in his words, "cringe", but liked the rest quite a bit. Dinner tonight was enchiladas. I made a couple with pulled pork and a couple with cheese. Both were good. I'll probably regret the carbs later tonight. This does get me close to having used the carb-heavy ingredients I have on hand, so I intend to begin a transition back to a very low carb and IF diet starting the end of this week. ---------------------------------------------------------------- 2023.04.25 Busy day at work, with most of the month-end prep work needing to be done. I finished my parts of this, so tomorrow should be a little more relaxed. I'm working on the little tool to translate a simple infix language to pali assembly. The initial parts have progressed nicely. I will be starting on handling conditional and looping forms next. Arland and I discussed optimization possibilities a little. I think I'll probably work on an IR that we can work with before generating the pali code. ---------------------------------------------------------------- 2023.04.24 Should I update my development box to OpenBSD 7.3? I've decided to attempt the update next Saturday, after I am done working on the first Konilo & ilo release. ---------------------------------------------------------------- 2023.04.23 Not a good start to the day. I'm having pain in my left foot and both hands. Also a headache and very tired due to lack of sleep. Spent the morning resting. I'll start preparing the final default block set today. This is probably going to progress quickly as it's mostly just removing all the non-essential stuff. ---------------------------------------------------------------- 2023.04.22 Arland rewrote `titles`; it now allows exiting early. I have started work on refactoring it to be merged into the standard system blocks. Konilo for ilo & napia will be released on May 1, unless I run into any significant bugs before then. I'm not expecting any problems. This initial release will just be the vm, image, and system & glossary blocks. A secondary release (probably in the fall) will add in more stuff to the default block set, as well as more tooling & documentation. Konilo/arks is not yet being released, but I will start sharing snapshots of it on patreon in the coming week. ---------------------------------------------------------------- 2023.04.21 I'm at work on an editor for unpacked blocks (so 64x64 chars!). I really have no idea how to make this practical. I'm crafting an extended version of the basic editor. Arland and I will need to discuss the user interface part of this a lot. I've thought on having an odd model where the block presents as code, data, commentary, and tests; with each view being separate but interleaved. We are talking about a Konilo system, sharing blocks across the three VMs. In this, Napia would serve as a backbone, handling various I/O operations and cron-style tasks, with ilo & arks based systems running the user facing programs. This could be interesting to explore. ---------------------------------------------------------------- 2023.04.20 A somewhat quiet day. I was able to work on a project for work (a quote management tool) today. Using Python and RetroForth. I have a lot of the functionality in place; working on a calendar view currently. Dinner was chicken, gravy, and mashed potatoes. Very pleasant, though not good for my bloood glucose; I really need to go back to IF & keto soon. ---------------------------------------------------------------- 2023.04.19 Arland & I are starting a block editor that will use byte packed blocks. (So 64 lines, 64 characters per line). This is a *lot* of space compared to the normal blocks. It'll be interesting to see how useful it becomes. I'm starting to look into possibly setting up a solar system for the computers. I'd like to start reducing my power usage and reliance on the electrical grid. I also want to see if I can work out a practical, affordable way to do a self-hosted server setup. Both of these will probably take a while to achieve. ---------------------------------------------------------------- 2023.04.18 The random number generator is now a basic xorshift model. It's faster, smaller, and works without any issues. I'm happy with it. I did a little example to play with this; it's kind of fun to watch. I'm tired today, but am going to spend a bit of time reading before I try to sleep. ---------------------------------------------------------------- 2023.04.17 We discovered some major bugs in Arland's random number code. I will replace it with something better. ---------------------------------------------------------------- 2023.04.16 I finished the Common Lisp implementation of ilo. This has been more frustrating than I would have liked, and I still need to go through and fix formatting and add comments. But that'll wait a bit until I can bring myself to stare at the parenthesis once more. I'm going to revisit the (termina) vocabulary sometime soon. I think I'll work on a lightly extended interface for konilo that takes a little advantage of VT100/ANSI capable terminals. RetroForth & nga will see the start of a new i/o device for handling stack (and other) error states this week. ---------------------------------------------------------------- 2023.04.15 Celebrated Ben's birthday today. It was a short (2 hours) event. PECO was supposed to turn off power most of the day, but did not. This was good. It's much cooler today. I was able to get the integer handling working, and found and fixed several other bugs. Interactivity is working, and I can read blocks now. Next will be saving blocks, and then maybe the image. In other news, Arland & I have started some early work on sound support for use alongside Graphica. And some looks at plain X11. ---------------------------------------------------------------- 2023.04.14 Still tired, but made it through the day with the help of a lot of coffee and tea. I started on an ilo in Common Lisp. This will probably take a few days to get working. I did an initial broken implementation; Arland and I then resolved a few issues. It's still not working though. I'll put it aside for a bit, then give it another look next week. The last three days have been quite hot. It's supposed to be cooler this weekend. Hopefully it will be. ---------------------------------------------------------------- 2023.04.13 Very tired. I'm just going to bed; will probably write a bit tomorrow. ---------------------------------------------------------------- 2023.04.12 My hands hurt today. I'm working on Makefile stuff. I think the next release will not have both a Makefile and a GNUmakefile, just one. ---------------------------------------------------------------- 2023.04.11 Last day of this mini-vacation. I had a couple of short work related activities, which took less than an hour to resolve. I chose to spend the day resting. I published an updated copy of the old repl.c for nga to my /temp files. It can serve to illustrate interfacing Retro from C. I will put together a better example before the next release. I did some minor source cleanups to the Retro device interface code. I'm starting to consider the issues w/non-Unu source file support. ---------------------------------------------------------------- 2023.04.10 Doing better today. My blood glucose is still high, but not as dangerously so as yesterday. Breakfast was potato rolls, with cheese sauce, ham, eggs, and onion. I enjoyed it, but Tammie didn't like it. (She's not a fan of onion) A brief visit w/Tammie's mom. We gave her some pictures of the kids. Heading home was uneventful, but long. I need to see about doing better routing for breaks next time. Home a little after 16:30. Tammie is sleeping, I'll do so too. ---------------------------------------------------------------- 2023.04.09 I had a lot of issues today. I couldn't sleep last night, and my blood glucose spiked high. After getting it down somewhat I was able to get the a pharmacy to get a glucose meter & test strips. With this, and a lot of insulin, I was able to get my glucose levels back to a managable (but still high) level by dinner. Tomorrow we head home. Hopefully the travel back goes smoothly. ---------------------------------------------------------------- 2023.04.08 Today started off cold, but it warmed up by afternoon. Breakfast was an egg quiche, with sausage, bacon, scallions, and croutons. I enjoyed it. We went to visit Cindy & Darrell. The drive up was uneventful, and we had a good time. (They do keep their house rather warm; it was just under 80F most of the time). Coming back, we let Glen practice driving for about 30 minutes. This was a little uncomfortable as it was his first time on the road, but he did a good job. He'll probably get more practice tomorrow and on Monday. I'm not planning to do much programming tonight. ---------------------------------------------------------------- 2023.04.07 Breakfast was pleasant. We had fresh fruit, waffles, and eggs. Afterwards Tammie & the kids headed to her mom's for a few hours while I spent the time reading and watching some educational videos I had queued up. This was broken up by a couple of short walks. We had dinner at the diner Tammie's brother works at. It was ok; the food wasn't bad, and service was great. Pricing was average. I had a meatloaf, mashed potatoes, gravy, and broccoli. I'd eat it again. It's now 22:03 and we're back at the B&B. Looking forward to a bit of sleep before tomorrow's activities. ---------------------------------------------------------------- 2023.04.06 While it was nominally the first day of vacation, I was working. We had about 6 hours of driving, and I was given a surprise chance at a very large bid, so I spent much of the drive on this. And to my suprise, I landed a PO shortly before arrival at the B&B we are staying in. And thus I completed the single biggest order I've ever received. I'm looking forward to actually having some rest time now. ---------------------------------------------------------------- 2023.04.05 I'm renaming RetroForth/ilo to Konilo. For those curious about the naming, this derives from Toki Pona: Kon being a spirit or soul; ilo being a machine. So Konilo is sort of the soul of a machine. ----- Tomorrow we leave for a short vacation. This will not be the only one I take this year. I have 2-3 hours of work in the morning, then will be off until Tuesday or Wednesday. I'm not sure of the connectivity where I'm staying, so might be offline for a bit. ---------------------------------------------------------------- 2023.04.04 I've setup FreeBSD, NetBSD, and OpenBSD systems (under vmware) with Arland. Of these, OpenBSD has been the easiest one for me in everything except partitioning (though in this case we just went with the defaults.) I still need to get X11 setup on the NetBSD system. (I can go without, but Arland isn't comfortable in a purely console environment). I've also built gxemul; I'm hoping to try to setup some stuff under less common CPUs to aid in testing overall portability. This probably won't be finished before the upcoming trip. Also planned: a Plan 9 system. ---------------------------------------------------------------- 2023.04.03 I updated the RetroForth/ilo for x86 native kernel today. It's now on the current image, and compiles with -O3. This required adding a memcpy() function. Tomorrow I'll see if I can add an io extension to allow access to the physical memory and x86 io ports. I'm evaluating a few options for progressing towards an actual computer dedicated to RetroForth. I'm still unsure if I should expand the Teensy system to support display and keyboard, or if I should maybe try crafting a system running on a RPi with a very cut down Linux or BSD. I lean towards the Teensy approach though. With this there are less layers, and everything should be simpler to replicate on other hosts in the future. ---------------------------------------------------------------- 2023.04.02 Arland helped identify a bug in the ATA code. I've pushed a fix for it. We have this working under vmware now. Arland wants me to make an ISO image. I'm going to start looking into this. Today is much cooler than yesterday. This is nice; I was able to get a little sleep. The birds are quite noisy today, I hope they quiet down soon as I'd like to try to get to bed earlier tonight. After this upcoming trip I'm going to get strict on my dietary limitations again. I've been having too many carbs lately, and I really need to scale way back. ---------------------------------------------------------------- 2023.04.01 Still hurting. I might try to minimize typing over the coming vacation. Arland found that a recent commit broke RetroForth/arks graphica words. He has pushed a fix for this, and we'll backport the changes to RetroForth/ilo graphica tomorrow. The `retro-ilo.sh` script now supports x86-64 Haiku. I'll be adding a binary for this to the collection. (The current Alpha4 has been quite stable on our test system; I'm hoping to do a permanent install on a similar box soon). We have a FreeBSD 13 system setup locally now. Working on also setting up current NetBSD and DragonFly BSD as well. ---------------------------------------------------------------- 2023.03.31 Nothing significant to report today. My hands are hurting, so I'm skipping any more writing for now. ---------------------------------------------------------------- 2023.03.30 I updated the RetroForth/napia image today. It's now based on my latest RetroForth/ilo image, so all the recent improvements are in place now. Starting to decide what I'll need to bring on vacation next week. I want to pack pretty lightly this time. Dinner tonight was tacos. I've been experimenting with fillings; most of the non-meat based ones I tried were good. I like rice, but more than a small quantity raises my blood glucose. ---------------------------------------------------------------- 2023.03.29 Starting to prepare for an upcoming vacation (April 6 - 11); I have a handful of small things I'd like to work on. I am writing an ilo+graphica that uses an actual bitmap (1 bit per pixel) display. This should be beneficial for a possible play.date port, and maybe getting a classic Mac version with the visual stuff. Tomorrow I'll try to finish the (std) wordlist data. Tonight my hands are starting to hurt, so it's time to stop for now. ---------------------------------------------------------------- 2023.03.28 It's been a long day. Just spending some time reading before bed. Earlier today I worked on RetroForth/nga, starting to split the device code into separate files. For the next release I'm hoping to reduce redundant code and make it easier to add new things. ---------------------------------------------------------------- 2023.03.27 RetroForth/nga (12).2023.3 is out. I've published the source and opened an issue for updating the FreeBSD port entry. Trying to get a pkgsrc tree working so I can try to update the NetBSD port. I'll be looking at the guix port as well. It's time to try to get it a bit more widely available. ---------------------------------------------------------------- 2023.03.26 My blood glucose spiked last night. Wasn't able to get to sleep until after 4AM, and woke up around 7AM. Very tired. Watched John Wick 4. As expected, not much plot, but plenty of fighting. The theatre wasn't crowded. I had an enjoyable time, but was *extremely* tired by the end. Going to bed early. ---------------------------------------------------------------- 2023.03.25 Just took it easy today. I wasn't able to get much sleep last night, but once I did fall asleep around 6am, I managed to sleep until almost 10; which is much better than normal. We picked up groceries for the next week. This ended up costing more than planned, but should be ok if we are careful. Arland started exploring programming for the Panic Play Date console. He & Alex have a small game in development now. I showed Arland some of the old RetroForth systems (11 & 10). He was disappointed in the small starting vocabulary of 10, and didn't like the dictionary system & heavy use of vectors in 11. I'm not surprised by this. ---------------------------------------------------------------- 2023.03.24 It was a long day at work, but I was able to wrap up my month end preparations as well as completing three more pages of the inventory counts. For the evening, I'm just relaxing and reading a bit. Tomorrow I start running through the various tests I have for RetroForth/nga. This will hopefully go smoothly. It would be nice to wrap up work on this so I can push out an update next week. ---------------------------------------------------------------- 2023.03.23 I finished the glossary data for RetroForth/nga; going to start the final rounds of testing tomorrow. Jim Butcher has started writing the next book in the Dresden Files (a series I've enjoyed reading over the years). Hopefully it'll be published around this time next year. In the meantime, he also completed the second book in Cinder Spires; a series I have yet to read. It might be a good time to start, possibly during a short vacation I have coming up in early April. ---------------------------------------------------------------- 2023.03.22 Still having a lot of hand pain. My left hand is stiff and feels swollen. Going to try to avoid doing too much with it tomorrow. I'm helping Arland with some ideas for a comic he's hoping to start working on. I roped in Glen to aid in some of the world design. We have a couple of character descriptions and an early skeleton for the first part of the story. Not doing much in programming today. I wrote an explaination of a few issues someone in the IRC channel was having. Other than that, I'm just writing a little under RetroForth/ilo on the Teensy. I think I'll be making much more use of the system on these little boards in the future. ---------------------------------------------------------------- 2023.03.21 Pain. I live with pain in my hands & wrists every day, but some (like today) are much worse than others. I'm going to have to figure out something to help with this as things aren't really improving. In spite of the pain today, I have finished the glossary data for all but one word in Retro/nga. I'll wrap that up tomorrow, along with the other thing I did today: glossary data for the (std) library blocks in Retro/ilo. This will get all the Retro systems closer to a release. Retro/ilo+teensy4.1 is working nicely. I'm making a few tweaks around line endings, but it's running fine and I'm actually using it now for some of my writing. ---------------------------------------------------------------- 2023.03.20 I was very tired today. But it was quiet at work, and I feel ok now, at the end of the day. Retro/nga is nearly done. I still have ~20 words to document, and then need to run through final tests & write the final release notes. But it should be doable by the weekend. On my goal of having a little portable Forth machine, I think I know how I'm going to proceed. Still need to source some stuff, but I have most of the parts other than the display. For Retro/nga (and the others), I'd like to offer a nicer interactive environment under ANSI terminals. Some work towards this has begun. ---------------------------------------------------------------- 2023.03.19 I did some work on RetroForth/arks. This was around loading sets and applications. Specifically: `needs` now creates an indicator and doesn't reload block sets that were previously loaded. And I added a * sigil that loads an (app:....) block and runs an application. This can be used to have an application loader that imports any dependencies, then runs the program in question. That's probably it for now. I'm pretty tired this weekend, so as it's almost 19:00 will begin winding down for bed now. ---------------------------------------------------------------- 2023.03.18 I'm interested in things like a Pinephone + keyboard, the MNT Reform Mini, and old palmtops like the HP Palmtop 200LX or some of the early Windows CE devices. Computing on the go has been something I embraced years ago, and I'm always looking at ways to do things better. I mostly work on an iPad mini and a Samsung Galaxy S22 Ultra. I find these to be portable enough to keep at least one with me throughout the day, but I do kind of prefer the iPad; mainly due to the applications, but also the size. What I like better on Android is the existence of Termux. I wish something like this was an option on iOS, but nothing I've found there really works for me. ---------------------------------------------------------------- 2023.03.17 I'm using this as a sort of journal. The notes here won't be long (just a block each), and there's not going to be any actual topics. Just whatever I feel like recording on a given day. Today I'm having some RSI problems. The pain is back, and it's the worst I've felt in the last month. This is somewhat worrying but I'll manage. The frequency with which the pain is returning is increasing, so I will probably need to reduce my typing again. It's hard for me; writing is a major part of my life and I hate not being able to pursue it. ----------------------------------------------------------------