I was able to side load nightly builds of xovi, appload, and koreader onto the re/Move, so I can now read books much more comfortably.

No real programming on my personal stuff today, but I did make a small QOL improvement to Para, and have some more concrete plans now for the OS for ilo.

We moved the NVR to the sound booth, and it remained up & not rebooting back there. Why? I have no idea. If it stays stable we'll probably redo the camera wiring to go back here.

My glucose was a bit unstable today. Not really high, but a bit annoying.

The new Amene vocabulary for Retro is going better than the first attempt. I stopped at 1730, with a working `ls` for the root directory. I'll do a little more after dinner, and then break for today.

I've also done a little on the RDML spec for use with Unu. It's not a high priority project, but I should have it done this month.

We discussed looking at bringing Para to a point where we can release as a product. So next week I'll see what's needed there.

Amene has Py versions of the C tools; I fixed the nested dir bug *and* a FUSE driver, so an amene volume can just be mounted and used on Linux.

I have a bit of code in Retro for this, but think I'll restart from the ground up tomorrow. The bits I have aren't working and just feel off.

I posted a Patreon update.

The Re/Move tests are going well. It's ok for reading, and good for notes. No advanced stuff like the Nomad, but it's way better than the Go7 for writing.

In the evening, I found a bug with nested directory handling in the C code. I'll look at it closer tomorrow.

Set up the new mattress in the evening, and later picked up some tea. The evening was spent listening to blues, and a bit of reading.

Around 2130, Dustin asked if I was free for lunch. So I have a meeting scheduled tomorrow.

A new mattress for my bed should arrive tomorrow. It'll be nice to have something in better shape & w/o broken springs.

The reMove: I like it. I've managed to get part of my library on it after converting to ePub. The device converts them to PDF on opening. It's not great with ebooks, so it'll be nice when I can run KOReader on it. But it's nicer to read on this display than on the Go7. I also have a daily scheduler PDF that looks good, and matches up well with my daily needs.

The reMarkable Paper Pro Move (too long; will use reMove). This is kind of fascinating. It's the same height, thickness, and weight as the Nomad. It's notably thinner in width. The color screen is much, much cleaner & brighter than the kaleido screen on the Go7. It's actually readable in normal light w/o the front light. Though using USI, the stylus it has is much more precise than the generic ones I've tried. Might be a custom software that builds over USI? Will look into it.

No KOReader yet, but I should be able to side load it on once an unofficial extension framework (xovi?) is updated.

In the afternoon I worked on Para-G, fixing a display bug in the orders listing.

Two cups of coffee, a cup of tea, lunch, and back home by 1425.

I've ordered a reMarkable Paper Pro Move, due to arrive sometime tomorrow. I'll be testing it for a few months to see how a slim color device can be used alongside the Nomad & Viwoods.

(I might have too many eink devices, will have to look at what to do with some of them in the near-ish future)

I had a lot of emails at work today, but it wasn't overly busy at the counter or phones. Next week we'll resume some cleanup & inventory projects.

My new dice pouch & the Nomad writing risers are shipping this weekend, so hopefully I'll have them sometime next week.

I'm researching reMarkable's new product a bit. It looks quite interesting.

I'm not going to write much more today. I'll hope tomorrow is a little better.

I have about 90% of the Amene spec done. I also have a prototype tool in C to operate on Amene formatted block stores. I'll take a brief look back at at extent based model instead of the inode based one, but I think I'll be sticking with inodes.

I've also begun sketching off ideas on the OS design.

I'm picking the filesystem stuff back up. It's time to finish up the specification, implement it, and write a custom OS for ilo using it. (This will *not* be a Konilo extension, though I may write a Konilo vocabulary for operating on the filesystem).

The incoming network went down during the morning service. I was able to get it working before the afternoon service. I'll look into better redundancies for this in the future.

I finished readinv 14 novels this month. I hope to maintain a similar pace next month.

Mg plans for a non-Forth OS for ilo are progressing. I should be able to start on this soon.

In landscape, with an external keyboard, the Go7 lets me run Konilo in a very readable size. It's pretty nice. (This has now become the main device I'm running Konilo on).

I decided to try to complete Amene and to build out the full Nonix userland on top. Think a light DOS/Unix style single task OS, but all coded in Forth. There will be a shell. My plan is to just distribute this as a standard Konilo ROM and a block set that initializes and runs Nonix. Maybe even borrow some ideas from DuskOS towards a C-like language on top in the future.

The initial ARKS spec work is done. I still need to add detailed descriptions of the ten new instructions, but the non-I/O part is probably pretty close to done. It's a minor expansion of ilo: adding a second core and interrupts, as well as additional RAM and more standard I/O devices. It also exposes the stacks via RAM. As designed, a konilo image will run unchanged on it, so we will have a solid software system until Arland writes his custom system on it.

I spent the evening resting.

Very light turnout at church in the evening. Hopefully we'll be back to normal after Sunday, with the end of summer vacations.

Dustin sent me a list of ideas for Para additions. I've begun organizing these, and may start some implementation towards them soon.

I spent much of my work day addressing billing issues. Then a trip to church to setup a stand for the ATEM, a visit to the chiropractor, and picking up a bit of food for dinner.

We had a few small technical issues, but serviceswent well.

We took Alex to get a new SIM card since the original died after a water related incident, and I ended up with a headache.

New software updates on the Boox Go 7 and the Supernote. The SN adds some digest improvements and text boxes, which might be very interesting. For Boox, there's a new infinite note type (don't care about this at all; I like pages), and (much more important for me), fixes to auto rotation.

I'll make sure to upload my WISP blocks tonight. It's been a few days I think since I last pushed them.

Ended up being ok. Walk, coffee, lunch. Reading. Some plans for Napia work. And a session of SoloDark with a new story. Now at 1820, going to do dinner, then read & rest.

I picked up my prescriptions ($280!) and groceries, then tried to rest in the evening.

At work we found a possible payroll issue that may have been ongoing for some time. I'm starting to investigate this now.

No programming on my projects today.

We landed the $73k order.

I should be landing a $70-80k order soon though.

Programming for work: para-branch, rolling out the wire random labeling support for NES, and some UI fixes. I also began work on a para-time update to add support for tracking schedules and automated reminders, but this won't be implemented today.

For Konilo, I have my initial blocks for unix:time and unix:fs ready to load into my personal set. I might do these tonight, or I may end up waiting until tomorrow. We'll see how I'm feeling in a few hours.

The wireless bridge was working today, so streaming via the ATEM went well.

I ended up with a severe headache and exhaustion, so didn't do any programming in the evening. This is a bit sad, but I might be up to a little more tomorrow.

Around noon I went out for a walk around the mall, and a stop at Uno for lunch. I only took the Go7 and a folding keyboard, so the mid-day dev work is being done on the eink tablet. I do note that running ilo under Termux uses more battery on this than I 2 would expect. I am still getting about 14 hours of use per charge.

Today's tasks: finishing the time device and starting to add in blocks for interfacing with these non-standard bits to my set.

Back to Forth. I published the HFS stuff for ilo today, and an update to the IRC logs, adding HTML output.

I now have more info on the power situation at the church, so have some ideas on what to look into to begin fixing things.

Having severe power issues at church. Highest voltage I see is 101V. Will dig further into this.

These are 20-22 hour days for me.

I finished the host file system write functions today. (For ilo)

The Go7 is pretty nice as a tiny tablet. Simple games like Stardew Valley and lots of reading-centric apps work well on it.

Trying to get some needed rest this evening. Back to Retro tomorrow.

Color is interesting. It's quite novel to see colors (even tho' they are faded/muted) on eink, and I like it so far. Battery use is ok; I am at 43% and have about 6 hours of actual use so far.

I'll stick with mono in the termina programs for now, but I may adjust my color choices to work better on these ink displays.

Church tonight, and I plan to go to bed early afterwards. I'll upload my WISP blocks today, and do a little more work on the Retro binary builds.

I picked up a split folding keyboard. It feels ok, but I have to remember that it's qwerty, not my custom split layout. A little annoying, but I should be able to adapt over the next few days.

I removed the wifi bridge at church for now. More on this later.

For Retro, I have built the amalgamation on amd64: Linux, Win11, FreeBSD, and OpenBSD. Still have to do an aarch64 Linux, a macOS build, and then try to get a few x86 binaries built. Release day is set for Friday.

Less tired, but still resting in the evening.

It looks like I may be getting a Boox Go 7 (Color, Gen 2) for testing. I'm kind of looking forward to that; I've never used a color eink display before, and it might be a little faster than the Go 6 I use as a tiny mobile computer for Konilo.

I am very tired, so going to bed early.

Minor glucose events during the day. Not enough to be a bother, but still concerning.

Tomorrow I'll start working on the binary collection for Retro.

I was able to finish the RetroForth updates, and have posted the release candidate to my patrons. Planning for a release next week, along with binaries of the amalgamation for a variety of host systems.

Now off to do a bit of website work.

I'm writing today's entry on the small Boox Go 6. Despite no pen support, this remains oddly compelling, and the screen has the beat quality of all my current eink devices. It still makes a great tool for on the go reading and programming in Forth.

As the month ends, I finished the billing at work, did some cleanups, and began to prepare for the weekend. I didn't end up programming again, but will get back into RetroForth tomorrow so I can get the new release done for the weekend.

We're having a sales meeting on the 11th of August. This will be my first time attending in a management role.

Dustin was in the store briefly, but will be out again until Thursday or Friday.

On Retro, I was able to get half of the remaining issues done. I'll try to finish the rest in the next day or two.

I spent my afternoon reading and resting. Back to work tomorrow.

Afterwards, we went to mom's and spent time with Joe and his family, who are briefly visiting from Grenada. It's nice to see them again. It'll be a few years until they're back here.

No programming yesterday or today.

Assuming a similar number of missing people, and a reasonable scaling of efficiency, we should be done by 17:00. Maybe as soon as 1600, but that'll probably depend a bit on the length of the breaks people take.

I have finished an initial read-only Para for iPad/iPhone. It's working, but I'll be refactoring it soon, before I start on the addition of editing functionality.

Didn't end up getting to Retro during the day. Maybe tomorrow...

Programming today was mostly on Para. I started a new iPadOS/iOS application for it. So that's a nice change of pace. It's using the REST API instead of embedding web views, and the basics for finding orders and transfers are working correctly.

Tomorrow I'll be trying to do some work on Retro.

We had a major issue with our server at work today. I was able to get things back running by around 14:00, but it was a very stressful time. I'm talking a bit with Dustin about making the setup more redundant in case of future issues.

Glen ordered a nomad for use in his D&D games. I think it's a great device for this, and he should be able to get some decent use of it. I'll help him setup character sheets and whatever other templates he may need for this.

Viwoods has shipped the new cover for Aipaper mini. Hopefully it will be nicer than the canvas one that ships with the unit. That one is ok, but holds dirt, and is getting frayed after about six months of use.

Before & after the walk I spent a bit of time on Para-branch. I was able to get a number of things done, including a log viewer, date based log rotation, updated API documentation, and an email based workflow for approving/denying PTO requests.

Spending some time in Konilo, working on my personal wiki, and dabbling a little with double-cell math. I might have a workable path for this. It'd be helpful for a few things, but is not a thing I'll need very often.

In the evening, I was able to complete 12 items on my todo list for the new Para-branch code, including three bug fixes, two new reports, and a number of UI improvements.

Not a bad day.

Programming on the new Para went well. Another ten items dones, and I should be able to fully retire the last bits of the old system by early next week.

I did a bit of work on backend stuff for the new church website, and a little coding on Para-branch for work.

For personal projects, I am making headway on the retro.c amal- gamation fixes, but won't be able to test the building on a Windows machine until next week.

We did a count on generators, and I am now cross referencing it against orders to see how off we are. This will take a few days to complete.

Still quite tired. I will need a rest day soon.

I asked people to help in the back, not everyone did, so I will assign those who didn't 1 hour slots there tomorrow. No one gets to slack off when there's work to be done.

I am tired again, and didn't read today.

What should I work on today? RetroForth. We'll see if I can fix the rest of the amalgamation issues...

I didn't finish the amalgamation fixes, but did get a few small Nga implementations done (ported from the C), did some exploring on the Makefiles, and then did an ilo using sqlite3 for block storage. So quite a bit of interesting things.

I added a notes section to para-branch, and fixed a few small bugs. There's still more to do, but it's already proving to be very nice to use.

I'm doing a lot more with the Supernote Nomad lately. Mainly in writing. It really is my favorite eink device these days; even though it's not good at running most applications.

More bug fixes tomorrow, probably. Tired today. so going to rest now.

Church wentfairly well, but Dr. Fedena had some confusion with the lapel mic & clicker. I'll try to make this a little more streamlined Sunday night.

Still hot and humid. I hope it starts to cool down soon.

Work tomorrow will be ... not fun. I have a list of things that need doing, so we'll see how well people do at completing the tasks I assign.

Not doing any Konilo coding tonight, but am sending Arland the ilo c# stuff and build instructions.

Programming: back to work on para-branch. I have about six items left to finish before we go live, so want to try to complete as many as possible today.

No programming today. Just some reading, finishing a book and starting the next in the series. I'll finish these this week.

Glucose came down after the walk. I spent time working on a set of tests for ilo implementations.

In the evening, I had some tacos, then my glucose spiked high and stayed up.

Today was answering questions for coworkers, doing a bit of remote troubleshooting on the label printers in Emmaus, and a lunchtime walk & getting a coffee.

I didn't end up doing much programming today. I'll wrap up here for now, and upload my blocks so WISP.txt will update.

Church was already cool when I arrived, thankfully. I finished the slides for several more songs, updated some notes on the status, and did a little work on the ATEM configuration. I'm starting to have issues with the Amazon ethernet dongle, so may need to replace it.

The church service was good. We had a guest speaker, Charles Hamilton, who is planning to start a church in Philly. His style is a lot more like the southern preaching I've enjoyed over the years.

I provided fans for the desks at work, this will help keep us a bit more comfortable. I set up a cleaning schedule for the bath rooms, and did a bit more work on the para-branch code.

Programming today? Improvements to form handling in para-gwh; more backend work on para-branch; and I started writing words to do proper double cell maths in konilo & retro. I have a working design for this, but it'll take me a while to get it typed in & tested.

My glucose remains running low. I don't know why, but it's lead to me not sleeping well. (Hard to sleep when my glucose is as low as 45!)

The only good thing with the lack of sleep is more time to read, but keeping a constant eye on glucose levels in the 50's is proving to be quite exhausting.

Glucose remains low. This is very odd; I've had points where it was low for a bit, but never this long.

Not doing much coding today, just rest & reading.

I implemented a software floating point system in C, for testing purposes. Not sure I'll actually use this, but it could be of some use.

At work I did a bit of programming on an updated label printing interface. This needs a bit more refinement, but will combine the two separate ones (transfers & orders) into a single one. I will have to do a bit of thinking on how to cleanly migrate it to use the new REST API for para-branch as we transition to it in July.

I did a little work on the PTO calendar stuff in para-branch, and worked on the amalgamation issues. Not quite there yet, but maybe another day or two if I'm lucky.

I've done a bit of playing around with Arland's qhautau32 stuff, but not worked on a Konilo implementation yet. This makes me curious as to how hard a floating point (for Konilo) would be.

The Samsung Book Cover Keyboard Slim (long name...) for the Tab S10 FE isn't too bad. The keys are a bit small, but I can type on it. I'll still use my ergo keyboards primarily, but this is going to be handy at times when I'm away from my desks.

For Para-GWH, I setup a separate application for submitting the warehouse forms, and debugged a bit of email related issues. It will form the start of the email part of Para-Branch later this week.

I committed an update ilo-aarch64-linux.s and the new node.js ilo vm. More to come in the near future.

And my glucose kept spiking high, even when I wasn't eating.

I got a keyboard for the Android tablet. This makes it much more useful to me now. With Termux, Retro, Konilo, Claude Code; I can do a lot of work locally for the first time in quite a while. (Other than just doing Retro & Konilo stuff).

Some work on Para-GWH, implementing digital versions of forms for the stores to use, and some refactoring on Para-Branch. Then reading in the evening.

I'm doing a little work on my wiki tonight. And then reading.

I'm now considering a UTF8 <-> UTF32 version of ilo. It might be nice to have a version that could support this, as it'd let me make a lot of use of Unicode in the blocks.

It's been a few days since I last uploaded the blocks. I'll do that tonight...

At work, I switched Gilbertsville to the production system and then had a slew of issues with using it on the iPads. These are now fixed, but it was pretty stressful.

It was Father's Day. I received a nice mug at church, and had a quiet afternoon.

I spent a bit of time working on the SoloDark tool. It's going a little slower than I'd like, but I should be done later this month.

I did a bit of research into some things relating to my portable computing setup. I might make some small adjustments to this in the near future.

At work, I did some stuff in the new Para-Branch code. This now has display & staging of transfers, and started the backend for managing the label printers. Most of what's left is building out the email, printing, and some backend stuff. I am hoping to get most of this done next week.

I've noticed that when actively using claude code ("cc"), I hit the limits after about 2 hours, then have to wait 2-3 hours for a reset. This is moderately annoying.

While the AI code gen tooling is proving helpful (esp. given my RSI issues), it's still time consuming to review the generated code. And it's not a replacement for handcrafted code. I love programming, and while I'll likely make use of things like CC for work projects, and things that aren't as interesting to me, I'll keep writing code by hand in Konilo & other languages as long as I'm able.

In the evening, I used claude to take some existing prototypes and build a web application for browsing an ilo block file. It works pretty well, which is nice. I also had it generate a view for browsing the manual blocks directly, which is working after a few small manual adjustments.

The new iPadOS multitasking updates might be useful. Not so much on my mini, but if I go back to using a larger one, this could be helpful.

In the late evening, some reading, and started claude on a task for tomorrow.

Tomorrow I'll see if it can help automate some creation of the reporting templates, which would be very helpful as my hands are not in great shape at the moment.

For most of the day, I was in Emmaus, scanning in packing lists, and doing a little work on the shelves.

I've resumed work on my own gauntlet-style horror themed story, which I hope to run later in the year.

Evening was quiet, I read a bit, and watched a few space themed lectures.

I have done a little more on the -impure ilo code. Hoping to test the basics tomorrow.

No work on ilo today, but I have implemented a number of fixes to old parts of Para, and begun the actual work on a complete rewrite.

For ilo, I started the -impure branch, which begins with adding a host file system device. It's similar to Retro's, but much more limited. The -impure branch is not currently planned to be included with the source tree, but will be provided to Patrons and might be published as a secondary option for those needing it.

(My interest here is to see if I can move some of my Retro stuff to a Konilo with a few extra devices for host interaction. But with a much, much simpler VM).

I ended up with a big headache, so just rested in the afternoon and evening.

Blood work: all values good, and A1C is down to 5.3.

Left work early to get blood work done.

No problems with the live stream setup at church. Assuming this this goes ok on Sunday, I'll move on to the sound eq bits.

No one at work know I'm doing this yet.

No work on Konilo or Retro today, but I did fix some bugs in the label printing programs, so they are now in full use. I need to complete a port of them into C# sometime soon; just want to make sure there's no issues left with the original version.

I didn't win, but a fair number of people were brave enough to try my entry, which combined beef, chicken, sausage, pinto & kidney beans, and a variety of peppers: bell, long hot, habenero ghost, and trinidad scorpian.

I'll let the left over portion sit for a bit, and expand it out to a full batch later in the week.

It's been a long time since I did a solo D&D style game, so it's fun to pick this back up with a simpler system and see what happens.

Today I began (in Python) a BASIC [subset] to Pali assembly compiler. I hope to have some non-FORTH stuff on ilo this year.

I've begun a ground up rewrite of Para, the in-house order managing system we use. Time to get it cleaned up, documented, and in shape to maybe start bringing in others.

I did a little work on the new block-backed file system design. It's still far from ready to share, but I'm getting happier with it.

I don't really have high hopes of things improving until more drastic steps are taken. But we've been lax for too long, and there's a general apathy/negativity about the place, and it's way more apparent the longer I've been away on the other projects. So, while I've never really sought to be a manager, I'll step up & try to fix things.

In the evening I did a bunch of research on the ATEM network stuff & found a firmware update that apparently adds support for tethering off a USB-C phone. I'll test this tomorrow.

Very tired by the end of the day. Hoping for a peaceful week to come.

I updated the second laptop for church use. It (an HP Spectre) gets uncomfortably hot during use. I might look at opening it & seeing if the vents are dirty, and maybe replace thermal paste.

This is not yet aiming for use on UEFI boot. That might come in the future though.

The native system is very fragile. I may look at rewriting it in x86-64 assembly...

I did some more work on label printing code in C#, but did not get much done on this front. Needing a bit of a break from it I think.

There are some fragile points. Not building with -O3 results in reboots, making changes to the display code seems to corrupt the output or result in reboots. I'll have to investigate this all as I have time.

But for now, I'm happy to see it running from a hard disk image.

For konilo-over-telnet, I'm doing a little more testing and have begun writing up a bit of documentation and instructions for my patrons to access. I'll hopefully have this ready to go by the end of the week.

I also began to clean up a set of blocks to serve as the new starter set included with Konilo. It would be nice to get this done so I can push a new release soon.

One more thing: I'm working on file system ideas again...

I'm doing a bit of server housekeeping today. I fixed bugs in the script that creates the binaries/ page on konilo.org, added the .NET executables, and reviewed my logs.

I also began setting up some stuff for prototyping a self-hosted church page.

I tried booting NomadBSD on the work laptop. It did start, but I was unable to actually use the keyboard. I'll do more tests of this in the future. I also tried Haiku, but it died during the boot process.

Glucose trending low today (as of 1845). Dinner was chinese, so it remains to be seen if I'll stay normal / low, or go high.

I started work on a desktop application for use w/Para. This may be a big step forward if I can get everything to work.

I didn't do any Forth work today, but did work on some C# code for printing project labels to network zebra printers. This is about 2/3 done; I still have some work to do on auto sizing the label text, and need to refine the interface a bit.

We finished the network cabinet, desks, and ran a new data line for the extra switch in warehouse C. Everything at the new ware- house takes longer to complete than expected.

We finished hanging the TV, which ended up taking a lot longer than expected. It looks pretty nice, but was much more work than anticipated.

At church, the ATEM & Wifi bridge setup wouldn't work, so I was not able to stream. On Sunday I'll try a different approach. I have a different wifi bridge device coming Saturday and a USB drive for doing offline recordings direct from the ATEM. Hoping this will work.

YouTube limits the number of uploads per day, so getting the old videos from FB up will take a long time to do.

I'm exhausted today, so will just rest tonight.

I have some concerns with the wifi bridge setup; will look at replacing it in the future. Maybe a better bridge, or pos. use powerline connections?

Later I took a walk (it's hot & humid today), had some lunch, then spent the afternoon & evening resting.

I worked on the label printing stuff for Gilbertsville. I can print the branch labels reliably now, and have worked on adding a visual UI.

For ilo.cs, I ported the terminal emulation code & graphica-0. This will make Konilo on Windows more useful.

I decided to do a bit of work on porting ilo-x.c to C# so I can offer Graphica support on another platform. This is partially working now. Input, drawing pixels, and basic text display is working, but I haven't added (termina) or scrolling yet.

I might be able to get this finished this week.

In other news, visited the chiropractor, picked up a script, and then did some grocery shopping.

Feeling slightly better than I did yesterday. I can at least eat a little now.

I did some work on the round labels. Printing is working, but I haven't tried it over Bluetooth yet.

Still sick. I wasn't able to actually eat anything until late, and even then only some apples & cucumber slices. Wasn't feeling up to programming today.

Arland drove part of the day. This was not fun; I'm not feeling well & the drifting made it worse. No accidents though.

Doing a little work on my blocks this evening. Not sure how far I'll get, but am sketching some ideas around a clock device, and the serial visual terminal project (part of the hardware ilo system)

I've developed a stomach bug; not feeling well at all as the day draws to close.

Discussed some parts of the Konilo hardware project with Arland, then ran a TTRPG in the evening. A nice day overall.

Tomorrow is travel. Not looking forward to this.

The konilo-over-irc code has been posted for my patrons, and the instance on the new server remains working well. Now on to the stuff needed to complete the konilo-telnet system...

One more day before a semi-vacation / trip to see family. I have work & church tomorrow, finishing up prep for a Shadowdark one- shot, and some programming to do for a new printer driver tomorrow.

Glucose came down around lunchtime and stayed good the rest of the day. I helped a new hire with the initial paperwork, and now have admin for adding users to the time clock. Maybe a bit odd.

Konilo-irc is now running on a separate VPS. I'm now doing the actual work on getting konilo-telnet to fully function. Hoping to have it done within a week.

I didn't do any programming today, other than shutting down my test daemon for konilo-telnet due to a bug causing severe server load. I'll investigate further and restart it once I can find & fix the issue.

After this we had lunch at Uno's, then I went for a walk. I'm quite tired, and won't be doing much programming today.

I'm going to push out the irc bot code this weekend, along with a patreon post on it. I'm also getting close to being ready to start the hosted konilo system. I have decided to set up a VPS just for this, to keep things separate from my main systems.

I didn't end up doing any programming today. My hands aren't in good shape right now. I did fix a typo on unu.retroforth.org (pointed out by @rickcarlino).

The live stream @ church worked without issues. I'm going to shift focus now to doing EQ work. Each microphone has a badly setup EQ, and all are different. I'll move towards something that is at least consistent, then begin tuning both the live & stream mixes to improve sound quality further.

Not much programming today, but I did a bit of reading on the Swift frameworks for printing to bluetooth for a work project.

No programming due to tiredness today. Maybe tomorrow?

With a little luck I'll be able to finish a lot of the initial stock layout tomorrow.

For church, I'm on to my next couple of tasks: recording for the choir and EQ for the camera audio if necessary.

Also Ben's birthday today. We spent the afternoon at mom's. It was a nice time.

My hands are in bad shape, so I won't be programming over the weekend.

In Emmaus we continued work on anchoring the pallet racks and finished most of the construction on the pipe racks outside. I am following up on various POs, trying to get everything in so we can finish with the shelf layout process.

We worked on anchoring some of the pallet racks, but ran into issues with the brackets the supplier sent. All of them had to be modified to work, which really slowed this down.

Was too tired for programming tonight. Maybe tomorrow...

For work, I have a new pair of label printer to write control software for. Hoping this is much better than the Brother ones from the last project.

For konilo-telnet, I'm doing more tests ahead of a soft launch to my patrons next week.

For konilo-irc, I have a version of the VM that limits runtime per query to 60s. I'll be doing much more testing of this next week.

At church, had some minor issues w/OBS & the ATEM. I'm looking into having the ATEM stream directly.

Due to tiredness, log entries here are likely to be short for a while. Sorry, but working 12+ hour days is draining.

In Konilo, I've continued working on the konilo-irc & konilo- telnet code. I have it functional, but still want to figure out restricting runtime & reducing CPU load.

I've started wiring up the SPI display, while I wait for the new Teensy w/headers to arrive. I note that SparkFun is now making the Teensy boards offically. This might be good for future as it might mean more reliable supply & production.

(Once we have a working model, I plan to put aside enough parts to rebuild/maintain it for the foreseeable future. I do want it as a long-term system)

Brought the iPad mini home; I'll do a little work on Konilo-iOS this weekend if not in too much pain. I did pull a muscle in my hip, so walking has been very painful, and my hands hurt, though less so than last week.

Back to using separate ilo.rom/ilo.blocks in my personal branch. I'll keep the combined ilo.sys code aside for future, but I like having a clearer separation, even though it makes things harder for installation.

On the hardware project: I ordered an SPI display, a few more breadboards, and another Teensy. I will try to have this wired up over the weekend, then can work on finishing the display code for the ilo vm.

Target display is 480x320, and I'll be running a 1bpp graphica-0 mode on it. This display can do input with a resistive touch screen, so might be doable for the limited pointer support in the future.

Evening update: work went well, and I have a bit of progress @ church. Finished another eight songs, and have all audio going to the ATEM, though still not being picked up by OBS. Will do more work on this in the next week.

Worked in Emmaus & Gilbertsville. We have most of the cameras up at Gilbertsville now, and will try to finish a lot of the Emmaus stuff tomorrow. Didn't get home until around 19:00, so had a late dinner and then read a bit before bed.

I worked in Emmaus today, doing more configuration on cameras and a little on the counters. We'll be there again tomorrow; in hopes of getting towards the end of this project.

The morning live streams worked ok. I'll set up an HDMI extract- er tonight, so I can try to listen in on the ATEM audio output. I'm hoping this will be helpful in the future to figure out some of the audio issues.

At home, we set up a monitor riser & new display for Ben's PC. I have an identical setup for the Mac Mini, but we've not set this up yet. Maybe later in the week.

With Glen's help, I have mostly finished a character for D&D 5e, for a one-shot story he'll be running in the future. This is a much longer process than Shadowdark.

No programming again as my hands are hurting a lot. I'll try to get back to work on the konilo-telnet & konilo-irc stuff tomorr- ow. But no guarantess; I really need the pain in my hands to re- duce, and can't work for too long without resting them more.

Glucose was solid all day, but I only ate one meal, a low-carb enchilada dinner. Was nice, though I do wish I had been able to have a lunch.

One or two more days in Emmaus, I think. We have to do two more cameras, build the front counter, and set up a shelf in a hall- way. Also putting on the tops of the main stock shelves. So I will likely be back on Monday.

No coding today - my hands are too painful & I'm exhausted.

I did start on the next part of the hosted Konilo project: using telnet to connect to a Konilo session. The initial process only needs a few tiny changes to ilo.c, and (like the IRC interface) is now ready for me to start on per-user blocks & state. This will probably be more useful than the IRC interface, I think, as it gives access to a full Konilo system w/o the UI limitations of IRC.

This is pretty nice. Each session is pulling around 480K RAM, so should be light enough to scale nicely on a small VPS.

I finished setting up the network rack, fixed a couple of connection issues, and then we started on the cameras. Three are now working, and we'll try to get things done for the others tomorrow.

At church, OBS crashed, and the ATEM went down. We lost audio, and the streams failed to be usable. After rebooting all of the equipment, it's back up. I'll try to do some testing on Saturday or Sunday morning.

I received a shipment with several things for the AV setup at church, which I'll work on setting up tomorrow. Some of this will be a big help in getting the audio mix to a much better state.

At work I spent the day doing data terminations for the server rack. Tomorrow I'll be finishing the camera wiring, then most of my part of the Emmaus wiring will be done.

I have a USB-C+HDMI monitor now. Using this with DeX is nice. I have a good size screen, keyboard, and am using the phone as a trackpad. This might be my travel computer setup for a while.

I didn't end up programming today, choosing to rest my hands after work. Tomorrow I'll probably do a bit of work on the bot, and then write up a status update on the Patreon.

Back to Emmaus tomorrow.

I was able to start doing some volume & eq work on the voice channels today, so the sermon audio is much cleaner now. I have done a multistream to three sources, which worked once I drop the video bitrate to 3k (running at 5k was dropping frames with three stream). This worked ok, so I'll run with the three for now.

I have access to the website, so corrected the YouTube link, and will start updating other parts over the next week as I begin preparations to migrate to a different, less expensive server. One other update: I have remote mixer access now.

I updated [bot]dpans to respond to its nick as well as dpans:, then spent a bit of time updating things in the manual.

After getting home, we did a little grocery shopping, then had dinner and went to bed. I'm very tired, so not going to do any programming tonight.

Quite tired this evening. And my hands are hurting a lot. This isn't unexpected, but is annoying to me.

Not going to do much programming tonight. Veltas pointed out a few typos in the manual, which I am fixing before an early rest.

After work, headed to church, where I typed up a lot of songs (now 40% done!), did some testing on the microphones & the mix channel, and finally the evening service.

For konilo-over-irc, I made a bit of progress, but am not done. It's likely going to take until the weekend before I have time to try to complete it, if my hands allow.

I started some early prep for another one-shot TTRPG story. I'll be doing a series of these as I get comfortable with a role as GM, and to help the players get used to actually playing their characters.

The work on state preservation went well, so I've been able to do some testing on a larger scale. Running ~100 copies at the same time is demanding on the server, but is doable. I think I will setup a VM for this when I'm ready to go live.

I was able to do a fair bit of rewriting on the konilo irc bot. It's now relaying (many) of the output lines to a private chat, though some are missed. It also fails to properly kill sessions when I shut it down. Very annoying.

I'll be in Emmaus tomorrow, so might not get to work on this for a bit. When I do get back to it, I'll do some VM work to support saving state, which might help in some ways. This might let me not keep running instances, just spawning & killing them as the requests flow.

The interaction model between Konilo and users on IRC is hard to get right. In a channel, it doesn't make sense to relay all of the output to the user as messages, since that would make it too hard to follow along if there's more than one user. So I have it writing output to a file and providing updates there. But it'd be nice to let a user directly chat with it, and get the output that way. This is something I'll be working on this week. I will be posting the current code for it to patreon tomorrow I think.

Did a bit of planning for another one-shot shadowdark adventure. This will be less horror-themed, and more focused on exploration and puzzle solving. Tammie should like it more than the one set in the swamp.

Hands are kind of numb, so not doing much coding today, but I am thinking over some things with the konilo-irc code. I'll try to work on the private/shared block setup tomorrow, and then do a bit to reduce CPU load on the sessions.

I helped Gene with layout on a mini skateboard ramp that he is hoping to work on over the weekend. It'll be fun to see if this actually works out for him.

I posted a few details on the konilo-over-irc project to Patreon though the code is not yet published. Maybe over the weekend?

As with yesterday, unstable glucose, with a huge evening spike. I'm hoping this stablizes out over the weekend.

My glucose was very unstable today. And my hands are hurting a lot. Going to bed early.

Very tired today, and having hand pain.

We did some prep work for tomorrow.

Doing a little programming today, mainly working on disabling the UTF32 path in RetroForth.

We had a ~ 2 hour "one shot" ShadowDark session (not part of the actual planned story), mostly to explore a bit of combat & some exploration stuff. As expected, Alex's character made some bad choices, and the party had to fight a bog zombie, but no one died.

I've pushed the ilo.sys variants of the vm to the repo & shared them on patreon.

Other things to note: I'll be refreshing the ilo/konilo for iOS soon, and will try to get it approved in the app store. I plan to have it modified to use an ilo.sys, and plan to provide some update/recovery/backup functionality in the application.

For RetroForth users, an update is quite overdue, but will come this year.

Blood glucose ended up running on the edge of low to low today. This made work more difficult than I would have liked, but I was able to get through the day.

After dinner I fixed the clear screen issue, so ilo-sdl-1bpp is now working nicely for me.

Tomorrow I'll probably merge it into the repository, and set up a directory structure for the variants using the ilo.sys file instead of the separate rom & images.

At church, I finished another 18 songs, and tested the audio. It now seems to be passing through the microphones, but I still am not picking up the pianos.

I wasn't able to apply the patches for ilo-sdl-1bpp, so am now waiting for a copy of the full file.

Very tired. veltas is looking at the ilo-sdl-1bpp code. It'll be interesting to see if he can make it perform better without a significant redesign.

I provided an update on Konilo on Patreon. It covers much of the latest changes in regards to ilo-x86-native, and some details on the expansions in (nonix) and changes to the startup blocks.

Didn't do any writing in the evening due to tiredness. Made some shirataki arrabbiata, which was good. Spent the evening resting.

Shortly before bed I was informed that I'd be going to Emmaus tomorrow, so am looking forward to a possibly less stressful day (though not necessarily an easier one).

Was still very tired, so didn't do programming today. Just the normal church, then reading, and a bit of writing for a one-shot Shadowdark campaign separate from the main story.

In the evening we had heavy wind and lost power. I went to bed early; hoping I can get some sleep.

I've started working on the boot sector from yesterday. I wrote it using nasm, which I'm more familiar with than as. But using as will mean dropping a dependency, so the work is to translate it to as. Given my tiredness, I might not get this done today. After I have this done, I want to see if I can make an adaption to boot from a hard disk. This would open (with some tweaks to the ilo vm code) a possibility of just providing an HDD image since floppies are getting pretty rare.

I also figured out building a flat, non-ELF kernel and wrote a boot sector, so I can create a bootable raw floppy image. I am very happy to say that it's possible to boot without needing GRUB.

Tomorrow I think I'll be working on the ilo-x86-native code. I still have a list of things to work on that I made before the #DecemberAdventure, so it will be good to try to get these done.

I'm very tired again. And today will be a long day for me. Maybe some sleep will be attainable tonight?

Today I decided to do a bit more on the ilo-x86-native code. I would like to get this closer to being ready to publish in the near future.

Good news: I have (finally) finished a new keyboard driver. This replaces the x86 assembly one from RETRO4 for the first time, so the code is now entirely stuff I've written.

Getting back to a slightly more active participation in ##forth IRC channel. There have been some issues in the channel with a few discussions venturing into questionable and objectionable directions. I have to watch more closely and deal with this if it continues.

Back to Konilo. I'm gstarting some documentaton updates, with 2 goals: getting it fully up to date with the latest changes and adding support for using it without needing (termina) loaded.

But it went well overall, and I now have the ability to live stream to YouTube.

The good news is that I think I'll be able to start doing some live streams to the YT channel.

Dinner at church. I didn't eat, but Dan & I setup an audio mix channel for me to use with the ATEM & a music channel on the main mix for audio from the computer. So I'm now tied into the mixer, which will (once I get things configured & EQ'd) make for much better sound.

I'm hoping to try a live stream soon.

For programs that will have a split termina/non-termina model, I am adding a :ui variant (e.g., wisp and wisp:ui) with separate LOAD blocks.

I've added `page` and `pages` words to (nonix), for displaying titles on a page by page basis (similar to catalogue, but in the dumb terminal mode).

I've merged in the update to retro, adding `block:set-reserved`

Glen has apparently entered in several more songs, so I should be able to merge those into my collection tonight. Tomorrow I plan to work on generating slides for all of the completed ones though this will probably take more than one day to finish.

I started the split of (termina) & non-(termina) in programs where this makes sense. I also replaced the old * loader with *: and have * take the name from the stack instead.

I'll likely switch back to the Atreus tomorrow. The keys on that are about 10g lighter, and I definitely feel the difference.

I borrowed another song book; Glen is going to help enter some of the songs. I have done 143 so far.

For Retro, I'm making a change in the blocks library, which will make it easier to use with the new ilo.sys file format. This will be merged in tomorrow I think.

I have updated the Konilo install on the Tab S6 Lite, and have begun testing some updates to `k.sh`, for my system updates. I hope to get the more intelligent sync working soon.

I have the Ergodox EZ back up and running. This is pretty nice for a change. Noisy (blue switches are loud), but feels good to use a proper split again.

I've continued to do a bit of work on tweaking some things in my personal Konilo system, and also did a little work on the x86- native branch of ilo.

I went on a long walk, had lunch with Jenni and Gene, then did a bit more walking before going to an afternoon doctor visit. The overall results are good. I'll be trying to reduce my insulin use further, and moving to the 12.5 strength of Mounjaro. I hop this works ok.

I fixed the WISP export. It's not uploaded to the server yet, but it should be running by the weekend.

The base system + (std) + (nonix) sits at 5,777 cells. I will probably add a few small things, but expect it to remain under 6k cells, and then another 6,200 or so for the extra programs.

I'm going to take a look at the (nonix) words, with a goal of cleaning them up to a core set, then moving them into (std). It should be enough to cover the basics of locating and organizing the blocks, leaving extra stuff for the other tools.

No programming today.

Continuing down the path of not using (termina), I'm looking to see if I can split some of the programs into separate basic & termina versions.

My day job was fine. Nothing too stressful, and I'm starting to see my glucose numbers return to a normal range as stress levels drop.

In any event, I much prefer reading & writing on either of them to doing the same on my iPad.

Other stuff: I'm testing use of the Viwood for my work notes, and doing some testing of superinstructions on my ilo-unix.c code.

I showed Arland the SuperNote Nomad today. Will try to show him the others soon. I might give him one of these.

I'm finding that I like these for my notes & Forth programming. They're not fast enough for video (though the Boox & Viwoods can sort of manage it), but are very good with text and simple images.

So Konilo is ok on the SuperNote. There's some lag with the display, but it looks great, and I'm happy to use it on this. But I don't think I'll be using many things on this other than the notes and KOReader, and some use of Konilo. For many things, the faster response times of the Viwoods mini is better feeling, even if there is more ghosting.

I am going to make some changes to the `k` script I use to keep my Konilo system in sync across devices. The big one will be to use precompiled ilo binaries when possible.

I spent some time with the SuperNote A6X2 this evening. I've setup some templates, and started several notebooks. It's very nice for writing. For Termux, it's somewhat slow, but does work.

For my personal wiki, I'm going to setup a gopher interface. I plan to try writing it in Konilo & shell script, and will run under inetd. This presents some interface challenges since there are no inline links in gopher.

Today I refactored the (termina) code for handling interface hints & the display. The code is a little less dense, and words are shorter than before. Overall it's slightly larger in memory due to the extra headers, but cleaner to read, so I'm happy with the tradeoff here.

My glucose is still a bit unstable & spiking when I eat. I'll be attempting to do a fast this weekend, to try to help reset a bit (though I'm not sure how well this will work).

Visiting w/my dietician today. Hoping it goes well. My eating of late hasn't been ideal, due to unstable work schedules. I hope to get back on track soon.

This does have me considering more on reducing the use of color in the (termina) programs. It's already optional, but I may just do plain output in more things going forward.

Notes for Viwoods AI Paper: I don't care about the AI part, but the hardware seems decent. It's fast enough for me, though not a major jump over the Boox Go 6. Konilo runs well on it.

I've updated my cron jobs to use the new combined rom+blocks variant of ilo, so the WISP feed should now be updating again.

Tomorrow I will be working on updating the (wiki) code to make use of (termina) prompt functionality, and bringing in faster entry lookups from (man).

I'm also going to start working on an "ilo.sta" format for saving system state (ram + stacks + registers), in a form that we might be able to use across hosts.

For songs, I'm now up to 108 complete, and Arland has another five ready to send to me.

I'm going to make a few changes in my eink stuff in the near future. (Consolidating a few devices, and am looking at more capable things to replace the Kindle Scribe).

No actual programming today, though I did spend a bit of time just reading through my blocks.

Going this way will open another option as well. I might expand ilo-unix to keep a "clean" copy of the ilo.sys in /usr/local/ and copy it to the $HOME if a copy isn't there. This would allow for an installable (via ports/packages) distribution. I need to think on this for a while.

This will be more consistent with the native & embedded variants though does break some existing tooling, and will break backward compatibility. I am leaning towards doing this though.

For ilo-native-x86: I now have a functional version which loads the ROM from mass storage. So physical blocks 0-63 for the ROM, and data blocks (mapped to 0) start at physical block 64. This needs more testing, but seems to work ok.

This is pretty nice as it gets me a good step closer to a ROM- able ilo. This might have some impact on a few things in the hosted systems. More on this tomorrow I think.

It's amazing to me how stress affects glucose levels. Now that we are back to mostly normal staff at work, and I can relax a little, my glucose numbers are back where I'd like them to be.

I decided to start adding a few new things to (catalogue). Later this week it will support copying blocks, erasing them, and possibly a few other operations.

Glucose control is good this morning. I took a short walk before work, which might have helped. Will try to do this more often.

I'll be working using the Go 6 all week I think. Apart from a bit of slowness, the only real disappointment I have with this is the lack of stylus support. It'd be nice if it could act as a digital notepad. But it's great for reading on, and the Konilo under Termux is running nicely.

Using the Boox Go 6; I note that due to only having 2GB of RAM, it's (mostly) a single tasking machine. Background programs need to be restarted when switching most of the time. This isn't an issue for me, but is worth noting I think.

I also added several more pages in my wiki today.

Using the Boox for today's programming. I'm setting up a few additional programs for IRC & Fediverse access. I switched to using a lighter launcher, and have swapped over to KOReader. I wish this supported an EMR stylus (or even an Apple-style active one) as it'd be a great size for a pocekt notepad. As I do more with my eink devices, I'm finding the benefits of reading on a non-LCD to be nice.

(I might look at the Boox Tab Mini C, or the Note Air 4C as a future option; the bigger screens might be helpful for reading).

I think part of the issue is the overhead of dealing with FAT, and actual file i/o. If I can get the raw SD access working this might be faster. A backup plan: for my personal block set of 4,096 blocks, I could solder on 16MB of PSRAM & cache the blocks in RAM. This would help, but feels wasteful. I'd prefer to not resorting to this.

As of today, I'm at 656 entries in WISP, having not missed a day since I started keeping this.

Arland sent me an updated ilo.py a couple of days ago. Today I began actually reading it, and will likely merge it in tonight. A note on this: he has it stripping out escape sequences; I'd rather this be made optional. I'll ask him if he can revise to support this.

Today is mounjaro day, so glucose control is pretty good. My average is up a bit, but I'm hoping that with the end of the holiday season, it'll start dropping down once more.

Today I didn't have work, so I spent time reading, resting, and took a nice walk. It is a bit cold, but I still enjoyed the day.

Tomorrow it's back to work for a few days, and probably back to a little programming.