How truly wonderful! I went out tonight and the first thing I noticed was the temperature drop. It felt actually quite pleasing. What a welcome surprise, I didn’t expect that at all. It was warmer in the forst than between the fields. The tiniest breeze helped to cool off the surroundings I think. Right now, the temperature shows 23°C. It’s supposed to reach 18°C at 5 in the morning before it rapidly shoots through the sky again.

When I left the house I even saw the very end of a nice sunset. A bat was around, too. The several thousand fireflies delivered a fantastic show. It’s such a pity that I cannot show this to you. :-(

There were many frogs or toads around. Luckily, the light tan gravel road made for a good constrast to the darker hopping amphibians. So, I spotted them just in time. No animals were harmed.

The moon was out and lit up the scenery. I was perfectly chasing my own shadow for several hundred meters on a forest road. I had the moon right in my back. That moon light shadow felt magical. <3

It must have set a new record on picking up spider webs along the way. The threads around arms and legs always feel quite yucky. People were blasting music somewhere in town. You could here that noise in the entire forest. I found that rather annoying. All street lamps are operational again, so I got already blinded right at the entrance to the town. But other than that, this was a very nice evening stroll. Totally recommended. Already looking forward to tomorrow. :-)

@movq@www.uninformativ.de Regarding https://movq.de/blog/postings/2026-06-16/0/POSTING-en.html:

In my opinion, the KDE 3.5 menu was organized way better than the Windows Start menu. Granted, a typical KDE installation had much more applications to offer, too. So, there was more need to get it right. And it probably was also later in time.

Isn’t Notepad++ and Python cheating!? :-D

Crazy story on the clock’s seconds. I never heard of that before. Neat.

Yeah, UI these days is horrible. (That’s why my own TUIs suck, too!)

In Magic today, the Phyrexian Invasion failed in the first game, but the second game was EPIC!

I played my (unlisted) Dragons 2: Draconic Boogaloo deck, and…

Turn 1: Nothing special
Turn 2: Miirym (when a dragon enters, copy it)
Turn 3: Tiamat (choose 5 dragons from deck, put in hand)
Turn 4: Klauth (when dragons attack, create mana equal to their total power)
I attacked with all 5 dragons, which made 28 mana x2 = 56(!) mana.
Then (still turn 4) I played Scourge of Valkas (when a dragon enters, deal damage to target equal to number of dragons) + 5 other dragons, dealing 6 + 2 x (7+8+9+10+11+12+13+14+15+16+17) = 270(!) direct damage (more than double enough to kill the other 3 players).

Damn fine win, if I do say so myself.

Every now and then, I think that I have carefully proof-read my message enough times and hit the “Add message” button in tt. But then, in the message tree, I spot another missed typo. My process is then to go to my twtxt.txt and fix it by hand. However, I still have to clean up tt’s cache. This is rather tidious:

1. Recall the sqlitebrowser ~/.local/share/twtxt/tt2.sqlite from my shell history.
   
2. Switch to the “Browse data” tab.
   
3. Go to the messages table and wait a second or two until it’s loaded.
   
4. Sort by the created_at column twice, so that I get descending order.
   
5. Select the first message, which is typically the one in question.
   
6. Find the “Remove currently selected row” button in the tool bar.
   
7. Commit the changes.
   
8. Close sqlitebrowser.
   

So, I finally implemented the removal of messages from the cache in tt. I can now hit d and confirm the removal. Bam! Should have done that ages ago!

https://lyse.isobeef.org/tmp/tt-confirm-message-removal.png

Next up is the search, I think.

A lot of questions keep me awake at night.

One of them - If I had only 5 more years to live, what would be my gift to the world?

Playoff Success of 8.5+ Regular Season SRS NBA Teams ?~L~X https://thenewleafjournal.com/b/E2i

@movq@www.uninformativ.de I only had 3.5” disks.

@movq@www.uninformativ.de Great find! Other than a couple of games I bought over the years, my 5.25” disks were always dark grey / black.

And 3.5” disks were another matter entirely… 😄

Happy 5.25 day to anyone who celebrates.

(“Colour diskettes”. Ha.)

https://movq.de/v/a84fbdc756/a.webp

https://movq.de/v/a84fbdc756/b.webp

Estoy probando el simulador Flightgear 2024.5.1. Muy bueno!

I just wanted to look up 9V block battery prices online and these automatically generated descriptions are getting dumber by the minute:

Datum der Erzeugung: Verfallsdatum minus 7 Jahre

(Date of manufacturing: expiration date minus 7 years)

Or look at this one:

Die leistungsstarke 9 V-Block E Batterie, auch 6LR61 genannt, eignet sich besonders gut für Taschenlampen, Radio oder Kinderspielzeug, die einen gleichbleibenden Strombedarf haben. Ihre max. Spannung beträgt 1,5 V.

(The high performance 9 V block E battery, also known as 6LR61, is particularly suitable for torches, radio or child’s toys, which have a steady power demand. Its max. voltage is 1.5 V.)

The battery is best suited for… devices where it fits. No shit, Sherlock! Has anyboy ever come across 9V block torches? O_o I haven’t.

@bender@twtxt.net It’s been a while (6.5 years) since I’ve done this. I’d do it like this:

• Boot some Linux from a USB stick on the new machine. Preferably Arch Linux, since that is what I’m running and that’ll make the upcoming chroot easier.
  
• Partition the new disk, create LUKS devices, filesystems, …
  
• Mount the new filesystems and copy all data (user data and the system itself – everything). Do this either over the network or by hooking up the old disk directly.
  
• chroot into the new system (Arch has an arch-chroot tool for that which is used during normal installation, if I’m not mistaken). Inside the chroot, install the bootloader.
  
• Do some fixups, like adjusting /etc/fstab or /etc/crypttab.
  

And I think that should be it. 🤔

My first game of Magic ended with a truly EPIC TURN yesterday…

It was a 5-player game, and I was running my (unpublished) Superfriends deck (mostly Planeswalkers and counter manipulators). After some ups and downs, I was able to pop the ultimate abilities on a handful of PWs all on a single turn, pumping my Bioessence Hydra to 110/110 (!) before tapping it twice to kill 2 opponents, and then following that by destroying all of the lands of a 3rd opponent and stealing all of the creatures from the 4th, at which point the survivors decided to quit. As I said, EPIC TURN!

Game 2 ran long, so I dropped out. But that first game…

Just saw the video. Can’t believe that ladder is that expensive. Even in AUD, it is almost $100. It is also 2.5 stars, with 13 reviews. Gulp. Engineering aside (and you are right, it is pretty interesting, and some, if not most of it went over my head), the ladder is rubbish. This is the one I have. Not super, but have been with me for a while, and used quite a bit, still as good as new.

I won our only game of Magic for this week with my (yet-to-be published) “Bolas Triumphant” deck: 5 players over 3 hours, including 4 board wipes (one of which came from my Nicol Bolas, God-Pharaoh), and I even got to cast Omniscience via a Fae of Wishes. I can’t speak for everyone, but I know I had a good time. 😁

What do the Gopher Troopers think of the following? The Gopher protocol is a nearly-forgotten network protocol from the early 1990s, designed to serve and navigate text-based menus and documents over the Internet. While it’s far less common than HTTP/HTTPS today, there are still some security risks associated with Gopher and Gopher space. Let’s break them down carefully: 1. Lack of Encryption Problem: Gopher was designed long before widespread use of SSL/TLS. All data—including credentials, file transfers, and menu selections—is transmitted in plaintext. Impact: Anyone intercepting traffic (e.g., via a network sniffer, public Wi-Fi, or a compromised router) can read sensitive information, including usernames and passwords. 2. No Authentication or Access Control Problem: Gopher servers rarely implement robust authentication; access control is usually limited or non-existent. Impact: Unauthorized users might browse sensitive directories or download private files, particularly if servers are misconfigured. 3. Server Software Vulnerabilities Problem: Modern OSes can still run legacy Gopher servers, but the software is often unmaintained. Impact: Old software may contain buffer overflows, directory traversal bugs, or command injection vulnerabilities that attackers could exploit. 4. Malicious Gopher Links Problem: Gopher menus can contain links that point to scripts or other servers, similar to hyperlinks in HTTP. A client following a malicious link could inadvertently: Download malware Access sensitive internal network resources (server-side request forgery) Impact: Could serve as a vector for attacks if a user opens content from untrusted sources. 5. Legacy Protocol Weaknesses Problem: Gopher lacks modern web security mechanisms like: Content security policies Same-origin policies Cross-site request forgery protection Impact: If Gopher is bridged to other services (like modern browsers via gateways), old vulnerabilities may be exposed. 6. Information Leakage Problem: Gopher servers often provide directory listings without restriction. Impact: Sensitive files, backup directories, and internal documents may be exposed unintentionally. 7. Bridging Risks Problem: Some modern browsers access Gopher via gateways (HTTP-to-Gopher proxies). These bridges may: Expose sensitive internal resources to the gateway Introduce logging or tracking that wouldn’t exist on pure Gopher Impact: Attacks could occur indirectly through insecure intermediaries. Key Takeaways Gopher is inherently insecure due to its design in a pre-HTTPS era. Main threats: eavesdropping, unauthorized access, malware delivery, and exploitation of unpatched server software. Safe practice: Use Gopher only in isolated, trusted environments, or through secure HTTP(S) gateways with proper sanitization.

This one comes from this years (now scrapped) April 1st DLSS 5 gag, that was originally supposed to use Microsofts AI - in ways similar to the Nvidia technology, which produced interesting, overly detailed results. I wanted to see if I could beat the AI thing at drawing something like that myself and many redraws later, this is my best result.

I’m supporting incremental SQLite schema changes to just upgrade from an older database version to whatever the current software version supports. In the past, I already noticed that this is quite expensive in unit tests when each test case runs through the entire schema patches and applies them one by one.

To speed up test execution I now decided that I finally go through the troubles of maintaining both a set of incremental patches and a full schema setup in one go. A unit test verifies that both ways end up with the same structure. This gives me a set of SQLs to check the structures:

SELECT type, name, tbl_name, sql
FROM sqlite_schema
ORDER BY type, name, tbl_name

Unfortunately, the resulting CREATE TABLE SQL queries are formatted differently, depending on whether the full schema was set up in one big step or the structure had been modified with ALTER TABLE. Mainly, added columns are not on their own lines but appended in one physical line. That’s why I wanted an SQL formatting tool. Since I didn’t find one that works decently, I’m now doing some simple string manipulation. Joining consecutive whitespace into a single space character, removing spaces before commas and closing parentheses and spaces after opening parentheses. This works surpringly good enough. Of course, if it fails, the “diff” is absolutely horrendous.

Now for the cool part, my test execution dropped from around 5:05 minutes to just 1:32 minutes! I call that a win.

I just stumbled across PRAGMA table_info('tablename') https://sqlite.org/pragma.html#pragma_table_info, PRAGMA foreign_key_list('tablename') and friends. I guess, I have to play with that, now. It’s probably much better to use than the SQL text approach.

Turns out, I even go down to only 50% quality for my thumbnails: https://lyse.isobeef.org/tmp/galres.txt The difference between 50% and 80/90% is just barely noticeable.

$ convert -strip -quality 50 IMG_20251106_035048_448_size_400.jpg 50-stripped.jpg
$ convert -quality 50 IMG_20251106_035048_448_size_400.jpg 50.jpg
$ ls -lh 50*jpg | awk '{print $5 " " $9}'
26K 50.jpg
25K 50-stripped.jpg

@falsifian@www.falsifian.org Correct, the two smaller versions are loading perfectly fine. The hickup is only for the originals. But in all reality, the middle ones are sufficient for me personally. Please don’t get me wrong, at least for the people photos, the subjects are large enough. The Japanese landscapes, however, would definitely benefit from a bit more detail. ;-)

I just tried it once more, and now, the tree with the sign (/photo/5Zy4pqVIt0oP/IMG_20251106_035048_448.jpg) fully loaded very quickly. Same with the Japanese dish (/photo/tJbmg8oleYbh/IMG_20251030_091719_086.jpg) and shopping center (/photo/qXG5ucIjpPju/IMG_20251029_045002_778.jpg). But the previous and next ones all ran into the same problems again. When I’m very lucky, I eventually get the upper half. Typically not even that much, a third, a fifth, or even less.

Waiting a bit before making an attempt, the wooden walkway through the forest or park (/photo/ojQpDLfBoGN4/IMG_20251023_043829_011.jpg) eventually also made it. But unlike the other successful attempts, it took a long time.

The more photos you add, the more beneficial it might be to separate the index into several different albums. I didn’t measure it, but it felt like 10 to 20 seconds for all the thumbnails to load. That traffic adds up.

Another idea would be to strip the EXIF data from the thumbnails and reducing quality to 90% or even 80%. Using the famous tree with the sign, I cannot tell the difference between the original thumbnail and the 80% quality one. I’m sure it depends on the subject. Here are the numbers:

$ convert -strip IMG_20251106_035048_448_size_400.jpg stripped.jpg
$ convert -quality 90 IMG_20251106_035048_448_size_400.jpg 90.jpg
$ convert -quality 80 IMG_20251106_035048_448_size_400.jpg 80.jpg
$ convert -strip -quality 90 IMG_20251106_035048_448_size_400.jpg 90-stripped.jpg
$ convert -strip -quality 80 IMG_20251106_035048_448_size_400.jpg 80-stripped.jpg
$ ls -lh *jpg | awk '{print $5 " " $9}' 
46K 80.jpg
45K 80-stripped.jpg
64K 90.jpg
63K 90-stripped.jpg
132K IMG_20251106_035048_448_size_400.jpg
127K stripped.jpg
$ ls -l *jpg | awk '{print $5 " " $9}'
46160 80.jpg
45064 80-stripped.jpg
65012 90.jpg
63916 90-stripped.jpg
135070 IMG_20251106_035048_448_size_400.jpg
129647 stripped.jpg

@movq@www.uninformativ.de I’ve got the same problem that you had the other day: finding past temperature data. But yeah, it looked much warmer than it actually was. Maybe 5°C? Possibly less when I found myself in the snow- and rainstorm in the end.

With the wind, my fingers were frozen. I should have worn gloves. Without them, I could only put my hands in the pockets of my jacket. That didn’t help much, though, because I frequently stopped to take yet another photo, so they cooled off again right away. :-D

Balancing the big/long, closed umbrella under my arm while I had my hands burried was also a little tricky.

First world problems. :-)

Bit of a remake, of a chubby-looking, quick meme doodle, from 5 years ago.

Wow, as I anticipated, this is waaay out of my capabilities to really understand it. But I’m quite happy to just have spotted a mistake in an explanatory comment in section 4.5.2 “The icode Array”. Of course, it should be /e + tc + /i + ni + t\0. Let’s hope that my e-mail with the patch actually makes it into Briam’s inbox. I fear GMail just hides it in the spam folder.

@eldersnake@we.loveprivacy.club

Steps to world domination:

1. “Invent” “AI” (by using other people’s data).
   
2. Get people hyped about it and ideally hooked on it.
   
3. Only provide it as a cloud service. But hey, if you want to, you can run it locally!
   
4. Buy all hardware available on the market, so that nobody but you can build more systems.
   
5. All PCs of consumers and competitors are too weak now and can’t be upgraded anymore.
   
6. Everybody depends on your cloud service! Win!
   

All of that is possible because corporations don’t have a “conscience” in capitalism. Nobody forces the RAM manufacturers to sell all their stuff to just one or two buyers, but since the only goal of that manufacturer is to make money, they do it.

@movq@www.uninformativ.de How about “Quongsi”? I generated the first five letters with pwgen --no-capitalize --no-numerals 5 and since that already showed up in DDG search results, I simply appended the last two, which yielded nothing on DDG and Google).

What kind of project is it? Maybe we can help you find a name or nudge you in the right direction.

@prologic@twtxt.net And I froze my ass off yesterday at -5°C and strong winds. 🤣

Jumping Frog Radius
⌘ Read more

Thinking about doing Advent of Code in my own tiny language mu this year.

mu is:

• Dynamically typed
  
• Lexically scoped with closures
  
• Has a Go-like curly-brace syntax
  
• Built around lists, maps, and first-class functions
  

Key syntax:

• Functions use fn and braces:
  

fn add(a, b) {
    return a + b
}

• Variables use := for declaration and = for assignment:
  

x := 10
x = x + 1

• Control flow includes if / else and while:
  

if x > 5 {
    println("big")
} else {
    println("small")
}

while x < 10 {
    x = x + 1
}

• Lists and maps:
  

nums := [1, 2, 3]
nums[1] = 42

ages := {"alice": 30, "bob": 25}
ages["bob"] = ages["bob"] + 1

Supported types:

• int
  
• bool
  
• string
  
• list
  
• map
  
• fn
  
• nil
  

mu feels like a tiny little Go-ish, Python-ish language — curious to see how far I can get with it for Advent of Code this year. 🎄

Apparently MTD of uknowingly toggling mouse on in tmux for the first time is 1.5 agonizing days

This post has been written with UW PICO 5.09 and yarnc

Better Technology, Worse Motivation: GenAI’s Mediocrity Trap

While generative AI (GenAI) promises productive efficiency, it can paradoxically lead to lower-quality work. We conducted an experiment with professional illustrators and found that AI assistance flattens the quality curve—it accelerates initial gains but sharply diminishes the returns on sustained effort. Faced with this, a significant number of professionals made a strategic choice: they sacrificed the final quality to save time.

From http://www.jin-li.org/uploads/1/1/4/5/114595093/ai_and_motivation.pdf

I haven’t read this and can’t vouch for it; seems vaguely AI-boostery. Still, the conclusions are interesting. This seems to be the picture that is emerging about generative AI generally: most people don’t like it and find that degrades the quality of work. Coders seem to like it and think that it helps them, but in fact it makes the slower, less productive, and more bug prone.

By all measures it’s a bad technology. We should just be honest about it. There is no need to make excuses for multi-trillion-dollar corporations.

@prologic@twtxt.net AI is slot machines for coders:

• “Before starting tasks, developers forecast that allowing AI will reduce completion time by 24%. After completing the study, developers estimate that allowing AI reduced completion time by 20%. Surprisingly, we find that allowing AI actually increases completion time by 19%–AI tooling slowed developers down.” https://metr.org/blog/2025-07-10-early-2025-ai-experienced-os-dev-study/
  
• “Stack Overflow data reveals the hidden productivity tax of ‘almost right’ AI code”: https://venturebeat.com/ai/stack-overflow-data-reveals-the-hidden-productivity-tax-of-almost-right-ai-code
  

The same intermittent reward operant conditioning that gets people addicted to gambling and thinking that if they follow certain rituals they’ll win “next time” drives people’s beliefs that AI tools are making them more productive when they’re making them less productive. I’m going to guess that a side effect of this is that people think they’re typing less when in the longer term they’re typing the same amount or more when you factor in the productivity loss (as far as I’ve read the studies don’t measure this so I’m only guessing).

People are also being rapidly de-skilled by this technology: the more they use it, the more their actual skills atrophy. “Continuous exposure to AI might reduce the ADR (adesoma detection rate) of standard non-AI assisted colonoscopy, suggesting a negative effect on endoscopist behaviour.” (science speak for saying that radiologists get worse at seeing tumors in scans once they’ve used AI): https://www.thelancet.com/journals/langas/article/PIIS2468-1253(25)00133-5/abstract

Nobody who cares about the future should be using this stuff for anything.

@lyse@lyse.isobeef.org Probably wouldn’t help, since almost every request comes from a different IP address. These are the hits on those weird /projects URLs since Sunday:

    1 IP  has  5 hits
    1 IP  has  4 hits
   13 IPs have 3 hits
  280 IPs have 2 hits
25543 IPs have 1 hit

The total number of hits has decreased now. Maybe the botnet has moved on …

Fark me 🤦‍♂️ I woke up quite late today (after a long night helping/assisting with a Mainframe migration last night fork work) to abusive traffic and my alerts going off. The impact? My pod (twtxt.net) was being hammered by something at a request rate of 30 req/s (there are global rate limits in place, but still…). The culprit? Turned out to be a particular IP 43.134.51.191 and after looking into who own s that IP I discovered it was yet-another-bad-customer-or-whatever from Tencent, so that entire network (ASN) is now blocked from my Edge:

+# Who: Tentcent
+# Why: Bad Bots
+132203

Total damage?

$ caddy-log-formatter twtxt.net.log | cut -f 1 -d  ' ' | sort | uniq -c | sort -r -n -k 1 | head -n 5
  61371 43.134.51.191
    402 159.196.9.199
    121 45.77.238.240
      8 106.200.1.116
      6 104.250.53.138

61k reqs over an hour or so (before I noticed), bunch of CPU time burned, and useless waste of my fucking time.

I just noticed this pattern:

uninformativ.de 201.218.xxx.xxx - - [22/Nov/2025:06:53:27 +0100] "GET /projects/lariza/multipass/xiate/padme/gophcatch HTTP/1.1" 301 0 "" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/112.0.0.0 Safari/537.36"
www.uninformativ.de 103.10.xxx.xxx  - - [22/Nov/2025:06:53:28 +0100] "GET http://uninformativ.de/projects/lariza/multipass/xiate/padme/gophcatch HTTP/1.1" 400 0 "" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/112.0.0.0 Safari/537.36"

Let me add some spaces to make it more clear:

    uninformativ.de 201.218.xxx.xxx - - [22/Nov/2025:06:53:27 +0100] "GET                       /projects/lariza/multipass/xiate/padme/gophcatch HTTP/1.1" 301 0 "" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/112.0.0.0 Safari/537.36"
www.uninformativ.de 103.10.xxx.xxx  - - [22/Nov/2025:06:53:28 +0100] "GET http://uninformativ.de/projects/lariza/multipass/xiate/padme/gophcatch HTTP/1.1" 400 0 "" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/112.0.0.0 Safari/537.36"

Some IP (from Brazil) requests some (non-existing, completely broken) URL from my webserver. But they use the hostname uninformativ.de, so they get redirected to www.uninformativ.de.

In the next step, just a second later, some other IP (from Nepal) issues an HTTP proxy request for the same URL.

Clearly, someone has no idea how HTTP redirects work. And clearly, they’re running their broken code on some kind of botnet all over the world.

Thank you for the encouragement and love and kind words, @lyse@lyse.isobeef.org @movq@www.uninformativ.de @bender@twtxt.net @doesnm@doesnm.p.psf.lt and others along the way I’m not sure of their feed uris 💕 I’ll keep at it, but for the time being I will keep my distance, mostly off IRC, because I don’t have the energy to spare in that kind of engagement (what//if the worst happens, it’s so draining). I need to remember what I ever did any of this for, it was back in ~2020 and I wanted really to build small interconnected communities that any non “tech savvy” person (more or less) could also benefit from ane enjoy. Even if there are aspects of the specs we’ve built/extended over time that aren’t “perfect”™, they’re “good enough”™ that they’ve last 5+ years (I believe this is 6 years running now). I want to spend a bit of time going back to why I did any of this in the the first place, and get a little micro-SaaS offering going (barely covering running costs) so encourage more folks to run pods, and thus twtxt feeds and grow the community ever so slightly. Other than that, I plan to get the specs “in order” to a point (with @movq@www.uninformativ.de and @lyse@lyse.isobeef.org’s help) where I hope they’ll stand the test of time – like SMTP.

Thank you all ! 🙏

Replacement wifi adapter for my main laptop arrived today…5 days early!

@prologic@twtxt.net Let’s go through it one by one. Here’s a wall of text that took me over 1.5 hours to write.

This section says AI should not be treated as an authority. This is actually just what I said, except the AI phrased/framed it like it was a counter-argument.

The AI also said that users must develop “AI literacy”, again phrasing/framing it like a counter-argument. Well, that is also just what I said. I said you should treat AI output like a random blog and you should verify the sources, yadda yadda. That is “AI literacy”, isn’t it?

My text went one step further, though: I said that when you take this requirement of “AI literacy” into account, you basically end up with a fancy search engine, with extra overhead that costs time. The AI missed/ignored this in its reply.

Okay, so, the AI also said that you should use AI tools just for drafting and brainstorming. Granted, a very rough draft of something will probably be doable. But then you have to diligently verify every little detail of this draft – okay, fine, a draft is a draft, it’s fine if it contains errors. The thing is, though, that you really must do this verification. And I claim that many people will not do it, because AI outputs look sooooo convincing, they don’t feel like a draft that needs editing.

Can you, as an expert, still use an AI draft as a basis/foundation? Yeah, probably. But here’s the kicker: You did not create that draft. You were not involved in the “thought process” behind it. When you, a human being, make a draft, you often think something like: “Okay, I want to draw a picture of a landscape and there’s going to be a little house, but for now, I’ll just put in a rough sketch of the house and add the details later.” You are aware of what you left out. When the AI did the draft, you are not aware of what’s missing – even more so when every AI output already looks like a final product. For me, personally, this makes it much harder and slower to verify such a draft, and I mentioned this in my text.

You, @prologic@twtxt.net, also mentioned this in your car tyre example.

In my text, I gave two analogies: The gym analogy and the Google Translate analogy. Your car tyre example falls in the same category, but Gemini’s calculator example is different (and, again, gaslight-y, see below).

What I meant in my text: A person wants to be a programmer. To me, a programmer is a person who writes code, understands code, maintains code, writes documentation, and so on. In your example, a person who changes a car tyre would be a mechanic. Now, if you use AI to write the code and documentation for you, are you still a programmer? If you have no understanding of said code, are you a programmer? A person who does not know how to change a car tyre, is that still a mechanic?

No, you’re something else. You should not be hired as a programmer or a mechanic.

Yes, that is “skill evolution” – which is pretty much my point! But the AI framed it like a counter-argument. It didn’t understand my text.

(But what if that’s our future? What if all programming will look like that in some years? I claim: It’s not possible. If you don’t know how to program, then you don’t know how to read/understand code written by an AI. You are something else, but you’re not a programmer. It might be valid to be something else – but that wasn’t my point, my point was that you’re not a bloody programmer.)

Gemini’s calculator example is garbage, I think. Crunching numbers and doing mathematics (i.e., “complex problem-solving”) are two different things. Just because you now have a calculator, doesn’t mean it’ll free you up to do mathematical proofs or whatever.

What would have worked is this: Let’s say you’re an accountant and you sum up spendings. Without a calculator, this takes a lot of time and is error prone. But when you have one, you can work faster. But once again, there’s a little gaslight-y detail: A calculator is correct. Yes, it could have “bugs” (hello Intel FDIV), but its design actually properly calculates numbers. AI, on the other hand, does not understand a thing (our current AI, that is), it’s just a statistical model. So, this modified example (“accountant with a calculator”) would actually have to be phrased like this: Suppose there’s an accountant and you give her a magic box that spits out the correct result in, what, I don’t know, 70-90% of the time. The accountant couldn’t rely on this box now, could she? She’d either have to double-check everything or accept possibly wrong results. And that is how I feel like when I work with AI tools.

Gemini has no idea that its calculator example doesn’t make sense. It just spits out some generic “argument” that it picked up on some website.

The AI makes two points here. The first one, I might actually agree with (“bad bot behavior is not the fault of AI itself”).

The second point is, once again, gaslighting, because it is phrased/framed like a counter-argument. It implies that I said something which I didn’t. Like the AI, I said that you would have to adjust the copyright law! At the same time, the AI answer didn’t even question whether it’s okay to break the current law or not. It just said “lol yeah, change the laws”. (I wonder in what way the laws would have to be changed in the AI’s “opinion”, because some of these changes could kill some business opportunities – or the laws would have to have special AI clauses that only benefit the AI techbros. But I digress, that wasn’t part of Gemini’s answer.)

Except for one point, I don’t accept any of Gemini’s “criticism”. It didn’t pick up on lots of details, ignored arguments, and I can just instinctively tell that this thing does not understand anything it wrote (which is correct, it’s just a statistical model).

And it framed everything like a counter-argument, while actually repeating what I said. That’s gaslighting: When Alice says “the sky is blue” and Bob replies with “why do you say the sky is purple?!”

But it sure looks convincing, doesn’t it?

This took so much of my time. I won’t do this again. 😂

@movq@www.uninformativ.de Wenn ich dran bleibe, vielleicht auf früher. Zum Glück kann ich die Seiten mit der Werbung für die 5-Minuten-Terrine überspringen.
Das stelle man sich mal vor: Damals™ gab es Werbung IN Büchern - mitten im Fließtext und inhaltlich leicht eingearbeitet. Das ist mir aber bisher aber auch nur in der Buchreihe begegnet.

@lyse@lyse.isobeef.org In my case it was a silver necklace, a hummingbird with a wing connected with the cold welding I mentioned using thin brass wires.

It made it in a goldsmithing class (I went to a private craftmanship high-school) so no phones allowed (no photos of it) and no “take home” of the works.

Here’s a rough sketch of it drawn by memory, the dots in the wing is where it connects to the body.

The technique is basically the same as i described, but the scale is much smaller, the whole piece was about 5-6 cm on the largest side.

The rivet was made by drilling a hole through the parts, than with a short and thicker drill you widen the hole on the surface to let the rivet settle flatter on the piece, then with a rubber hammer you hit it to flatten the head until it’s snug on the hole, lock them together by doing the same on the other side.

Note that widening the hole with a thicker drill head won’t make a difference with bigger holes, mine had holes of about 1-2 mm of diameter maximum.

Here’s a sketch of what is going on for clarity.

@zvava@twtxt.net @bender@twtxt.net At first I added it without thinking when planning the possible fields based on other UIs I was researching.

I was about to discard it but after thinking about it a bit I noticed that the services allowing to have a separated nick and display_name could unlock some good uses.

For example some added context or at-a-glance information like pronouns or statuses (like Artist [Accepting commissions] or App Name (v2.5)) while other used a more readable version of the nick (blog.domain.com became Person Name's Blog).

Of course it is absolutely optional and it can be safely ignored, but with my vision of being able to build more that a pure twtxt clients, giving it a first-class support just like the other known fields felt right to me.

@prologic@twtxt.net I checked a while a ago and there were, like, 3-5 collisions or something like that. Not that many. 🤷 I have to specifically look for them – I don’t notice it in normal operation.

And I need to make something absolutely clear as well here. Twtxt was completely and utterly dead back in {Aug 2020](https://yarn.social/about.html) when I came across the spec and its simplicity and realised the lost opportunity. Since then we’ve continued to grow a small but thriving community. The extensions we’ve built over time have stood and lasted the test of time for the past ~5 years. We need not break things too badly, because what we have today and was designed years ago actually works quite well™ (despite some flaws).

@prologic@twtxt.net I know we won’t ever convince each other of the other’s favorite addressing scheme. :-D But I wanna address (haha) your concerns:

1. I don’t see any difference between the two schemes regarding link rot and migration. If the URL changes, both approaches are equally terrible as the feed URL is part of the hashed value and reference of some sort in the location-based scheme. It doesn’t matter.

2. The same is true for duplication and forks. Even today, the “cannonical URL” has to be chosen to build the hash. That’s exactly the same with location-based addressing. Why would a mirror only duplicate stuff with location- but not content-based addressing? I really fail to see that. Also, who is using mirrors or relays anyway? I don’t know of any such software to be honest.

3. If there is a spam feed, I just unfollow it. Done. Not a concern for me at all. Not the slightest bit. And the byte verification is THE source of all broken threads when the conversation start is edited. Yes, this can be viewed as a feature, but how many times was it actually a feature and not more behaving as an anti-feature in terms of user experience?

4. I don’t get your argument. If the feed in question is offline, one can simply look in local caches and see if there is a message at that particular time, just like looking up a hash. Where’s the difference? Except that the lookup key is longer or compound or whatever depending on the cache format.

5. Even a new hashing algorithm requires work on clients etc. It’s not that you get some backwards-compatibility for free. It just cannot be backwards-compatible in my opinion, no matter which approach we take. That’s why I believe some magic time for the switch causes the least amount of trouble. You leave the old world untouched and working.

If these are general concerns, I’m completely with you. But I don’t think that they only apply to location-based addressing. That’s how I interpreted your message. I could be wrong. Happy to read your explanations. :-)

@prologic@twtxt.net I can see the issues mentioned, but I think some can be fixed.

1. The current hash relies on a url field too, by specification, it will use the first # url = <URL> in the feed’s metadata if present, that too can be different from the fetching source, if that field changes it would break the existing hashes too, a better solution would be to use a non-URL key like # feed_id = <UNIQUE_RANDOM_STRING> with the url as fallback.

2. We can prevent duplications if the reference uses that same url field too or the client “collapse” any reference of all the urls defined in the metadata.

3. I agree that hashing based on content is good, but we still use the URL as part of the hashing, which is just a field in the feed, easily replicable by a bot, also noting that edits can also break the hash, for this issue an alternative solution (E.g. a private key not included in the feed) should be considered.

4. For offline reading the source would be downloaded already, the fetching of non followed feeds would fill the gap in the same way mentions does, maybe I’m missing some context on this one.

5. To prevent collisions there was a discussion on extending the hash (forgot if that was already fixed or not), but without a fallback that would break existing clients too, we should think of a parallel format that maintains current implementations unchanged, we are already backward compatible with the original that don’t use threads at all, a mention style format for that could be even more user-friendly for those clients.

We should also keep in mind that the current mention format is already location based (@<example https://example.com/twtxt.txt>) so I’m not that worried about threads working the same way.

Hope to see some other thought about this matter. 🤓

@zvava@twtxt.net Not much of a known fact these days, but thereused to be a Yarn phone app (https://git.mills.io/yarnsocial/app), last version released 5 or so years ago, but it still suggests, it has to be somewhat feasable, to make another one. I don’t think anyone tried since, because the web version works well on phones, but I’m still hoping, we get a more native phone experience, one day.

Which is good, because Zone 5 is killing me! ( okay, not really but … )

For some time I’ve been trying to spend a fair amount of my running or jumping time in Zone 5, the hardest. After a little research I’m learning that maybe I should be concentrating more on Zone 4 instead.

Here’s an interesting thought/angle on this topic:

gemini://gemini.conman.org/boston/2025/08/21.1

A further check showed that all the network blocks are owned by one organization—Tencent [4]. I’m seriously thinking that the CCP (Chinese Communist Party) encourage this with maybe the hope of externalizing the cost of the Great Firewall [5] to the rest of the world.

@movq@www.uninformativ.de Thanks, glad you like it, but sadly I’m not sure, if there’s still a way, for this particular project, to continue.

Reducing 38 pixels (previous smallest) to 27, inside of a 7x7 square canvas, is a result I’m really happy with. Now it seems I can only shave off single pixels and get a lot worse looking results - to the point it doesn’t even look like my mascot, to me.

There doesn’t seem to be a hard cap for drawing tiny dogs. It’s possible to arrange 5 pixels, in a way someone recognizes them, as some kind of a dog. The record for cats, is currently a single orange pixel: https://youtu.be/gzeK8NKuzmg

The only way to beat that, is either a monitor, with just a single red diode lit, inside one of its pixels, or an image file that’s broken and empty, on purpose.