(#6ozbn3a) @lyse@lyse.isobeef.org That is brilliant! 🤣

(#6e546wa) FTR, I see one (two) issues with PyQt6, sadly:

1. The PyQt6 docs appear to be mostly auto-generated from the C++ docs. And they contain many errors or broken examples (due to the auto-conversion). I found this relatively unpleasent to work with.
   
2. (Until Python finally gets rid of the Global Interpreter Lock properly, it’s not really suited for GUI programs anyway – in my opinion. You can’t offload anything to a second thread, because the whole program is still single-threaded. This would have made my fractal rendering program impossible, for example.)

(#6e546wa) @prologic@twtxt.net Hm, same startup delay. (Go is not an option for me anyway.)

It’s hard to tell why all this is so slow. Maybe in this particular case it has something to do with fonts: strace shows the program loading the fontconfig configs several times, and that takes up a bulk of the startup time. 🤔 (Qt6 or Java don’t do that, but they’re still slow to start up – for other reasons, apparently.)

To be fair, it’s “just” the initial program startup (with warm I/O caches). Once it’s running, it’s fine. All toolkits I’ve tried are. But I don’t want to accept such delays, not in the year 2025. 😅 Imagine every terminal window needing half a second to appear on the screen … nah, man.

(#6e546wa) Be it Java with Swing or PyQt6, it takes ~300 ms until a basic window with a treeview and a listbox appears. That is a very noticeable delay.

Is it unrealistic to expect faster startup times these days? 🤔

Once the program is running, a new second window (in the same process) appears very quickly. So it’s all just the initialization stuff that takes so long. I could, of course, do what “fat” programs have done for ages: Pre-launch the process during boot, windowless. But I was hoping that this wasn’t needed. 😞 (And it’s a bad model anyway. When the main process crashes, all windows crash with it.)

(#3neip4q) @lyse@lyse.isobeef.org Yeah, I noticed that too. I haven’t double-checked my code, though. Maybe it has something to do with selecting the correct URL? I mean, these feeds don’t have any # url = fields, so maybe that’s it?

(#oos3ufq) @lyse@lyse.isobeef.org Ah, there it is. 😃 Never gets old. 👍

(#tqjqvwa) @arne@uplegger.eu … I still haven’t watched that show. 🤦

(#kf2fo5q) @prologic@twtxt.net And none of them use Yarn-style threading. I don’t think they’re aware of us, they’re probably using plain twtxt. Other than one hit by @threatcat@tilde.club a few days ago, I’ve seen no traffic from them. 🤔

Speaking of sunsets … https://movq.de/v/753ab5f9e5/sunset.jpg

(#gcaqtmq) @threatcat@tilde.club Let me guess, sl? 😏

(#sxlpyva) This looks like a botnet, to be honest. The IPs are all over the place. Ethopia, Brazil, Kenya, Lebanon, Netherlands, … I mean, that’s the logical thing to do, isn’t it? Do your web crawling on infected PCs. Nobody will block those, because those are the same IP ranges as legitimate requests. And obviously you don’t have to pay for computing time.

… and they all send invalid HTTP requests, all answered with HTTP 400 … How silly.

(#onzfgpa) @bender@twtxt.net Better safe than sorry, I guess. 😅

My goodness, a new level of stupidity.

The bots are now doing things like this:

GET http://uninformativ.de/projects/lariza/feednotify/datenstrahler/slinp/countty HTTP/1.1

1. That URL does not exist.
   
2. By including http://uninformativ.de in that request, this instructs the webserver to do an HTTP proxy request. Of course, this isn’t allowed on my webserver (and shouldn’t by allowed on any normal webserver), resulting in HTTP 400. And even if it were, the target would be the exact same server, making a proxy request unnecessary.
   

And of course, it’s not just 50 hits like this or 100 or 1’000 or 10’000. No, it’s over 150’000 in the last 2 days. All from vastly different IP ranges of different cloud hosters.

This almost looks like a DDoS attack, but it’s just completely stupid. This feels more like some idiot vibe coded a crawler.

I used Gemini (the Google AI) twice at work today, asking about Google Workspace configuration and Google Cloud CLI usage (because we use those a lot). You’d think that it’d be well-suited for those topics. It answered very confidently, yet completely wrong. Just wrong. Made-up CLI arguments, whatever. It took me a while to notice, though, because it’s so convincing and, well, you implicitly and subconsciously trust the results of the Google AI when asking about Google topics, don’t you?

Will it get better over time? Maybe. But what I really want is this:

• Good, well-structured, easy-to-read, proper documentation. Google isn’t doing too bad in this regard, actually, it’s just that they have so much stuff that it’s hard to find what you’re looking for. Hence …
  
• … I want a good search function. Just give me a good fuzzy search for your docs. That’s it.
  

I just don’t have the time or energy to constantly second-guess this stuff. Give me something reliable. Something that is designed to do the right thing, not toy around with probabilities. “AI for everything” is just the wrong approach.

(#ulrmviq) @lyse@lyse.isobeef.org Well, they say you have to build up stocks, don’t they? 😅

The font is fiamf3 (scaled up 2x, it would be too small when printed). It’s the same one that I use in my terminal and the status bars. 😃

(#wswlm2q) @lyse@lyse.isobeef.org Yeah, it feels broken. It often needs a couple of retries and a lot of patience. It’s been like that for months. 🫤

Lol, YouTube supports increasing the playback speed, but when you want to go to 4x, they want you to pay extra:

https://movq.de/v/a1c79485fb/s.png

(#5ara5ka) @lyse@lyse.isobeef.org There’s a couple of new users on https://tilde.club, but since this is a shared host, I doubt that they have access to their access.log files. Hence they’ll never see followers, unless we notify them out of band. 🫤

Android shopping list apps disappointed me too many times, so I went back to writing these lists by hand a while ago.

Here’s what’s more fun: Write them in Vim and then print them on the dotmatrix printer. 🥳

And, because I can, I use my own font for that, i.e. ImageMagick renders an image file and then a little tool converts that to ESC/P so I can dump it to /dev/usb/lp0.

(I have so much scrap paper from mail spam lying around that I don’t feel too bad about this. All these sheets would go straight to the bin otherwise.)

(#36db6ya) @lyse@lyse.isobeef.org Yeah, I’m glad I’m not the only one who didn’t get this right. 😅 You never had to configure a systemd timer? Lucky. 😅

(#gk5t5mq) @bender@twtxt.net No plus-aliases, just aliases. The mailserver runs on my OpenBSB box and is managed using BundleWrap (we use that at work), so to create a new alias, I push a new BundleWrap config to the server.

(#zxchmeq) @prologic@twtxt.net Glad you’re back. ✌️

(#o67gqfa) @lyse@lyse.isobeef.org It’s possible to run the validator locally (my blog generator scripts do that):

https://validator.w3.org/nu/about.html

That way you don’t forget. 🥳

(#5dyjtqa) @prologic@twtxt.net FWIW, I love the idea and I do the same with my email domains. It’s the most effective way to fight spam, IMO. 🥳

(#s62fiaa) @bender@twtxt.net All good. ✌️ It’s just that I’ve been through several iterations of this (on other platforms), AI output back and forth, pointing out what’s wrong, but in the end people were just trolling (not saying that’s what you had in mind), because apparently that’s “fun”.

(#oa65m7q) This is formatted poorly on twtxt.net, so here’s a plain text file: https://movq.de/v/971c5a125d/wall-of-text.txt

(#f4xwmia) … and now I just read @bender@twtxt.net’s other post that said the Gemini text was a shortened version, so I might have criticized things that weren’t true for the full version. Okay, sorry, I’m out. (And I won’t play that game, either. Don’t send me another AI output, possibly tweaked to address my criticism. That is besides the point and not worth my time.)

(#f4xwmia) @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.

The criticism of AI as untrustworthy is a problem of misapplication, not capability.

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.

Skill Erosion vs. Skill Evolution

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.

3. The Technical and Legal Perspective (Scraping and Copyright)

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.)

tl;dr

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?

Never again

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

(#kspztjq) @bender@twtxt.net It’s sad. Remember that Munich once ran the LiMux project. 😞

We could build a strong IT sector in Germany or the EU, but we just don’t want to.

(#34cy36q) @lyse@lyse.isobeef.org @bender@twtxt.net I’m not very knowledgable regarding the two points you mentioned, hence I didn’t include them in my list. But, yeah, from what I’ve heard, it doesn’t look good.