Thanks, @movq@www.uninformativ.de! Oh no, get well, mate!

Yes, our singer is a male I’m pretty sure. Of course, it’s hard to tell after sunset whether our blackbird wears a black or brown feather coat, but during daylight I’ve always only seen black ones sit on this roof ridge. It appears that Wikipedia is backing this up a little bit: https://de.wikipedia.org/wiki/Amsel#Reviergesang

I just added a video. Hmm, filtering the background camera noise also makes the audio rather squeaky. :-(

@lyse@lyse.isobeef.org Those are some very colorful shots. 👌 It was pretty warm here as well, health issues prevented me from going out, though.

(Have we established that Azabache is male? 😃)

@klaxzy@klaxzy.net I should cancel Netflix as well. Back when they started their streaming service, it was a revelation: Finally, I could watch interesting shows in English, without having to wait for years, and legally (I like to be a paying customer, if it’s good). But this is long over. The interesting shows are gone or, once again, I have to wait for years until they’re available on Netflix. So, why bother anymore? 🤷‍♀️

@rnlog@yarn.girlonthemoon.xyz Well, welcome back. 👋

@lyse@lyse.isobeef.org Right. :(

@lyse@lyse.isobeef.org it is lots of work, but things do not need to happen overnight! Take it easy, baby steps. I think it will be worth it!

¡Muchas gracias @bender@twtxt.net! I was also thinking about categorizing them a few years ago. But it’s so much work. I would have to tag every photo on its own. My use case goes more towards “give me all albums with squirrels”, though. Let’s see. I would need some tooling for easy tagging first. And then, the question is, which categories do I want to have to begin with?

@bender@twtxt.net Absolutely!

@bender@twtxt.net Trust me, so do it. I certainly gained some weight.

@bender@twtxt.net You got me there. Mine is mildly passive-aggressive, but ultimately supportive. Yours sounds… less helpful. 😀

@lyse@lyse.isobeef.org hmmm, pizza! With that amount of food (a) I wouldn’t complain, and (b) I will be obese on my return. LOL.

They are all very nice @lyse@lyse.isobeef.org! You truly need some sort of easy to navigate website with them all, categorised, and such… A robot can dream, right? :-)

@lyse@lyse.isobeef.org so beautiful! Azabache is so faithful too! That bird deserves some unsalted, organic seeds!

@movq@www.uninformativ.de more to read, but I read between your words that you are starving for more: https://aphyr.com/posts/411-the-future-of-everything-is-lies-i-guess :-D

@movq@www.uninformativ.de But even the same model behaves differently if you upgrade to a newer version. Or it hopefully should, why otherwise would one want to update in the first place? A fixed, predictable output would be the result of us oldtimers doing something with inferior, ancient technology!

@bender@twtxt.net Thanks, I’ll read it – once I have the energy. 😅

@lyse@lyse.isobeef.org Oh, yeah, right, I hadn’t even considered that (we mostly use one model). Choose a different model and it does something completely different. Cool stuff.

@movq@www.uninformativ.de I fully agree. And bonus points for different versions interpreting the same intructions in other ways. My collegues reply: Sure, but it just works so good, most of the time.

@klaxzy@klaxzy.net Wow, that’s crazy! Good move on your part. :-) Terminating the subscription was certainly the right thing to do.

@kiwu@twtxt.net I returned home from an on-site week at work. Commute was an adventure every day. It started off with a canceled train on Monday morning. Luckily, some very good mates granted my asylum. But even with shorter rides, I faced delays due to fuckwits on the tracks, then the train was terminated early due to the large delay, so we had to change trains. On the bright side, they then sent an entirely empty one, but I don’t get why they just didn’t continue with the first one instead. Due to another delayed train I didn’t catch my connection and the next one was canceled, so I had to wait for the following one. Super great fun. I’m very exhausted now and am very glad that I had already filed in flex time for tomorrow before the on-site event was scheduled.

Meeting my workmates in person was actually nice. It’s okay to do that once a quarter, I don’t need to do that more often. We should have had more meetings, though, trying to work in the office was expectedly incredibly inefficient. We certainly would have had more topics to actually discuss and think about. And most of them would have really benefited from nearly everybody being in the same room. Anyway.

Today, I even met my workmates from past projects in the office, too. So, the socializing was great.

@movq@www.uninformativ.de LOL. I think I get the idea. I am concerned about AI too. Managers starting with “I don’t know anything about this, but here is what says”. Infuriating.

I came across this one today, here is a gift link: https://www.nytimes.com/2026/04/15/opinion/art-artificial-intelligence.html?unlocked_article_code=1.bFA.XNiu.ZukFfdNl3Al1&smid=nytcore-ios-share

@bender@twtxt.net Or maybe I’m just shitty at communication and maybe that’s why nobody at work understands my “arguments” against AI/LLMs. 🤪🤣

(I’m too tired to rephrase the OP. Maybe some other day. Actually, rest assured that I will complain about this again. 😅)

@movq@www.uninformativ.de it went over my head, sorry. Someone wanted you to vet their instruction files, correct? People writing them should know what they are doing, otherwise they should engage with LLMs like that at all (unless it is a hobby, outside the enterprise).

@bender@twtxt.net … that was not my point. 🥴

@movq@www.uninformativ.de that’s the way large language models work, with a prompt. Rather than entering the prompt, most inference providers allow for specific files to be created that define the scope of what’s been requested, the skills the model is supposedly to posses, the stuff to “remember”, etc. Some will “learn” and add that “knowledge” to the proper files.

@itsericwoodward@itsericwoodward.com at least that encouraging in a gentle way. Imagine how do I feel waking up in the middle of the night to take a piss, and within seconds get a message from my watch “It is time to stand up!”. Like, wth?! 😅

@thecanine@twtxt.net so cute, and happy!

@prologic@twtxt.net that would work too. Not just for this adventure, but for the ones to come! Do it, do it!

@bender@twtxt.net Nah, i’m gonna start a blog for our adventure i think instead 😅

@prologic@twtxt.net maybe that’s your next project! :-P Welcome back from Vacationland!

Drew one in the old Pokémon battle angle and resolution.

@lyse@lyse.isobeef.org I’m afraid my ability to share photos like you do aren’t quite up to scratch 😅

@lyse@lyse.isobeef.org Clearing legally? You must have an amazingly efficient legal team – there’s like 10 new tools every week. 🤣

@prologic@twtxt.net Back to slaving away, now that your batteries are fully recharged. But we might be in for some little picturesque treats of QLD and NSW. :-)

@bender@twtxt.net Thanks! :-) Nature explodes everywhere at the moment, so pretty.

@movq@www.uninformativ.de No, just the damn training. For the tools, they always want the latest shit available. The company isn’t quick enough in purchasing and legally clearing the latest models, services, etc., they say. Other than that, they seem to be happy.

@prologic@twtxt.net Welcome back 👋

@lyse@lyse.isobeef.org

even our hippest AI enthusiasts found it absolutely terrible

Does this refer to the training course or to the tools themselves? 🤔

@lyse@lyse.isobeef.org one my most favourite smells too! I also love the rain. Absolutely love it!

@lyse@lyse.isobeef.org totally in love with these flowers, and bee!

@movq@www.uninformativ.de I couldn’t agree more! I also have the feeling that it causes more people to just accept “it’s a software problem, there’s nothing that can be done about it”. Which is very frightning to me.

Up until now, I was successful in refusing to actively use that crap. I had to do one mandatory AI training, but even our hippest AI enthusiasts found it absolutely terrible. Probably also nailed together by the same rubbish they want us to now use everyday as much as possible.

Code reviews are the part that I have to deal with most. And I believe that the code quality is degrading.

Let’s hope the bubble bursts sooner than later. It will definitely burst at some point. That’s for sure.

@lyse@lyse.isobeef.org Yes, and that’s why I’m 100% convinced that we’ll see a massive brain drain in a couple of years. This will affect young people even more, because they don’t have all the “old” knowledge to fall back on.

It’s concerning, I’ve warned about it many times, nobody listens.

I think the best thing one can do is explicitly not use any AI tools but keep your actual skills intact. Might be out of a (good) job for a while, but once this bubble bursts, this is who is going to get hired again. (I think.)

And considering how insanely expensive all this is, I’m still (mostly) convinced that the bubble will actually burst. This stuff just isn’t sustainable.

… or I might be wrong. And if so, I see an even darker future that I don’t want to put into words right now.

@xuu@txt.sour.is, what’s going on with y’all up in the mountains? The mouse has been mighty quiet for a while!

@movq@www.uninformativ.de “out of morbid curiosity” LOL. That draw a laugh out of me, so easily! 🤭

@movq@www.uninformativ.de Yup, I’ve also seen the floating point conversion happening with (1 << 63) - 1 yesterday night. But instead of pausing to think about it for a second, somehow all I had in mind was “give me a better representation, ain’t gonna have time for this shit”, so I turned it to hex. Beyond my comprehension what I was thinking there. O_o That’s embarrassing, unbelievable. Well, I blame late o’clock where my brain had already quit on me and went to bed.

Very interesting data point you raise there. The fun part didn’t cross my mind yet or at least I couldn’t pinpoint it. In hindsight it’s totally obvious, though. Past experience also tells me the exact same. Dealing with a problem and researching something myself is a so much more better teacher. The longer I faced up with a topic, the higher the chance to really manifest in long- or at least mid-term memory. If I just get told something, the odds are that it’s completely erased from memory in a matter of days if not hours.

@lyse@lyse.isobeef.org AI result ahead, feel free to ignore.

I “asked” the AI at work the same question out of morbid curiousity. It “said” that SQLite converts that integer to floating point internally on overflows and then, when converting back, the x86 instruction cvttsd2si will turn it into 0x8000000000000000, even if the actual floating point value is outside of that range. So, yes, it allegedly actually saturates, as a side effect of the type conversion.

I couldn’t find anything about that automatic conversion in SQLite’s manual, yet, but an experiment looks like it might be true:

sqlite> select typeof(1 << 63);
╭─────────────────╮
│ typeof(1 << 63) │
╞═════════════════╡
│ integer         │
╰─────────────────╯

sqlite> select typeof((1 << 63) - 1);
╭──────────────────────╮
│ typeof((1 << 63) ... │
╞══════════════════════╡
│ real                 │
╰──────────────────────╯

As for cvttsd2si, this source confirms the handling of 0x8000000000000000 on range errors: https://www.felixcloutier.com/x86/cvttsd2si

The following C program also confirms it (run through gdb to see cvttsd2si in action):

<a href="https://yarn.girlonthemoon.xyz/search?q=%23include">#include</a> <stdint.h>
<a href="https://yarn.girlonthemoon.xyz/search?q=%23include">#include</a> <stdio.h>

int
main()
{
    int64_t i;
    double d;

    /* -3000 instead of -1, because `double` can’t represent a
     * difference of -1 at this scale. */
    d = -9223372036854775808.0 - 3000;

    i = d;
    printf("%lf, 0x%lx, %ld\n", d, i, i);

    return 0;
}

(Remark about AI usage: Fine, I got an answer and maybe it’s even correct. But doing this completely ruined it for me. It would have been much more satisfying to figure this out myself. I actually suspected some floating point stuff going on here, but instead of verifying this myself I reached for the unethical tool and denied myself a little bit of fun at the weekend. Won’t do that again.)

@movq@www.uninformativ.de Now, it’s official, I AM an idiot. Yeah, thank you, that’s it! Apparently, I’ve quickly unlearned to use my brain with great success. :-(

@lyse@lyse.isobeef.org

Disclaimer: Can’t guarantee that I’m fully awake and I’m being trained at work not to use my brain anymore, so maybe this is complete bullshit. 😪🧟‍♀️

It says here that SQLite uses signed integers:

https://sqlite.org/datatype3.html

In pure bits, 1 << 63 would be 0x8000000000000000, but as a signed value, it gets interpreted as -9223372036854775808. Subtracting 1 yields -9223372036854775809 – but that doesn’t fit in 64 bits anymore. It’s possible that SQLite doesn’t want to wrap around but instead saturates? Haven’t checked. 🤔

With 62 bits, there is enough room.

With 1 << 64, I have no idea how SQLite wants to handle this, because this should immediately trigger a warning, because it doesn’t fit right away. Maybe it gets truncated to 0?

sqlite> select printf('0x%x', 2 * (1 << 64));
╭──────────────────────╮
│ printf('0x%x', 2 ... │
╞══════════════════════╡
│ 0x0                  │
╰──────────────────────╯
sqlite> select printf('0x%x', 0 - 1);
╭──────────────────────╮
│ printf('0x%x', 0 ... │
╞══════════════════════╡
│ 0xffffffffffffffff   │
╰──────────────────────╯
sqlite> select printf('0x%x', 0 - 2);
╭──────────────────────╮
│ printf('0x%x', 0 ... │
╞══════════════════════╡
│ 0xfffffffffffffffe   │
╰──────────────────────╯

@lyse@lyse.isobeef.org I don’t axtually k ow what the incline was we went up! Haha 😅 Honestly just guessing hmm must be documented somewhere 🧐

@prologic@twtxt.net Awwwwwwww! I love these stripes, very cool!

Oh, I bet these inclines are no joke. I also know one about 200 meters long terribly steep dirt path up a hill around here. Climbing that is super exhausting. I just looked it up on a map. And it’s just ~17° or ~30% incline. Okay, that’s absolutely nothing compared to your adventure. :-D

But you got your exercises for the day then. Which will make for an even greater sleep tonight. ;-)