Even as a speaker of the language I haven't a single bloody clue what a "hazzy" is. Is ChatGPT normally this bad at being prompted in one language to generate text in another, or is this an operator fuckup, I wonder?
It's a paywall. The text is supposed to have a blur over it so it's not readable, and it's "smart" to use fake text for something like this. I guess the blur didn't render for you?
What a truly bizarre article. Arguments about pre-existing randomness, temperature and whatnot aside, I simply cannot comprehend what the author here really thinks the "best word" is. There's no such thing. We humans fall on familiar patterns of writing ourselves, so we may forego something with a flourish in favor of a more commonly-used word unless we put in effort to be "special", which should be used sparingly. That is to say, human writers are likely to choose a "worse" word in far more than the supposed 51% of cases, and that has no effect on the actual quality of writing in the end.
But even if there were such a thing as a truly "best word", for some context, what are the examples here? Mango vs pineapple? Gray vs overcast? In what case is one of these better, that AI would normally infer but would suddenly be "perverted" by SynthID? Do you think your emotional state and preferences are being evaluated if they aren't explicitly in memory? And if they are there, do you think that the generator will bypass those instructions in favor of the watermark instead of placing it somewhere you won't care? I just. Genuinely don't get it. There may be words that matter in specific contexts or to you as a reader, so you should bloody well put them there.
The images are pretty, but good lord what horrendous UX. Why are they still stuck with this? Also -- why would someone else be allowed to tweak a user's file manager settings? Insane.
The first definition of this phrase per Merriam-Webster is "to gloss over, explain away, or patch up (differences, disparities, etc.) especially in order to maintain a semblance of unity or agreement". I was also not familiar with it until I went to look.
You're not wrong regarding your first paragraph. However, capitalism is not going anywhere ;) as such, moral objections to the use of tooling in this system amount to objections to the tools themselves. I don't see "people might use AI more if we dismantle and rebuild the entire world order" as a particularly realistic scenario, sad as that is.
And in any other case I would agree with your second paragraph as well! But AI is very, excessively loud. Even if you never wanted to -- especially if you never wanted to -- you were already overexposed to it ages ago (when it was at its worst, really, but that's probably beside the point). You would have made your choices then: use now, never use, or use later if it improves. I'll grant the technicality that "everyone" is hyperbole of course, but when you have grandmas and children both talking about the thing, yeah no. You'd be hard pressed to find someone without a stance. And someone remotely interested in Codeberg not having one? Come on now.
"Not going anywhere" is such a strange argument. Tomato juice isn't going anywhere, does it mean I'm going to be guzzling it at some point in my life? What?
If you've made your choice about LLMs based on morals or ethics rather than the quality of the tooling, you're not going to change that choice because the models haven't disappeared or have improved, unless you're forced to (but that forced bit applies only to commercial open source which really just isn't on Codeberg at all in the first place).
It has been long enough that just about everyone has made that choice by now. Those who are okay with this won't magically flip and start using LLMs and grow a reason to leave.
I don't know the answer to this, but I have to wonder if, for source files specifiically, .git is the culprit here... It's not part of your project, it's part of your repo. Which maybe makes sense if people ever divorced their source code from the repo but that's not a thing anymore. Others probably just copied it.
This is why Linux needs a sensible, universal way to hide files that doesn't involve bloody renaming them (since that will just cause the original one to be re-created! What is that!?). Many file managers respect the .hidden file, which is a start! But it's not enough, shells and `ls` don't care. It's one of those many strange failures that make Linux on the desktop still painful to use.
Victimized, not that many. Generally if you do domain squatting like this you try to appear "legitimate". Sell actual paint or whatever. Because the moment you try actually pretending to be the thing you aren't... well, this happens. Quick takedown, domain gets transferred to the actual owner of the trademark.
I haven't installed Maria on Windows in forever, but does it not do that anymore? Yes, this is definitely true -- I remember seeing it several times, that's how I found out HeidiSQL existed!
Every time someone claims LLMs "talk like real people", I have to wonder what kind of people they talk to, what kind of conversations they lead, and just how boring their life must be. Like. What? No, they do not. No person actually talks like this while they're being a person. At work, sure. But that's not "people mode".
> Make the agent sound clinical, robotic.
It literally already does. I don't know how you'd make it sound less natural than this, at that point, without making it literally go "beep-boop" every sentence.
I am extremely confused by the fact that... you basically can't make a Google account without providing a phone number anymore, can you? The article even says as much. So what is this? In what scenario would this actually manifest?
The SQLite parts that exist in Firefox right now are no less fiddly. If you crash the thing and the journal files are still there, it starts back up and presents you with a blank profile. Absolutely horrifying if you don't know about this quirk (and about the fact that you can go in and delete those even after such a restart happens).
Personal experience here -- every now and again GitHub will slow to an absolute crawl. All of it. Page loads, pulls and pushes, will take minutes to complete. And you can forget about trying to download a release when this happens...
I haven't really looked for other complaints so I can't provide much in ways of sources/corroborations.
Wayland, Flatpak and Gnome are dead set on trying to turn my computer into a weird oversized smartphone. AppImage tries to recreate the "just download it and run" experience that Windows and macOS have, but hasn't quite gotten there yet (AppImages will sometimes just randomly link to system-wide libraries anyway, forgetting to bundle dependencies).
Meh. I guess I'll check back in some years, as it always goes.
Do you genuinely, truly believe that "4 digits then "-" then 2 digits then "-" then 2 digits" is more readable than "\d{4}-\d{2}-\d{2}"? Do you think your walls of text won't make everyone's eyes glaze over a few lines in? More importantly, do you believe it's more writable? A lot of regex operations are ad-hoc, and being able to type them out quickly and look at them in a small text field at a glance is very important. And as far as coding is concerned, well, your thing is useless outside of the JS ecosystem, is it not?
Dead? Replaced?... Your toy is cute, but it's more suited as a regex generator, or perhaps parser/"explainer" (but we already have several well-known projects there) than a standalone tool.
Discord does not require a smartphone for MFA. If you are thinking TOTP, there's a ton of desktop apps that generate those. Passkeys are supported for the purpose as well.
Replacing E's with O's is, somehow, also an issue with OCR as old as time. Remember seeing a bootleg Family Guy (?) DVD going around some years back where ovory songlo vowol wos o on tho synopsos on tho bock.
Appears to be running on plain HTTP, and trying to access it over HTTPS presents a bad cert and then redirects somewhere incomprehensible. No idea what the domain owner is doing here.