Well, sure, with an image classifier, the bird identification is doable. I’m sure I could implement that if I went looking for some open source thingamabob that does that. But it’s still not something I could actually understand. That part definitely hasn’t changed over the years.

That must have been frustrating when the user base responded “but I already got my Blåhaj”

Frankly they should have nuked “OneNote for Windows 10” long ago and quietly replaced it with the Office version. Or better yet, not launch a separate version to begin with under the same name. But this is Microsoft, having multiple apps with the same name is just the norm.

AI business is owned by a tiny group of technobros, who have no concern for what they have to do to get the results they want (“fuck the copyright, especially fuck the natural resources”) who want to be personally seen as the saviours of humanity (despite not being the ones who invented and implemented the actual tech) and, like all big wig biz boys, they want all the money.

I don’t have problems with AI tech in the principle, but I hate the current business direction and what the AI business encourages people to do and use the tech for.

In SMITE’s case, the characters come from mythological sources and those sources are public domain. However, the way they’re depicted was chosen by the game developer and their depictions are copyrighted by them.

If someone copied the list of characters and made their own game with their own artwork and gameplay and everything, SMITE’s creators could do absolutely nothing about it. But if they copied any substantial elements from SMITE directly, then it starts to go in the direction where lawyers start rising eyebrows. At that point it’s no longer making original stuff based on the same PD material.

I have no idea why the makers of LLM crawlers think it’s a good idea to ignore bot rules. The rules are there for a reason and the reasons are often more complex than “well, we just don’t want you to do that”. They’re usually more like “why would you even do that?”

Ultimately you have to trust what the site owners say. The reason why, say, your favourite search engine returns the relevant Wikipedia pages and not bazillion random old page revisions from ages ago is that Wikipedia said “please crawl the most recent versions using canonical page names, and do not follow the links to the technical pages (including history)”. Again: Why would anyone index those?

GIMP (at least in v2) does have a vector path tool and stores the paths with the image! Thing is, they kind of work like selections and you have to explicitly stroke the paths on bitmap layers. It’s a bit more complicated than necessary and not easy to grasp at first.

For illustration work, having good support for both vector and bitmap elements is pretty damn convenient. For example, in comics, you draw the comics themselves in bitmap layers, while panels and speech bubbles go in vector layers. Having the ability to edit the speech bubbles easily is pretty neat.

(Optimally inking/outlines would be vectors too, but most people prefer to do that with bitmap tools anyway, or vectorise later.)

Krita actually does these pretty solidly - vector tools are there and they’re pretty easy to use. In GIMP 2, the vector path support actually is there and the editable texts are actually pretty great, but it has the air of “power user trick, for those in the know” rather than something people actually discover easily. You also need to update the vector strokes manually. (Haven’t tried GIMP 3 yet.) The fact that people still assume you can’t do this stuff really says it all.

Oh the term has rich history! First used in modern tech sector sense in 1975 about IBM.

[stern stare]

Forgotten the history, have we? I was referring to Microsoft’s tactics in the early 2000s.

Crypto bros have watered down the term and made it a laughing matter. They ruin everything they touch. Goddamn it.

The cheaper model will be called Espressy. It’s like the Espressigo, but without the Go.

Wait a second. You’re expecting Google to not FUD? Ha ha ha oh wow. I mean I didn’t actually expect them to do so, but yeah.

I run ad blockers. As a security measure. Ad companies collect insane amount of data and do a bunch of shady stuff whenever they can get away with it.

I want to support websites whenever I’m able, but the way ad companies operate just ain’t it.

If they clean up their act, maybe then I could stop using ad blockers, but it’s been decades and I don’t have high hopes.

Also using ad blockers for performance and usability reasons. For example, I used to use a bunch of Fandom wikis and couldn’t understand why people hated the UI. Then I saw how Fandom looks like without ad blockers and holy shit how can humans live like this

Mastodon, Pixelfed, Lemmy and Bookwyrm. They all seem to cover most of my social media needs which (in all other cases beside Lemmy) can be described as shouting in the void and being happy if someone else is there too.

There’s always the old piece of wisdom from the Unix jungle: “If you write a complex shellscript, sooner or later you’ll wish you wrote it in a real programming language.”

I wrote a huge PowerShell script over the past few years. I was like “Ooh, guess this is a resume item if anyone asks me if I know PowerShell.” …around the beginning of the year I rewrote the bloody thing in Python and I have zero regrets. It’s no longer a Big Mush of Stuff That Does a Thing. It’s got object orientation now. Design patterns. Things in independent units. Shit like that.

And Calibre, a third party software for managing ebooks, has a plugin to crack Kindle files.

Unfortunately currently broken for the latest version of Kindle for PC, which switched to a different encryption scheme. It also uses KFX file format that nobody likes, which fortunately can be converted to EPUB with another plugin, but de-DRMing doesn’t seem to work right now. It still seems to work for titles in AZW3/MOBI that didn’t get DRM update or didn’t have DRM in the first place.

No, that’s martial arts. Martial law involves less improvisation and adaptability. However, it loses to martial science.

Voyager.

Used Sync, but it’s not maintained. Liked Connect a lot. But, well, if the platform is open source, why not use an open source client too? Switched to Voyager and I’ve been very happy with it, actually.

Previously, you could just download the books on the Kindle for PC, use a random decoder software or install a plugin for Calibre, and boop, decoded books, readable in Calibre, can be converted to EPUB.

For ssssssome reasonnns I’ve been looking at how to do the same thing again, but apparently you need an old version of KfPC because the new one uses new encryption/file format that hasn’t been sussed out yet. Weirdly enough, even with the newer app, I’ve still been able to download a bunch of books that didn’t have DRM to begin with, but of course Amazon doesn’t exactly advertise if a book is DRM-free anywhere on the store page.

Also weirdly enough this quest of mine actually started last year when one Finnish ebook store was closed down, but that was less of a problem. I just downloaded all of my purchases as unencrypted EPUBs. Guess the local publishers are less dickish, worst thing they asked for was watermarking.

Long ago, I used Debian on my main PC but decided to go with Ubuntu on the laptop because it was easy to set up.

Later I switched to Debian on the laptop, too, because ultimately I felt Ubuntu was just Debian with extra steps.