Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)GE
Posts
1
Comments
1,006
Joined
2 yr. ago

  • I'm not really familiar with Automattic or any of their acquisitions (I know Tumblr and Pocket Casts, but I'm not a regular user of either). https://en.wikipedia.org/wiki/Automattic#Products

    What's their track record here? Should we expect anything they acquire to be gutted and squeezed like they're Broadcom, or do they actually develop the things they acquire in a way that serves their users?

  • I mostly agree. I usually recommend Mint to new users, largely because their web site and defaults are very beginner-friendly. Mint is the modern version of what Ubuntu used to be 10-15 years ago. At this point I don't think Ubuntu has tangible advantages over Debian for new users.

    I really like Slackware's site. It's not sexy, but it's functional, organized, and easy to navigate. The Zsh site is counterintuitive to me with that sidebar-that's-not-really-a-sidebar, and hyperlinks whose text requires the context of a header that is not aligned with them.

    I just checked out Ubuntu's web site for comparison, and...uh...now I feel like I owe Debian's web site an apology. I guess the consumer desktop Ubuntu distro doesn't actually have its own web site anymore? I mean, you can get to it from there, but it's hidden under menus, and seems almost like an afterthought. Ubuntu's main web site is bizarre right now, with a prominent green "Download Now" button that does not lead the user anywhere close to downloading Ubuntu, but rather directs them to one of a rotating selection of signup forms to download various technical whitepapers like "A CTO’s guide to real-time Linux". That's a radically different target audience than the last time I went to their web site (and also a weird design anyway).

  • It's so silly but I love multiple reactions in Slack and Discord. Signal's reaction system feels half-baked in comparison.

    I'd also like custom reaction, though I'm not sure how that would work on a direct messaging platform.

  • Why not? It's a great general-purpose distro.

    My point is that 9front's user-unfriendliness is a feature (explicitly intended), whereas I think Debian's is a bug (not intended or desired). I'm not psychic, though, so I could be wrong about the Debian team's goals.

  • The Debian web site needs a good UX overhaul. Prioritize the things people are most likely to want, make them prominent and uncluttered, and present a logical flow from one task to its follow-ups.

    Just a quick glance yields the simplest example: the download link is not the first or most prominent thing on the main page. Clicking "download" gives you the netinst AMD64 ISO, which is reasonable enough, but there is no indication of how to install it. Clicking "user support" takes me to a page with extremely verbose descriptions of IRC, usenet groups, and mailing lists. I think the fastest way to get installation instructions is to click the tiny "other downloads" link (after I've already downloaded the one I want!), and then a link to the manual from there.

    This is not a good UX. This is a demographic filter. You can argue that's appropriate for a technically-oriented OS. 9front explicitly makes itself unapproachable to dissuade casual users, but I think Debian can and should be more appealing to mainstream, casual newcomers.

  • That's one way of looking at it.

    Another way of looking at it is that we've had image macros and memes for a long time, and part of the charm and success of the format is precisely that it requires very little effort, anyone can do it, and there are a million variations. I don't have a problem with the same idea in audio form.

  • The way I see it, this is an addition, not a replacement. I mean, the MIT license has existed for decades. Plenty of time for any interested human singer to have done it.

    We're going to see a lot of amusing ideas brought to life that would otherwise have simply not been worth the person-hours.

  • Pretty happy to axe the dock and put the Firefox search widget right at the bottom. I don't understand why so many browsers, launchers, and apps only let you select from a few hardcoded search engines instead of just letting you add your own with a URL pattern.

    Widget resizing actually works now, whereas it used to be weirdly dependent on your home screen grid size settings.

    Only issue I've found so far is that increasing your home screen's row count completely removes all your icons. That's a big bummer. It was always pretty flaky with arrangement shifting but this is a new one on me.

  • I wouldn't say Apple disregards backwards compatibility, but they certainly don't prioritize it to the degree Microsoft does, or that the general open-source community does. For Microsoft, backwards compatibility is their bread and butter. Enterprise customers have all sorts of unsupported legacy shit, and it dictates purchasing decisions and upgrade schedules.

    Apple gave devs and users a ton of lead time before dropping 32-bit support. The last 32-bit Mac hardware was in 2006 (the first gen of Intel Macs); it wasn't until Catalina's release in 2019 that 32-bit apps stopped running, and Apple continued releasing security updates for older OSes that could run 32-bit apps for a couple years after that. So that was basically 15 years of notice for devs to release 64-bit apps.

    That was much more time than they gave Classic Mac apps under OS X, or PowerPC apps on Intel. I was much more annoyed when PowerPC support was axed. Only a matter of time until Intel apps stop running on Apple Silicon, too. That's gonna be the end of the world for Steam games. Ironically, it's already easier to run legacy Windows and Linux games on Mac than it is to run legacy Mac games.

  • In the context of video encoding, any manufactured/hallucinated detail would count as "loss". Loss is anything that's not in the original source. The loss you see in e.g. MPEG4 video usually looks like squiggly lines, blocky noise, or smearing. But if an AI encoder inserts a bear on a tricycle in the background, that would also be a lossy compression artifact in context.

    As for frame interpolation, it could definitely be better, because the current algorithms out there are not good. It will not likely be more popular, since this is generally viewed as an artistic matter rather than a technical matter. For example, a lot of people hated the high frame rate in the Hobbit films despite the fact that it was a naturally high frame rate, filmed with high-frame-rate cameras. It was not the product of a kind-of-shitty algorithm applied after the fact.

  • There are plenty of lossless codecs already

    It remains to be seen, of course, but I expect to be able to get lossless (or nearly-lossless) video at a much lower bitrate, at the expense of a much larger and more compute/memory-intensive codec.

    The way I see it working is that the codec would include a general-purpose model, and video files would be encoded for that model + a file-level plugin model (like a LoRA) that's fitted for that specific video.