The issue is it’s an LLM. It puts words in an order that’s statistically plausible but has no reasoning power.

Hilarious to have to look this low for it, but who want to stand up and declare themselves mainstream.

Polished, reliable, and solid, and snaps are not a big deal or an insidious evil, and neither is Canonical. They make missteps for sure. But with containers etc stability is more important than immediate updates and it’s excellent about kernel updates for new hardware. It’s slick Debian, and if the fuckery ever gets real switching to Debian is easy.

Really? I disliked Perl for 3 decades on unix and Linux and I’ve never felt like I have been held back by not knowing or using it. I don’t remember the last time I saw a Perl script, let alone needed to understand one.

Yes - videos. I don’t recall latency problems since many years ago with some cheapo external speakers. FWIW I just tried a latency test on Youtube to check (currently on Shokz) and it seemed good. Frankly I have no idea if some low latency tech is being automatically used but I certainly didn’t take any steps to ensure it was (Ubuntu, these days using Pipewire).

How common is it that they don’t have that? because it’s a long time since I had latency issues in years of Bluetooth headphones. Anker, Phillips, Sennheiser, Shokz, all sub $100 headphones and I haven’t had latency.

deleted by creator

And for no charge.

However, having had a voice chat for an hour this weekend with someone 200 miles away, I can tell you that 30 years ago it worked so much better it’s not even funny; it was just expensive.

My phone provider (Fi) gave me an internet connected call rather than use the cell voice network (proudly telling me it was encrypted). It was full of dropouts and there was a serious latency that really inhibited conversation. I switched to a few other options like WhatsApp and the audio quality improved but the latency did not, and even got worse. Young people may be barely aware that a 200 mile phone call had tiny latency - you would not know there was any - because there was a literal wire connection between each end and communication was at the speed of light. Even transatlantic calls had minimal latency unless it went by satellite.

Sure today we can do it with video, but frankly, for a chat, I don’t even see much benefit. I’d certainly choose voice-only if it meant zero latency, and sadly I seem to have chosen a mobile provider that does its best to prevent that.

Yes, that’s why we use typing, to get better working code more easily. That’s why I use type annotation and enforced checkers in Python. It makes it so much easier and quicker to create good systems of any significance.

Seems pretty reasonable that if you want to tell the world you hate something, you might want the world to understand why you hate it, or else perhaps we might assume it’s not a reasoned position. That’s certainly the conclusion I’m coming to.

You must truly loathe vscode etc then if you hate emacs for overhead. I can’t really see why you should hate something just because it uses a slightly less small amount of resources. I don’t even know how you’d notice on any machine from the last 20 years.

said every young generation since the dawn of civilization

Bizarre that you’ve hated it for 30 years yet didn’t know one of the earliest things users learn about it (that actually is fine to use). Perhaps you should examine why you hate something you’re almost completely ignorant of.

Though most jokes and criticisms about Emacs betray complete ignorance of it, so you’re hardly unusual.

I feel like this would be a question best addressed where the online blind community hangs out (er… I don’t actually know where that is), and there may be plenty of blind users here (?) but I also wouldn’t be surprised to hear there are not. I certainly would tear my hair out if I had to work through comments audibly to try to work out what might be worth reading. But I don’t know and I’m not trying to speak for blind people.

The one blind developer I knew was heavily into Emacs, which has historically had a speech interface and can operate very well purely text-based, that I think made it a good choice. His ability to be productive was awe inspiring. I don’t know if anything else has taken over, but I expect if I lost my sight it might become my entire environment since I already know it. But I wouldn’t be surprised (and hope) there are more suitable options for less technical users.

Most often it’s a case of minor patches to the audio system, or recognizing a card reader (both of which are weirdly often unique per model, often even on sub-models), and these are patches that do make it into the older kernel. That process just takes time, and they want a usable image right away, so they get something specially patched that will shortly be replaced by a mainstream kernel update. For extreme cases there’s dkms, where you can have a package that replaces a kernel module from source that is rebuilt on kernel upgrades.

Canonical used to manage PPAs and packages to handle this - keeping them updated and then eliminating them as the fixes appear upstream. The PPA packages also contain trivial things like tweaks to get the keyboard hotkeys to work right.

I know this lives on because I have a Thinkpad and the vanilla Ubuntu installer adds “sutton” packages to tweak its install, and sutton is the internal name for Lenovo enablement (although the package has nothing active in it, so presumably vanilla just works by now, which I’d hope, as it’s a 3 year old machine).

Hardware enablement. A factory install where all the hardware worked, meanwhile all the fixes are upstreamed. So backported fixes to the current LTS.

Just that work, and frankly doing it near the factories makes sense. We were “offshore” doing it from the US. Canonical is global, headquartered in the UK.

Just me. It was my job for a while at Canonical, until the work was moved to China.

Lol, no mention of the fact that Ubuntu was already shipped on almost the entire Dell range, but only in China and developing world markets. This was because they had sold millions of laptops without OS in those markets, which immediately were flashed with pirated Windows, and Microsoft were pissed off. They pressured the Chinese govt to require computers must ship with an OS, so Cannonical/Ubuntu stepped in, did it for cheap (~$1/machine) and… they were still of course flashed with pirated windows immediately.

They didn’t ship to the US or Europe etc., because in those markets Dell got more kickback-money than they spent, from Windows and the various crapware they shipped pre-installed. So shipping Ubuntu in the US actually cost Dell money.

I’ve really appreciated running i3 within gnome-flashback. I keep hitting things that just work, that I didn’t even realize I wanted. I hit the pause/mute button on my headphones when watching youtube and it paused… Things like ssh agents, hotplugging monitors and having it remember your preferred config, the main gnome settings GUI, the compose key, etc… I’m just not interested in reinventing all that, even though each individual step is usually not hard.

I’m glad to see Regolith is making a DE with Sway. One day I’ll migrate from i3 and I may let Regolith give me a DE.

Sway is a WM not a DE. So you create your own DE? Or, I see Regolith is integrating sway, I think with Gnome Flashback as with i3. Not sure if there are others.