sarnold | man what's the deal with so many more folks running into problems that sure feel like they've run their systems out of memory? it felt like that was a huge deal 25 years ago, fifteen to twenty years when it wasn't a big deal, and now it's a huge deal again the last year or two | 01:46 |
---|---|---|
ItzSwirlz | sarnold, chrome tabs scary | 01:47 |
arraybolt3 | sarnold: Everything became a browser. | 01:50 |
arraybolt3 | Think about it. Most of what you do is in your browser, the heaviest app on your computer. Your email client is a browser. Your chat client is a browser (via Electron). Sometimes even your ISO flashing utility is a browser (balenaEtcher, which uses Electron). Electron is taking over the world and taking our RAM with it, and browsers have consumed our use of computers. | 01:51 |
arraybolt3 | Between a heavyweight browser and tons of apps-that-are-really-browsers, no wonder our systems can't cope. I have 32 GB RAM and am sometimes startled to see more than 8 GB just in use when I'm not doing all that much. | 01:51 |
arraybolt3 | There's also memory leaks - KDE seems to just gradually lose track of stuff the longer you have it on, and GNOME is a power-hungry behemoth. | 01:51 |
arraybolt3 | One silly hope I have is to make my own distro one day that has a bunch of tools that allow one to use the Internet *without* a browser. | 01:54 |
sarnold | ItzSwirlz: lol | 01:55 |
arraybolt3 | (You could also just use Lynx and FrogFind but that's pretty limited.) | 01:55 |
sarnold | arraybolt3: yeah :/ it's pretty brutal. I keep thinking that it'd be nice to actually try out and contribute to gemini browsers, and then smack together silly little web<->gemini proxies for the content I care about, and try to get back to the internet of thirty years ago, but contents of today.. | 01:55 |
arraybolt3 | Gemini browsers? | 01:56 |
arraybolt3 | Like right now I have nothing more than KDE Plasma, Element, WeeChat, Chrome, Thunderbird, Xiphos, and a couple of terminals running, and I'm idling at ~4.9 GB RAM usage. | 01:57 |
arraybolt3 | I've had it get as bad as 8 GB, I think. | 01:57 |
sarnold | https://gemini.circumlunar.space/ | 01:57 |
sarnold | more gopher than web, but new :) heh | 01:57 |
arraybolt3 | sarnold: :O That looks so awesome. | 02:00 |
arraybolt3 | Just use APIs to port common websites over to that and boom. | 02:00 |
arraybolt3 | That may play into what I'm intending to do. | 02:00 |
sarnold | there used to be a 'weboob' package in debian that I think had a bunch of that kind of tooling already written | 02:01 |
sarnold | but the very juvenile names for everything was a bit much, I think debian kicked em out | 02:01 |
arraybolt3 | Ugh, I feel they took things too far removing image support. | 02:05 |
arraybolt3 | Even browsers from before the days of Windows 95 support images. | 02:06 |
rfm | once you start adding media types you've got the camel's nose in the tent and you'll be using gigs of ram again before you know it | 02:07 |
arraybolt3 | I suppose images can be transferred via links though. | 02:07 |
arraybolt3 | rfm: Heh, you're not wrong. | 02:08 |
sarnold | heh yeah | 02:10 |
sarnold | but I'd so much rather load a dozen images all in one go rather than request them one at a time by hand | 02:10 |
lotuspsychje | good morning | 02:11 |
arraybolt3 | One thing I really like about Gemini so far is it looks so simple that it should be reasonably easy to make a client without security vulnerabilities. | 02:11 |
arraybolt3 | That would be amazing. | 02:11 |
arraybolt3 | Now I want to make a graphical client :P | 02:12 |
sarnold | :D | 02:12 |
sarnold | just don't go building it with electron! | 02:12 |
sarnold | that'd spoil the whole point of the thing | 02:12 |
arraybolt3 | rofl | 02:13 |
arraybolt3 | No, I'd probably do it in C++ and Qt. | 02:13 |
arraybolt3 | Though... hmm, that's getting dangerously close to a web browser :P | 02:15 |
arraybolt3 | But this is giving me some hopefully really good ideas. Thank you for sharing that! | 02:20 |
sarnold | :D | 02:20 |
ducasse | morning | 06:48 |
ice9 | why the main repo archive is not using https? | 10:07 |
ravage | because it would be hard to distribute the certificates for an official country mirror to many people | 10:08 |
ravage | so the decision was made not to use it at all. https is supported in general. you can find a https mirrors with https at https://launchpad.net/ubuntu/+archivemirrors | 10:09 |
ogra | ice9, the Release files are gpg signed anyway (and get verified by apt) so https wouldn't be that much of an improvement ... | 12:05 |
ice9 | ogra, they are gpg signed by every archive independently or signed by canonical and distributed to the archives? | 12:20 |
ogra | the latter | 12:21 |
ogra | they are signed by the (unique) ubuntu archive key | 12:21 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!