|
Just a program that's just as terrible as any other solution, except now package managers need to deal with it this instead. And accidentally deleting a normal filesystem path potentially breaks an absolute ton of apps, with no way to tell other than broken symlinks. Way, way better than trying to work out some kind of modern, maintainable method of deploying applications. rpath can do it. But it shouldn't.
|
#
?
Jun 24, 2015 03:10
|
|
|
|
| # ? May 15, 2024 17:46 |
|
|
i havent checked out nix in a while is it still going to fix the world y/n?
|
|
#
?
Jun 24, 2015 05:38
|
|
|
The correct way to distribute applications (which is used by all modern platforms) is via bytecode anyway.
|
|
#
?
Jun 24, 2015 05:44
|
|
|
Breakfast All Day posted:i havent checked out nix in a while is it still going to fix the world y/n? linux has taken over everywhere but on the desktop it's truly the best irony for the neckbeards
|
|
#
?
Jun 24, 2015 05:50
|
|
|
i meant nix the purely functional package janitoring sperg army knife
|
|
#
?
Jun 24, 2015 05:56
|
|
|
well it's not at the core of rhel/centos or Debian/Ubuntu and isn't going to be used in an embedded system so it's maximum potential usage is limited to hyperspergs
|
|
#
?
Jun 24, 2015 06:02
|
|
|
evol262 posted:Also, .app containers basically being chroots with full deps inside them helps an awful lot. except an app bundle isn't a chroot and also doesn't have full deps inside it, only that stuff the developer chose to include (and no OS components because those aren't redistributable)
|
|
#
?
Jun 24, 2015 06:36
|
|
|
static deps don't solve the problem because binary interfaces as as much a social contract as a technological one and you have to solve both to find a working solution
|
|
#
?
Jun 24, 2015 11:10
|
|
|
eschaton posted:except an app bundle isn't a chroot and also doesn't have full deps inside it, only that stuff the developer chose to include (and no OS components because those aren't redistributable) 10/10 pedantic response Breakfast All Day posted:i meant nix the purely functional package janitoring sperg army knife guix is the one true sperg distro
|
|
#
?
Jun 24, 2015 14:54
|
|
|
evol262 posted:Way, way better than trying to work out some kind of modern, maintainable method of deploying applications. *shrug* I disagree. I think this is the maintainable way. the basic point is to just fix libraries so that multiple versions can be installed at the same time, by fixing them all (and probably adding a new feature to ld.so) to use relative paths to find their associated data. on top of this, just about any "install an application that wants specific versions" scheme at all can be implemented.
|
|
#
?
Jun 24, 2015 15:09
|
|
|
the real solution is to ditch all this terrible 1970s technology and rewrite the linux in something that reflects the modern state of the art, eg node.js
|
|
#
?
Jun 24, 2015 19:13
|
|
|
hobbesmaster posted:linux has taken over everywhere but on the desktop 2015 year of linux on the desktop Ok, maybe we should cut our losses and just shoot for 2016.
|
|
#
?
Jun 24, 2015 20:19
|
|
|
evol262 posted:10/10 pedantic response a++ why 2015 is still not tyolotdt response "you were wrong about how this important feature of a successful desktop os works and also how it is used in the real world" "pedantic!!!" (keeps on pushing various dysfunctional non solutions)
|
|
#
?
Jun 24, 2015 21:12
|
|
|
maybe 2015 will be the year of wayland on the linux desktop
|
|
#
?
Jun 24, 2015 21:24
|
|
|
Mr Dog posted:maybe 2015 will be the year of wayland on the linux desktop a legendary forger, an appropriate name
|
|
#
?
Jun 24, 2015 21:30
|
|
|
a few years ago one of the big name linux kernel devs had a big goog+ post about how the linux distribution model is fundamentally broken and will never scale beyond where it's at today without massive change. basic prob is that by optimizing for a kajillion pluggable core OS components so everyone can ~customize~, distros take on a ton of extra work integrating said core OS comps w/ thousands of upstream packages this arrangement has hugely negative ecosystem value in that it results in massive duplication of effort across distros and also greatly complicates the job of upstream developers. in the open sores world where a significant chunk of software is done on volunteer effort and/or chronically underfunded corporate sponsorship (underfunded bc u can't make much money selling open sores software) all this needless extra labor is deadly. there is simply no way to compete with closed src when you're budget is tiny and it takes way more work and politicking to get anything done until linux nerds divest themselves of the attitude that one must not settle on a single ui toolkit, init system, etc etc, prune down distros to be nothing more than the bare canvas on which apps are painted, and let app authors worry about distributing they are software they own selves, linux will continue to be a bad joke
|
|
#
?
Jun 24, 2015 21:31
|
|
|
tl;dr: if you have to use a linux use redhate/centos and ignore anyone who suggests otherwise.
|
|
#
?
Jun 24, 2015 21:41
|
|
|
Breakfast All Day posted:i havent checked out nix in a while is it still going to fix the world y/n? yes, it's p good. we had a 'discussion' about it a few pages back.
|
|
#
?
Jun 24, 2015 22:33
|
|
|
BobHoward posted:a few years ago one of the big name linux kernel devs had a big goog+ post about how the linux distribution model is fundamentally broken and will never scale beyond where it's at today without massive change. basic prob is that by optimizing for a kajillion pluggable core OS components so everyone can ~customize~, distros take on a ton of extra work integrating said core OS comps w/ thousands of upstream packages in other words, desktop linux today is an excellent fit for the desires of pretty much everyone who cares about desktop linux except for a vocal minority who still cling to the fantasy that somehow if it was more like windows or osx then normal people would suddenly start using it instead of windows or osx for some reason, and this is important because
|
|
#
?
Jun 24, 2015 22:45
|
|
|
can't wait for the poettergeddon
|
|
#
?
Jun 24, 2015 23:04
|
|
|
okay, so I've looked at xdg-app a bit more, instead of doing real work. I'm more convinced it's a good idea, but it still has a couple critical flaws. First, this is a system for writing sandboxed "gnome apps" or "steamos apps" not for writing "linux apps". Or at least, this is the only way the runtime approach doesn't become a clusterfuck. I imagine there'll probably be only like 3-6 major runtimes (prob. steam, rh, ubuntu, gnome, kde in decreasing order of my expectation that they'll actually succeed at maintaining ABI compatibility) however, writing "gnome apps" will be like writing "android apps" in the sense that this is very very much not writing ordinary linux programs. the xdg-app devs talk about firefox and libreoffice in their high-level future examples, but I don't think these apps will ever fit well into the runtime model. possibly they could be forced to fit in, but nobody will be happy and there will be more unruly things, like python or sublime text and so on that still refuse to fit. These app developers don't want to be put into the position of "oh, now you have to choose sides in the desktop wars, pick the gnome runtime its the only sensible choice you heretic." So instead of vendors providing apps that work on any distro, it'll be gnome devs wrapping firefox for the gnome runtime, so you can install it in yet another still not from the vendor way that's pretty much a distro within another distro. however, I think it'll be a great system for, like, linus's dive log program or i dunno, evemon. just pick gnome runtime who cares and boom, it works on all distros. great. Second, I think bind mounting custom /usr for the runtimes is a technically inferior approach compared to using rpath together with libraries that aren't broken (due to absolute paths.) It solves the problem just as well, and it's adaptable to making it possible to install "linux apps" from vendors regardless of distro in a much more general and useful way than the runtime concept. it also makes my life easier as a developer, as I can much more easily develop against versions of libraries are aren't installed system-wide
|
|
#
?
Jun 24, 2015 23:23
|
|
|
crazypenguin posted:okay, so I've looked at xdg-app a bit more, instead of doing real work. I'm more convinced it's a good idea, but it still has a couple critical flaws. lol
|
|
#
?
Jun 25, 2015 01:41
|
|
|
You can have multiple runtimes installed on a machine and there will only be a few of them. Say, GNOME 3.14, GNOME 3.16, and "Freedesktop.org 2015031300" (just the Wayland client libraries, pulseaudio client libraries, freetype, but no toolkit and definitely nothing except libraries and some shared static data, like, the window manager comes from your distro, not a runtime, and in turn things like the WM and coreutils run from your distro, not from within an xdg-app sandbox, same as something like Android RIL or ART is not an app in its own right). Desktop-agnostic apps like Firefox or Libreoffice would be built on a particular release of the FD.o runtime, GNOME apps would be built against a recent-ish version of the GNOME runtime. Each runtime version gets security patches but it doesn't get upgraded from like Gtk 3.14 to 3.16. When there is one canonical (heh) GNOME 3.14 distro that means there's a lot more concentrated manpower available to maintain it and provide security updates. xdg-app provides the infrastructure to make this happen efficiently in terms of disk space and bandwidth and trust, mostly by borrowing a bunch of ideas from Git. It is a good thing imo but the initial explanations were really fuzzy and it seemed like it was some big bloated Ubungu-VM-per-application thing. Sapozhnik fucked around with this message at 02:01 on Jun 25, 2015 |
|
#
?
Jun 25, 2015 01:59
|
|
|
hahaha gently caress no you can't trust any of these upstream attention deficit idiot fuckheads to debug their own poo poo
|
|
#
?
Jun 25, 2015 04:06
|
|
|
Soricidus posted:the real solution is to ditch all this terrible 1970s technology and rewrite the linux in something that reflects the modern state of the art, eg node.js cobol pls
|
|
#
?
Jun 25, 2015 06:05
|
|
|
BobHoward posted:a++ why 2015 is still not tyolotdt response BobHoward posted:a few years ago one of the big name linux kernel devs had a big goog+ post about how the linux distribution model is fundamentally broken and will never scale beyond where it's at today without massive change. basic prob is that by optimizing for a kajillion pluggable core OS components so everyone can ~customize~, distros take on a ton of extra work integrating said core OS comps w/ thousands of upstream packages BobHoward posted:this arrangement has hugely negative ecosystem value in that it results in massive duplication of effort across distros and also greatly complicates the job of upstream developers. in the open sores world where a significant chunk of software is done on volunteer effort and/or chronically underfunded corporate sponsorship (underfunded bc u can't make much money selling open sores software) all this needless extra labor is deadly. there is simply no way to compete with packages when you're budget is tiny and it takes way more work and politicking to get anything done  capitalism, thx. package darwinism. you don't need to politik when you're on top. the only duplication happening is debian-ish distro's utter poo poo patchsets on top of solved problems BobHoward posted:until linux nerds divest themselves of the attitude that one must not settle on a single ui toolkit, init system, etc etc, prune down distros to be nothing more than the bare canvas on which apps are painted, and let app authors worry about distributing they are software they own selves, linux will continue to be a bad joke that canvas is systemd. game, set, match. crazypenguin posted:First, this is a system for writing sandboxed "gnome apps" or "steamos apps" not for writing "linux apps". Or at least, this is the only way the runtime approach doesn't become a clusterfuck. I imagine there'll probably be only like 3-6 major runtimes (prob. steam, rh, ubuntu, gnome, kde in decreasing order of my expectation that they'll actually succeed at maintaining ABI compatibility) crazypenguin posted:sublime text crazypenguin posted:linus's dive log program or i dunno, evemon. just pick gnome runtime who cares and boom, it works on all distros. great. crazypenguin posted:Second, I think bind mounting custom /usr for the runtimes is a technically inferior approach compared to using rpath together with libraries that aren't broken (due to absolute paths.) It solves the problem just as well, and it's adaptable to making it possible to install "linux apps" from vendors regardless of distro in a much more general and useful way than the runtime concept. crazypenguin posted:it also makes my life easier as a developer, as I can much more easily develop against versions of libraries are aren't installed system-wide
|
|
#
?
Jun 25, 2015 07:48
|
|
|
your posting is really bad evol262.
|
|
#
?
Jun 25, 2015 07:56
|
|
|
evol262 posted:linux thread, not the "hip editor right now that actually sucks and nobody will care about in a year thread". go dig up a textmate thread from 4 years ago to rehash all the reasons why ${editor_of_the_moment} will never be as glorious as emacs or vim. linux is better without sublime text. hmmm its almost as if successful ideas of textmate carried on in sublime text, and now ones from st are influencing a new gen of middleweight editors also st is like 6 years old now, itll prob get usurped by one of the open source clones soon but its not exactly a flash in the pan vim is a language for interacting with editors that happens to have a reference implementation
|
|
#
?
Jun 25, 2015 08:06
|
|
|
lol 6 whole years I guess it's a lifetime in the cadt world, but real editors are older than I am
|
|
#
?
Jun 25, 2015 08:26
|
|
|
evol262 posted:stuff tl;dr Also, what does docker and containerization and jails and whatever have to do with linux desktop anyway??
|
|
#
?
Jun 25, 2015 08:37
|
|
|
BobHoward posted:until linux nerds divest themselves of the attitude that one must not settle on a single ui toolkit, init system, etc etc, prune down distros to be nothing more than the bare canvas on which apps are painted, and let app authors worry about distributing they are software they own selves, linux will continue to be a bad joke
|
|
#
?
Jun 25, 2015 10:32
|
|
|
crazypenguin posted:it also makes my life easier as a developer, as I can much more easily develop against versions of libraries are aren't installed system-wide let me tell you about this great compiler flag -isysroot
|
|
#
?
Jun 25, 2015 10:35
|
|
|
ruby idiot railed posted:hahaha gently caress no yeah, anyone who thinks the xdg poo poo is going to only have gnome 3.14 and 3.16 and not also every possible variant of 3.14.00001..3.15.99999 because some terrible dive log app "needs" it and will break with any other version is loving delusional the solution is not xdg, it is a vendor saying "we only support ___ and to put up a UI on our system you will use ___, we will not expose any other public API." I was hoping Ubuntu would finally be the people to do that for Linux and they've certainly come the farthest but everything they've done is still a laughably pale imitation of the consistency and compatibility that NeXT brought to Unix twenty seven goddamn years ago
|
|
#
?
Jun 25, 2015 10:41
|
|
|
hey guys i just finished playing the 2 metro reduxes on my fedora GNOME with nvidia GEFORCE on max settings and it went pretty fine thanks for reading!
|
|
#
?
Jun 25, 2015 10:50
|
|
|
yeah Gtk has "won" apparently even though more apps are moving from Gtk2 to Qt than are moving from Gtk2 to Gtk3. Or because people are fuckheads a huge body of GUI apps is still built against Gtk2. You can tell which programs are Gtk2 by the fact that they look like hot diahorrea on a hidpi display.
|
|
#
?
Jun 25, 2015 13:58
|
|
|
eschaton posted:let me tell you about this great compiler flag -isysroot yeah, it's not enough. now I need to run things, not just compile and link. now what about those libraries having hard coded paths to /usr/share and not finding their data? and of course, there's still the "how do i distribute this to users" which was really the main topic
|
|
#
?
Jun 25, 2015 14:56
|
|
|
sincere lol at anybody who doesn't use vim what are you even doing with your life???
|
|
#
?
Jun 25, 2015 16:34
|
|
|
|
|
#
?
Jun 25, 2015 17:15
|
|
|
Mr Dog posted:You can tell which programs are Gtk2 by the fact that they look like hot diahorrea reminder that gtk was created by extracting a bunch of lovely c code from gimp into a library because qt wasn't free enough 
|
|
#
?
Jun 25, 2015 17:23
|
|
|
|
| # ? May 15, 2024 17:46 |
|
|
so loving future posted:sincere lol at anybody who doesn't use vim mostly editing my .emacs tbqh
|
|
#
?
Jun 25, 2015 18:27
|
|