After a rumour circulated on Reddit about Valve using Flatpak in future for Steam, it was suggested by user "gutigen" that I reach out to Valve for an official comment. I now have an answer.
I asked a simple question: "People have been claiming that Valve is moving Steam to being a Flatpak package, would you be able to confirm if this is happening?"
The reply was this:
Quoted with permission.
To put that into simpler terms: they won't be using it for how they deliver the games directly, so you won't in future get games downloaded and installed as Flatpak packages, but how Steam itself is packaged and interacts with your Linux distribution may change in future.
So essentially, nothing is happening right now as it's still at the research stage. However, it is great to hear that they are actually looking to further improve Steam on Linux. Article taken from GamingOnLinux.com.
I asked a simple question: "People have been claiming that Valve is moving Steam to being a Flatpak package, would you be able to confirm if this is happening?"
The reply was this:
Pierre-Loup A. Griffais, ValveHi Liam,
Not quite; we're looking at some of the underlying technology to see if it would be a good fit to improve the Steam runtime environment interactions with the host system. If we went forward with it, we would be using some of the same kernel functionality Flatpak/bubblewrap is, and hopefully reusing some core code, but we have no plans to change the cross-platform distribution and packaging method at the core of Steam.
Quoted with permission.
To put that into simpler terms: they won't be using it for how they deliver the games directly, so you won't in future get games downloaded and installed as Flatpak packages, but how Steam itself is packaged and interacts with your Linux distribution may change in future.
So essentially, nothing is happening right now as it's still at the research stage. However, it is great to hear that they are actually looking to further improve Steam on Linux. Article taken from GamingOnLinux.com.
Some you may have missed, popular articles from the last month:
The comments on this article are closed.
All posts need to follow our rules. For users logged in: please hit the Report Flag icon on any post that breaks the rules or contains illegal / harmful content. Guest readers can email us for any issues.
Does anyone know why they chose Flatpak over Snap?
Flatpack has one advantage over snap, like the possibility to bundle a runtime containing common dependencies shared between Flatpack packages. Thus Flatpack packages tends to be much smaller than snap ones. The runtime can be updated separately providing benefits for all other Flatpack apps without breakage.
2 Likes, Who?
Snappy have the same functionality ^^ .Does anyone know why they chose Flatpak over Snap?
Flatpack has one advantage over snap, like the possibility to bundle a runtime containing common dependencies shared between Flatpack packages. Thus Flatpack packages tends to be much smaller than snap ones. The runtime can be updated separately providing benefits for all other Flatpack apps without breakage.
1 Likes, Who?
I made the mistake of adding Snappy to my system a while back, it took me two days of fighting to get rid of it, every time I removed it a stray piece would cause the whole thing to reinstall, it's worse than a bloody virus & I want no part of it again thanks. If Flatpak is anywhere near as bad as that they can keep it off my system.
FlatPak is a hell of a lot better, very self contained with few external deps.
And thus begins the flamewar... Or do you want to back up your "a hell of a lot better" with some facts? (Even in post-factual times.)
2 Likes, Who?
Mountain Man Nov 23, 2016
A few things come to mind:
1) Isn't this essentially what Steam already does with its built-in runtime library?
2) Isn't it up to the developer and not Valve how they wish to distribute their software? If a developer wants to use Flatpak instead of the Steam runtime library then why can't they? For that matter, developers have always been able to bundle individual libraries if they wished, so what advantage does Flatpak really offer?
3) I find it ironic that Flatpak is supposed to make it easier to install software on multiple Linux distros... but they only officially support Fedora and Ubuntu.
1) Isn't this essentially what Steam already does with its built-in runtime library?
2) Isn't it up to the developer and not Valve how they wish to distribute their software? If a developer wants to use Flatpak instead of the Steam runtime library then why can't they? For that matter, developers have always been able to bundle individual libraries if they wished, so what advantage does Flatpak really offer?
3) I find it ironic that Flatpak is supposed to make it easier to install software on multiple Linux distros... but they only officially support Fedora and Ubuntu.
0 Likes
Why not AppImage?
0 Likes
speaking of... valve hasnt said a word about wayland or mir... any clue what mr steam might do when that bridge comes?
0 Likes
A few things come to mind:
1) Isn't this essentially what Steam already does with its built-in runtime library?
2) Isn't it up to the developer and not Valve how they wish to distribute their software? If a developer wants to use Flatpak instead of the Steam runtime library then why can't they? For that matter, developers have always been able to bundle individual libraries if they wished, so what advantage does Flatpak really offer?
3) I find it ironic that Flatpak is supposed to make it easier to install software on multiple Linux distros... but they only officially support Fedora and Ubuntu.
ok, instead of googling random internet opinions lets do the Archwiki chalenge
https://wiki.archlinux.org/index.php/Snapd
https://wiki.archlinux.org/index.php/Flatpak
my conclusion from the above:
snap: PLEASE NO!!!
flatpack: sounds good and better than the situation we have now.
0 Likes
orochi_kyo Nov 23, 2016
You know, when I see this Snap vs Flatpack war, what I see is that a reliable software distribution for all linux distros just have a long way.
If this happens along users, I cant imagine how harsh is the situation between software devs, thinking they have the best option for linux instead of accepting that maybe others hit the nail ...
Last edited by orochi_kyo on 23 November 2016 at 6:47 pm UTC
If this happens along users, I cant imagine how harsh is the situation between software devs, thinking they have the best option for linux instead of accepting that maybe others hit the nail ...
Last edited by orochi_kyo on 23 November 2016 at 6:47 pm UTC
1 Likes, Who?
Don't see the problem myself. But Flatpak is a RedHat thing, and RedHat don't like Canonical, so a lot of RedHat people are negative about anything that comes from Canonical.
It's not only Red Bat, there is tons of reasons not to like Canonical :
- They don't contribute much to the Linux community
- They have the "not invented here" philosophy, meaning that if a Linux project is successful they'll always create a concurrent project just for the sake of having one. This takes developers away from the main project and slow down the evolution of Linux (Unity, Mir, Snap...)
- They aggressively patch the packages in their distribution, making it so that bugs in the upstream project are not in Ubuntu but often other bugs are here, making it a nightmare for the original developers (I don't really know if they still do this)
- They partenered with Amazon to put ads on Ubuntu
- They partnered with Microsoft to have Ubuntu running on Windows
I could probably find others but I think that's good enough.
1 Likes, Who?
Snappy have the same functionality ^^ .Does anyone know why they chose Flatpak over Snap?
Flatpack has one advantage over snap, like the possibility to bundle a runtime containing common dependencies shared between Flatpack packages. Thus Flatpack packages tends to be much smaller than snap ones. The runtime can be updated separately providing benefits for all other Flatpack apps without breakage.
As far as I know only Flatpack provides this feature...
Another drawback for snappy is you cannot set up a third party repo for it (only the one from Canonical is allowed), which is probably the biggest issue for Valve.
0 Likes
Nop Snappy propose this function : https://developer.ubuntu.com/en/blog/2016/11/16/snapping-qt-apps/Snappy have the same functionality ^^ .Does anyone know why they chose Flatpak over Snap?
Flatpack has one advantage over snap, like the possibility to bundle a runtime containing common dependencies shared between Flatpack packages. Thus Flatpack packages tends to be much smaller than snap ones. The runtime can be updated separately providing benefits for all other Flatpack apps without breakage.
As far as I know only Flatpack provides this feature...
Another drawback for snappy is you cannot set up a third party repo for it (only the one from Canonical is allowed), which is probably the biggest issue for Valve.
And you can add use different repo and add various stores. In fact it's a nice solution for Steam because they can really easily integrate a proper Steam Snapp repo in steam.
Last edited by seb24 on 24 November 2016 at 9:11 am UTC
1 Likes, Who?
Nop Snappy propose this function : https://developer.ubuntu.com/en/blog/2016/11/16/snapping-qt-apps/
And you can add use different repo and add various stores. In fact it's a nice solution for Steam because they can really easily integrate a proper Steam Snapp repo in steam.
Is this really the same ?
Flatpack's runtime have a namespace, and they can be installed in parallel, you can have different version for different apps. The system automatically remove duplication, so you only have what you need installed.
Well, for the store it seems you can setup your own, but the whole publication process seems a lot simpler with Flatpack (Flatpack repo is more like a PPA).
0 Likes
autonomouse Nov 24, 2016
I'm not the best placed person to defend Canonical here, seeing as though I work for them, but I've been a fan for longer than I've been an employee, and I have nothing to do with making the OS or snappy so I'm going to anyway. It's just that I don't like to see people bashing any FLOSS company - Canonical or anyone else - they're giving away their stuff for free, without lock-in or restrictions for you to do what you want with. Even if it doesn't benefit you directly, it will benefit you by the other organisations you do like being able to take advantage of their work too.
In-fighting solves nothing, especially when there is a much more important things to be doing, like trying to limit the influence of proprietary technologies. You might not like what a particular FLOSS company is making, but if so, just don't use it. There's no reason to dislike them just because they don't give you exactly what you want, all the time.
Well, apart from making and releasing the most popular linux distro, rigorously tested and provided on a regular basis to you, free of charge, twice a year. But I take the point that they have historically made relatively few contributions to the kernel itself. They contribute to a lot of upstream projects (Openstack, etc), and they provide a *lot* of FLOSS tools: In addition to snappy, there's Juju, MAAS, LXD, the fan - all good stuff, especially for server side. Then they do lots of hardware certification - talking to manufacturers and making sure ubuntu just works on lots of laptops, etc. This has tangible benefits for other distros. And that's not to mention Upstart, Launchpad or Bazaar. Some things are better received than others, but to say they don't contribute to the community seems a little disingenuous.
If Canonical had a real NIH syndrome then they would have stuck with Upstart and deviated from Debian. It didn't. When Debian moved to systemd, it followed, even though upstart actually predated systemd.
There's bzr I suppose, but they're starting to move towards git.
Launchpad is still heavily used, but it didn't quite live up to its promise. Because they chose bzr not git, github kinda won that one. Most people are moving to github now (including Canonical for some things).
At the end of the day, they're engineers and liek building stuff. If something isn't exactly what they need and they don't feel it's worth it to change it, they'll create something new. But really, since when has the open-source community been about everybody rallying behind a single thing? Surely the idea is to throw as much against the wall as you can and see what sticks?! If we all wanted to live in a monoculture, we'd been on www.gamingonios.com. In hindsight, you can look back and say everyone should have invested in technology X and not bothered with technology Y, but at the time, who knows which is right? Who's to say that Wayland has it right and Mir has it wrong? Canonical investigated Wayland, then decided against using it. They had their reasons, it was their choice. Sure helping Wayland would have moved everybody along, but if it wasn't the right tool for the job, then they weren't going to do it.
What's the problem with having both and see which is better in the long run? We all know that competition drives innovation. Looking back at the KDE/Gnome split, do we really think one of those projects is evil and the other saints? No, they have taken different approaches, and the community is richer because of it. Ideas are borrowed from different places. Codebases are built upon shared components wherever possible.
Think of it like a game engine. We frequently hear that by extending an engine to also build for linux, things are better compartmentalised, many bugs are found along the way and the whole engine benefits. Choice is good.
Don't know much about that either. Maybe they could have made more effort to push things upstream when they were less well established. Dunno, sorry.
They tried it as as a means to generate income to help fund the project. Nothing insideous. The company is funded by a sugar daddy, and one day that money will run out, so in the interests of being around long term, it's gotta make enough money to stay afloat.
This was one thing they tried. People didn't like it though, so they removed it as soon as it was appropriate. The data was never identifiable anyway, but I think the message was received loud and clear with that one. They're allowed to make a mistake or two.
I think they are dealing with this a lot better than many of the other not-yet-profitable companies that will sell out their users at the first opportunity without even telling them.
You say that like it's a bad thing?
Anf while I'm at it:
Here you go: http://blog.dustinkirkland.com/2016/06/howto-host-your-own-snap-store.html
I don't mean any offence here, but I wanted to set the record straight. It's a company made up of some of the most intelligent, talented, passionate and idealistic people I have ever met. Oh, and me.
I like what they stand for and what they do and I don't like seeing them framed as the root of all evil when all they ever tried to do was build something for people to use.
I like Red Hat too. What I don't like is the way the companies seem to fight like an old married couple. They have far more in common than they have differences. I wish the two companies would get over it amdcollaborate more, but I guess seeing as though they're competitors in the server space, some of the debates spill over to the desktop world.
In-fighting solves nothing, especially when there is a much more important things to be doing, like trying to limit the influence of proprietary technologies. You might not like what a particular FLOSS company is making, but if so, just don't use it. There's no reason to dislike them just because they don't give you exactly what you want, all the time.
It's not only Red Bat, there is tons of reasons not to like Canonical :
- They don't contribute much to the Linux community
Well, apart from making and releasing the most popular linux distro, rigorously tested and provided on a regular basis to you, free of charge, twice a year. But I take the point that they have historically made relatively few contributions to the kernel itself. They contribute to a lot of upstream projects (Openstack, etc), and they provide a *lot* of FLOSS tools: In addition to snappy, there's Juju, MAAS, LXD, the fan - all good stuff, especially for server side. Then they do lots of hardware certification - talking to manufacturers and making sure ubuntu just works on lots of laptops, etc. This has tangible benefits for other distros. And that's not to mention Upstart, Launchpad or Bazaar. Some things are better received than others, but to say they don't contribute to the community seems a little disingenuous.
- They have the "not invented here" philosophy, meaning that if a Linux project is successful they'll always create a concurrent project just for the sake of having one. This takes developers away from the main project and slow down the evolution of Linux (Unity, Mir, Snap...)
If Canonical had a real NIH syndrome then they would have stuck with Upstart and deviated from Debian. It didn't. When Debian moved to systemd, it followed, even though upstart actually predated systemd.
There's bzr I suppose, but they're starting to move towards git.
Launchpad is still heavily used, but it didn't quite live up to its promise. Because they chose bzr not git, github kinda won that one. Most people are moving to github now (including Canonical for some things).
At the end of the day, they're engineers and liek building stuff. If something isn't exactly what they need and they don't feel it's worth it to change it, they'll create something new. But really, since when has the open-source community been about everybody rallying behind a single thing? Surely the idea is to throw as much against the wall as you can and see what sticks?! If we all wanted to live in a monoculture, we'd been on www.gamingonios.com. In hindsight, you can look back and say everyone should have invested in technology X and not bothered with technology Y, but at the time, who knows which is right? Who's to say that Wayland has it right and Mir has it wrong? Canonical investigated Wayland, then decided against using it. They had their reasons, it was their choice. Sure helping Wayland would have moved everybody along, but if it wasn't the right tool for the job, then they weren't going to do it.
What's the problem with having both and see which is better in the long run? We all know that competition drives innovation. Looking back at the KDE/Gnome split, do we really think one of those projects is evil and the other saints? No, they have taken different approaches, and the community is richer because of it. Ideas are borrowed from different places. Codebases are built upon shared components wherever possible.
Think of it like a game engine. We frequently hear that by extending an engine to also build for linux, things are better compartmentalised, many bugs are found along the way and the whole engine benefits. Choice is good.
- They aggressively patch the packages in their distribution, making it so that bugs in the upstream project are not in Ubuntu but often other bugs are here, making it a nightmare for the original developers (I don't really know if they still do this)
Don't know much about that either. Maybe they could have made more effort to push things upstream when they were less well established. Dunno, sorry.
- They partenered with Amazon to put ads on Ubuntu
They tried it as as a means to generate income to help fund the project. Nothing insideous. The company is funded by a sugar daddy, and one day that money will run out, so in the interests of being around long term, it's gotta make enough money to stay afloat.
This was one thing they tried. People didn't like it though, so they removed it as soon as it was appropriate. The data was never identifiable anyway, but I think the message was received loud and clear with that one. They're allowed to make a mistake or two.
I think they are dealing with this a lot better than many of the other not-yet-profitable companies that will sell out their users at the first opportunity without even telling them.
- They partnered with Microsoft to have Ubuntu running on Windows
You say that like it's a bad thing?
Anf while I'm at it:
Another drawback for snappy is you cannot set up a third party repo for it (only the one from Canonical is allowed), which is probably the biggest issue for Valve.
Here you go: http://blog.dustinkirkland.com/2016/06/howto-host-your-own-snap-store.html
I don't mean any offence here, but I wanted to set the record straight. It's a company made up of some of the most intelligent, talented, passionate and idealistic people I have ever met. Oh, and me.
I like what they stand for and what they do and I don't like seeing them framed as the root of all evil when all they ever tried to do was build something for people to use.
I like Red Hat too. What I don't like is the way the companies seem to fight like an old married couple. They have far more in common than they have differences. I wish the two companies would get over it amdcollaborate more, but I guess seeing as though they're competitors in the server space, some of the debates spill over to the desktop world.
5 Likes, Who?
autonomouse Nov 24, 2016
^^ there goes my lunch hour
2 Likes, Who?
- They have the "not invented here" philosophy, meaning that if a Linux project is successful they'll always create a concurrent project just for the sake of having one. This takes developers away from the main project and slow down the evolution of Linux (Unity, Mir, Snap...)
Please forgive my poor English, but...
As far as I know Unity began its development because GNOME team didn't accept Canonical's propositions at the beginning of GNOME 3 development, so they decided to go their own way and developed their own shell corresponding to their vision. At this time GNOME 3 design was far from where it is now (now it even shares some design decision with Unity). You may dislike Unity as a shell / DE, and it's absolutely fine... but I don't think that Unity's development has slow down the evolution of Linux in any way (I use Unity and I'm happy with it: it's stable and efficient even if it's not a highly "customisable" shell / DE). :)
As far as I know Snap and Flatpack take their roots at approximatively the same time... I don't see how and why Canonical should have taken the Flatpack road at this time, Snap was first developed as an answer to Canonical's needs in phones and IoT fields before being proposed as an "agnostic package". Both Snap and Flatpack being under heavy development it's difficult to tell which one will ended being technically superior, if any of them (and you actually have the freedom to use one, both or none of them if you want). :)
- They partnered with Microsoft to have Ubuntu running on Windows
Canonical are hardly the only one with a partnership with Microsoft which is one of the "strategic alliance partners" of SUSE since 2006 (source. Most of the work (if not all of it) for bringing bash on Windows has been endorsed by Microsoft employees and I don't see how and why Canonical should have attempted to stop them as Ubuntu is a free software (if Microsoft respects the licence I don't see why they couldn't propose those functionalities for their customers). :)
Last edited by riusma on 24 November 2016 at 3:33 pm UTC
2 Likes, Who?
Here you go: http://blog.dustinkirkland.com/2016/06/howto-host-your-own-snap-store.html
I don't mean any offence here, but I wanted to set the record straight. It's a company made up of some of the most intelligent, talented, passionate and idealistic people I have ever met. Oh, and me.
I like what they stand for and what they do and I don't like seeing them framed as the root of all evil when all they ever tried to do was build something for people to use.
I like Red Hat too. What I don't like is the way the companies seem to fight like an old married couple. They have far more in common than they have differences. I wish the two companies would get over it amdcollaborate more, but I guess seeing as though they're competitors in the server space, some of the debates spill over to the desktop world.
I already acknowledged that as I mentioned above... I wasn't aware of that because they provided this solution after the community complains about Snapcraft.
I also don't think the issue with Canonical is about a NIH syndrome, I think it's more about project governance.
They want to control the big projects they rely on, their controversial CLA is another evidence of their stance on that matter.
0 Likes
Patreon
PayPal
See more from me