What Subset Games (FTL, Into the Breach) think of Valve's Steam Play

I'd be curious to see if any games decide to bootstrap DXVK or similar without wine directly into their game / engine. I am curious if it would make porting easier, or even be a functional idea to porting to Linux.

Most people on Linux are pretty BIG on maintaining as much control as possible -- programming is no different -- why license and deploy proprietary dependencies when a in-house alternative can be a better fit.

In web programming it's pretty much the same -- why include 100 different javascript libraries with conflicting dependencies written by some dude across the planet and unsupported when you can just go minimal and maintain as much project control and independence.

A good example is there used to be a WordPress plugin -- NextGEN Gallery, that got sold and bought, and come version 2.0, the JSON API completely changed thereby ruining weeks of work due to API version incompatibility.

These are the kinds of problems you get when you rely on other people's software. Being Store Agnostic is a pretty smart move.

/rant


Last edited by ElectricPrism on 18 September 2018 at 12:29 am UTC

QuoteOn top of that, our contributor Cheeseness also posted his own Cheese Talks article on his blog about giving some of his thoughts.

Wow! I just read the article...
That is not an article; THAT IS A BOOK!... I am waiting for the movie adaptation.

The Cheeseness article is interesting. Lots of good information and some interesting (and worrisome) perspectives.

But I think he is a bit too cavalier in dismissing the issue of sheer quantity of users. He sets out the possible good things from Steamplay/Proton like this

QuoteAssuming that more developers supporting Linux natively, and treating Linux users as equally valid/valued customers are the most positive outcomes, let's consider a few of the possible ways that something like Proton could be used to leverage that.

Note that more Linux users is not in there. I'm pretty sure that one of Valve's major objectives in all this is to facilitate more people playing games on Linux/SteamOS, one way or another--whether through adopting Linux as their PC desktop or via new Steam Machines or in some other way as yet undreamed of, the point is to remove a barrier to adoption. Cheeseness doesn't so much dismiss that idea as simply not consider it a desirable objective.

QuoteBeyond this, chasing a higher userbase is a flawed goal in and of itself. It is far better to focus on increasing the quality, accessibility, and robustness of experiences, and in that way create a more attractive platform, than work to artificially inflate the userbase in the hopes of hitting some kind of critical mass. The alternative results in skewed priorities that revolve around things that don't really benefit any users or the platform in tangible ways.

His ideal it seems would be a Linux that continually improved in quality and which was treated as a first-class OS citizen, but without actually growing much in number of users so as to stay a congenial group of communities. Unfortunately, I think that ideal is a bit naive. Numbers count. If you don't have the sales you won't be treated as a first-class citizen. Network effects are important, and we also live in a time dominated by large corporations with weak antitrust enforcement where big commercial concerns are predatory on small. Open source acts as something of a hedge against this, and has allowed Linux to survive in a hostile environment where an equivalent proprietary offering would have died. But ultimately, many of the objectives Cheeseness wants cannot happen without a bigger slice of the pie. Ignoring Proton's potential impact on growth in Linux use ignores the main issue and makes the rest of his analysis a bit moot.


Last edited by Purple Library Guy on 18 September 2018 at 3:10 am UTC

Thanks Liam, I've been curious about Subset's plan for Into the Breach since Proton was announced. I'm waiting for Steam Play to come out of beta before I try this, but I'm pretty tempted to just go ahead and get it already.

I also enjoyed the Cheese article.

Quoting: Purple Library GuyHis ideal it seems would be a Linux that continually improved in quality and which was treated as a first-class OS citizen, but without actually growing much in number of users so as to stay a congenial group of communities.

That's not what I perceived from your quoted text. The idea isn't that we shouldn't grow much, but that we shouldn't grow too fast. Notice, in particular, "work to artificially inflate the userbase" and "create a more attractive platform" (emphasis mine of course). I think his point is that if you just invest on putting a half-baked product into more people's hands to inflate your numbers, you'll ultimately shoot yourself in the foot by showing potential returning customers a bad experience. Whereas by prioritising your product's quality, you may not get as fast a short-term adoption as with an equivalent marketing budget, but your retention rate will be much better and favor steady growth.

We don't want everyone to use Linux so we can make it better.
We want to make Linux so good everyone wants to use it.

Quoting: Salvatos

Quoting: Purple Library GuyHis ideal it seems would be a Linux that continually improved in quality and which was treated as a first-class OS citizen, but without actually growing much in number of users so as to stay a congenial group of communities.

That's not what I perceived from your quoted text. The idea isn't that we shouldn't grow much, but that we shouldn't grow too fast. Notice, in particular, "work to artificially inflate the userbase" and "create a more attractive platform" (emphasis mine of course). I think his point is that if you just invest on putting a half-baked product into more people's hands to inflate your numbers, you'll ultimately shoot yourself in the foot by showing potential returning customers a bad experience. Whereas by prioritising your product's quality, you may not get as fast a short-term adoption as with an equivalent marketing budget, but your retention rate will be much better and favor steady growth.

We don't want everyone to use Linux so we can make it better.
We want to make Linux so good everyone wants to use it.

Yeah, maybe, but it's still a naive perspective. It implicitly buys into the idea that every platform has, or will ultimately have, the number of users it in some sense deserves. Either a belief in the efficient working of markets or of the efficient workings of the "marketplace of ideas". Neither of these work in real life.

Linux doesn't have few users because it's poor quality. For one thing, it's quite good quality. But to the extent that there are de facto quality problems in Linux, the causality runs the other way--graphics and other device drivers, for instance, are poor or lag in time because Linux has few users; if it were a platform big enough that vendors had to support it, those problems would vanish like the morning dew. But that's not the problem. Linux has few users because other platforms have locked up preinstalls, so they come with the computers people purchase. It has not grown because of various engineered barriers to entry--advertising budgets, engineered incompatibilities, payola-dominated sales channels, et cetera et cetera. To think all we have to do is build it and they will come as we contemplate the wonders of the perfect platform under the shade of the bodhi tree is unrealistic. We are the underdog here, the big boys are fighting with every dirty trick going and have been since the 90s, and Cheeseness is saying taking one hand out from behind our back would be "artificially" inflating the user base. No, sorry, the whole basis of who uses what OS is "artificial" by Cheeseness' standards and has been since Bill bought QDos, or before.

I was trying not to come down too hard on that perspective because in general I really respect Cheeseness, but I just can't. It's not just wrong, it's not even just dumb, it's dangerous. It's like we're headed to a gunfight and Cheeseness is saying "Put that thing down and bring this pair of garden shears instead" because for some reason he thinks it's about trimming a hedge. You do not defend yourself against the mafia by showing the world that your garden is more beautiful than theirs. I'm in favour of the garden being beautiful, don't get me wrong, it's just not a good reason not to try to win.


Last edited by Purple Library Guy on 18 September 2018 at 6:56 am UTC

A shame that their "storefront agnostic" Take doesn't include itch.io ^^
Nevertheless, intersting to know and does for sure make them more sympathic!

I agree with Purple Library Guy 100%. I also thought of the "if you build it they will come" phrase as I was reading Cheeseness's article and I think he is wrong there. They will not come because they're already somewhere else and they _see_ no reason to move.

I think Linux is already good enough and has been for years now. It has not been enough. The general population does not care about Linux's advantages and they just want to use what they know and what will run the software they know (how many times have I heard, yeah, ok, but does it run <insert arbitrary piece of Windows-only software here>, oh, it doesn't, but I need it) and to some extent what came installed on their computer - even if they themselves say "Windows 8 is awful I hate it".

The situation can only be solved by marketing and easy access to hardware with Linux pre-installed. The perception of Linux has to change. People have to learn they can do all they need to do with it. This is exactly the thing Apple had been doing for years - telling people they sell fully-functional computers which have all the software people will ever need. I remember when Macs were in the perceptual position Linux is in now some 15 years ago - seen as expensive, useless toys for a very specific few. Linux isn't seen as expensive ofc, but the rest stands. Apple put millions into marketing and more millions into software.

Someone needs to do that for Linux if it's to succeed.

I believe we don't need all people who use Linux to subscribe to the philosophy behind it. We do, however, need the numbers so various entities that create hardware and software find it worthwhile for their business to support Linux properly. I believe both proprietary and open-source software have their place and it would not be good if either of them disappeared.

I'd say a Linux market share of about 20% would do the computer landscape a world of good.

QuoteIt's not a final solution for Linux for us because we like to be storefront agnostic as well as OS agnostic.

This was always my main concern with Proton (especially as many act like Wine doesn't exist anymore, oddly enough). Proton is Steam client only and thus it's not ideal for general use or for the long-term.

"Proton is Steam client only and thus it's not ideal for general use or for the long-term."

Proton is not Steam only or did i miss here something, you can use proton without steam, but most of the time it doesn't make sense to do so, because wine + dxvk,...... does the same.
And who said that others are not allowed to use proton in there projects, i couldn't find anything which says only steam is allowed to use proton. But maybe i am totally wrong here :)

Quoting: toojaysThanks Liam, I've been curious about Subset's plan for Into the Breach since Proton was announced. I'm waiting for Steam Play to come out of beta before I try this, but I'm pretty tempted to just go ahead and get it already.

I e-mailed Subset with some questions as soon as I saw that Into the Breach had been whitelisted for Steam Play. I quickly got a very nice and detailed response - they're good at communication.

What an incredible mindset! I just wish more developers thought like this and tried to do things this way as much as possible.

Quoting: Guest(I know people will knee-jerk and think I'm trying to say there shouldn't be more users, but no, I'm saying there shouldn't be more for the sake of more - there should be more in tandem with technical efforts to properly handle more).

I understand what you're saying and it's defensible. But I have to say I do think I disagree even with what you're actually saying. There should be more for the sake of more. Heck, I don't even like having to say that, but it's the brutal reality. From both a technical and social/political perspective, it's really important that there simply be more Linux users.

From a technical perspective, Linux will never get the love and support it needs at 1-2%. I would be willing to bet that even now, there's a huge difference in how well Linux supports primarily desktop/gaming hardware and how well it supports server-oriented hardware. Server-oriented hardware Just Works on Linux, you can bet, because Linux has a huge server marketshare. If there was a thing like Flash in the server space, it would run on Linux or it would disappear. Server-oriented software supports Linux or it loses to offerings that do. Same same with cloud stuff and mobile stuff and IoT stuff and so on. But stuff that's needed for the desktop or for gaming languishes, simply because there are too few of us.

The software end starts to shade into the social/political side. If Linux was at 20%, there would be no debate or worry about whether Vulkan would catch on or whether Microsoft would manage to lock everyone into DX12; nobody would be willing to lose that market by picking DX12 over Vulkan. If Linux was at 20%, Apple might even be playing ball instead of rolling their own. But there are other sorts of things--anyone remember libdvdcss? Think nobody will ever come up with a scheme like that again and leave Linux out in the cold? I think it's impossible to make such schemes bulletproof, but they can and have made it illegal to circumvent them. And they can shut Linux out of them because Linux users are few; it is politically possible to systematically marginalize Linux because Linux has few users, making it politically vulnerable.

There's another technical perspective. Much software on Linux, and Linux itself and its graphical environments and such, are Free Software. That means more developers using Linux implies more software development of Linux. Now, Linux users skew technical currently, and much Linux development happens because of non-desktop Linux anyway. So multiplying the Linux desktop and gaming user base by 10 wouldn't multiply the developer base by 10, or anything like it. But I'm sure it would at least double the desktop/gaming relevant development. A lot of what's wrong with Linux IMO is simply that there aren't quite enough developers to get the software really solid and competitive. A massive increase in the user base would bring a big impetus to polish up the Linux desktop and its open source desktop software ecosystem. Gnome, KDE, Cinnamon, Mate, LibreOffice, the Gimp, etc. would benefit hugely. Wine itself--it's often been said that Wine can never become really good and reliable because Windows is a moving target. Recently with a bunch more development on Wine and Wine-related stuff, it seems to be turning out that if you have more developers you can move significantly faster than Windows and its "moving target" status don't matter so much any more.

So yes, there should be more for the sake of more. More users is not just a valid target to aim for, it is one of if not the most important target to aim for. Most other remaining Linux problems cannot really be solved without, but will shrink to minor obstacles or disappear entirely with, a larger user base.

Some interesting thoughts here, people. Thanks for the kind words. I had hoped to write a little summary for GOL readers as I usually do, but unfortunately, I've been absolutely swamped (with working on games that support Linux directly, for whatever that's worth ^_^).


I see some concerns about my perspective that it's healthier for the Linux userbase to grow organically and slowly. It's fair to have different perspectives, but I'm disappointed by the assumption that I'm being cavalier rather than considered here.

FWIW, I consider the growth we've had over the past 5 years to be notable and positive. By many measures, that in itself can be considered a "success." I look forward to a world where Linux is the dominant desktop platform across the world, but I'm also in no rush to get there. I feel that there's more room for a positive future from grass roots driven growth (which I spend a huge amount of effort on) than most alternatives, and the other nice thing is that barring any crazy shifts in copyright perception/law that threaten F/OSS licences, Linux, BSD, and other F/OSS licences will by nature outlast their proprietary competitors.

Beyond that, unsustainable growth in any domain is typically unhealthy. Bubbles suck. We don't need one here, and we certainly shouldn't see one as essential to the survival of our platform - if any perception in this is naively dangerous, it's that one. As I said in the article, big influxes really aren't a great thing to specifically strive for.

In my eyes, it's naive to view this metaphorical race as a sprint that we can afford to exhaust ourselves in instead of a marathon that we can't.


Also, "who're" is totally an acceptable contraction :D

Quoting: Werner"Proton is Steam client only and thus it's not ideal for general use or for the long-term."

Proton is not Steam only or did i miss here something, you can use proton without steam, but most of the time it doesn't make sense to do so, because wine + dxvk,...... does the same.
And who said that others are not allowed to use proton in there projects, i couldn't find anything which says only steam is allowed to use proton. But maybe i am totally wrong here :)

It is possible to build Proton yourself and/or run it outside of Steam. Whether distros decide to package it, and whether other distribution platforms decide to ship it remains to be seen.

The same goes for the "Steam Runtime," and maybe here's another point that the ups and downs of names come into play. If the SR had been called something else, would more distros have shipped their own metapackages that made distro-specific versions of those libraries available? I often wonder whether that was the initial intention behind the SR and Valve's approach toward allowing/encouraging distro-specific packaging of the Steam client.


Last edited by Cheeseness on 19 September 2018 at 7:24 am UTC

Quoting: CheesenessSome interesting thoughts here, people. Thanks for the kind words. I had hoped to write a little summary for GOL readers as I usually do, but unfortunately, I've been absolutely swamped (with working on games that support Linux directly, for whatever that's worth ^_^).

That's worth much to me. (Just by the way and totally unrelated, how I would love to have another insult duel game next to Before the Storm...)