It's pretty basic.

The os already has to have all the code to deal with the filesystem. Offering it for image files as well is not much work, and fairly obvious.

Except it didn't have all the code to deal with .ISO and .VHD images, where did this idea that it was there but just not enabled come from?

Do you think that reading a CD is the same as reading an ISO? Do you think that the reason why all mounting software on windows used to emulate a CD/DVD/BR-drive which would then "read" the "disc" was because they just wanted to do things the hard way when there was code already in place to handle disc images?

Yes, it does. An ISO image is just the bytes that would normally be on the ISO 9660 filesystem on a CDROM. It's exactly the same, minus a bit of complexity regarding the framing of the filesystem.

VHD is quite another matter.

Any code to "emulate" a drive was a combination of two factors. 1 - windows isn't designed for adding filesystem support or device support in a reasonable third party way. 2 - support for pirates. Antipiracy schemes would rely on drive behaviors beyond filesystem content.

Good Practice != Common practice, I know many users (some would even consider them knowledgeable users) that don't do any kind of backup of any media. :( sad but true.

This, even people who should know better often don't, which argues for a better and easier/more automated system.

Going back to the original point ...

I don't find myself surprised that they hesitate to support Linux in general for the simple fact that there are a lot of different distributions and quite a few of them are structured much differently than others. While I don't personally think that this is as big a deal as most people seem to make of it, I could see someone being a bit overwhelmed by it. I just attribute this to the corporate mindset that seems so pervasive in the world today, even to the extent that it gets applied to entities that are nothing like a corporation.

Plus, the number of linux users in the world is something of an unknown quantity. How do you quantify the number of users of a system that's freely downloadable and installable on as many machines as you want? There's no good way to quantify the install base even without dealing with issues like dual-booters or people who bought a machine that came bundled with Windows because it cost $50 to $100 less that way. I know I did, my laptop and my current desktop rig are both counted as windows machines because of that, even though they both spend a much higher percentage of the time booted into linux rather than windows. I actually dread booting either of them into windows most of the time simply because I know the first thing that will happen is ... windows update, reboot, update, reboot, update, reboot. I know, if i were to start windows more often than every few weeks that probably wouldn't be quite so bad. But the fact that I can get my game on in linux much more than I used to be able to makes it possible to go weeks without messing with windows anymore. It's been said that there are lies, damn lies and statistics and I definitely find that to be true in this case. Everyone throws around the 1% figure like it's gospel and no one even knows where that number comes from. Even within the linux community we really don't have a good way to keep track of it. I've seen estimates from Microsoft themselves that put linux systems at around 4% of desktops. Still not a large percent overall, but still 4 times the amount most people seem to estimate.

On the other hand, there is nothing that would stop them from choosing a single distribution (or even a particular version of said distribution ie: Ubuntu 12.04 which is the current LTS version). I know that just playing with a few games, I've gotten most of them to run as smooth as glass on my Opensuse 12.2 system. In most cases much better than they run out of the box on Windows 7. But don't ask me about Windows 8, I won't know anything about it. I honestly can't see a reason to spend my hard-earned money on upgrading a system that, to be honest, I sometimes go to additional effort to avoid using.

As for wider support, there are ways to do that too, one I can think of off the top of my head is the mojo installer. It is especially nice because it embeds everything it needs into the installer itself, so it doesn't have to care which libraries are on your system to make the installer itself work. But, even if you don't want to get that involved, there are things like sharutils that let you make "shell archives" that run from the shell, contain a compressed data payload and allow programming logic to be embedded into them. I doubt that most linux users would care how pretty the installer is as long as it's functional. Plus, you don't need to worry nearly as much about the filesystem structure if you install the games somewhere like /opt so that they don't interfere with the rest of the system.

That said, it really wouldn't surprise me if most linux users would be satisfied by being able to simply download a binary that would use the data they can extract from the existing windows installer. Truth be told, I'd personally be fine if all they did was put a hyperlink on the game's page that pointed at where to download the binary from the publisher's website (if it's freely downloadable). I'd be appreciative of that if only because it would save me a few minutes of web searching time.

Still, the way it is now I can live with. I'm used to making things work myself and I still get a little bit of an achievement high when I get something running like clockwork. But, it would be nice to be able to get the binary to run the game at the same place I get the data set to run it on. Even if it were unsupported.

So, to answer the initial question. Yes, I think the linux ecosystem is ready for them. The real question is, are they ready for it?

Well, if you mean with "corporate mindset" a focussed mindset which wants/supports/allows the creation of reasonable supportable platform, than, yes, I want more "corporate mindsets".

In contrast to what you are (indirect) indicate, that there is a polarization between
"commercial-corporate-propritary-mindset" and "open-source-community-mindset", I'm pretty sure that open-source is very fine and compatible with commercial & platform models.
E.g. the OSS Android platform, as soon as someone created a (more or less) stable platform, the linux kernel was a success in a commercial market with many companies, they love android. Or see the mozilla foundation and their premium product firefox, a success, widely commercial used and supported, "despite" being community driven and open source. Another example for a commercial successful and non-fragmented community driven project is wikipedia. So, no you are wrong, the reasons for the problems of the linux ecosystem in software and game distribution are not mindset of someone else, but their own mindset. To credit is only a encrusted mindset deeply rooted in the last century and a inherent culture of unix and fragmented distro infrastructure.

So, don't mix in things like the innocent OSS and community integrated processes , they work fine & are no problem for gog.

And you mean with "ready", the usual "We don't move, you have to play after our rules."-readiness?

PS: also, gog is already pretty community driven and focussed! It's especially off to call GOG a faceless and not consumer-careing company :(

PPS: you mentioned the mojo installer by Ryan Gordon, a great programmer and linux gaming supporter. 2 facts: he was also at Loki where the loki setup was created, more or less similiar concept as mojo, a installer for all linux variants. Loki tried his best to make it robust and introperable, but as usual linux distros and community was caring not at all for loki setup... ultimately they broke all packages of them on several levels. :( Also, Ryan was so annoyed by the hostility and conservatism of the linux distros when he tried to improve the linux gaming with Interview with Ryan C. Gordon about Linux Gaming

Ubuntu Linux is another such product. So are Red Hat Enterprise Linux/Fedora, SUSE Linux and Open SUSE and plenty of others.

These tiny fragments are only limited or not all commercial succesful, accepted and supported: almost no commercial software, adobe flash left, HW-manufactures hate supporting them, even GOG says they are too hard to support, loki software died (while targeting this market), etc

The corporate mindset, probably not the best way I could have phrased that. It was more a reference to judging the value of something based only on market share, profit margins and other stuff like that. While a lot of the participation in open source is in fact done by corporations, a lot of it is also done by non-corporate entites and individuals as well. We all have our metrics and it seems the ones that are bandied about the most are the ones that don't necessarily translate into success or failure for a lot of the community driven projects.

I also did not mean to imply that GOG is big, faceless corporation as I also recognize the more community-minded approach they take to things. If that were my belief, I would not have spent even a single dollar with them. Again, it was a generalization of the all-or-nothing approach that people in general tend to take, not an indictment of anyone in particular.

I also realize that GOG wants to make the experience for their customers as seamless as possible which I also appreciate. I just don't see that as incompatible with things as they sit. Whenever you're going to support something, you have to place limits on what you'll support. Going completely open-ended is suicide, hence the suggestion to specify a supported configuration even down to a specific point release if that's what you feel you need to do. What people do outside of that is their own issue, the same as right now when I buy the version that's guaranteed to work on Windows and I take it and put it on my linux installation and make it work without their help.

As to the reference to mojo, yes, I'm also aware of the history there. However, that was simply an example of a tool that could be quite useful for distribution-independent packaging. I've met Mr. Gordon and I very much admire his talents and skills. I also remember being a bit disappointed that no one jumped on the idea of fatelf because that could simplify and streamline a lot of things. I don't know if you've ever had a chance to see any of his presentations on open source game development tools, but if you haven't there are some up on Youtube (and linked from his site on icculus.org for that matter) and they address a lot of the things that people call "issues".

I really think you're reading between lines that aren't even there for some of what I said, however. Yes, you have your groups of fiercely independent people on both sides of the issue and yes, you have your rabid fan boys on both sides as well. However, since I no longer live in the middle of the war-zone (I stopped dealing directly with all of those issues in 2004, now I just follow it with some interest, but it has no net effect on my daily life), I've found that I'm sort of ambivalent about all of it, I just do what works for me. So if you're looking for rabid fanboiism from me, I'm afraid you'll be disappointed. Life has taught me to be a bit more pragmatic in the way I approach things, however I'm definitely not a PR or Marketing guy, so I appreciate that I might not always phrase things the best possible way. The point was definitely not to be confrontational or cast blame, although it seems like people are ready to read into that.

And for the record. When I say ready, I mean "You offer products and I'll buy what I'm interested in buying." "You offer your terms and I'll decide if it's something I can live with." ... and ... "If it's not something I can live with, then obviously I won't be buying." Speaking for myself, which is the only person I'm actually qualified to speak for anyhow, I don't see that as bad.

As much as I hate DRM, I would rather 'rent' games I can play then not play them at all. Steam is getting all my money at the moment, and with so many games in the pipeline and the convenience of Steam, when GOG finally get their act together they may wonder why it was so hard for them to just let us download a tarball or the packages that the developer had already created.

Current Linux users do use Linux after all, we are not a retarded bunch that need a big glowing button just to install something, we can manage to get by without the need for GOG to hold our hands every step of the way, just slap a big beta or testing graphic on the Linux version for now with no support available & use the windows version if problems arise etc, until GOG are happy with their own client/package management at which point newbies to Linux can get in on the action. At least that way they would not have lost the momentum, they should have announced Linux at that BS Press Conference the other month.

some update:
State of readiness of linux for steam... scary, the issues are legion :/
https://github.com/ValveSoftware/steam-for-linux/issues?page=1&state=open

I'm really curious if valve will be able to sort this mess out....

Please post the bugzillas for Steam for Win and Mac OS X so we are able to compare. Thanks.

already looked but didn't found a windows or macos steam client tracker ... think is not public. Would be very interesting indeed :)