Canonical is looking into selling proprietary software like Adobe's Photoshop and Apple's iTunes within its distribution, Ubuntu. This would undoubtedly be helpful for certain end users wanting to switch to Linux, but is it good for free software in the long run?
A Question of World View
All of this does once again raise the issue of proprietary software in the Linux space. Naturally, Canonical is free to do whatever they like to their operating system, but what effects might it have on the rest of us?
There’s no doubt that if Linux could seamlessly run those “essential” applications like Photoshop, it would become a much more attractive operating system to a majority of the existing market.
But do we care? Well, it depends on your world view.
Free software exists because some see it as a better way. We use it because it provides us with specific freedoms, we trust it, we can hack it, we like it. This is what Linux and free software has to offer. If you don’t want to use it, then don’t. If you prefer to use Windows and proprietary applications, then use those.
The Linux desktop got to the way it is today thanks to free software and the open development model. The popularity of Linux on the server and super computer market has not been a result of proprietary software, but rather outstanding free software. It did so by providing a better option over existing Unix systems, with which it has much more in common.
For companies like Canonical however, it’s not about free software but rather gaining market share and ultimately making money. Ensuring that popular Windows software can run on Ubuntu gives it a better chance of securing more market share, regardless of the philosophical arguments.
You see, the desktop is a different ball game. It’s less about services and more about front-end software, most of which is designed specifically for Windows. Linux is already at a massive disadvantage.
However, supporting closed source software is tricky, especially when it’s running on a system other than the one it was designed for. Proprietary video drivers enable users to have decent 3D acceleration, but they also introduce numerous issues of their own. Due to their closed source nature, these issues can’t be resolved by the free software community. By introducing support for proprietary software, Canonical might be digging a hole too deep for itself to climb out of. If users are paying for software, they will expect it to work properly. How can Canonical ensure this happens when they can’t see the code base? Would a company like Adobe be willing to fix issues in a timely fashion? Photoshop CS under Codeweavers is only supported at a Bronze level which means that it will install and run, and can “accomplish some portion of their fundamental mission,” but expect a bumpy ride.
And anyway, this whole idea simply goes against the grain. Supporting proprietary software in Linux, are you nuts? Wine has been in development for 15 years and yet it still can only install a small number of applications well.
It’s a tough call, but obviously one that Canonical is willing to take. It’s true that by supporting proprietary applications, Ubuntu would become a more popular and “usable” for Windows users.
Perhaps there is room for both world views. After all, each distribution has the right to its own goals and agendas, Canonical and Ubuntu just as much as any other. Those who choose to run free software can continue to do so, while those who need Linux to run their Windows software can find it.
Make Better Free Software
Don’t get me wrong, I think the Wine project is great. We should have open implementations of all languages and systems. There is a difference however, between having a project which enables some Windows applications support, and going all out to encourage proprietary software on the Linux desktop (I’m not suggesting that Canonical is doing this, either).
For Linux to be able to entice users away from Microsoft, does it really need to support their Windows only applications? Why do users prefer Photoshop over GIMP? If it’s a user interface issue, then skin it to look like Photoshop. If it’s shortcut keys, then map them too. If it’s features, then start hacking.
Instead of putting effort into getting proprietary software running, a game at which Linux will be forever playing catch up, can’t we just work together to improve existing Free software?
Yes, Free software takes time to develop into a powerful alternative. It generally doesn’t have money thrown at it like proprietary applications, and so develops at a much slower pace. Those applications which do develop quickly are often backed by large corporations which have employees developing the software, OpenOffice.org for instance. It might also not be possible to do this, but if we could address the reasons why users simply “must have” their Photoshop type applications over free alternatives, perhaps we could create a viable enough alternative over time.
Ubuntu has supported proprietary video drivers. While this has been extremely helpful to those who want to make use of the drivers, has it been detrimental to the cause of free drivers? Could all the commercial Linux distros get together and work with a company like NVIDIA to develop a decent open source driver? Perhaps they could, but it probably wouldn’t work. Linux needs to become a big enough player in the Desktop market so that proprietary software companies simply have to release free software in order to be successful on the platform. This is happening, slowly. To some degree, it’s a chicken and egg problem. How do you become powerful enough on the desktop so that you can influence companies like these, without first using their proprietary applications to gain popularity?
The question is, by supporting proprietary software in the operating system and kernel are we less likely to see a free version in the long run? And if we never see a free version, do we care? Will we get to a point where closed source is just as acceptable as free software? If proprietary Windows applications can run on Linux, how much motivation would there be to develop outstanding free applications? Will running proprietary software under Linux keep the development focus squarely away from free software?
Does the integration of proprietary software provide a short term benefit and a long term handicap?
A Fork in the Road
Again, essentially it’s all a matter of world view.
Are we pushing freedom, or are we trying to create an operating system which can compete with Windows? If its the latter, then supporting Windows applications might be important for those “must have” applications which prevent users from migrating. If it’s the former however, we could be doing ourselves a disservice by putting energies into running proprietary applications rather than developing free competing products.
Are we trying to convert the world and steal as much market share as possible and therefore need to integrate Windows software to provide us a faster rate of success? Or are we just making and using the best free software possible and those who choose to join us along the way are welcome?
Why should we care that proprietary Windows software doesn’t run on Linux? Can’t we just say, “Use free software, or don’t. The choice is yours.” Well, that might fine when you’re content within yourself to just use free software, but when you’re an evangelist out to convert people to Linux, or if you’re a company trying to gain market share, it doesn’t always work. When someone won’t use Linux because “Photoshop” doesn’t run, somewhere a little penguin gasps its final breath and rolls over dead. Sure, we would love them to use GIMP instead but in the end it’s their choice. This is what Free software has to offer, all of these riches. Unfortunately, you can’t bring that piece with you, now what will it be?
I don’t have the answer. Personally I’d love to just say, “Hey, this is Free software, this is Linux. Use it or not. Up to you,” and not care whether they use it or not. But I do care. I care because I want Linux to succeed on the desktop and as long as the proprietary, vendor lock-in model prevails, it’s not that simple.
No doubt, this move by Canonical is beneficial to a number of users. It will also further push Ubuntu into the realm of Linux newbies and help it attract more Windows users over. Somewhere along the line however, should there be a migration path off proprietary software? Surely it’s better to have users on free software in the long run, which benefits everyone.
There are three choices. Either, we convert users over to Free software entirely. Or we convert them to free software plus some “essential” proprietary applications, or we don’t convert them at all. So what’s worse? Having users on Linux and running their proprietary Windows applications, or not on Linux at all?