Alexandre Julliard (English)
Introduction
Wine has, for the last fifteen years, been under development with the dearing goal of allowing you to run your Windows applications in a Unix-like environment.
For a better insight on the path taken fifteen years ago to reach its current usable state, we've talked with the project's creator: Alexandre Julliard.
This interview is also available in Norwegian.
Introducing Wine
Could you please introduce yourself for our readers?
My name is Alexandre Julliard, I'm 38, I'm the maintainer of the Wine project, and have been for 14 years now. In my day job I'm the CTO of Codeweavers, where I mostly work on Wine too.
For the portion of our readers who haven't heard of Wine, could you present the software?
Wine is a reimplementation of the Windows libraries on top of Unix; it lets you run existing Windows programs on a Unix system, mostly Linux but also Mac OS, Solaris, FreeBSD. Unlike virtual machine solutions, since Wine is a complete reimplementation it doesn't use any Microsoft libraries and doesn't require a Microsoft Windows license.
Do you only work with Wine? Is it possible to earn a living working with open-source?
Yes, essentially all of my work is Wine development, and all the code I write is released as open source; and that's true for all of the Codeweavers developers. So it's clearly possible to earn a living with open source, even if it's not always the easiest way...
You also work at Codeweavers, who got their own commercial version of Wine called Crossover. What would you recommend new Linux users to get? Crossover or Wine?
For new users, Crossover is probably the best choice, since it's easier to use. Wine has made a lot of progress in that area, but it's still more complicated, mainly because it isn't restricted to supporting a small subset of applications.
Do you use Wine yourself?
No, the irony is that at this point I have no real use for Wine myself, there are Linux applications for everything I need to do. So I only ever run Windows applications in order to test Wine, not to actually use them.
After all this time creating a compatibility layer for Windows applications, what do you think of Windows?
The quality of Windows is really bad; most of the APIs are badly designed, full of bugs and inconsistent behaviours, and cumbersome to use. The documentation is poor; it has improved a bit after the legal trouble that Microsoft has been through, but it's still far from good.
Most Windows applications are also full of bugs, mainly because Windows tries hard to hide application bugs to give the impression that everything is working fine, even when it isn't. Running an application under Wine is in fact a good way to find bugs in it; of course we then have to add workarounds to hide the bugs, like Windows does.
In the beginning...
What gave you the idea to start a project like Wine?
The project was started back in 1993, shortly after the birth of Linux. At that point Linux was showing great promise as an OS, but of course there were no end-user applications; so with a group of early Linux users, we got together and decided to see if it would be possible to make the myriad of available Windows applications run somehow. We found that it was possible, even if it has been a lot more work than expected at the beginning.
Wine stands for, ironically, Wine Is Not an Emulator. Why did you pick such a name for your project?
The name is of course a joke, along the lines of Gnu's Not Unix, but it also makes the point that there is no hardware or CPU emulation going on in Wine. Windows programs run under Wine exactly as they would under Windows, with no performance impact. This means that unlike with virtual machine solutions, applications can run under Wine as fast (and in some cases faster) than under Windows.
What kept you motivated? Did you really expect to get as far as you have today?
Part of the motivation is of course the challenge of doing something that most people would have considered impossible. For me another motivation is working with all the contributors, receiving new patches and new ideas every day and having to keep up with that. And of course the fact that many users are able to do useful things with our work (even if we mostly hear from users when things break...)
I can't say that when starting this I expected to still be working on it 15 years later. I had always believed that we could get as far as we are now, but I probably didn't expect that it would take that long. But as long as it's fun, I don't see any reason to stop...
How many people are currently working on the Wine project?
There are probably around 50 active developers at the moment. Over the course of the project there have been over 1000 contributors altogether.
You are employed by Codeweavers, who also sponsors the Wine project. How does this cooperation work out? What does the sponsorship from Codeweavers allow you to do now which you couldn't do before?
The cooperation is mostly that Codeweavers pays Wine developers to work on improving Wine; this way they can hack Wine all day long instead of having to do it only in their spare time. This means of course that Wine can progress a lot faster. We now have about ten full-time Wine developers working at Codeweavers.
One thing I'm careful about is to not let Codeweavers have any control over my decisions as Wine maintainer. I view my work at Codeweavers and my work as Wine maintainer as completely separate, and my decisions as maintainer are based purely on what I consider best for Wine, even if it's sometimes not what would be best for Codeweavers. In general of course the interests are aligned, since it's in Codeweavers best interest that Wine become as good as possible.
If I've understood this correctly, Crossover includes some proprietary code which makes it run some applications better and faster than regular Wine. What kind of code is this, and is it code that eventually will end up in Wine as open-source?
Actually there's no proprietary Wine code in Crossover, all the Wine code is LGPL. The only proprietary part is the GUI that manages the installation of applications. The only difference between Crossover Wine and the public Wine is that we have a few hacks that are not suitable for inclusion in the main tree, mostly because they help one app but break another; but we try to keep the code bases as close as possible. There is certainly nothing magical about Crossover that cannot be done with plain Wine, it just makes things easier.
Looking ahead
Do you think Wine will play an important role in the future?
My hope is that someday Wine will stop playing a role, because Windows will be dead and nobody will need Wine anymore... However, that is far in the future, and in the meantime I do think Wine has a very important role to play, by making it easier for people to start using alternative to Windows without having to give up all their applications at the same time, which for most people is not an option.
Jon Parshall, COO at Codeweavers, once told me that Crossover has about 150.000 users, and that the number of Wine users is about eight times that. Are you proud of the popularity of the project? Do you think the number of users will continue to grow?
Certainly, it's very gratifying to hear that the work we do is useful to so many people, that's a big motivation to continue. I definitely expect the number of users to continue to grow, as alternative operating systems like Linux and Mac OS become more popular.
This might be a bit overdue, but congratulations on your stable 1.0 release of Wine. What has 1.0 ment for the development?
For the development, 1.0 marks the point where all the basic infrastructure is in place, and provides a solid foundation for the future. The release process itself, with the code freeze and the testing period, has been an opportunity for us to fix a number of long-standing bugs, and I'm fairly happy with the end result.
For users, 1.0 is signalling that Wine is now at a point where it can actually be useful. We are now out of the "promising but not working" state that we've been in for so long. Of course there are still bugs, and applications that do not work, so our job is not quite done yet, 1.0 is only a beginning.
Do you ever think Wine will "catch-up" with Windows' features? To clarify, do you ever think Wine will get to a point where it will support all the features of Windows X by the time Windows Y gets out?
In general we are not very concerned about Windows releases, since we don't use any part of Windows itself. What matters for Wine is what features the applications use, and new features only trickle slowly into applications since they need to run on older Windows versions for a long time after a new version is released. So in general we have plenty of time to catch up with new releases.
Do you ever think Windows will lose enough users that Windows eventually will need something as Wine, to run Linux apps in Windows?
In a sense that's already happening, there are many applications that have been developed for Unix and then ported to Windows, and there are tools like Cygwin to make the process easier. It's not a strict equivalent to Wine however, since most Unix applications come with source, so it's a lot easier to port them by recompiling than by trying to run actual Linux binaries on Windows.
Google has been implementing Wine in some of their products for Linux-support. Do you see the use of wine in this fashion increase? Is this a good or bad thing?
That's definitely a good thing, Wine has improved a lot as a result of Google's involvement. We are certainly seeing more companies using Wine this way, it's a good way to quickly release a Linux version of a product to test the market and see if it's worth investing into a full Linux port. A good example is the Chrome browser: using Wine, Codeweavers was able to release a Linux version of the browser within days of the Windows release, while the native Linux port is still months away.
Is there anything else you can think of that hasn't already been covered in this interview?
No, I think that's all. Thank you for reading this far!
We thank Alexandre for his time and patience and wish him the best of luck on his future endeavours.
