A Fantastic Interview with Wine's Man Alexandre Julliard

By Fernando Ribeiro Corrêa and Luis Strano

Originally published at http://www.olinux.com.br/interviews/10/en

Olinux: Tell us about your background?

Julliard: I've studied computer science at the Swiss Federal Institute of Technology. I've been working in software development for the last 10 years, mostly in embedded software (routers, payphones, etc.), and I've been working as a hobby on the Wine project since its start in 1993.

Olinux: Explain shortly the Wine's History and Organization? What�s its main purpose?

Julliard: Wine started in 1993, and I've been the maintainer since 1994. The main purpose is to allow running Windows binaries under Linux (and other Unix systems), as well as to allow people to recompile their existing Windows source code to build native Unix binaries without having to modify the existing code. The organisation is along the "benevolent dictatorship" model; all the changes are submitted to me, and I have the final say on what gets or doesn't get into the source tree.

Olinux: What is your main work in real life? Are you full time worker for wine?

Julliard: I'm working for the company CodeWeavers; part of my job is to work on Wine, the other part is to work on custom development jobs for customers. At the moment this includes doing Wine development on behalf of Corel, so yes I'm mostly working full-time on Wine, at least for now.

Olinux: What you do in Wine?

Julliard: A little bit of everything... my speciality is mostly in the low-level areas like memory management, threading, etc. Plus of course studying incoming patches and merging them into the source tree.

OLinux: Are there business companies or any other type of organization that finances Wine development? How did Corel investment and support change wine�s development?

Julliard: There is no organisation putting funds directly into Wine, but there are companies having people work on Wine; of course the major contributor in this area is Corel, with a lot of people working full-time on Wine. Corel's involvment has done a lot for Wine, mainly by addressing areas of the code that had been a bit neglected until then, and also by doing extensive testing of all their office suite under Wine and fixing all the problems they encountered. Wine is definitely much better now than it would have been without Corel's help.

Olinux: Everyone in Wine's staff is a volunteer?

Julliard: No, since there are people at Corel working on it as part of their job.

Olinux: How people are organized and what are tools are used to control the results of the work being done in different projects and parts of the world?

Julliard: People communicate through mailing lists, and all the code is in a CVS repository accessible read-only by anybody. All the changes are reviewed by me, and then stored in the CVS tree where anybody can test them.

Olinux: How many people are working for wine nowadays? Are you satisfied with the results?

Julliard: It's hard to say exactly how many people are working at any given moment, but I'd estimate it at 30-40 active developers. I think the results are very impressive, particularly when compared to the amount of resources Microsoft is putting into Windows.

Olinux: Are there anything that can be done to improve productivity?

Julliard: Having more people of course; better documentation of the Windows API would help a lot but I don't think there is much to hope for from Microsoft in this respect.

Olinux: Describe the active projects and their core activities? How are the tasks divided and co-ordinated in terms of content and staff?

Julliard: There is no formal organisation in sub-projects. Everybody works on what he wants to, and coordination is done through the mailing list and CVS tree.

Olinux: How would you answer Bill Gates statements that Linux wasn't any danger to Microsoft monopoly because it was decentralized and uncoordinated?

Julliard: Linux is a danger to Microsoft precisely because it is decentralized. There isn't one company that you can buy or put out of business, so the usual Microsoft tactics do not work against Linux.

Olinux: Do you see any problem regarding quality of software development and maintenance due associated to the volunteer work?

Julliard: I think the quality is usually better with volunteer work, since people take the time to do things the right way, and also take more pride in their work since it is published for the whole world to see. The main problem with volunteer work is that the parts of the code that are less fun to write get less attention; this is why it is a good thing to have both volunteer and paid developers on the same project.

Olinux: How do you describe wine achievements in 90's and what are the prospects and goals for 2000? When users will have a perfect version of wine?

Julliard: Wine has come a very long way, from the initial 16-bit support to now nearly complete support of both 16- and 32-bit APIs, OLE, DirectX, etc. We are getting to the point where the core of Wine is complete, which should lead us to the first general public release, hopefully sometime this year. The perfect version that runs 100% of the applications with 100% compatibility is probably never going to exist, but a version that is perfect for a certain number of tested applications is certainly possible; and in fact today's Wine is already perfectly good enough for certain applications.

The OLinux site also has more Linux interviews.

English: http://www.olinux.com.br/interviews/
Portuguese: http://www.olinux.com.br/entrevistas/

Talkback: Discuss this article with peers

"Linux Gazette...making Linux just a little more fun!"