Linux Today: Linux News On Internet Time.

Editor's Note: Coping with Success

Mar 11, 2005, 23:30 (8 Talkback[s])
(Other stories by Brian Proffitt)

By Brian Proffitt
Managing Editor

So, I am writing this column in Day 2 of my OpenOffice.org 2.0 beta experience, and I must say, I am a happy little camper. The tools are solid, the document interoperability is excellent, and, best of all, this puppy is fast, fast, fast.

It finally breaks the one bugaboo that has seemed to plague OO.org all these years: an interminably slow startup. From this user, a big thanks to the OpenOffice.org development team.

The scuttlebutt around LinuxWorld last month dealt a little bit with OpenOffice.org. Nothing very negative, but there seemed to be a consistent observation that the OO.org team was woefully understaffed. Given the quality of this beta product, I have strong doubts about that. But enough people were mentioning it that I have to wonder.

Then, more recently, there's been squawking about the Firefox development team. Mike Connors posted a strong blog entry that seriously questioned the current status of the Firefox team and a number of pundits have stepped up an in varying degrees have wondered aloud if Firefox isn't in real trouble.

First off, there's trouble and then there's trouble. As open source projects, there is very little, if any, change for OO.org or Firefox to die. Even if every OO.org developer were to quit tomorrow, there would be more developers out there who would eventually pick it up. Sun Microsystems' involvement in the OO.org project would assure that.

The Firefox situation is a bit more problematic, as it has less corporate involvement. Yes, there's the Mozilla Foundation as an overseer, and presumably AOL. In the case of the latter, however, the financial and development support connections are a little vague. At least to me.

In the case of the former, today's news about the Foundation's shift of focus away from the now-defunct Mozilla 1.8 development tree to the Firefox and Thunderbird apps makes it clear that the Powers That Be at Mozilla recognize that allocation of resources is a critical issue, and I am glad to see they are responding to it.

Wednesday, Brian Behlendorf of the Apache Software Foundation touched on this very same issue when speaking to an analyst conference out in California. In that talk, he emphasized that it is important for strong open source projects to stay vendor-neutral and not be dominated by one person or company. He cited, of course, the ASF as an example of how this works well.

Curiously, he apparently cited OO.org as another good example, though in the same article he was quoted as describing that project as having an 80/20 split, with Sun holding 80 percent and the community 20. Even in Indiana, 80 percent seems dominant to me. But, articles have been known to be wrong.

Still, I wonder if on a practical scale, Behlendorf's argument is true. On a theoretical level, I agree with him 100 percent. But in terms of day-to-day commitment, I think there has to be at least one strong vendor/company involved with certain open source projects—if only to act as a safety net when the project gets very successful.

While I do not think Firefox is in trouble, I do think that its development team got caught way off guard by the huge success it continues to have. Suddenly the outcry for new features and bugfixes have accelerated way past what was expected, and Firefox' team has to adapt. Now that the Mozilla Foundation is devoting resources to the project, I think this is less of a concern, but there's no denying that it could be a lot easier if a big vendor got involved with the project to help them organize and support even more developers to lend a hand.

Personally, I'd suggest Novell, but that's just me. Google seems to be trying out the idea, and that would be a welcome addition, too. Let's see how the reallocation of Mozilla's resources will happen.

I will not go completely opposite from Belhendorf's assertion that better open source projects are community-driven. I wonder, though, if there comes a point where sheer volunteerism may not be enough to keep a massive project delivering at the pace it needs--that there needs to be an over-arching group that can step back and view the larger picture to see where the project needs to go and keeps the right resources dedicated to it.

I would contend that such an organization could be non-profit or commercial. Corporate does not always equate to evil. While some have privately told me of minor concerns about the overall health of the OpenOffice.org project, no one has indicated that this is Sun's fault, and--despite Sun's ever-changing attitude towards Linux--many have praised Sun's involvement in the OO.org project.

Behlendorf cited a project's method of organization as one reason for its success. Avoid project leaders, and build communities. With that, I agree on every level. The more distributed a project is, vendor involvement or no, the better chance it has for continued success.


This brings me to a program note: this coming week I will be on vacation in North Carolina, leaving the site in the capable hands of Rob Reilly.

I'll be back on March 21, hale, hearty, and hopefully rested.