February 14th, 2008
The scaling problem and open source
Most trends in the open source market come down to one word.
Scaling. (The picture is from a Cisco report on scaling VOIP, but the problem is general.)
To stay worthwhile a project must grow. It must attract new developers, new community members, and new money. It must scale in complexity and functionality.
I feel your pain here. The same trends underlying open source also underly blogging, which as a writing activity is just about the same age.
In the beginning blogs were individual projects, just as the great open source projects started with one developer. You might think of Dave Winer, then, as the Linus Torvalds of blogging.
Linus dealt with this problem through a succession of jobs, culminating in his present work for the Linux Foundation. This gives him the resources to keep the kernel growing, although he has written often of the technical challenge in doing just that.
Bloggers haven’t had it so easy. If you look at the Technorati Top 100 most popular blogs, you won’t find Winer’s name there anymore.
Instead you’ll find communities like DailyKos (no. 11), publishing companies like TMZ (no. 12), group efforts like the Huffington Post (no. 4), and important corporate efforts like the official Google blog (no. 17).
This is all due to the problems of scaling. To keep getting great content you need more people. To feed the servers or buy new ones you need a business model.
Blogs aren’t really blogs anymore. They are publishing exercises.
The same problems exist in open source. Projects need people, they need organization, they need users and beta testers. In other words, they need money.
Each of us has the right to fork any project, but as a practical matter that is seldom done. Successful forks require wrenching a large group of people from the sponsor’s direction. It requires that you find new funding sources.
It’s no longer, I’ll take my code and go home. A successful fork is a new business. Programmers who want to spend their time programming, like journalists who just want to report, have to work for someone else who brings in money from their efforts and organizes the work.
This is true no matter what you do. Cooks who start restaurants become businessmen or women. It’s no longer about doing what you do, but finding a way to help other people do what you did.
In this scaled environment, corporate projects have an advantage. Corporations are built to make money, to organize large work — to scale in other words. Individuals aren’t.
Even if you can create a non-profit foundation around your project, as at Mozilla, someone must still find the money. Even a .org has a .com.
So when some journalist says, if you don’t like it fork it, you have a right to be skeptical. A good fork must be organized.
All you want to do is have some fun, as a programmer. All I want to do is have fun, as a writer. ZDNet has given me that freedom. Perhaps your employer has given you that freedom. Quitting gets harder as salaries grow and bills pile up.
But it can be done. Thanks to the Internet you can organize a fork from many places all at once. This is an important check on corporate power over software, as is every customer’s ability to do their own support, or outsource it to a third party.
The modern world of blogging and open source isn’t perfect, but compare it to the old world of newspapers and proprietary software.
Thanks to the problems of scaling we don’t have all the power we’d like. But thanks to the power of the Internet we have a lot more than we did.
And every once in a while we need to exercise it.
Dana Blankenhorn has been a business journalist for 30 years, a tech freelancer since 1983. You can follow Dana on Twitter. See his full profile and disclosure of his industry affiliations.
Subscribe to Linux and Open Source via Email alerts or RSS.
