The tragedy of open source

Published Monday, 3 December 2001 4:00AM CST by in Technology

0

Learning new systems is never easy. The hardest part is learning to think like the people who designed the new system, because even when they try to think like us, the users, they really can’t. The best systems designers engage human factors experts to serve as an advocate for users, but it’s almost impossible for the human factors folks not to drink the designer’s kool-aid and start thinking like the systems designer. All systems are designed to please those who designed them. And there’s nothing wrong with that.

UserLand‘s Frontier and Manila (Manila is part of Frontier) is the content management system to which ARTS & FARCES internet has migrated from Zope. I’ve been trying to get my head around Frontier and Manila, off and on for the last month or so, and intensely for the last week or so. Zope is an open source web application server that’s available at no charge; Frontier/Manila is a closed-source content management system with an integrated web server, from a small developer, that costs $900 per year.

Frontier better meets my needs, and that’s why I switched. It’s also representative of the tragedy of open source.

The best candidates for using open source software are large organizations with large IT departments. Why? Because they’re best positioned to provide the resources necessary to support projects based on open source products. Unfortunately, they’re also the least likely to use open source software. Why? Because, with a few notable exceptions, they can’t call up a vendor on the phone and get support. And they’re scared that support will disappear. And they’re scared that their IT people will spend too much time tweaking and putzing, er, optimizing. Whether or not any of this is true doesn’t matter. That’s how large organizations feel. Every one of my Fortune 100 clients has an open source policy. Every one of them prohibits its use. Every one of them also has a lot of skunkworks projects that are based on open source products that manage to stay under the corporate radar. Large organizations are not only in the best position to use open source software; they’re in the best position to make major contributions to open source software.

This is another tragedy of open source, but not the one I want to focus on here.

The tragedy of open source that I want to focus on is the fact that it will never be built for people like me, the technically-literate, non-programmer. Think about it. Open source tools are built by experienced programmers whose income comes from either 1) supporting their open source products (rare); or 2) coding for someone else (more common). Now just who do you think these guys (and they’re invariably guys) are building software for? Others like themselves, that’s who. Certainly not for people like me. Remember that all systems are designed to please those who designed them. The minute they release a product that someone like me can use to its fullest extent, they’re out of the selling-support-by-the-pound business.

But isn’t this just as true for closed-source products? Certainly, we’ve all heard the horror stories of paying for product support by the minute with little or no result. Those situations are mere anomalies and market opportunities for savvy competitors. That’s one thing that the free marketeers are right about. If you sell a closed-source product and fail to support it, count on being out of business. Look, I despise Microsoft’s business methodologies more than most, but you have to admit that it makes products that people can use.

UserLand’s support staff, who are exceptionally knowledgeable because they’re also the ones who designed and wrote the software, have been very patient with my questions. They’ve either solved or are working on most of the problems I’ve had.

Frontier is a mature product, starting out as a scripting environment on the Macintosh, but Manila is only two years old. As a result, some of the documentation is incomplete and even wrong. For example, support for directories like this is hidden in plain sight and incomplete (I’ve written a process for working with directories for large publications). But at least it exists, and it’s bound to get better.

[Editor’s note 31 July 2002: Both of the above links are deprecated as a result to a move to yet another content management system; this one being open source and commercial.] I have strong hope for open source; after all, the Internet wouldn’t exist without it. Only time will tell if the tragedy of open source is irreversible. But for now, my software of choice comes from small, independent developers. Open source is fine, but give me something I can use.

0 responses. Comments closed for this article.