What is Up With Content Management Systems?

Category: Random Observations Bin

Wednesday, October 9, 2002

So I've recently been asked to evaluate various content management systems (CMSes, for short), and after exhaustive analysis, I can sum up the marketplace in two words:

They suck.

No, seriously. All of them. They ALL SUCK. What is the matter here? Why can't the collective intelligence of thousands of bright engineers come up with a CMS that DOESN'T SUCK?

The problem is that the CMS market has fragmented into two parts. One is dominated by "enterprise-level" players like Vignette and Interwoven. These guys offer products that have feature lists long enough to choke a horse. The problem is, their prices are just as astronomical as their feature sets -- installations of these products can run into the hundreds of thousands of dollars. Plus even then, they don't actually do anything out of the box -- you've got to pay programmers (yours or theirs) to customize the thing for you. And half the time they don't work even then.

The other part of the market is the low end. In this space there are literally dozens of products, ranging from low-cost (ala Userland Manila, $899/year) to free. Problem is, at this end the solutions all suck too. So far I've looked at OpenCMS (couldn't get it to INSTALL on our Apache/Tomcat setup), Red Hat Content & Collaboration Manager (Oracle-only, which we can't afford, and won't support PostgreSQL until next year), and Zope (cool architecture, but only lets you code in Python, a dead-end language).

Oh, and then there are the ten million Slashdot-alikes: PHP-Nuke, PostNuke, Xoops, and the like. Great if you want to run a site that's a long list of links with threaded discussions -- useless if you want to do anything else.

Now Microsoft is muscling its way into this moribund market with Microsoft Content Management Server 2002. At a cool $42,000/processor, it ain't cheap -- but since the competition here is so insanely weak, what do you think the odds are that they own 80% of this market within 5 years?

Look, all I want to do is help people run their Web sites. There's got to be something out there that doesn't suck that I can use to do this. Maybe you know what that something is -- if so, educate me. Otherwise -- ugh, I don't want to think about "otherwise".