Revolution postponed

Infoconomy has an article looking at the slow growth of Native XML Databases. It's pretty obvious at this point that the technology hasn't really caught on and based on the way the database companies market, I can't say it's surprising. It's really too bad, as when I look at projects I see so many places where Native XML Databases could massively simplify things while at the same time performing better. Unfortunately, developers simply don't know anything about them and don't even consider the technology.

In the current project I'm on there are two key components of the system that simply cry out for an XML based solution. What do we have though? Relational and a tremendous amount of complexity and instability as a result. Relational was plain and simply the wrong solution for the problem, but as far as I can tell alternatives were never considered. This is an education issue and is an area that current Native XML Database vendors have done an astoundingly poor job of handling.

I think a lot of the problem stems from the vendor's desire to target enterprise markets by entering the business through the frontdoor. This has meant they had to sell to CIOs and as a consequence they've had to price their products in a way that has put them out of reach for use in niche applications. This marketing focus has also pulled their attention away from the people who will really make a difference, deveopers. If developers don't know about the technology, selling to all the CIOs in the world isn't going to actually get the technology used and makes any argument about return on investment irrelavent. It's not a formula for long term success, nor has it proven to even be a workable formula for the short term.

Posted by Kimbro Staken