Latest News

Informed about Informix?

How many heard the fall of Informix? Enterprise Pursuits May 8, 2001 Nelson King Hed: Informed about Informix? dek: How many heard the fall of Informix?

If a tree falls in the forest and there’s no one to hear it, does it make a sound? Informix Corp. has just been sold to IBM, and I wonder how many are left who heard that news and felt some reverberation.

I’m an old database guy, cutting my teeth on the early versions of Oracle and dBase (in quite different worlds of computing), and yes, Informix. Informix was a high flyer and a major player at a time when it seemed like there were more database products for the enterprise than any other kind of software.

Back in the mid-’90s there was a flurry of worry on the part of the big relational database companies that object-oriented databases (OODBMS) might be on to something. Heck, they might even become competition. The general response was to co-opt as much of the OODBMS technology as it took to make it look like relational systems could do the same thing. This movement was led by Informix, whose “Universal Database” was the first of many similarly named database engines. It featured “object-relational” database technology and a series of adapters, called DataBlades, for various kinds of data.

Tech people ate this stuff up. The Web was happening. Multimedia was going to rule the Web, and object-oriented databases did multimedia. So did object-relational databases. It was a good spiel. (I know; I did a few myself.) However, it wasn’t reality. As we’re learning–the hard way–it isn’t flash that makes the Web, it’s transactions. For transactions you don’t need fancy object-style database capability.

Unfortunately for Informix, its commitment to future technology left a lot of its bread and butter clients complaining about the lack of attention. While the R&D for object-relational technology cost a bundle and the marketing cost even more, there was only attrition at the customer base. Informix ran into financial trouble. It survived by dropping most of the DataBlade technology into the quiet bin and refocusing on its traditional clients. But these efforts were too little, too late. I’m sure Informix employees are thanking the computer gods for a Deus ex Machina in the form of IBM and a billion-dollar buyout of the Informix database division. It could have been a lot worse.

Is there a moral to the story? You bet. When it comes to mission-critical software, I believe corporations are inherently conservative, even reactionary. If left to themselves, IT departments seek to find software that works; is familiar; is supported; and looks like it will be around a while. They tend to find a vendor and stick with it. Over the years, this behavior marginalizes many products and favors only a few mainstream products. Now, for all practical purposes, there are four relational database management systems that dominate (excluding the mainframe market): IBM, Oracle, Microsoft, and Sybase.

I don’t think there’s necessarily anything wrong or unusual about either the process or the outcome. Similar things happen in the consumer world. However, you have to be sanguine about some of the consequences. Concentration of a market into so few products limits choice. The bigger, more complex, and more important the type of software (be it database, office suite, CRM, or whatever) the more likely an enterprise will be faced with only a few choices, leaving it very little leverage. It’s very difficult to switch database management systems, and vendors know it. This is not an ideal situation for corporations.

Now consider an extrapolation to e-commerce. For many if not most corporations, e-commerce is mission-critical. Only a year ago, the software that drives e-commerce (various servers, Web page builders, and business components) was available from a horde of companies. The horde is dwindling. Even major e-commerce companies such as iCat and Open Market are in death throes. Where do you think this will wind up? Any bets that some of the names left standing will be Microsoft, Oracle, and IBM? Will there be anyone else? And whose software is your enterprise e-commerce built upon? I wonder how many Informixes are out there in the land of e-commerce?

Maybe it’s time to start listening to the forest.

Editor at Large Nelson King also writes Pursuits monthly for ComputerUser magazine.

Leave a comment

seks shop - izolasyon
basic theory test book basic theory test