Remember the browser wars? Netscape versus Internet Explorer was once the most hotly contested topic in the computer industry. But now, with Microsoft having soundly cornered the marketplace, the battle for Internet browser dominance seems like ancient history.
Don't think all's gone quiet in the software marketplace, though. There's another war brewing already, and the stakes this time around may prove to be even higher than those of the browser wars. This time, the scuffle is over databases.
I know what you're thinking: databases? Bo-ring! Besides, isn't the real debate about the operating system? Linux versus Windows versus the Mac OS, and all that?
But if that's the case, then why is Red Hat, already the largest commercial Linux vendor, planning to release a new database product? It announced its less-than-imaginatively named Red Hat Database in late June, explaining that the project would build upon code from the existing Open Source database PostgreSQL.
The decision to enter the database market now seems a little odd, considering the competition they'll be up against. While Red Hat does command more than two thirds of the Linux market, in the database market they'll have to contend with big money from the likes of Oracle, Microsoft and IBM.
It's true that Red Hat plans to compete primarily with other Open Source database projects at first. But even company representatives don't deny that they're interested in a piece of the larger pie.
What's going on here? After seeing what happened to Netscape, why would a company like Red Hat expend R&D money going up against the database heavyweights — especially in this soft economy? Why not concentrate on the OS? Far fewer of us use big database packages than use operating systems or Internet browsers, after all.
But the Web has changed all that. Microsoft's Web site, Amazon.com, local community bulletin board Craig's List, the search engine Google and the file-sharing service Napster all rely on databases for nearly every request they serve.
Each of these sites serves to demonstrate a growing trend. While at one time software was written to exist solely on your PC, these days some of the most popular applications are being designed as Web-based client/server systems.
On the front end, or client side, is the browser (or sometimes a custom application, as in the case of Napster). On the back end are the databases and application servers that handle the heavy lifting.
As this division of labor becomes ever more prevalent, operating systems have grown more diversified as well. Vendors market some products as client-side operating systems (like Windows 98 and Mac OS 9), while others are targeted at the server market (like Sun's Solaris and Windows NT).
But no matter what the market, almost no one chooses an OS out of simple brand loyalty. Many times, the features of one operating system versus another are too nebulous for the layman to understand. Usually, what we want to do with our computers is what drives our choice of OS.
This is true of both server and client-side operating systems, but it's most readily apparent in the consumer market. Graphic designers, for example, often prefer the Mac OS, because its feature set — and the applications written for it — support graphic design well. By contrast, business users tend to choose Windows for maximum compatibility with the majority of existing software.
The odd man out is Linux, which doesn't compare favorably on either score. So far, the Open Source OS has enjoyed relatively little penetration into the desktop space, except in academia, and it doesn't seem likely that it will take great strides toward consumer acceptance any time soon.
This deficiency is exactly why the Red Hat Database announcement is so crucial. If Red Hat can't count on Linux to capture a significant portion of the client side OS market, they still have the server side to shoot for. But to succeed, they'll need a database.
That's because, while operating systems have received a lot of attention in the press lately, a pure server operating system doesn't give you much, in and of itself. It's the applications you can create and serve with it that matter.
Just as graphic designers favor the Mac OS for the wealth of visual arts programs available on that platform, server operating systems often succeed or fail on the merits of the development tools they support. Key among these tools is the database. After all, the majority of what client/server applications do boils down to the storing and retrieving of information.
Modern databases try to differentiate themselves from their competition in various ways, each claiming to be faster, more reliable or more cost effective than the next. They also typically offer their administrators advanced query languages, to allow data to be stored and retrieved in complex ways.
Building on this basic functionality, many vendors provide application servers to run on top of the database. These provide entire programming interfaces to the stored information. Together, these components combine with the operating system to form a complete server-side platform on which custom applications can be built.
Of course, the purchasing decisions for these platforms all happen behind the scenes, from the perspective of the consumer. Internet browsers are products we use directly — and so, for most of us the browser wars hit much closer to home. And yet the database wars could potentially have every bit as much impact on the future of the industry — perhaps even more.
In the not-too-distant future, it's foreseeable that the current methods of accessing the Web could disappear. Once they mature, network appliances or mobile devices could eventually replace PC-based browsers, rendering the whole Netscape-versus-IE debate effectively moot.
Databases and server-side operating platforms, on the other hand, will still be around. In fact, with the PC replaced by a lightweight client like a mobile phone, they'll handle more of the processing burden than ever before.
Because of this, a well-marketed Open Source database (like the one Red Hat proposes) could potentially do more to cement the hold of Open Source software in the enterprise than the Linux OS itself. In effect, it would transform Linux into a complete Open Source Web applications platform, to compete with commercial offerings from the likes of Sun, Oracle or Microsoft.
And unlike the client OS market, where Mac OS and Windows users are separated along well-defined lines, the server market is less divided into OS "camps." A number of platforms have carved successful niches, and so far there's no clear winner. Red Hat is banking that this means there's opportunity for still more platforms to enter the field.
But the really refreshing thing about the database wars is that no one vendor has to win. There's room for solutions at various levels of price, speed, reliability and function. And so long as no one vendor can control both sides of the computing equation — both client and server — then no single vendor can control all of computing; an overturned verdict in the Microsoft trial notwithstanding.
Looking at it that way, the database "wars" start to seem not so much like a bloody grudge match. Rather, the whole thing starts to sound suspiciously like good, honest competition. And that's something this industry could use a whole lot more of.
