Anant Jhingran's Musings: Database Blog

My friends and colleagues, Mike Stonebraker (well, my advisor at Berkeley) and Don Haderle (my predecessor at IBM) have started a new collective blog, The Database Column, alongwith several other database folks. I applaud their intention. Their challenge, as for all of us who belong to companies or are in general vocal and opinionated, is to avoid being a marketing shill, as they also acknowledge. I would love for this to become a good forum for discussions in this major industry, but...

First, the very first posting, One Size Fit All -- A Concept Whose Time Has Come and Gone, does not live up to the promise of not being a marketing shill. It is a blatant ad for Vertica. [And the fact that I clearly disagree with its conclusions, but more on that later...]
Second, and I will have to see how it evolves, architectural evolution of databases, while academically interesting and critical for all our customers, is not the sole issue (I would say, not even the primary one) for at least the customers I deal with. It is how applications get written, it is tools, it is the cost of skills, it is TCO, it is content management, it is information as a service etc. So if the focus is narrow around databases, then I think the interest in this forum will wane.
While others might feel that this club is exclusive, I have no issues with that -- after all, all blogging is by one or few to express their opinions. So as long as the group does not try to represent the voice of all of us, they are free to express their opinion, and I look forward to reading it.

Now for the substance of the first post by Mike Stonebraker. The death of relational databases has been predicted several times in the past, and this is not the last time when someone will say this. Remember OODB's? Remember XML only DB's?

Death of its architectures has also been predicted in the past. Instead, the relational architectures have evolved, without forcing applications to be constrained. Yes, relational systems have become big. But I would say the following: new architectures are successful when they attack a different problem. Providing an equivalent bit replacement (arguably at a lower cost, as Mike would say) is hardly the dominating argument that would cause people to switch. Quite simply, as my colleague Curt Cotner points out, database engine technology is a drop in the bucket in terms of the investment it takes to be a full player in database these days. You need APIs/drivers for all the application environments (JDBC,ODBC, OLE, ADO, .NET, Ruby, PHP, Perl, etc.). You need integration with the application servers (J2EE, persistence layers, XA protocols). You need system integration (workload management, etc.). You need tooling to save people cost on the admin side. And the list goes on. Ultimately, these AD/Admin issues consume 70% of the IT budget...

So would an arguably better, specialized database really win? Not in Curt and my view...

Looking forward to agreeing with my advisor sometime in the future, especially if he writes about streaming :) :)