ManagementSpeak: We’ve suffered from several serious vendor occurrences, but a fix is coming.
Translation: They almost have more patches than bugs now.
Dave Thomason, this week’s contributor, shows how to patch the management explanation.

This week, InfoWorld celebrates the best technologies of the past year.

But that’s way too obvious: It’s XML — simple, elegant, and its adoption has been fast, broad and deep.

Boooriiinng. It took no insight to figure this out. It barely required a pulse.

So we’re going to focus on worst ideas of 2001. InfoWorld’s focus is on technologies this year, not individual vendors and products, which means I won’t get to (for example) include:

Palm for its complete lack of innovation. When all interesting innovation for this platform came from Handspring and Sony, for shame.

Novell for its Al Gore-like ability to snatch defeat from the jaws of victory. Never has such superior technology (NDS that is, although Gore’s animatronics were phenomenal …) suffered under such consistently inept marketing.

Microsoft, of course, also deserves opprobrium for releasing Active Directory as catch-up to NDS rather than leap-frog technology, but that’s really a 1999 carry-over.

Apple, which still hasn’t figured out that when you’re a market underdog you need marketshare and mindshare, not margins. Macs need to cost less than equivalent Wintel PCs, not twice as much. Despite the superiority of Mac OS X, Apple still deserves to fail.

UCITA’s backers, for stuffing it through an egregiously biased process. If you have any doubt that we’ve entered an era where big-lie propaganda is business-as-usual, read the press releases defending this bag of muck, and try not to gag.

But our focus is on bad technologies, not individual products or companies. The clear winner would be Fat Network Architectures (the more accurate name for “thin client”), only that’s an old issue. Fat network architectures provide convenience to providers while diminishing usability. They’re only appropriate for applications whose deployment is broad but whose use is occasional.

Another candidate is Enterprise Application Integration (EAI) security. EAI itself is incredibly important right now. But with few exceptions, EAI systems must have superuser privileges, which means they add yet another, independent layer of security administration. (Incredibly, Microsoft seems to have gotten EAI security right in Biztalk, which reportedly passes through already-established access rights.)

The runner-up is the Application Service Provider (ASP). ASPs, which were never anything more than rehashed timesharing service bureaus retooled onto fat network architectures (not a promising start!), never figured out that technology is now built into the core of the modern enterprise. Integration — a critical business priority (the reason EAI is so important) — is hard enough when a business runs all of its systems on its own computers. The ASP model makes it nearly impossible.

The clear winner, though, is Internet Attached Storage (IAS). At least fat network architectures are sound engineering. IAS, on the other hand, separates data storage from data processing, connecting them through the Internet at speeds and latencies that are a tiny fraction of LAN, bus or channel attachment.

I’m quite certain IAS is a winner when viewed through the distorted lens of Total Cost of Ownership (TCO). I’m equally sure that if you move a production database to IAS, what used to be overnight batch will turn into something quite different.

Between now and the time it’s finished running, you’ll have had time to fix the Y3K bug.