Well shucks Bill, how many hours of staring off into the middle distance did it take your brain trust to come up with that paradigm-changing strategy? Was there ever a point at which your user focus groups or market research or anything else indicated that your customers were more interested in features and functionality than in security and reliability?
Sarcasm aside, I do have to admire Microsoft for putting its collective hand on its heart and saying “we gotta do better”. Surfing through the press stuff on Microsoft’s website I came across a Microsoft paper prepared for submission to the 31st World Economic Forum in New York this year. This paper goes well into Trustworthy Computing and makes good reading for anyone involved in the software business.
But none of this is really that new. While you’re on the web, check out this August 1998 Interim Report from the US President’s Information Technology Advisory Committee. Section 3.1.1 and the bits about being dependent on “fragile software” and about how the “technologies to build reliable and secure software are inadequate” are particularly pointed.
Imagine if engineers made buildings, bridges, cars and aeroplanes the way that we build software. It’d be disastrous. And, by the way, that old chestnut about computing still being in its Wild West phase is complete bollocks — NASA has built bombproof software since the 60s. So why has it taken the rest of us so long to catch up?
Sure, it was all a lot simpler then (after all, men went to the moon on just two million lines of code) and economic and market forces — which don’t typically apply to aerospace projects in the same way as they do to commercial software — play a major role in the weakness of much of today’s systems.
The Microsoft paper goes into a lot of detail as to what the issues might be and how they might be resolved but, at the end of the day, it’ll take cash. And the software industry needs to accept some responsibility and reinvest some of its vast wealth into ensuring its own future acceptability to its end users.
Thin again …
As the editor mentioned in his February 4 Computerworld editorial (2002: The year of guessing wildly), I’ve been doing some more work on the thin client thing. While I still haven’t managed to catch up with the elusive man from Citrix (see All over bar the shouting) I have been speaking to a number of resellers, users and advocates among my circle of contacts.
The news is all good for Citrix. I have to admit, I was extremely wary about adding another layer of software into the already very complex Windows client-server environment, but all my research, the evidence and the experience of others suggest that it’ll (a) work (my kind of solution) and (b) save me a big chunk of my IT infrastructure budget. Compelling stuff.
Swanson is IT manager at W Stevenson & Sons. Send letters for publication in Computerworld to Computerworld Letters.