Great post, Perry.
What is the cause of this? Well, I think blame can go around a lot. In no particular order, I would 'blame' the following:
Accelerated release of new technologies. There is a great fear across the board that a long-term or even middle-term project will be obsolete before it is done, hence a rush to completion.
Accelerated project definitions. In the 'old days', sometimes a project was spec'ed out over a period of months, if not years. There is too much pressure in the IT world to produce.
Accelerated release of new tools. Developers have little opportunity to become proficient in a new tool version before the next version comes out. This creates two problems: Half-way decently engineered applications on state-of-the-art platforms and Really well engineered applications on soon-to-be-obsolete platforms.
And, no, we are not going to have to pay for this in 10-20 years because the very nature of application lifecycles has changed. Most in-house applications evolve over time, with some major revisions sprinkled during the lifecycle. Rarely is a situation where it's all going to fall down and go boom, especially in enterprise applications.
Users are also far more involved in the development cycle than they used to be: This creates a catchall for stupid coder tricks and actually helps an application become, by inference, more robust.
>There was an interesting article in the LA Times business section yesterday about the current state of IS departments. It was actually kind of scary to see in print what our industry has come down to.
>
>The gist of the article was how money has skyrocketed to the top of the list of job requirements for programmers. Creating good code, or the learning experience has plumetted in the ranking. There was a term coined by McConnell in one of his latest books that basically describes how the current state of technology is to get a few programmers together and code. No design spec necessary. No more 9-5 type stuff, this is burning the midnight oil heads down coding.
>
>Of course you could expect boatloads of errors, but since people are working all day long, no problem to fix them. The goal is not to produce award winning software. It's to get a program that works good enough to show off to potential investors.
>
>Management was partially to blame as there are more and more Dilbert managers coming on-line who have no clue about the developement process (or anything related to programming for that matter).
>
>The article signed off by saying how we are probably going to have to pay for this type of environment in 10-20 yrs, much like the Y2K histeria of yore.
>
>Is this really what it's all come down to. Or is there just another world out there populated by the hotshot GenX folks that we are just not a part of?
>
>Pf
------------------------------------------------
John Koziol, ex-MVP, ex-MS, ex-FoxTeam. Just call me "X"
"When the going gets weird, the weird turn pro" - Hunter Thompson (Gonzo) RIP 2/19/05
