April 23rd, 2001

self portrait

The rites of IT

So finally on Friday, the project I've been working on pretty much non-stop for three weeks was finished. What a relief. After all the bogus posturing by one half of our team, things finally got to where they needed to be. We were also able to get it done at a reasonable time so everyone involved didn't have to start the process at 5:00 p.m. on a Friday and stay until whenever.

Really, I understand some of the reasons behind what the IT/back end half of our group was getting at. Things should be tested. Changes should be made when they'll affect as few people as possible.

The manner in which they communicated and applied these concepts, however, was hypocritical. They wanted us to write up and distribute widely a test plan for this project. It was a stinking set of HTML pages, for crying out loud. Never mind that the development and staging environments they set up for the project had vastly different parameters from each other, not to mention the production environment. We could easily check the behavior of the pages to verify that it was doing what we wanted it to on the various browsers and/or platforms. A test plan? Where was the widely distributed test plan for the project they turned up earlier in the week?

So I wrote up a test plan anyway. It was basically a document that said, "Click all the links. Here's what you'll get." Mind you, I included notes that various environment differences meant that sometimes the URL you'd get would be a 404 error. All would work fine in the production environment, however, Nevertheless, narrow-minded engineers followed the plan to the letter and sent off a flurry of emails reporting "bugs" in the new page. The only one of these that was remotely valid was problem with logging in to the registered user portion of our site; depending on browser caching settings, you might log out and not be able to log back in without quitting your browser. This problem, however, had nothing to do with the pages I'd created and, in fact, existed on the current site! So much for the rigorous testing they espoused.

But all that is behind me now. I did actually have some fun on this project as it was the first time I've been able to actually code a set of pages to the standards set up by the W3C. For so long, I've had to make sure pages would be at least reasonably functional in Netscape 3.x era browsers that the funner, cooler parts of things like style sheets just didn't come up. Couldn't use them, no sense spending time learning much about them. As the good folks at the Web Standards Project kept telling us: without standards support in browsers, people making web pages were having to work too hard; coding multiple versions of pages and/or using scripts to work around the differences in implementations. Thankfully, the latest versions of both major browsers offer excellent support of the basic standards. Now all I have to do is wait the six months to a year it'll take our base of Netscape 4.x users to migrate to compliant browsers and I'll be happy. Or at least content.