I just couldn’t pass on the Trek reference. I’m really not looking for a Klingon. What I’m looking for is an IT person who is actually helpful, who actually believes that someone with a PhD in a humanities discipline might have the first damn clue about computers, who actually does not actively participate in Bill Gates’s master plan for taking over the whole universe.
I guess while I was at Capital, Barb spoiled me.
Are you supposing, by now, that there’s a story of some sort here? Why, yes! Yes, there is.
Part of my job is to examine writing in the curriculum (campus-wide) and develop a writing across the curriculum program. One of my first steps in this regard is to implement a survey of attitudes toward and practices of writing in courses and of faculty across campus. Not scientifically valid, I know, but a good base-line starting point, nonetheless.
I wrote the questions, decided what kinds of answers would be appropriate to them, wrote the xhtml to code the form, wrote the php to control the form (pretty elegant, if I do say so myself—seven nodes to the survey, all bundled in to one php file, with one massive data dump at the end), did the MySQL set up to store the data, debugged code (for like three hours), play-tested the form on my home test server, play-tested it some more on my “live” server, and—yesterday—sent the link out live.
Here’s what I know:
- I tested my survey app on the following OS/browser combinations: OSX/Firefox2, OSX/Safari, WinXP/IE6, WinXP/IE7, WinXP/Firefox2, WinXP/Firefox1.5
- In my tests, after the extensive debugging, everything went perfectly—the app worked just like I wanted it to.
- Last night, I logged in to my MySQL dbase to see if anyone had actually done it yet; I had no survey data, but three people willing to participate in the follow-up interview, which you can’t get to if you don’t submit your survey data.
- I called a halt to the survey, because something was FUBAR.
- I ran through my testing protocol again (all six combinations); everything worked, picture perfect.
- I contacted IT (here’s where the fun begins), knowing that I’d wait overnight for a response.
- IT was no help.
What do I mean, exactly, by “no help”? I’m glad you asked. What I mean specifically is:
- First suggestion: We offer survey services; why didn’t you have us do this for you? (This is called We need job security.)
- Second suggestion: It can’t be a problem on our end; it must be a problem with the survey; what app are you running? (This is called: Not my job, man!)
- Third suggestion: You wrote the app yourself? You must have messed up in the coding. (This is called: You’re an idiot who couldn’t possibly know enough about computers to pull this off.)
- Side note: I’m going to correct everything you say about coding, to feel better about myself; even if what you say is correct, I’ll find a technicality. (This is called: See previous.)
- Fourth suggestion: What you’re saying is not possible. (This is called: Observable phenomena do not matter, my word is law.)
At this point I stopped bothering to try. I made some changes to the code. I’m pretty sure I know what’s happening and my “friend” in IT doesn’t believe me. It has only been an issue on campus with the WinXP/IE7 combination (WinXP/Firefox1.5 has worked as scheduled; I don’t think I’ve seen a Mac-derived response, yet). I’m thinking that this is a result of the overzealous Microsoft can’t plug its security holes so it builds a big wall around them that stops all sorts of useful things too computer security craze that is sweeping the country (WinVista: Allow or Deny?), and has firmly taken root on this campus, too (the Micro$oft Exchange Server is the only email server on campus—there is no POP3 email…grrrr! But I think I’ve ranted about this before).
I’ve asked some colleagues whose survey data I know vanished to try again using the Winblows/Exploder combination, hoping that I’ve managed to write code sufficient enough to fool whatever “security” device is causing the problem. I’m thinking at some point, my session cookies are getting erased and all of my $_SESSION[] variables are going bye-bye. I can rewrite the whole bloody app without them, but that would be a pain.
I think I’ve said before that putting Micro$oft in charge of security is like that medieval mistake made by the Chinese: hiring the Mongols to guard the Great Wall. Some people say there’s never a Mongol around when you need one.
So maybe I am looking for a fierce Klingon warrior, ’cause Mongols? Mongols I got.