A New Experiment

Hello, internet!

Earlier this week there was an issue at work. Some software patch wasn't patching what it was supposed to patch or something -- I'll admit I stop paying attention almost immediately.

But there I was, early in the morning with some developer leaning over my desk talking about issue numbers and missing subdirectories or something.

I had not had coffee, so while we were staring at the same screen (mine) it took a while for me to mentally condense what the guy was saying into "there have been new changes to the old patch, please produce a new version." I added the "please" part, too, because my internal tirade translation algorithm adds politeness.

Our "process" includes an initial build request form which is filled out by the Project Manager with full issue number details. I requested one of these.

The developer looked stricken. The Project Manager would not be in the office for a few hours and he really needed to test the busted functionality under a fresh patch.

Against my better judgement, I agreed to build the latest version of the broken modules and reconcile the paperwork after the Project Manager showed up.

The developer stood there as I checked the files out of the code repository. The screen filled with lines of scrolling text and I made the comment that a lot of changes had been made.

He wandered off and I built the patch and made it available to him.

Is anyone surprised that it was screwed up? If so, I suggest reading any of the posts archived on here. That feeling of surprise should soon be replaced with wry amusement or perhaps a feeling of smug superiority.

Anyway, the Project Manager and the developer were both standing over my shoulder, questioning my methods (which are, with admittedly millions of environmental variables the same every freaking time) and trying to determine where I had somehow screwed everything up and potentially caused to company to lose trillions of dollars or something. Again, I confess that my attention was wandering quite a bit. I started to scroll through my "Cute Kitties" iPhoto folder.

Those little guys are adorable!

I remembered the wall of new code from earlier and asked about potential changes other developers may have made to the same data. After some discussion and a meeting, it was discovered that indeed some other developers (multiple!) had made changes to the same code for other issues and checked the changes back into the repository without testing. Or spell checking. Or any reasonable reliable faith in a higher power. And those changes, the ones after the changes by the guy who was spending enough time in my personal space to start paying rent, had busted everything.

However, I built them into the patch, so (after another meeting and some time spent yelling) it was decided that I would delete the new build, check out the changes only up to the ones made by the early morning developer and build the patch again.

So, to summarize what was probably a very long and annoying story only once punctuated by the thought of adorable kittens, I did someone a favor and was punished with additional work and multiple meetings.

Never again.

From this moment on, I'm ignoring every email not sent by a Project Manager with a detailed and complete official build request form. In fact, I've configured a rule on my inbox that makes even looking at emails from anyone else a total chore.

And the issue of people wandering up and requesting work in person?

I'm hoping that posting this sign over my desk may reduce the chances of that:

 

Feel free to make a copy and post it in your own work area.