This one’s a long one folks. Click the “Read More” link below to view the full play-by-play.

Monday, Day 1:

1. Things are fuzzy as of this writing [Because I haven't been taking journal-style notes.], but: Unfortunately, one of my testers was taken away in order to do development work. I saw this coming [The tester expressed interested in going back, and he had other hats to fill.]. When asked by a stakeholder if this was okay, I reacted from-the-gut and said “I’ll do what I can to manage.” I recognized that the tester was needed elsewhere, so I let him go and prepared to simply take the hit. [This was a bad move.]
2. Today was Monday, so given the process envisioned last week, today was regression. I started the tester off doing simple verification for fixed bugs, and surprisingly, things went very quickly.
3. If we were to strictly follow the process envisioned last week, then next up should have been regression charters. At the time, I didn’t think there were any regression charters worthy of more exploration [I'm not sure much got fixed.]. Even so, there might have been more pressing issues: We really wanted to start testing our mapping application, but there were things blocking the testers from actually testing [The login functionality was borked. This is a big blocker.]. I sent the testers off on some lower priority charters that didn’t have to do with maps, and I started trying to get the logging-in issue fixed.
4. The login issue was caused by multiple faults in multiple parts of the software. We have three systems: The backend — the part that holds all the data; the frontend — the user facing portion; and the authentication system — the part that makes sure users have the right credentials. There were three problems: a) the authentication system was returning the wrong values, b) the frontend wasn’t responding correctly, even if the authentication system was returning the right values, and c) the backend was saying, “We won’t accept the user’s credentials even if the authentication system and the frontend respond appropriately.” [Yes, the backend can speak. ]
5. Because my role is to facilitate my testers so they can actually do some testing, it was my job to make sure this problem got fixed. I went to the developers of the three parts (of which I was one of them), and pleaded to make this issue a high priority. Graciously, all accepted, after a bit of nagging and following up over the next day. [Whew. A lot was riding on that fix.]

Tuesday, Day 2:

1. I realize that having only one tester is not working well. Things are fuzzy, but only a little testing got done between my single tester, my more manegerial style work [Trying to get the login issue fixed.], and the general needs of the day.

Wednesday, Day 3: “The Meeting Day.”

1. I had a discussion with my only tester about some more recent priorities. Some of these came from “on high,” and others I perceived as important to the testing group. Full-on testing of our mapping application could finally be done, so we had to make sure we hit those things as soon as possible.
2. Since I recognized that one tester

UPDATE, 3/9/08: This was a very, very long post. Unfortunately — and I’m not sure why — this was cut off at the sentence you see above, maybe a third of the way in. I don’t have time to recreate it, and I hadn’t backed this one up. It looks like this week is cut short.