[Author's note: Yes, I'm backdating again. I'll write two entries tomorrow. Or the day after.]

Okay, I'm done with my test pass.

I'm a software tester for my company. With as many black eyes as it gets for software quality (and usability, and security, and just about everything else possible) I expect some barbs. For what it's worth, I think the more intelligent ones are well-deserved. If I didn't think the products couldn't be refined and improved, I would be looking for other work, no?

It's taken me over a year, but I finally 'get it.' I understand what I'm to do. Or, failing that, I have deluded myself into believing that I do when I don't.

At any rate, I've been killing myself this past week. I am sure I worked more than 50 hours. I would lay down major money that I worked more than 60, and my betting threshold for majory money is 2/3. I would not even bet against my working 70.

But I'm done, for now. My product is off to the final beta test, during which point, I'm to do more test. However, as far as our team goes, it's "shipped." It's off to the final beta site, and, if it doesn't crash and burn in beta, they'll deploy it to the other users besides the beta ones.

Software test is generally seen as either a formality, a nuisance, or a cost center. No, not here, I'm talking about in general. For software test to have a chance at success, we either need to be in on the planning stage, or we need to say, "Whoa!" once the developers have finished coding the product and then familiarize ourselves with it, etc.

Well, we could be playing catch-up while the developers are working on the product. That rarely works well. To start with, the specifications for the product always change during development. Yes, I would love for them to be written in stone, but that's unrealistic. The problem is not that they change, but that no one tells test when they do...

As Rodney Dangerfield deadpans, "No respect."

I've been beating myself silly this past week sure that my labour would not be noticed. Actually, I was telling myself that it'd be a good thing, too. I feel it took me 50+ hours to do what a 'real' tester would have done in 40 or less. Yes, I know that some of you think I hold myself to unrealistic expectations, but I've found that it's better to be a little scared and run faster than be complacent and ... get eaten.

Oh, and some good news: I'm mildly surprised that I received some recognition for my efforts. Some of the bugs I've found led to mildly serious issues. Obviously I can't go into details here. I just file them, and every so often, the developers say, "Hey, good catch." I figure that's just par for the course, but the other day one of the team leads specifically called out one bug in talking to me and said it was a good catch.

(To clarify, team leads in my company are 'first line managers' elsewhere. They have technical expectations like the rest of us, but, to add to the fun, they need to also write performance reviews for their subordinates. My rule of thumb is, anyone who writes my review is my manager. Too bad this one lead isn't my manager in this one case, but good thing otherwise, because I don't know his area at all. :) )

It was a good way to end a really grueling week. I'm out from it and hope to evolve back to humanity over the next few days.