It’s Like Moving In All Over Again

Monday nights are our usual get-together nights around here. We sometimes gather together in person, or online if we’re A) sick, B) lazy, or C) all of the above. Last night the group was gearing up to head to NYC in The Division 2, but because of my renovations, I wasn’t immediately prepared to join in. I quickly ran some cables, set up one monitor, and had to rely on headphones for audio, but everything was go at Chez Scopique, and fun was had by some/many/most/all. The only issue is that my 4K monitor doesn’t seem to play well with the supposed 4K cable I have. When I originally bought this display, I had a heck of a time getting things “not blurry”. Well, they are blurry again, and I have no recollection about what I did to de-blurry things last time.

My shopping list is now Primed with 2 sets of legs, three monitor mounting arms, and some cable channels. Once I get those parts, I can install the other two desk parts, and that will be that for surfaces (he says with the certainty that ensures that “that it will not be that”).

WomBot In The Wild

WomBot is now officially in BETA. It’s been invited to the Wombattery server, and a few folks were playing around with it last night…much to my horror.

“No plan survives contact with the enemy”, they say, and no product — no matter how long in development or how rigorously tested in-house — survives contact with a user base. The thing about development is that developers make shitty QA specialists. You’d think that because we understand the systems and know what kind of feedback makes the best feedback that we’d be perfectly suited to ensuring that only the best product makes it out the door, but it’s really the exact opposite. We know the exact conditions needed to get everything to work because we’ve been testing in increments as we go. When we get a system to work, we often deem it “done”…until it ends up in the hands of people who have no prior contact with the product and who therefor have absolutely no preconceived notions on what is the right way to use it.

This is really the best case scenario as it shines a light on shortcomings, but it’s also horrifying because pushing a product out the door requires a certain level of confidence that the product is good enough to push out the door. Watching that product suddenly crash and burn because the first action of the first person to lay hands on it was to use the product in a way that was never envisioned during development…that really calls competency into question no matter what side of the source code you’re on.

Right now I’m waiting for some clog in the repl.it environment to clear up so I can fix some things in the development version of the bot, but the beta version is still up and running on the Wombattery server. I don’t anticipate having to make any changes to the data or data structures, so I’m hoping I won’t have to wipe the data store in the future as I go about shoring up what’s wrong with the product.