Counter argument. Never leave things up to assumptions. You tend to not get what you want when that happens. Sabrina can make mistakes and oversights, to the point im not comfortable leaving the safety of allies to brina pilot.
What comes off as common sense for you doesn't mean others hold the same standard. Better safe then sorry.
My vote specifies "fast-setup", takes care to avoid all but the simplest of pure-physics effects, repeatedly mentions maintaining evasive actions and alertness, and even proposes low-tech suggestions or handing tasks off to Mami if we're still not being efficient enough. Past that, we leave the implementation up to Brinapilot. You will note that I don't instruct Sabrina to not run into walls. You will note that I don't remind her that not getting hit is a good idea. I also don't tell Sabrina how to organize that pincer movement, nor do I suggest a search pattern or any mental patterns that would help her avoid tunnel vision so she can follow the directions about coordination, updating allies, or watching for traps. I don't suggest any way to handle hailing our opponent other than saying to do it and what our goals are with that action. I barely specify anything, in fact, past the high-level goals and some illustrative examples to show Brinapilot the available possibilities in case none of our listed options are perfect. There are a
million things that we could micromanage in this vote. Why don't we do any of them? All of those are just as important, and just as likely to be screwed up, as what you're worrying about here.
The answer is that we
shouldn't be trying to micromanage to that degree, nor with that mindset. Asking "What if we fail to specify this minor detail" guarantees that vote-crafting is driven by fear and that failures are answered with vicious scapegoating and vitriolic finger-pointing, leading to stress and rage and toxic thread climate.
For that reason alone we have been
instructed by the QM to provide more general votes and to let Brinapilot handle the details. And beyond that, micromanagement isn't
effective. For all my recent annoyance about vote-points that seem to have been missed, when Brinapilot does execute on something, she tends to do so more effectively than we ever could. She can act based on the information that's available to her senses where we're stuck with what the words of the story convey to us, she can act on a moment's notice instead of having to wait for a full round of voting, and in many cases - social interaction being a large one - she's just plain
better than we are.
So no. I reject the position that we should "never leave things up to assumptions". I think that we need to spend our time and our sanity efficiently and sustainably and that fear-driven micromanagement is one of the least efficient and most stressful things we can spend our scarce intellectual resources on.
Then it's a good thing Veb's vote specifies "fast-setup" for grief sensors, isn't it?
And even if that doesn't work, there's a reason that it's just one of several totally divergent alternative approaches that I'm suggesting for the task.
To flesh out "fast-setup", while I'm thinking about it - the hyperspectral scanners we built earlier were high-performance monsters, spectacular sensitivity way up and down the EM spectrum with high resolution and a wide field of view. We don't need any of that. We want a couple frequencies of infrared for "night vision" and thermal tracking (maybe try to pick up
footprints?), throw in a second array for microwaves around the 100MHz band that'll penetrate walls, both with just enough resolution and FoV for us to keep a lock on our target as she maneuvers. Needing four or five fewer OOM on the sensor count won't do much, we're already leaning on parallelism hard, but the reduction the number of
types of sensors we're building - three or four instead of a full spectrum - is a dramatic complexity reduction and should cut the time significantly.