A long conclusion
We did it, another “long experiment” in the books. Total time? Twelve hours. Twelve, non-stop. Okay there was exactly one bathroom break, one coffee drank and one protein bar eaten, but I kid you not, that was all there was time for. When I say long experiment, I mean LOOOOOONG experiment. 98% of that is standing, well moving around doing things, so let’s say “not seated” since that’s more accurate. So, was it worth it? Well now, that’s the real question isn’t it?
The first time we did this experiment it was a mess. I’m not even sure we got useful data. For the record, this experiment involves dozens of experiments within the larger experiment, three sets of equipment (one from another lab), a few dozen people, and a whole lot of luck. The second wasn’t much better, in fact afterwards I was in pain, literal couldn’t sleep pain, for about a week. Sometimes we learn best by doing and an idea is forged into reality by trial and error, the earlier experiments were more error than trial, but we’ve gotten better at it.
Honestly, I can’t complain since long experiment was my fault. So a cautionary tale before we talk about today.
A year ago I had an idea, a “big idea,” it was so big we’re in the process of trying to get data together for a nature paper, like nature, nature, the top journal nature. Personally I’m excited and it’s going to (hopefully) change a lot of things. Because it’s such a big idea, a lot of other labs that we collaborate with also saw potential for other projects. Thus, long experiment was born, fully funded by a collaborator.
While the experiments will help with our (hopeful) nature paper, it’s not the purpose I had intended for big idea. The good news is that I’ve gone through three different iterations of big idea and the third version, which I haven’t even talked about until now really (I think anyway, I’m tired, go easy) is a huge change, because we can always do better. So all this trialing big idea in other ways has really helped cement different aspects that could be improved and equipment that could be built. I think tomorrow we’ll dive more into that, assuming I have time. Tomorrow is unfortunately going to be busy, both for personal reasons, i.e., more home repairs and for school reasons i.e., my DARPA deadline is just four weeks away (eeeeeek!).
So in short, big idea has a lot of interesting applications, not all of which I had considered when I came up with it. But it’s been a mutually beneficial thing and I’m thankful to everyone who’s helped big idea become a big reality (hospital-PI, collaborators, all of you for your support, etc.).
Twelve hours up and about made for a very long day, I think I said that, but I don’t think I’ve made that clear enough, because wow was it long. The experiment kicked off bright and early at, let’s say 7am ended twelve hours later at 7pm and aside from hospital-PI and one of our new lab hires, I saw absolutely no one from our lab. In fact, I barely saw the sun today, because that’s the kind of day it was. Of course I got like zero sleep (again) wracking my brain about the stuff I needed to complete for today, but we managed anyway, mostly thanks to an afternoon coffee that I basically downed quickly between setting up for anther part.
After roughly two weeks worth of prep (in my free time because nothing is ever that easy) I managed to get everything done and then some for the experiment. Everything was organized, labeled, and ready to go. It went as smooth as it could on our part (not so much on parts where other people were involved, but that was more bad luck than anything else).
Overall I think we managed to get it done as fast as we could. Hospital-PI said I earned two PhD’s for my effort, bringing the total number of PhD’s I have to zero. For now anyway, but that’s another post altogether. Mostly he was impressed with my labeling system and part of it was inspired by how the OR sets up everything for surgery and I thought, if the system worked well enough for them, why not try it for our experiments too?
While we did have some minor computer problems, to the point that I’m slightly worried about the computer we use. It literally started sound like a cell phone on vibrate. I’m not even sure what was causing it and it stopped as quickly as it started. In fact it only did it a few times at the start of the experiment and then the computer behaved until we finished. Unfortunately we need it, so we either use it until it breaks and we get a new one, or we get a new one and I try to fix it. Lab gremlins, they’re very real, but it looks like they took pity on us because we managed just fine.
Now if you’ll excuse me I’m going to sleep for a few years. Wake me up when I graduate.
Sounds like you’ve come a long way from the first attempt at “long experiment.” Might be time to stop being surprised when things go well 😉
I’m not sure why I’ve never thought to ask this, and maybe it’s a question you can’t answer yet, but … why is long experiment so long? What about it stops you from doing the work in multiple sessions across different days?
Off-topic, but I watched part of the Tesla “AI Day” presentation today and thought of you. They were showing off the latest humanoid robot prototypes, displayed a model of their knee and talked about the four-bar linkage. But their knee looked like yet another “floating” variant; I don’t think it has a large load-bearing contact surface like yours.
LikeLiked by 1 person
October 1, 2022 at 12:52 am
Haha, thank you! Yeah I guess, but we’re researchers and we’re superstitious so we cannot expect things to go well!
I’ve been wracking my brain for a way to explain it without going into too much detail, but the sort answer is there’s a very specific time window we have to do what we want, it can’t be put off for other days. Cost is a huge factor as well.
Oh I heard Tesla was planning on showing off their robot, I wasn’t sure it was actually going to happen though, that’s kind of cool. The solutions to the four bar linkage problem with contacting surfaces are very difficult to find, so I’m not surprised. That’s also probably why my design is still one of a kind years later.
LikeLiked by 1 person
October 1, 2022 at 10:36 am