And no, letting the robot hit the wall 10,000 times in a trial is not a good result if it can be avoided.
Recently I was working on a robot in a simulator which was supposed to navigate a wall maze, follow a line, then go onto a platform. I got it to go through the maze just fine but for some infernal reason there was a problem with the line following. The sensors kept missing the line it seemed. I must have run it hundreds of times trying to make it work, I'd spent about 20 hours straight on it just failing over and over and over. This wasted a lot of my time and taxing my brain to that extent makes me very stressed and irritated. It makes my blood pressure go up and I can just feel the heat coming out of my ears! And the worst part of it is that since I spent so much time trying to get the line portion to work over and over, I never got to work on the platform bit. My goal didn't advance due to errors, it was halted due to errors!
This is time that I can't get back! There's less things that I can innovate now simply because I wasted so much time trying to fix something which should have been trivial. It is NOT ok to let something like that happen repeatedly and just sit there and say "gee maybe it's supposed to do that."
Edit:
And it ends up looking a lot like
"Let's see if this works" *clonk*
"Let's see if this works" *clonk*
"Let's see if this works" *clonk*
"Let's see if this works" *clonk*
"Let's see if this works" *clonk*
"Let's see if this works" *clonk*
"Let's see if this works" *clonk*
[...] *clonk*
Eventually it stops being "Let's see if this works" and sounds more like "Are you fucking kidding me"
Okay, I'll" bite " though my earlier comment wasn't intended to wound or irritate, just observe and report.
I usd to be a CAD tech working with 2D CAD. Now this was fine until my colleague sneaked a solid modelling program and started to include isometric views of the finished article which really did help visualise the resulting tool. I wanted to do the same but the solid modelling program wasn't there for me. So, confident that as the result was 2D then there must be a way of creating the view.
I spent a whole day and most of the next batting my head off the desk trying every angle of reflection I could think of and getting close but jot getting it right. In exasperation I spoke to my colleague who was much lore experienced in design and asked for help stating that I'd tried every angle I could think of. He replied simply "What makes you think it's just one angle of reflection?". Blew my mind at the time and made the job a lot harder but I eventually figured it out.
Do I consider my time wasted when the answer could have been simply relayed? Do I consider my failed attempts as nothing but dead ends? No. I consider the whole process worthwhile and not because I achieved my goal so much as I endured.
Okay so I hate the sappy line about "learning about myself " (yeuch) but that's pretty much what happened. It want what I was aiming for and it won't in the plan but it happened.
If I was more awake I'm certain I could think of examples where the robot hit the wall when it shouldn't and lead to something new being found. All from a "failure".
I just think we get too tied up in delivering right now to save costs and improve profits and to squeeze more in. It lacks... Appreciation. Something about the journey being worthwhile on its own merits.
And I'm sorry if it comes across as preachy or contrary to your point. I believed it a valid point and so posted it. I engage in no competition here. Just scrolling past and posting thoughts.