Quite often my students will come to me and say “I have a really cool idea for an algorithm”. Most of the time, after hearing the student out, I will make suggestions about something similar but different to try.
The only reason I can do this is that, in most cases, I’ve already tried something similar to what the student is proposing — sometimes quite a few years earlier. Rather than make the student go through a month or six of pain, frustration and failure, I draw upon my own past experience to steer the student toward an approach that is much more likely to produce a positive initial result.
None of these experiences upon which I am drawing are publishable. You won’t find them in the literature under my name or under anybody else’s name. These are the research failures, the approaches that at first seem plausible but which prove, after hours of hard work, to be dead ends.
It’s a shame that there is no academic forum for sharing this valuable lore. You don’t get published for reporting things that do not work.
On the other hand, the very fact that you cannot find out these things on your own means that we professors are a valuable part of the process. I don’t know if this is good in the larger sense, but it certainly helps me to feel useful.
I consider research / implementation failures to be as valuable as successes. It would be very valuable to have a record of experiments that did not work in various fields.
I develop statistical / ML algorithms for trading (though my real passion is CS). Across a career, a large % of these sort of experiments ended in failure, though a few have performed spectacularly. This is especially true in the financial markets where all problems are open problems, however, we have these sort of problems throughout the sciences.
There is actually more value in knowing about the failures rather than the successes, because the amount of time and cost associated with these is far greater.
Yes, exactly! Well said.
There is a place, it’s called The Internet. Specifically, sites like StackOverflow allow newbies to ask the question publically, and then other newbies (and oldbies) can come and see an answer that answers more than the question they asked.
Increasingly, I’m inclined to think something up, then go search for it. 99% of the time, someone has asked the question before, had the idea before, and the resulting discussion is an education in itself.
So go, Ken, and be the answer that teaches; you might need to filter for ‘algorithm’ and ‘math’ but you’ll find someone trying to optimise an inverse square root in javascript with trig, or pondering some sort of tree datastructure for 3d rendering.
This is the wonder of the internet. When it is asynchronous and many-to-many, the tacit is worth discussing. For me, it works like an enormous ‘I wonder what would happen if’ system.