No, I like to stay agnostic about the underlying technology. How good the tests would turn out would not only depend on my ideas, but also on the underlying technology, and I just don't want to play the game of determining which. I prefer to stick to the theory, and leave it up to the engineers and practitioners to figure out the practical implications of my theory.Would you find this answer appropriate or snobbish?
The perverse incentives of academia
6 years ago
10 comments:
Thumbs down.
Given I work with implementing a certain formal theory, I know his comment is certainly valid. Implementation has a much greater effect on performance than formal complexity proofs. If he is a pure theorist, I guess it's a reasonable point to make, but definitely not the best way of phrasing it. Not having heard it in person, it comes across as defensive - a rehearsed rebuttal to a frequent criticism.
it sounds like a glorified cop-out to me....
CC - interesting that you point that out. Now that I think about it from the context you provide, I do have an officemate is entirely a theorist, i.e. he does proofs all day long. When we began talking about our respective work, he realized that a model used in his world to approximate a system and provide a framework for his theoretical work was entirely not based in reality and engendered gross simplifications.
He said, "so you mean, the model is basically useless?" I said, not exactly....the theories and proofs and bounds he does probably do provide some insight but nothing *directly* translatable to reality, and I suppose I would never expect him to then experimentally show that his bounding proof was true.
Still....the answer still sounds like a cop-out. I'm sure there's a better way to say it than, "let the engineers figure it out."
AA - the audience was not amused.
CC - he is a theorist, although in an area of research where many people (even his own gradschool labmates - I know some of them) do test their theoretical ideas in practice. I don't think anyone expected a full evaluation of the theory when it comes to performance, but a proof of concept of whether it would work to begin with wouldn't hurt. He was definitely being defensive in his response, but rather than put down the "practitioners," maybe he should think of a response that briefly explains on how one would go about testing it in practice. Even a "this is outside of scope of this work" would come across better.
PD - I think this is something pure theorists struggle with a lot, especially when presenting to non-theorist audiences (which an average CS audience is likely to be). He definitely needs to think of a way to answer that question better, without sounding offended and being offensive. That, plus if he was in fact to show that his theory works, people might actually try to use it - to me that would be a good motivation to make a proof of concept. Is that what makes me a "practitioner"?
In my field, numbers are everything and neither engineer nor practitioner would be offensive/patronising. I'm guessing that in that situation/tone of voice it was. I'm more likely to hear the same sentiments expressed in terms of self-deprecation, eg "Hopefully this will be useful to those who actually make things work".
I'm going to have to give it an eye roll and a snicker. (the laugh, not the chocolate)
CC - it is amazing how much difference wording and non-verbal communication makes! This was a great example of a foot-in-mouth.
UR - didn't expect you to share actual chocolate ;) An eye roll was hard to hold back after that answer!
seminar answer FAIL
Cath - too bad we can't submit it to the FAIL blog ;)
Post a Comment