Small world

[maradydd]

There's a post up on BoingBoing today (ok, yesterday for me) about open vs. closed search algorithms, suggesting that the search algorithms used by Google, Yahoo et al are bad because of their lack of transparency. It invokes a comparison to an important concept in computer security: "security through obscurity" is dangerous because an effective encryption scheme should be equally hard to break whether you know the internals of the algorithm that generated the ciphertext or whether you don't.

I think comparing this to search is a bad (or at best misleading) idea, and expounded on this in the comments. But I'm far more entertained by the fact that the two best comments on the post so far come from two sources with whom I am tangentially familiar, albeit from totally different directions: jrtom and radtea. Small damn world!

Comments

[jrtom.livejournal.com]

I think you're overstating the case, or at least overgeneralizing. From personal experience I can state that _some_ academics, at least, do in fact care about practical considerations. I have been one of them. :)

(If nothing else, you might consider that quite a few academics have ties to industry.)

[enochsmiles.livejournal.com]

From personal experience I can state that _some_ academics, at least, do in fact care about practical considerations.

Yeah, me too. I can actually one-up you: I *am* one. But hell if I'm going to stay one, and I gather from your comments that you're in industry now yourself.

Real academics don't leave industry: they go on to get their PhDs, do a few postdocs, become a professor, find a tenure-track position, get tenure, and during this entire process train other baby academics to grow up and be paper-producing machines who don't care about how things work "in practice" because they're concerned about "the bigger picture".

Academics who care about making things feasible wind up not publishing a lot of "good" work because they waste their time on mere implementation details, and thus have to choose: worry about things working in practice, and wash out of academia (perhaps go find a job at a company that is actually doing neat theory things in practice, and be happy as a clam, except for the whole not publishing anything business), or give up on that practical consideration angle, at least most of the time, and fight the academia battle.

Yes, of course there are exceptions to this and any generalization. But I think I am characterizing things pretty well: in the end, you either have to pick, have to not care, or have to get really lucky when it comes to "papers vs. practical work."

[jrtom.livejournal.com]

Hey, if you want to do one-ups, go right ahead. I'll just sit here. :)

You are correct that I'm now in industry. I may go back and finish the PhD at some point, but for now it's on indefinite hold.

I do know other academics that are strongly concerned with practical considerations, however. Some fields promote this more than others. For example (while this is outside my personal experience), I suspect that you'll find very few civil engineering professors that aren't concerned with practice. Ditto for most engineering disciplines.

If you want to constrain this to fields in "computer science", I would still say that there are areas in which academics are likely to be deeply interested in applications. Data mining, for one. Machine learning. Compiler design (I conjecture). Software methodology. In any of these fields, of course you'll find the theoreticians, but the fields themselves put a fair amount of emphasis on practical demonstrations and meaningful experiments and metrics, and if your papers don't reflect that, then you may have trouble publishing them.

In lots of other areas, I completely agree that academics tend to be less concerned with practical details--which, IMO, is as it should be (someone needs to do basic research that isn't immediately concerned with practice, if for no other reason than that we don't always see applications immediately).

As a point of interest: the on-site reviewers for the government grant on which I worked the longest while a PhD student were consistently most impressed and happy with the software library (JUNG) that I co-wrote while on that grant, although they thought the theoretical/algorithms/methods stuff was good too. (My supervisor, who's done a good job of nailing a foot in both the theoretical and practical side of the fence, initially thought that I shouldn't be spending as much time on JUNG. He changed his mind once the reviews started coming back. :) )

[maradydd.livejournal.com]

I would still say that there are areas in which academics are likely to be deeply interested in applications. Data mining, for one.

If you know of a university that has data mining professors who pay more than just lip service to giving a damn about applications, please let me know in case I ever decide to go back and finish my own PhD. (I left Iowa largely because of an advisor who flat-out ordered me not to work on OBELisQ because it "wasn't useful and couldn't be done anyway.")

[jrtom.livejournal.com]

Padhraic Smyth, UC Irvine. (My former supervisor.) He'd also be a good person to ask about professors elsewhere with a similar mindset; he's got excellent connections in both academia and industry. (As an amusing side note, he also used to be one of the people for whose names Google bought job ads on their own search page, along with people like Vapnik. :) )

U of Washington has a few, I think, but I'm a bit out of touch there. I suspect that UC Berkeley and other SF-area departments also have some such.

What you want to do, I suspect, is find a department in which the professors have a tendency to have strong ties to industry--this may be manifested by consulting jobs, or part-time appointments, or oscillations between industry and academic positions, or students that go on to work in industry.

[jrtom.livejournal.com]

Oh, and in DB stuff at UCI, Sharad Mehrotra. (Sometime collaborator with Padhraic.)

[enochsmiles.livejournal.com]

I do know other academics that are strongly concerned with practical considerations, however.

They won't last as academics. ;)

I suspect you're right about civil engineering and engineering having a higher percentage of people interested in practice -- that said, handwaving-away a problem for the sake of publishing a paper isn't beneath any academic in any field, and the problem continues to exist because the program committees for conferences put up with it.

As for experiments and metrics, sure, emphasis is placed on them for certain fields and certain conferences. That said, "meaningful" is up for debate. What's meaningful to an academic is not necessarily meaningful to an implementor.

[And as a side note -- you've never driven in Brussels, I assume, or you wouldn't have been so certain about civil engineers and practice. ;) ]

He changed his mind once the reviews started coming back.

Oh, definitely -- if "practicality = more grant money" then they change their minds in a heart-beat. Sometimes they even try to take credit for the whole idea in the first place. But I ask this: what percentage of students listen to their advisors, vs. the percentage that stick with the practical work? I'd like to think most of them stick to their convictions and plow ahead, and if I sample my friends in academia, that seems to be the case. But that could also tell me that the academics I am friends with are stubborn people who care more about practicality than finishing their PhDs.

I wonder if a sociologist has done any studies on this.

Nah, too practical.