Hacker News Reader

Comment by bcrack | Hacker News Reader

>"could it possibly be … that the best way to get good research and publications out of scholars is to hire good people, pay them the going rate and tell them to do the job to the best of their ability?"

Sounds very reasonable. Does anybody have good examples of this approach? I have the impression that this is similar to the environment of Bell Labs during the "golden years".

View on HN

[Google, Facebook, Microsoft] Research are probably similar in many ways -- well funded by rich companies that are self-interested in moving the field forward. We'll only know their true impact decades down the line, of course. MapReduce has certainly been very influential.

Bell Labs wasn't just a playground with infinite cash, they were very much supposed to figure out telecoms stuff for the benefit of Bell -- they just had a very wide berth for determining what that something was. I suspect the same is true for the aforementioned research departments.

(That said, I don't think either, nor Bell Labs, is a meaningful model for academia in general)

You're going to need some sort of controls, if only to kick out the folks who retire in place.

If one were being truly radical about this, a possible solution is to replace researchers' salaries with pensions -- remove the incentive to retire in place by allowing you to retire at home if that's what you want.

And then you have a pool of professors made up of people who can afford to teach for free, or perhaps create a system where some finance leech buys your future pension for NPV payouts now.

a system where some finance leech buys your future pension for NPV payouts now.

Isn't that called a loan?

I imagine that smart finance-leeches will tend to avoid loaning against 100% of the value of the pension, for much the same reason it's hard to get a mortgage that takes 100% of your income to repay. Beyond that -- where is the problem?

I said NPV.

Beyond that, the problem would be getting the same results as paying someone directly, while reducing efficiency by inserting a middleman to take a piece for... why, exactly?

And there it starts. Don't do that. In Germany we had this discussion in the 90's. All kinds of measures where taken into place to catch the slacking professors. Now we are left with a system the most talented don't want to work at.

Most academic people I know love their work. They do overtime although they are paid much less than a comparable position outside academia. Just live with the 5 percent or so slackers. The rest of the pack will squeeze out much more when not tortured with bureaucracy.

Have the teachers rate each other anonymously. The cream and dross will emerge, as will hard workers and shirkers.

If you define "the cream" as those who play the politics game best, then yes.

You must rate teachers. Using explicit external criteria, which as the article indicates, is inherently manipulative and biased, and thus counterproductive. Or you can use implicit internal criteria -- feedback from the participants who were hired to embody the objective function, to walk the walk. Given the tradeoffs, what better way is there to judge than anonymous peer review?

Retiring in place is actually quite boring, and the feeling of uselessness that results is depressing. People won't actually do that unless the alternative is worse. If you hire people who are passionate about whatever it they were hired to work on, they might not do work that is profitable, but they'll do work that is interesting and has some benefit.

I believe you undevalue the passion some people have for not working.

Those people will still have a passion for something, or they're going to be profoundly unhappy. Consumption (whether it be TV, video games, drugs, etc) can only provide some distraction from the pain, it doesn't make it go away. In fact, the longer you try to deal with the pain by distraction, the worse it gets.

Most of Polish professors are actually retired in place. They do the minimum required amount of teaching and put their names on (usually completely unoriginal) papers of their subordinates. Most of them didn't do any significant original research in decades. To pass the time (and to gain power to protect themselves from being kicked out etc.), they take on administrative functions such as deans, heads of units etc.

Um, no, now you are playing the wrong game.

See, we have two ends to optimize the system and we have to choose one. Either the system encourages the most talented to thrive or it's designed for mediocre individual performance where the worst performers are kicked out. In the former high performers thrive while in the latter the best we get is guaranteed non-zero output.

There is a beautiful dualism in rewarding and measuring. It's actually quite hard to invent good measures for creative work, so whatever measure you come up with probably does not measure what you want it to measure. The second problem is from intrinsic versus extrinsic motivators. Extrinsic motivators like pay rises or the risk of losing ones job work fine only if the task is simple and mappable to linear performance - like, say, logging. For any creative efforts extrinsic motivators kill performance.

Thus, it's very hard to measure creative work, and even if you could, you would not really want to use that data for anything because that would kill performance.

A smaller-scale UK based example (which gets a mention in the blog comments) is the MRC's Laboratory of Molecular Biology in Cambridge. Still very much a going concern although I suspect it's culture has changed somewhat over time.

How do you know whether someone is "good"?

Heh, if that's all there were to it, I'd be back in academic right away! Go Speed, Go!

I would agree ... nowadays ... not sure the hiring function of any institution is capable of doing this anymore ... it's sad that the good old Bell Labs days seem over ... how this impacts the future remains to be seen though ...

Reply on news.ycombinator.com