Assorted links


6. "They block young people." I'm continually amazed how unself-aware people are about their problems. I'd like to tell people "ya, I could have told you your solution 10 years ago, and also save you all the irrelevant words you are decorating your inevitability with. And as I remind my wife, this is so frustrating precisely because I'm not smarter than anyone else. Tenure never should have protected 6-figure salaries for professors.

With a dirt detection camera the challenge doesn't seem spotting the speck but avoiding false alarms.

It's pretty trivial to tune such a system to the appropriate level of alarms, once you get the resolution and speed high enough. And in the case of inspecting cars, speed is irrelevant. Visual inspection systems for packaging tins have to handle conveyor speeds of greater than one per second (sometimes quite a bit greater). Car lines increment at 1 car per 30 to 90 seconds.

Yes, OTOH a tuna tin can probably tolerate a blemish more than a new expensive car.

The alarm set-point problem ain't trivial at all: In some cases no matter what your set point you'll never get an acceptable rate of Type 1 and Type 2 errors. You can't just through more speed and resolution and hope the problem solves itself.

Certainly you don't want a system that causes an excessive amount of Type 1 errors (aka false alarms), but the previous standard was a manual inspection of every body panel for each and every car within the cycle time, typically 30-90 seconds. My point was that 30-90 seconds is not a lot of time to have a human inspect an entire car skin for minute blemishes, however that's a very long time for a high speed, high resolution video system to check the same skin.

So the current systems are designed for a high level of alarms (multiple false alarms), so they tend to catch everything that might be a blemish, then the operator manually inspects what the system has flagged. This drastically increases the success rate of the inspection process by having the machine identify the less than 1% of the cars surface that has blemishes on it.

6. If tenure is about protecting academic freedom, then it should be uncontroversial to give university administrations the power to decide when eligible faculty must shift to emeritus status. Emeritus status almost always allows for library access and guaranteed pension benefits. It often also comes with office space. Becoming emeritus does not meaningfully restrict anyone's ability to speak or write on issues of interest.

6. Let's get ''block young people'' out of the way as a classic lump-of-labor fallacy. That leaves two issues: old professors cost more, and some old professors are deadwood. For cost, keep in mind that tenure track contracts are voluntary agreements. One can argue just as well that when these old folks were young asst professors they were underpaid. The agreement is for the long run. You cannot pick and choose parts of the contract. Moreover, in most places pay raises are merit-based, so it is quite likely the old professor will not get a raise for a decade if he is not very productive (given lean budget times).

OK, finally this leaves the deadwood. Many universities ask their less research productive faculty to teach more, to make up for it. Also, deadwood don't get pay raises so over time their real salary is less impressive and costly for the university. Or the university can buy them out by offering them a year or two of extra pension income.

The autism is strong with this one.

This being Marginal, keeping in mind Prof. Cowen's blogposts on the subject, I'll take that as a compliment!

6. Tenure track contracts are voluntary agreements. Maybe they are overpaid when they are old, but they were underpaid when they were young. You can't pick and choose bits of a contract you like.

And the claim old folks take away jobs from young folks is just the lump-of-labor fallacy. If anything, retirements are replaced with piecemeal lecturers.

Faculty salaries are not the reason higher ed budgets have exploded. Asking them to teach more is a reasonable request, but look at where the money goes.

This boils down to deadwood, old or middle-aged. I have not seen evidence that there is a lot of deadwood among older faculty. Being lazy starts younger. Anyway, offer old faculty a couple years of pension income and most will take the buyout.

(Previous comment did not show up. Apologies if it does later.)

Yeah, a couple weeks ago a few hours of delay was introduced in posting comments. I don't know why, nor whether it was delibrate.

Delibrate or not, sure is annoying.

That soccer-field picture is cool. But why is there a road between the field and the grandstand?

And whose job is it to fetch all the balls that end up overboard?

Because it's the F1 Grand Prix stand as well.

See also: the rising value of land in Singapore!

4. Perma-smile surgery.

A solution for Bitchy Resting Face

#4 there reminds me of Joo Dee from Ba Sing Se.

The floating soccer field isnt a product of land prices. The platform is used for concerts and events with the bay and city as a backdrop. When not in use they lay some artificial turf and let local soccer clubs kick the ball around.

I remember reading somewhere that the government built 75 artificial turf fields around the city at a million a pop. Made me think of my Canadian hometown that agonized over building just one because of the cost.

Re: 6. Given that the purpose of tenure is to protect academic freedom:

(1) Most people, including many with lower incomes than professors, retire at about age 65. Thus, there is a reasonable expectation that professors accumulate sufficient retirement savings by age 65 to live reasonably comfortably, even without working after age 65. A person with sufficient retirement savings does not need tenure to express themselves freely because they are not financially dependent on their job.

(2) Suppose tenure had some finite length, say 35 years, instead of lasting one's entire life. It seems like 35 years should be enough time for someone to push out plenty of their most controversial ideas. That's at an individual level. At a societal level, since tenure periods would be staggered, there would always be enough professors that are far enough away from the end of their 35-year period to ensure lots of open debate and generation of controversial ideas.

(3) Even if one objects that not all 65-yr old professors are financially secure, certainly at least some professors (of any age) are financially secure enough that they don't need to work to live comfortably and, hence, don't need tenure for academic freedom. We already require that students and their parents disclose detailed financial information to be eligible for financial aid. Therefore, it's reasonable to ask professors to disclose detailed financial information to demonstrate "financial need" for tenure. Once a professor no longer has demonstrated financial need, then his or her tenure could lapse.

Length of tenure and financial need for tenure correlate with age, but not perfectly. Even if ending tenure for seniors doesn't make sense, we could still protect academic freedom with finite-length and need-based tenure.


This sounds exactly right to me. Keep in mind that retirement for tenured professors is called "emeritus" and comes with sweet retirement perks, including things like library access, pension payments, and perhaps even some sort of office space. Just because you're not "tenured" any more doesn't mean you can't keep on publishing whatever you want!

Tenure is going away anyway. Tenured and tenure-track faculty are now less than a quarter of teaching staff. Meanwhile administrators increase.

To be precise, professors are being replaced with lecturers. There are fewer professors and more lecturers, but among professors, tenure is just about as common as ever.

Comments for this post are closed