Scheduling

One of the most important but least regarded skills for a technical person is the ability to let go of something you’re working on. I’m not talking about the micro scale of being able to accept interruptions, though that’s important too. What I’m talking about is the macro scale. A senior engineer typically has many assignments queued up. Obviously a good engineer will work on the thing that’s most important instead of the one that’s most fun. However, even good engineers often seem to overlook the fact that the priority of the task changes as you work on it. Sometimes the priority changes because in a sense the one task was actually several and only some of them deserved the higher priority. Other times it happens because some other task appeared, or had its priority bumped. Either way, it’s important for people to recognize when this happens and “shift gears” accordingly. I’m no advocate of dumping incomplete code on people (though I know at least one person who could accuse me of having done that to him more than once) but sometimes it really is more important to start something new than to finish polishing the apple you’ve been working on.

Sometimes I think the way that we schedule our own tasks could use a bigger dose of knowledge from the way we’ve learned to make computers schedule processes. Concepts of preemption and interrupts are obvious, though the nuance that runtime should be inversely proportional to priority might not be. Many people have also latched on to the concept of priority inheritance; if something you’re working on is a prerequisite to something that’s high priority, it’s only common sense that you should “inherit” that high priority. My favorite candidate idea that needs to be imported into the project-management world is priority bumping. In some scheduling systems, every time a task “loses” in a scheduling comparison its priority gets bumped a little, so that it eventually becomes able to compete with higher-priority tasks instead of being starved forever. Some network-routing protocols use a similar idea called “battle scars” to adjust packet priorities for similar reasons. It’s a valuable concept, but one that generally seems absent from the way people schedule people instead of processes or packets. Its application would seem pretty straightforward, and would allow “important but not urgent” tasks to get done in a timely fashion instead of the ad hoc approach of putting it off again and again and again until it becomes a full-blown crisis.

New Term

I’m a little surprised, given the outrage that is being expressed across the political spectrum over the Republicans’ determination to use Terri Schiavo as a political payoff to their religious wing, that the term “Ghoulish Old Party” has not yet made an appearance on Google. Mark my words: it will by the time the next election rolls around.

You Too

Lately I’ve been noticing that there seem to be two kinds of people who participate in online political conversation: those who recognize tu quoque (literally “you too” a.k.a. “two wrongs make a right”) as a fallacy, and those who rely on it constantly. Some examples might include:

  • It’s OK that Bush supports rendition or private social-security accounts because Clinton did too.
  • It’s OK that civilians are needlessly killed in Iraq because they (actually not the same “they” but that’s a different fallacy) killed civilians on 9/11.
  • It’s OK to torture people because you know they’d saw your head off if they had the chance.
  • It’s OK to discriminate against liberals in government because liberals discriminate against others in academe (yeah…against the dumb ones).
  • It’s OK that women are sexually assaulted in the military because it happens in colleges too.

I know people are going to accuse me of being partisan or elitist, but it does seem like there are distinct patterns regarding who’s likely to fall into which category. People who took a logic or philosophy class in college (who are also more likely to be liberal) tend to recognize these kinds of statements as fallacious. People whose “debate” style was developed by listening to talk radio (more likely to be anti-liberal) are more likely to see that such tactics are easy and often effective, and not recognize that they’re fallacious.

Strawmen and ad hominem attacks are pretty universal and not confined to any particular ideology. Tu quoque, on the other hand, seems to be a characteristically anti-liberal failing. It’s a particular instance of the survival of the meanest dynamic I’ve alluded to before, except that this time it’s survival of the stupidest instead. Can you say “adverse selection”?

Hayek on Social Security

After having encountered it one too many times in the chain of who cites whom, indicating that it is a seminal work in its area of laissez-faire/libertarian political philosophy, I picked up Friedrich A. Hayek’s Road to Serfdom at the library about a week ago. I’ve been reading it in my limited spare time since, making copious notes that I expect will form the foundation for a series of articles here at some point in the future. I’ve been pleasantly surprised by how much I agree with this idol of people with whom I typically (and often strongly) disagree. The explanation is very simple: his views and arguments have been most egregiously misrepresented by those people. In at least three places he very explicitly states that his arguments are directed specifically against socialist “planning” and not against more general leftist beliefs, but those who invoke his authority almost invariably misapply his statements to the latter anyway. If I were him I’d be livid.

What I really didn’t expect to see from Hayek, though, was commentary on a very current issue, and on the opposite side from where most of those who pose as his intellectual descendants seem to be. That issue, as you can guess from the title, is social security. On page 120 of this edition (hardback, with a preface from 1976) I found this:

There is no reason why in a society which has reached the general level of wealth which ours has attained the first kind of security [security against severe privation] should not be guaranteed to all without endangering general freedom.

Then, in the very next paragraph:

Nor is there any reason why the state should not assist the individuals in providing for those common hazards of life against which, because of their uncertainty, few individuals can make adequate provision.

Those two paragraphs neatly sum up the “safety net” and “insurance” aspects of social security (Hayek even refers to “insurable risks” while discussing the second) which are frequently ignored by his supposed followers. Many people see social security simply as a retirement plan, from which everyone should expect to receive direct benefits. If it were such a thing, they would be right to oppose it as an ill-justified form of wealth transfer from young to old, but that’s not why it was put in place. Relying on social security instead of sound financial planning is like relying on an emergency brake instead of the regular kind. Both are in place to keep bad situations from getting worse, but they are not designed to be used constantly and they lose their effectiveness if people do so. In my opinion, social security reform should consist mostly of discouraging people from abusing the emergency brake, not in making it less effective when needed or doing away with it altogether.

The icing on the cake is when Hayek says this, on page 132:

Some security is essential if freedom is to be preserved, because most men are willing to bear the risk which freedom inevitably involves only so long as that risk is not too great.

This is almost the same as my own favorite point about social security, which is that in addition to serving goals of justice etc. it can also create a stimulus for entrepreneurial activity among those who might choose a more conservative path if the risks for business failure were starvation for them and their families. It’s my favorite point not because it’s strongest (I don’t have any strong evidence that any such effect really exists) but because it seems least often considered. All of the rhetoric about social security tends to be about its economic down side, but hardly anyone even stops to consider that it might have an up side as well. Providing a safety net is not only the right thing to do; it’s the smart thing too.

It should be interesting to see how those who claim to respect Hayek try to disavow these rather unequivocal statements supporting something that anyone can see closely resembles the social security system they oppose. My guess is that, even though Road to Serfdom is the work people cite most often and therefore clearly the one from which they have drawn the greatest inspiration, some less notable statement in some less notable work by Hayek will be offered up as evidence that his real beliefs were other than those presented here. I’m not buying it. There’s no honest reason to believe that an author’s real beliefs are anything other than those stated in his most seminal work.

Uh oh

The big news today is that Amy has figured out how to crawl. She has been able to get into and out of the crawling position for a couple of weeks, and she has been taking single “steps” for a few days, but today she seemed to put it all together. On at least a couple of occasions she was clearly trying to alternate moving one leg with the other, and succeeding well enough to move at least a couple of feet at a time. As every parent has told me, now things get much more interesting. I’ll try to get pictures/video some time soon.

Schooling Scheule

Possibly in response to my comment cataloging his own fallacies in his previous response to Locke’s Marxism of the Right, Scott Scheule has written a more serious response. However, I don’t think the new response is actually any more convincing than the old one.

Lockeâ??s hypothesisâ??that â??libertarianism is basically the Marxism of the Right”â??is propped up with several assertions. By far the most prominent in the article is the idea that libertarians deny â??that there are other goods than freedom.â?

at this point I find Mr. Lockeâ??s foundations begin to give way. If freedom is the only important thing, then it stands to reason any libertarian will sacrifice any other good for any chance at an increase of freedom. The street libertarianâ??s every finance, his every waking minute, should be dedicated to the increase of his own freedom.

Such isâ??I assertâ??not the case, neither for â??culturedâ? libertarians, nor for the â??streetâ? libertarians whom Locke attacks. Libertarians do eat, they do attempt to find romance, they do have hobbies, they do have careers; they do, in short, pursue other goals besides the enlargement of their own freedom.

That is an entirely valid point, but not necessarily one that “deflates” Locke’s reasoning as Scheule later claims it does. For one thing, the above acts could be interpreted as the exercise of freedom, without which that freedom would be meaningless. More importantly, Locke’s error could be merely one of poor phrasing and not poor reasoning. If he had said that libertarians deny other values than freedom, he might actually be quite close to the mark (though “undervalue” instead of “deny” would be better still). Libertarians take their name from one and only one moral value, and in practice they often do seem to consider others beneath notice. Freedom, often of a very narrowly defined and mostly economic kind, does indeed seem to be the sine qua non of their philosophy…much as Locke claims.

Scheule’s examples do not “pop” Locke’s thesis if we’re considering values instead of goods, and therefore become cousin to disproof by fallacy. Scheule bases his entire counterargument on what might be merely a poor choice of words – literally one word: “goods” instead of “values” – and then seems to consider that an adequate response to the whole of Locke’s essay. I disagree.

Bad Site Design

I’ve finally figured out why I keep having password problems at my broker’s website. For a while now I’ve been absolutely sure that I used the correct user name and password, but most of the time my login would still fail. It turns out that the site is using one more piece of information to decide whether or not to allow my login: my IP address. No, not a cookie; I checked out. As long as I keep logging in from the same place (home or work) everything’s fine. As soon as I switch between the two, my logins will fail. Naturally, there is no mention anywhere of this being the case. I guess the brain-damaged intern who wrote the code only had access to one network connection, and it never entered his dim little brain to consider that others might not be so restricted.

How Many Bad Apples?

Apparently a couple more bad apples have caught torturing prisoners in Afghanistan.

Two Afghan prisoners in U.S. custody in Afghanistan in 2002 died after being chained up, kicked and beaten by American soldiers, The New York Times reported Saturday

The newspaper cited Army criminal investigative reports obtained by Human Rights Watch that have not yet been made public.

Maybe the Pentagon and CIA should hire some real apple growers, who would surely know the basic statistical techniques to determine how many of the apples are really bad instead of just assuming – as the hawks apparently think we should – that those we have found are all that exist. They might even figure out why so many apples are going bad, and try to do something about it. Of course, this all assumes that the Pentagon and CIA actually see this as a problem that requires a solution. As long as they don’t, and as long as they receive political support from the similarly immoral, there’s no chance that such a rigorous investigation will ever occur.

Globe Editors Asleep Again

The last couple of days, it seems like the Boston Globe has been trying out a new policy of using the left side of page D1 as a platform for personal attacks and general spleen-venting. Yesterday it was Alex Beam in Living/Arts, taking a rather personal shot at author Joe Queenan:

Sod off, Joe, before someone starts printing your pathetic sales numbers – 6,124 copies of “Country” sold, according to Bookscan…
…Wait until Katherine sees those awful columns you’ve been phoning in to Smart Money magazine. Then you’ll be in real trouble.

Yes, I know it’s opinion, but it’s not quite the kind of opinion I as a subscriber thought I was paying for. It’s just childish trash-talk, not deserving of publication. Get a blog for that kind of dreck, Alex.

Today’s offense by Steve Bailey in the Business section is even more egregious. One of the big news stories around here lately is about a couple of dogs who have been killed by electric shocks from the local utility’s equipment. The father of one dog’s owner is trying to sue NStar for rather large amounts of money – originally $1.4M (the CEO’s bonus) and now “only” $740K (his salary). Bailey’s column is about how the father was convicted for drug dealing and implicated in a series of home invasions, with passing mention of how his lawyer also defends sex offenders. He winds up with this gem:

Is he an aggrieved father or is he looking for the biggest score of his life? You be the judge.

Ick. Now maybe the father really is some sort of gold-digging bad guy. That was certainly a thought that ran through my mind when I saw a previous story about the suit. It’s such an obvious thought that hardly anybody needs Steve Bailey to waste an entire column on what amounts to an extended ad hominem attack. The question should be whether NStar is at fault, and what amount would be sufficient to make them take safety issues more seriously. Exploring those issues could have made for an interesting column. What we got instead was just another mean-spirited pot shots from one of the Globe’s loudmouths-for-hire.

As in my previous encounters with hypocritical idea thief Scot Lehigh, I’m left thinking that the Globe editors (not to mention subscribers and advertisers) are wasting their money on “professionals” like this when better content is available for free in a thousand places on the web.

A Way With Words

Catallarchy is fast becoming my new favorite blog, with two excellent posts in two days. One of my well known passions is for logic and the rejection of logical fallacies, so it was quite rewarding to read Libertarians Do It Laissez-Faire which dissects another article with wit and precision.

From my admittedly biased reading, Mr. Locke has gone out of his way to include a member of every genus of logical fallacy. Not only is the reader provided with commonly seen ad hominems, but also an encyclopedic collection of exotic and rarely seen strawmen, and strange species of circular reasoning. Here, one can spot a non sequitur quietly slithering through the paragraphs, there, a gaggle of post hocs, and even lovely begging-of-the-questions, in full plumage no less! And several varieties of outright strange points.

Of course, Scheule’s reading is indeed quite biased and he commits more than a few fallacies of his own even as he tries to refute someone else’s (I might get around to posting my own field report on those some time this weekend) but it’s a fine piece of writing nonetheless.

Another of my well known passions is the Prisoners’ Dilemma, so I found Prisonerâ??s Dilemma In College Admissions even more interesting.

For the campuses, itâ??s what game theorists think of as a security dilemma or an arms race. Whatever you think of early decisions, you canâ??t afford to slip in the rankings as other campuses take advantage of those programs. Your prestige will go down and youâ??ll attract inferior students. Then too, your improvement matters only insofar as itâ??s relative to what others do. They can match or exceed your improvement with their own, forcing you to struggle back with further expansion of early decision. For students, itâ??s what game theorists think of as a tournament competition. Rightly or wrongly, many students believe their life chances depend on going to the best campus they can.

Perhaps Scheule and Gertner should talk more, since it’s exactly this kind of game-theoretic quandary that in my opinion best illustrates the conceptual brokenness of libertarian/laissez-faire non-solutions…but that, again, is probably best left as fodder for a later post. For now, I only have time to recommend these two fine articles to others’ attention.