» moving the blog

My blog is moving. I will be posting new entries over at method-combination.net. I have been thinking about doing this for a while; free-loading off of cs.rose-hulman.edu doesn't seem quite fair to them (lifetime accounts to alumni notwithstanding). And ssh access to the CS machines has been quite flaky as of late, so now seems like a good time to switch.

posted by Nate @ 7:33 PM [ 2 February A.D. 2007 ]

» planet rgsb trouble

Planet RGSB readers, beware: Planet RGSB has been eating some of my posts when I make multiple posts per day...so if you are getting your dose of irateness through Planet RGSB, you may occasionally want to pop over to my actual blog to see if there's more irateness to enjoy.

posted by Nate @ 9:40 PM [ 25 January A.D. 2007 ]

» numbers pseudo-science

Neil Postman, in Technopoly:

What we may call science, then, is the quest to find the immutable and universal laws that govern processes, presuming that there are cause-and-effect relations among these processes. It follows that the quest to undertand human behavior and feeling can in no sense except the most trivial be called science. One can, of course, point to the fact that students of both natural law and human behavior often quantify their observations, and on this common ground classify them together. A fair analogy would be to argue that, since a housepainter and an artists both use paint, they are engaged in the same enterprise and to the same end.

The scientist uses mathematics to assist in uncovering and describing the structure of nature. At best sociologists (to use one example) use quantification merely to give some precision to their ideas. But there is nothing especially scientific in that. All sorts of pople count things in order to achieve precision without claiming they are scientists. Bail bondsmen count the number of murders committed in their cities; judges count the number of divorce actions in their jurisdictions; business executives count the amount of money spent in their stores; and young children like the count their toes and fingers in order not to be vague about how many they have. Information produced by counting may sometimes be valuable in helping a person get an idea, or, even more so, in providing support for an idea. But the mere activity of counting does not make science.

Nor does observing things, though it is sometimes said that if one is empirical, one is scientific. To be empiricial means to look at things before drawing conclusions. Everyone, therefore, is an empiricist, with the possible exception of paranoid schizophrenics. To be empirical also means to offer evidence that others can see as clearly as you. You may, for example, conclude that I like to write books, offering as evidence that I have written this one and several others besides. You may also offer as evidence a tape recording, which I can supply on request, on which I tell you that I like to write books. Such evidence may be said to be empirical, and you conclusions empirically based. But you are not therefore acting as a scientist. You are acting as a rational person, to which condition many people who are not scientists may make a just claim.

This quote comes at the beginning of a chapter on “Scientism,” which, as Postman defines it, consists of three ideas: one, “the methods of the natural science can be applied to the study of human behavior.” Two, “social science generates specific principles which can be used to organize society on a rational and humane basis.” Three, “faith in science can serve as a comprehensive belief system that gives meaning to life, as well as a sense of well-being, morality, and even immortality.” It's pretty clear which side of the fence Postman (and I, for that matter) fall on.

I read this book on the way back from Palm Springs and enjoyed it. I think his conclusion falls into the same category as that reached by Samuel Florman in The Existential Pleasures of Engineering: “we ought to study and enjoy what we do because, well, it's what we do!” Which doesn't really get at why. I do appreciate his emphasis on the history of things as a means of understanding the present situation--that's definitely something I'd like to explore as a means of education as Becca and future children are homeschooled.

posted by Nate @ 9:20 PM [ 25 January A.D. 2007 ]

» business phone

One of the unexpected pleasures of having a business phone line is being able to tell telemarketers that they have called a business line. They are usually quite shocked, apologize, and then pledge to take the number off of their lists immediately. I have contemplated doing the same thing for our main phone line, but cannot convince myself of the honesty of it.

posted by Nate @ 9:05 PM [ 25 January A.D. 2007 ]

» zwingli the cat

Our friend Josh (who some of my readers, if I have any left, may know as the Fearsome Pirate) is going to a Lutheran seminary and left his cat, Fang, with another one of our friends, Jerry. Upon learning of this fact, I immediately decided that the cat needed a new name: Zwingli. (That would be Ulrich Zwingli, for those of you who are not conversant with the Protestant Reformation.) Ever since, whenever I go to Jerry's house, I make a point of calling the cat Zwingli and reminding everybody who interacts with the cat that the cat's name is Zwingli.

I got the chance to tell Josh of his newly-renamed pet at the New Year's Eve party Jerry had at his house. The conversation went something like this:

Me: So, Josh, we renamed your cat.

Josh: Oh, really? What did you name it?

Me: (stifling a giggle) Zwingli.

Josh: (blank stare)

Josh: Hey, Jerry, what are these called? (motions to the taquito he is eating)

Jerry: Taquitos.

Josh: Right. Taquitos. Anyway, I'm standing here, and this is a taquito, and I'm eating it. But in my mind, it's a knife, and I'm stabbing you in the heart with it.

The joke is complete. I can stop reminding people now.

posted by Nate @ 9:09 PM [ 6 January A.D. 2007 ]

» counting to pi

Becca can now say the value of pi. (3.1415926)

Actually, that's not quite true. What is more accurate to say is that Becca knows the numbers from 1 to 9 and can recite them given small amounts of prompting--we say one number, she says the next one in sequence. I think this is pretty impressive for a child slightly over a year and a half.

When I told Tricia about this post, we decided to try to get Becca to actually say the digits of pi after us. She said three, but then decided that when we said “point,” we meant for her to point at me. Becca thought this was very amusing. We gave up on “point” and proceeded to say, “one,” to which Becca promptly replied, “two!” (Or rather, as she says it, “cwu.”)

So we don't quite have pi down yet. But we're on our way there.

posted by Nate @ 9:05 PM [ 6 January A.D. 2007 ]

» lisp is a black hole

James Gosling in a post about closures in Java writes:

I have somewhat mixed feelings about closures: they are pretty complicated. But they're an instance of what I think of as the Black Hole Theory of Design. I have a really strong memory from years ago of Guy Steele saying roughly “Lisp is a Black Hole: if you try to design something that's not Lisp, but like Lisp, you'll find that the gravitational forces on the design will suck it into the Black Hole, and it will become Lisp”. [Guy doesn't actually remember making this remark, but he does say it sounds like the kind of flip comment he would make; I could also be totally mis-remembering who said it, but I haven't found any quote like it through Google].

I suppose you could also interpret this in a negative light and point out that any programmer who spends enough time in Lisp will get sucked in--so much so that said programmer eventually stops seeing the light from other programming languages.

posted by Nate @ 10:31 PM [ 21 October A.D. 2006 ]

» working from home

A few miscellaneous links that are somewhat related to working at home.

One of the things you hear people talk about a lot with regards to working from home is how to keep yourself motivated without somebody “looking over your shoulder.” Another thing you'll hear about is the lack of contact with other people. I guess it shouldn't be surprising that people have come up with coworking: several people with unrelated jobs come together to create “a shared office and add cafe culture.” I can see the benefits to this...

I thought I was doing pretty good in the comfortable office category when I got a used Steelcase Leap chair and a nice desk from Target. On the contrary! What I should have purchased was a 1966 23-foot Silver Stream trailer. Nice.

Related to the second paragraph above is this Joel on Software discussion thread about working from home. Lots of tidbits in there, but it seems like all the advice about working from home comes down to:

Is that really all there is to it?

posted by Nate @ 10:08 PM [ 20 October A.D. 2006 ]

» underhanded c contest

The Underhanded C Contest is similar in spirit to the Obfuscated C Contest, but with a twist: instead of writing obfuscated code, you write straightforward code that cleverly conceals algorithmic or security pitfalls. The past year's challenge was to write code that runs fast on one machine but slow on another--in straightforward, can't-hide-anything code. Very interesting.

posted by Nate @ 10:14 PM [ 11 October A.D. 2006 ]

» still kicking

Yes, I'm still here. I think Ryan's recent post will be enough excuse for me. I could talk about the sorry state of feed aggregators on Linux, or why stable sorts are good for you and pointer hashing is bad for you, or about my new job, or our new church, or any number of things.

Instead, I've modified my information over there to the right. You'll have to be content with that for now.

posted by Nate @ 8:22 PM [ 21 September A.D. 2006 ]

» leaving rice

Yesterday was my last day at Rice. I forgot to turn in my keys yesterday, though, so I had to go back in today. The department admin told me that I need not have bothered. She was probably right, because the Rice Police saw fit to tag my car with a ticket for a parking violation. Oh well.

I have been (seriously) thinking about withdrawing from graduate school for several months now. (“Withdrawing” sounds so much better than “dropping out,” doesn't it?) I couldn't settle on a thesis topic and the prospects that were out there didn't line up with my interests--or I was unwilling to figure how they might be coerced into lining up. Getting out now seemed like a better idea than continuing on for a year, possibly finding a topic, only to go through this same crisis stage next year and have “wasted” a year's worth of work.

There is also the matter of careers after the doctorate. I came to Rice expecting to teach after I finished my Ph.D. My undergraduate degree was from a school which was not a “publish or perish” sort of school and that was the sort of school at which I wanted to teach. After seeing other people graduate from the compilers group, search high and low and in vain for an academic position (teaching or research), I realized that such a position was extremely difficult to find.

Thinking about it for a while, I came to believe that the sort of university I want to teach at will probably not be around much longer. My reasoning on this was as follows: Universities make their money in one of two ways: a) incoming tuition money and b) research grants. (Ignoring endowment funds for the moment; I don't think they play a role here, for reasons that I hope will become clear.) The cost of college eduction in this country is rising rather rapidly. Colleges have a problem if their primary source of income is a), i.e. the sort of college at which I would prefer to teach. Such colleges cannot raise tuition indefinitely, which leaves them with several options: a) stop hiring new professors, b) cut professor salaries, and c) start to accept research money. Endowment money does not help you in the long-term, since you will probably deplete the endowment faster than it can be built up and building endowments to the levels that would be necessary to avoid these choices takes quite a bit of time. All of these are unpalatable options for a aspiring professor such as myself, to say nothing of the administrators of such colleges.

While I acknowledge that it is possible for a university to be an excellent teaching university and an excellent research university (from what I have seen in the computer science department, Rice is evidence of this fact), such universities are few and far between and competition for positions at such places is fierce (see above). Furthermore, while I really enjoy teaching, I have also confirmed to myself that I really enjoy programming as well. “If I'm coding, I'm happy,” is a pretty good motto for me. Coding seemed somewhat lacking in the job description of a professor. So, I decided to start hunting for a programming job.

I think the most amazing thing for me to have realized during this process is that yes, there are people who do not complete their Ph.D. I had this mental block that everybody who enters a graduate program eventually staggers out with a doctorate. If you didn't, I thought, you must be a “loser.” Well, I am now a “loser,” I have friends who are “losers,” many people I have talked to about this have children or friends who are “losers,” and I am in fact going to work at a company founded by a “loser.” Realizing that I am, in fact, not a loser for doing this has been quite liberating.

I mentioned that I do have a job. After Labor Day, I will be starting work at CodeSourcery, where I will be doing various compiler-related work, mostly on the GNU toolchain (GCC, binutils, glibc). I went to school for compiler work, I ought to make use of that knowledge, right? Talking about the company to my wife after I was invited to send in my resumé, she remarked that my excitement level about this work was far above anything she had seen concerning my graduate work. I am extremely excited about the work I will be doing for them.

The work in itself is cool, but one of the other cool opportunities this job affords is that I get to work from home. All of the employees work from home, communicating via IRC, email, telephone, and the occasional all-hands face-to-face meeting. When evaluating the offers from CodeSourcery and the other companies I considered, the prospect of working from home was one of the key factors in the decision--it meant we didn't have to buy a second car, didn't have to pay for car insurance, repairs, gas, etc. on that car, it meant eating three meals a day with my family, going to take walks with my family during the day, etc. etc.

The prospect of working from home is a little daunting--productivity, motivation, interacting with “virtual” colleagues, etc.--but I'm confident working for CodeSourcery will be quite satisfying.

My sister recently informed me that I needed more “personal” posts and less “computer” posts. Perhaps this one is a step in the right direction.

posted by Nate @ 7:46 PM [ 17 August A.D. 2006 ]

» better vista than microsoft

Remember all the features Microsoft announced that its next-generation OS, Vista, would have? Remember how many of those promises have been rescinded, scaled back, or postponed indefinitely? Fear not, Apple is here to save the day: Apple is making a better Vista than Microsoft. I first had this thought a while ago when I first heard about the possibility of GC in Objective-C in Leopard, Apple's next major OS release. Major things that OS X has which Vista was supposed to have:

I'm sure there are other things that I have forgotten about, and there are probably cool things that Vista will have that will be missing in Leopard. But really, everything cool in Vista is going to be here now in Leopard. Why bother with Windows?

posted by Nate @ 7:13 PM [ 17 August A.D. 2006 ]

» research restrictions

I was talking with a friend of mine who is also a graduate student in computer science and he divulged some interesting information. Seems that part of the grant money that is funding his multicore compiler research is coming from Microsoft and that money comes with strings attached--namely, that using GPL software is not allowed. (Well, he said they can, but doing so involves paying several hundred thousand dollars to Microsoft.) Which is unfortunate, as one of the natural platforms on which to build this research is Open64, an advanced, GPL'd compiler suite.

Clearly, Microsoft wants to protect their “investment”--i.e. if something useful is developed with their money, they want to be able to use it in their products. I believe that research funded with government money (i.e. public funds) should be free and open to the public that paid for it. At a minimum, this implies free access to papers describing the research, to software written as part of the research, to data sets produced by the research, etc. etc. So it's not completely unreasonable for Microsoft to dictate limitations on what somebody can do with their (private) money.

However, I also think they're shooting themselves in the foot by doing this for two reasons. One is that I can count on one hand the number of commercially viable software packages that began life as research projects. (For the record, my short list is Google, LLVM, Coverity, and possibly DTrace. Granted, I have limited knowledge, but I think the viable/non-viable ratio is rather tiny.) Two is that Open64 provides a fantastic platform on which to develop this sort of stuff. Hand-rolling your own system to avoid licensing problems means that a good chunk of your research budget is sucked dry by simply implementing the infrastructure to get to the point where you can actually start to do the research you really want to do. And knowing the system my friend and his group is going to be forced to use instead, I can say with some confidence that this is exactly the case.

posted by Nate @ 10:27 AM [ 13 August A.D. 2006 ]

» in god we trust

JWZ posted a little protest against the recognition of the 50th anniversary of the national motto, “In God We Trust.” What I found more interesting was a comment stating, in part:

The real revolutionary was the justice who figured out that putting the phrase 'In God We Trust' on all of our money actually ended up stripping out any real religious content and made it just another empty collection of words with no real meaning or weight. Now *THAT* is revolutionary shit right there.

Right on the money.

Current Music: Dirt Devils -- The Drill (Evacuate remix)

posted by Nate @ 8:23 PM [ 30 July A.D. 2006 ]

» dictionary of christian terms, part one

The Interpretative Dance Theocrats strikes me as something that Josh might write. Its stated purpose?

To be fair to these perplexed and terrified people, Christians are not easy to understand. To begin with, there are roughly 2,000 years of history to grasp, and certainly more denominations and subdivisions than that to take on board. For people who were raised secular, I imagine it's like trying to understand an opera after coming in halfway before the end: the stage is crowded with people, two of them seem to be dead, a woman is wearing a hat with horns, and everyone is making a terrible racket.

The time has come for some kind of crib sheet for the confused and frightened, a handy easy-to-use reference guide for identifying some of the key denominations, terms, and concepts in Christianity.

I wish I could quote the whole thing, because it's just that funny--it's so hard to pick just one entry to give you a flavor. I think I might have to go with the definition for “Baptism”:

Baptists are Christians who believe God can only be accessed by means of a swimming pool or, in some cases, a shallow outdoor stream. The first Baptist was John the Baptist, who was said to eat locusts and honey, although contemporary Baptists generally prefer barbecue. “Baptism” is also the term used to describe a key Christian ceremony, in which prospective members of the church are either initiated actually (Catholics, Orthodox, confused Protestants) or symbolically (Protestants, confused Catholics, religious studies professors). Catholics believe that anyone can perform a valid baptism, Orthodox believe that any Christian can, while Baptists, paradoxically, believe that only they can.

Read the whole thing.

Current Music: Armin van Buuren featuring Ray Wilson -- Yet Another Day (Riva remix)

posted by Nate @ 8:13 PM [ 30 July A.D. 2006 ]