Monday, September 27

Burn the disk packs!*

Buried in Adam Rifkin's post on the Google Labs Aptitude Test is a request that I answer the one of the test's questions:

5. What's broken with Unix?

How would you fix it?

I think it was unwise of Adam to ask for my response to this question — after all, the name for this blog was originally supposed to have been Don't Get Me Started (turns out that had already been taken, several times). It's pretty difficult to take this question seriously, though, so the danger isn't all that great. And to be fair to the author of the GLAT, it's fairly clear that this question isn't meant any more seriously than the rest of the "test".

For Adam's pleasure I could try to answer the question, but to keep the answer's length reasonable I'd have to choose between Unix the thing and Unix the idea. Even that's not a clear distinction.

So, unless something else compels me, I won't answer, except with a parable. Uh, not actually a parable, but more like a rave. Not my rave but one I once heard. Uh not really the one that I heard, really my version of what I heard. At least I think I heard it. OK. Somebody (John Wharton?) explaining why he hated Intel's processor:

No, I won't use the freaking Pentium 4, 'cause inside a Pentium 4 is a Pentium III, and inside that is a Pentium II, and inside that is a freaking Pentium nothing, and it doesn't stop there. No, inside a Pentium is a 80486 which I won't use because inside that is a freaking 386, and inside that is a 80286. You know, of course, that inside the 286 is an 80186 and inside that is the freaked freaking 8086 which has an 8085 inside, and inside of that is an 8080, and as we all know, inside that is an 8008 and inside that is a 4040 and inside that is the freaking 4004 and inside that is a freaking japanese freaking calculator

Since I heard the original version in the mid-80's, the rave presumably started from the 386. And, of course, the speaker didn't actually say "freaking".

[*] Stoney again, I think; but Alan claims the slogan too.

Friday, September 24

Capitalist Pig

I've already posted about becoming an advertiser. Now I've discovered a new (to me) way to monetize my success in building Recondite's readership to its currently impressive levels.

I've already made the changes. This commercial activity is less visible than AdSense, though — can you figure out what it is?

Principia Securita

I found another book at Digital Guru: UC Davis Prof. Matt Bishop's Computer Security: Art and Science. Apparently, it's been out since early 2003, but this is the first copy I've seen. Bishop is well-known for the practical advice he's published on writing secure privileged programs.

I've only read the table of contents and dipped-in here and there, but it seems incredibly comprehensive. At a little less than 1100 pages, it's 58% the length of Russell and Whitehead.

It's got the obligatory chapter on Bell-LaPadula1, and three chapters on crypto, but that's out of 35 chapters. Less usual, and more welcome, is the chapter on identity, and the four chapter-long section on assurance. There are five chapters in the "End Matter" section that seem like padding (I mean, a does a chapter on symbolic logic belong in the book?), but there may be requirements in textbook publishing I don't understand.

Unfortunately, it doesn't have any material on security and economics or human factors; of course, that's a fault shared by almost all other books on computer security.2

As I said, it's quite comprehensive and it looks like a great combination reference and textbook. I'm glad I bought it. Now I've got 900-odd pages to go...

[1] The Bell-LaPadula Model was developed in the mid-60's as a formal description of the necessary properties for computer systems supporting military-style classification of information. There's a hierarchy of classification labels (e.g., Unclassified ← Confidential ← Secret ← Top Secret) for data, and system users have clearance to corresponding levels. Sigh. Most thinking about computer security has started from this point — too bad it has nothing to do with the requirements of real systems.

[2] With the exception of Ross Anderson's Security Engineering: A Guide to Building Dependable Distributed Systems. But Anderson's book actually has surprisingly little material on economics and the book is far less comprehensive than Bishop's.

HAKMEM redux

While I was in the Bay Area last week I went browsing at Digital Guru bookstore in Sunnyvale, not far from where the old Computer Literacy used to be on Lawrence Expressway.

One of the books I found there was Hacker's Delight, a book very much in the spirit of the venerable MIT AI Labs memo HACKMEM.1 Whereas HACKMEM touches on mathematics, circuits and even cosmology2, Hacker's Delight pretty much sticks to (computer) mathematics.

The book has a nice website that goes along with it.

[1] Guy Steele's forward to the book also notes the resemblance.

[2] Check out Item 154, where Gosper proves that the Universe is two's-complement.

Wednesday, September 22

What is this feeling?

I feel guilty that I haven't been blogging. The problem is I only feel like blogging about politics right now, but my words seem inadequate to express how I think and feel. In listening to the Wicked cast album, I realized there was a song that says it very well:

What is this feeling, so sudden and new?

I felt it the moment I laid eyes on you.
My pulse is rushing, my head is reeling,
My face is flushing — what is this feeling?
Fervid as a flame... does it have a name?

Loathing! Unadulterated loathing!
For your face, your voice, your clothing!
Let's just say... I loathe it all.
Every little trait, however small,
Makes my very flesh begin to crawl,
With simple utter loathing.
There's a strange exhilaration,
In such total detestation.
It's so pure and strong!

Though I do admit it came on fast,
Still I do believe that it can last.
And I will be loathing, loathing you
My whole..... life...... long!
©2003 Wicked A New Musical

Monday, September 13

Virtually Madison Avenue

Check it out: Recondite now has Google AdSense towers in the sidebar. I would have done it from the founding of the blog, only Google wouldn't have me until now. First, they said that their topic analyzer didn't give good results (i.e., couldn't choose appropriate ads) for the blog. Next they said Recondite didn't meet their policy standards, but I contested that decision and they later reversed themselves. I suspect it was the sentence "I hate W." which was classified as hate speech rather than political discourse...

For those who don't know about AdSense: it's Google's program to place ads on the pages of independent web-sites. Google selects and places ad content on a participating website, and pays the publisher for click through.

You have no idea what satisfaction it gives me to be living in the world I imagined fifteen years ago. Of course, I thought micro-publishing would involve micro-payments, not Google aggregating the payments until it was worth writing me a check.1 But that's a detail.2

So here we are; the future has arrived — every man his own (commercial) publisher at less effort than, say, writing a letter. An opinion utopia, Gentlemen, if you can keep it.

[1] In Recondite's case, I suspect they may never have to write one.

[2] A detail, that is, to someone who wants to make predictions. For someone like me who tried to create micro-payment protocols, it was more than a detail. But let that one pass.

Saturday, September 11

Misunderestimating the problem

Angry Bear has a nice post on how Bush's "medical tort reform" doesn't address the explosive rise in our health-care costs.

According to the CBO, malpractice lawsuits cost about $24 billion in 2003. Their estimate is that tort reform could save around 25%, making for a $6B savings. But 2003 spending on health-care in 2003 was almost $1.6 trillion — so we're talking about less than 0.4%. Since health-care costs increased about 7% that year, it's clear that "tort reform" doesn't help.1

As Angry Bear says, Bush isn't interested2 in policies that lower health care costs, but I know someone who is.

[1] A subtle point nonetheless worth making is that the CBO study shows the growth rate in malpractice awards is half the growth rate of health-care as a whole, so tort reform is even less relevant than the above analysis indicates.

[2] Bush's plan doesn't lower costs, but it is good for HBOs, insurance companies, and the AMA. Of course, it's tough on the poor slob who gets the wrong leg cut off. Compassionate conservatism strikes again.

Thursday, September 9

Give Until it Hurts

Make a significant contribution to the Democratic National Committee and I promise I'll try to do something nice for you.

To participate:

  1. Give money via that link.
  2. Tell me how much you gave and who you are via email.
  3. Suggest something you'd like me to do for you (e.g., cook you a meal, write you a testimonial, fix your computer, write your business plan, etc.)

If it's practical for me to do and it is commensurate* with your contribution, it'll be more likely that I'll do it. If you gave $10,000, for example, I think it'd be fair for you to ask me to clean your house for a week...

[*] You're even more likely to get me to perform if the request seems small in comparison with the donation!

Too Good to be True Department

Notice of Predisposition: I hate W. Okay? Believe me. Ask anybody.
Disclaimer: I don't have any special talents regarding document forensics.

Right. The Boston Globe and 60 Minutes Bush national guard service stories were based on materials allegedly from the personal records of the now-deceased Colonel Jerry B. Killian. It seems quite obvious to me that the August 18th document is a crude forgery.

I say a crude forgery because it was done with MS Word using out-of-the-box defaults for font, font size, margins and other settings. If you type the text into Word (I did, look here) you get an identically-appearing result, minus the scrawled signatures and the (trivially simulated) photocopy/fax artifacts.1

The giveaways are:

  1. Alignment of the first character of the date line with the period in the first paragraph.
  2. All four line breaks in the first paragraph.
  3. Superscript "th" in "187th", fourth line, first paragraph. This is what you get with Word's default settings in the "Autocorrect Options..." menu dialog "Autoformat as you type" tab "Ordinals (1st) with superscript" setting.
  4. As was immediately observed, the whole damn memo was in a proportionately spaced font, pretty freaking good for a typewriter issued to an Air Force Reserve desk in 1973.2

I've just noticed that both Powerline and Ratherbiased pretty much cover the same ground.

Message to Democrats: Given the forger seems to be either very stupid or alternatively, hoping it will be detected, this could easily be a Republican dirty trick. DROP THE TOPIC, NOW

Update (9:20pm): Wow, there's a lot of stuff in the Blogosphere about this. The most convincing, showing an overlaid fax and Word reconstruction like mine is Little Green Footballs.

[1] Formatting details: Five tabs before typing the date line, three returns after. Double returns after "Memo", "Subject" lines, and each paragraph.

[2] Yeah, sure, the IBM Executive series typewriters. Give me a break.

Saturday, September 4

On the Unreliability of Human Bodies

A recent article in the September issue of IEEE Spectrum is titled Why We Fall Apart. It analyses human aging using reliability theory, simplifying the same authors' 2001 paper in J. theor. Biol. The Reliability Theory of Aging and Longevity.

Reliability theory is pretty mundane stuff1, although for complex systems the calculations are complicated (hence the software tools). It's used by engineers to make predictive models of failure. Among other uses, this is how designers calculate the MTBF figures you see quoted in disk drive specifications.

The essential point of the article is this: you can model the human body as being comprised of a combination of irreplaceable, redundant, non-aging parts, some of which are defective to begin with. Doing so predicts the salient characteristics of human mortality:

  1. The infant mortality period, which, not ironically, is a term from both reliability theory and population dynamics.
  2. The normal working period.
  3. The aging period with exponentially increasing failure probability described by Gompertz 200 years ago. That is, after age 25 or-so, the probability that you'll survive another year declines very rapidly.2
  4. The post-aging or late-life mortality period with linear failure probability. That is, after age 95 or-so, your probability of surviving for another year is bad, but pretty much the same as it was the year before.

The post-aging characteristic of mortality statistics is related to convergence, the fact that an 80 year-old Indian has a similar life-expectancy to an 80 year-old Dane, although the life expectancies of Indians and Danes are quite different. This reliability theory model accounts for convergence and late-life mortality nicely.

A consequence of this explanation of aging is that one way to control it might be to avoid developmental damage in vitro. Don't skimp on those anti-oxidants, mothers-to-be!

[1] See an overly simple tutorial here.

[2] Actually, mortality increase during aging is not the same in living things as in redundant machines. Failure rates in such machines follow the Weibull distribution. But that's because classical reliability theory assumes that when machines are created, all their parts are working. If you assume that there's a high probability that components are faulty, you pretty much get a Gompertz distribution.

Friday, September 3

Another anniversary

Recondite is three months old. One change I've decided on today: the tagline at the top will change fairly often.

What better way to celebrate than to talk website statistics?

  1. 56% of visitors use Netscape, the other 44% IE. I would think that means recondite has readership skewed much more towards Unix (and Mac) than 'net users in general.
  2. Current visitors per day average just below 50. High-water mark was August 11th, the day after I posted about Peter Deutsch's pycore, almost 1200 visitors that day.

I have no idea, really, what people like me to write about. In the absence of requests (hey, that's the ticket! radio show formats for blogs!), I guess I'll continue the random walk through topic space.

Keep those cards and letters coming in, folks.

Thursday, September 2

Iraq's not like Vietnam at all...

Let's see.
  • Country with no strategic value to US. Nah, Iraq's got oil. (Vietnam had fish sauce, but US had no Pho shops in '62).
  • Country mostly jungle and mountainous, providing cover for enemy movement. Nope, not much jungle in Iraq. Enemy seems to be able to move pretty easily anyway, though.


  • US government claims general population welcomes military presence; check.
  • Determined, indigenous enemy utilizing guerilla tactics; check.
  • US forces rely upon heavy armor and aerial bombardment, creating resentment in local population; check.
  • Enemy possesses source of supply and sanctuary in neighboring countries; check.
  • US government delusionally claim signs of progress everywhere, doubters are called unpatriotic; check.
  • Enemy is a coalition of nationalists and holders of a global triumphantalist ideology; check.
  • US forces isolated from population by culture and language, dependent on local allies of dubious loyalty; check.
  • Scenes of widespread violence and death on nightly news; check.
  • Conventional wisdom is "no alternative to victory", elites are secretly pessimistic; check.
  • American intervention universally deplored, even by close allies; check.
  • Concern over deepening commitment causes US strategic shift to "localization"; check.
  • Widespread suspicion of profiteering by companies with ties to administration; wait — did the Vietnam war have that?

Nah. The comparison's ridiculous.

Lies, Damned Lies, and Convention Speeches*

Sometimes I despair at the asymmetry between offense and defense. I'm not talking about the "war on terrorism", I'm talking about liars.

To protect commercial airliners from a couple of nuts with boxcutters we have to spend billions on baggage scanners and add an hour or more of delay for each of the million plus airline travelers each day.

Similarly, I could write a book exhaustively refuting just two sentences from Laura Bush's convention speech:

I could talk about the fact that my husband is the first President to provide federal funding for stem cell research. And he did it in a principled way, allowing science to explore its potential while respecting the dignity of human life.

I could talk about the fact that the $25 million in federal stem-cell research funds her husband allocated is less than 0.004% of the federal healthcare research budget. That it's less than 1/10th the funding put up by little Singapore, an economy 1/100th the size of the US. That it's about the amount of money being spent on political advertising for California's Prop 71 (which proposes $3 billion of state money for stem-cell research).

But I may as well rely on Michael Kingsley excellent LA Times Editorial:

It is true indeed that Bush's predecessors, from George Washington to Bill Clinton, failed to fund embryonic stem-cell research. Even Abraham Lincoln. Not a penny for stem-cell research from any of them. Historians believe this might have been because it didn't exist yet. But that's just a guess.

George W. Bush gave this nascent research a tiny sliver of money and piled on a smothering load of restrictions. As Laura Bush did not note, that makes Bush the only president to ever authorize federal rules against stem-cell research.

It is characteristic of Bush that he would not see, or have no patience for, the irony of justifying a policy on moral grounds and then, when it comes under attack, claiming that the policy is not having the very effect he is supposed to want. Meanwhile, it is characteristic of the Bush political machine to be utterly fearless about insisting that things are the way it would be convenient for them to be, despite the evidence that things are the way they really are.

The purpose of Bush's stem-cell policy is to discourage medical research using embryos. Bush supposedly thinks that these clumps of a few dozen cells are every bit as human as the people who will suffer and/or die from diseases that stem cells could cure. He had better believe that, because stem-cell research uses embryos being discarded by fertility clinics and doesn't actually add to the embryonic death toll at all. Only a deep conviction about the humanity of these microscopic dots (which have fewer human characteristics than a potato) could justify sacrificing real human lives to make the purely symbolic point that the dots are human too.

Scientists are in agreement that Bush's policy is succeeding. Stem-cell research has been drastically slowed. Yet Bush surrogates now pretend that the policy's real success is its failure to stop this research completely. Hey! You're supposed to think all those embryos being used in privately funded research are human victims, remember? It's a huge tragedy, remember? Stop bragging about it.

You should read the whole thing.

I could go on about this, but I'd never be able to get to Laura's next sentence, let alone her husband's speech.

[*] (Updated 10pm) Damn it, I just noticed that Fred Kaplan of Slate posted an article earlier today with the same title as this one. Well, that's just too bad, I'm not changing it.

Wednesday, September 1

Ich bin eine Republikaner

The Gubernator's convention speech is so easy to parody that only someone with no self-restraint would stoop to it. OK, here goes:

My fellow body-builders, my fellow Americans, how do you know if you are a Republican? Well, I tell you how. If you believe that government should be accountable to the people who make the big political contributions, not to any random idiot, then you are a Republican.

If you believe a person should only be told the things they need to know, not every detail of what their leaders do, then you are a Bush Republican.

If you believe claims of improvement in America's strategic, environmental and economic prospects more than you believe the evidence all around you, then you are a Republican.

If you believe our educational system should be accountable for our children's progress (learning abstinence and creationism), but cabinet officials needn't be accountable for the performance of their departments, then you are a Bush Republican.

If you believe that your leaders should bombastically praise our military while ignoring their advice, shortchanging them in men and material, all the while cutting soldier's benefits, then you are a Bush Republican.

If you believe that the best way to protect our way of life requires keeping American citizens jailed for years without access to courts or council, to strip-search Teddy Kennedy when he goes to the airport, and cutting the budget of the Air Marshal service, then you are a Bush Republican.

If you believe that our country is governed best when its laws and initiatives are ironically named ("Clear Skies" and "Healthy Forests" are my favorites), then you are a Bush Republican.

And, ladies and gentlemen, if you believe that we must be fierce and relentless in promoting Haliburton and other companies in the oil business, then you are a Bush Republican.

Now, there's another way you can tell you're a Bush Republican. You have faith in big business, faith in the resourcefulness of rich people and faith that losers deserve their fate. And to those critics who are always harping about the unchecked greed of our President's corporate cronies, I say: Don't be economic girlie-men.