Monday, 21 October 2002
9:38 - Thanks to everyone who has subscribed or renewed recently. I think I'm all caught up up processing subscriptions. If you've subscribed or renewed recently and haven't heard from me, please let me know.
For now, I've left the expired usernames and passwords active, but in the next couple of days I'll be disabling accounts that expired 9/30/02 or earlier and purging expired accounts from the subscriber mailing list. If you want to remain a subscriber, now is the time to renew your subscription.
I have some new products on the way from Intel, most notably samples of their new motherboards based on the 845PE, 845GE, and 845GV chipsets. These new chipsets are essentially updates of the 845E, 845G, and 845GL chipsets respectively, but are still worth looking at pending the release of the Springdale and Granite Bay chipsets next year. I'll also be updating the recommended system configurations based on what I learn.
For now I'm working hard on the Motherboards chapter, with several others now in the pipeline.
10:05 - I just sent the following to my subscriber list:
(If you are a subscriber and didn't get this message, please let me know.)
11:42 - Tom's Hardware has posted a fascinating review of 21 power supplies. It's worth a read, although in my opinion the article has a few omissions and a strange conclusion. For example, not one PC Power & Cooling unit appears among the 21 power supplies tested. PC Power & Cooling power supplies are generally regarded to be the best consumer-grade power supplies available, so this is a strange omission.
The conclusion is also strange. Those who read only the conclusion will come away with a different view than those who read the entire article in detail. The article rates only three power supplies "good", those made by Antec, Fortron, and Verax. The article recommends three power supplies, but not the three you might expect. Antec is missing from the recommendations, although it is presented as a more readily available alternative to the three recommended units. In its place, Tom recommends a Herolchi unit, which received only a "satisfactory" rating in the detailed results.
We stand by our own recommendations. If you want a solid mid-range power supply, buy an Antec unit. If you want the best consumer-grade power supply available, get a PC Power & Cooling unit.
Tuesday, 22 October 2002
9:17 - More than 30 years ago, as a college freshman, I wrote a paper entitled On the Historicity of Christ. My conclusion then was that there was no direct physical or documentary evidence that Christ existed, but that we could infer his existence from the behavior and writings of indisputably historic contemporaneous or near-contemporaneous figures. I got a D+ on that paper, the lowest grade I ever received before or since, but I suppose that was not surprising given that I was at the time attending a Presbyterian school.
And now CNN reports that an ossuary from the period has been discovered with intriguing words engraved upon it in Aramaic. I don't read Aramaic, so I'll take the word of the scholars who translated the engraving as "James, son of Joseph, brother of Jesus". My first observation is that those words are ambiguous. Clearly, the bones in that ossuary belong to someone named James, who was the son of Joseph. But was the Jesus mentioned here the brother of James or of Joseph? It could be interpreted either way, although had the names been less familiar many people would assume that Jesus was the brother of Joseph.
Obviously, many people will choose to believe that the Jesus mentioned here is the Jesus of the New Testament, but I think that's unlikely to be true. Accepting the statement that the practice of using ossuaries was common in Jerusalem from 20 BCE until 70 CE and that Jerusalem was then a city of about 40,000 people, let's extend that to modern times. Assume an American city of 40,000 people, and further assume that the use of ossuaries was common from just before World War I until 2002. In the year 4,000 CE, archaeologists discover an ossuary with the graven words, "James, son of John, brother of William". They have no other records to go by, but they do know that a man named William supposedly lived from about 1928 until 1961, and that his father's name was John. What probability could they assign that the William referred to on the ossuary was the particular William of whom they'd heard? A very small probability, obviously, because the names James, John, and William were very common ones in 20th century America, just as the names James, Joseph, and Jesus were very common ones in 1st century Jerusalem.
Nor will the argument hold water that only a famous Jesus would have been mentioned. In the first place, they have a grossly inadequate sample size to make that judgment. In the second, surely the words graven on an ossuary were a family matter? For all we know, the Jesus mentioned was a wealthy brother (or favorite uncle) of James. Perhaps the family included his name in the hope (or the promise) that Jesus would pay for an expensive funeral. Or perhaps that Jesus was the patriarch of his clan, and his name was included from familial respect. But the strongest argument against the assumption that this Jesus must have been famous and therefore the Jesus we know is that the Jesus of the New Testament, if he even existed, was certainly not famous at the time. He would have been just one of the itinerant holy men and preachers who were commonplace at the time.
So are the bones contained in this ossuary those of a kinman of Jesus of the New Testament? Maybe, but probably not.
12:53 - I was sitting with my mother discussing the latest fatal shooting in the DC area. She brought up the Richmond school closings, which she thought were a good idea. I told her I thought that closing schools was a terrible idea, on several grounds. She said I'd feel differently if I had a child in one of the schools in that area, but I told her I didn't think I would.
Consider this: if the sniper were targeting children at schools, that'd be one thing. He's not. One of his victims was a child at school, but so far the shootings appear random. There's nothing to say that children would be safer away from school than they would be at school. If the sniper's targets are random, dispersing the children is of no benefit. The downside is that those children are being relocated from a centralized, protected environment to their homes. But what of those children whose parents both work? What provision can those parents make on zero notice to see that their children are cared for at home? Inevitably, many of the older children will be left at home alone. Will they stay at home, or will they be out wandering the streets--or even their own yards--where they will be potential victims of the sniper? In what way does closing the schools make the students safer? None that I can see.
If I were in charge, I'd call for parent volunteers to patrol their children's schools. Ask parents who own deer rifles to bring those rifles to school and walk the perimeter, a dozen or so on duty at all times at each school. By all accounts, the sniper is shooting from very close range, perhaps 150 yards or so. Put yourself in the sniper's place. Would you shoot at a child at a school with a dozen armed people patrolling the grounds? I certainly wouldn't. I'd look elsewhere for a safer target.
Wednesday, 23 October 2002
9:29 - The Register reports yet more vulnerabilities in Internet Explorer, and suggests turning off Active Scripting as a work-around. In my opinion, anyone who uses IE and hasn't already turned off Active Scripting (as well as just about everything else) is either uninformed or a true optimist.
Frankly, I'm not sure why anyone would continue to use IE as his default browser anyway. I relegated IE to ocasional use more than a year ago. At first, I used Opera as my default browser, but its poor rendering on many sites eventually drove me to try Mozilla. I've been using Mozilla as my default browser for six months or so now, and I see no reason to change. It's fast, stable, renders well, and is not subject to the plethora of exploits aimed at Internet Explorer. Mozilla also improves with each release, while IE arguably hasn't seen any real functional improvements since the initial release of IE 5.0.
I consider IE a legacy browser, and wonder why so many people continue to use it. If you're still using IE, do yourself a favor. Download and install the latest stable release of Mozilla and give it a fair trial. Make it your default browser, and live in it for at least a week. Once you get the fonts, themes, and other preferences adjusted to your liking, I think you'll find as I have that there's no reason ever to fire up IE again, except when you hit a site that requires IE for proper rendering.
I also relegated Outlook 2000 to the gulag a couple of months ago, replacing it with Mozilla Mail as my default mail client. Mozilla Mail is clumsier to use than Outlook 2000 and less functional in some respects, but it's a competent mail client and does provide several nice features that Outlook does not. Mozilla Mail completely lacks one critical Outlook function, and that is support for Outlook viruses. I don't have Norton Antivirus set to scan incoming mail for Mozilla because it's not necessary. Messages infected with Klez and other viruses arrive in my mailbox, but starve to death for lack of anything to chew on. My mail also downloads in about half the time now.
I won't say that if you use Mozilla and Mozilla Mail you'll never be infected by a virus, worm, or Trojan, but the simple fact is that probably 99.999% of all infections occur through using Internet Explorer, Outlook, Outlook Express, and other Microsoft applications. If you don't use those, you cut your exposure by several orders of magnitude.
Thursday, 24 October 2002
10:42 - Many questions are as yet unanswered, but at least it seems that Washington, DC area residents can breathe a bit easier this morning. With the two presumed shooters now in custody, what remains is to find out why they did what they did. Assuming, of course, that these two men are in fact guilty of the shootings, which seems likely but is not yet proven.
The early reports made much of the fact that one of the suspects is surnamed Muhammad, but later reports made clear that Mr. Muhammad is an American who converted to Islam rather than a native follower of Islam. Mr. Muhammad apparently made statements after 9/11/01 in support of Islamic terrorism, but no connection has yet been made between him and Islamic terrorist organizations. So the question remains as to the motivation for these shootings. Were they the actions of an "ordinary" serial killer or an attempt to extort money? Or were they perhaps sanctioned or unsanctioned acts of terrorism in support of Islam? We may never know.
I was never convinced that the sniper shootings were organized acts of Islamic terrorism, simply because Islamic terrorists have sufficient resources to place multiple sniper teams on US soil. If these shootings had been sanctioned by Islamic terrorist organizations, I'd have expected them to employ many sniper teams, widely dispersed geographically. I'd also have expected clusters of shootings, with multiple victims over a period of a day or two, and then relatively long periods without activity. Nearly daily shootings increase the risk of the shooters without greatly increasing the level of terror, which could have been maintained at lower risk with less frequent shootings. None of that came to pass, which leads me to believe that these shootings had nothing to do with Islamic terrorism per se.
I'm sure we'll find out more over the coming days and weeks, so it's useless to speculate in advance of the facts.
12:20 - I really hate rude automated anti-spam mechanisms, which are becoming increasingly common. Here's an example. I've disguised the email address of my correspondent to prevent address harvesting software from grabbing his address:
On what basis is Mr. Magda's ISP's mail server determining that I'm a spammer? I see nothing spammish about his message or my reply. The only thing I can think of is that perhaps Mr. Magda's ISP subscribes to a service that has wrongly added one of the pertinent domain names or address blocks to a blackhole list. I wonder how many other legitimate messages Mr. Magda fails to see because his ISP uses overly-aggressive anti-spam measures. I also wonder how many of my own messages are not received by the people I send them to. This time, I caught the auto-reply, but such messages often end up in my own Trash folder, where they may be deleted without being read.
I'd want to be pretty sure of myself before I blocked email to one of my customers, let alone before I sent an automated "Go away Spammer" accusation to possibly innocent people. I'm not picking on Mr. Magda, or even his ISP, because this problem is becoming commonplace. Ultimately, of course, spammers are to blame for problems like these, but it is annoying to get such a message in response to a perfectly legitimate message that I've sent.
Friday, 25 October 2002
10:08 - The problem with bouncing mail has been explained. Mr. Magda writes to say that his ISP admins had blocked the domain rr.com because they'd gotten some spam from a Roadrunner account in August. Way to go, folks. Shut out a few million Roadrunner users because you got spam from one RR account. I have no problem with ISPs choosing to use well-managed blackhole lists, but this is simply stupid. Their priority seems to be blocking spam at all costs, so I'm surprised they didn't take things a bit further and block all email from the .com top-level domain. In fact, by their logic, they should block all email from all top-level domains, thereby ensuring their users a spam-free environment.
Bob Walder was one of the people who took my advice to try Mozilla, but alas his trial didn't last long. He emailed me to say that he'd run into a showstopper with Mozilla. It didn't import his IE Favorites. I replied that I found that odd because Mozilla had imported mine automatically during installation. Bob replied that he'd downloaded the latest beta, so perhaps that was the problem. He then downloaded and installed the latest stable release, Mozilla 1.1, but found the same problem existed with it. He speculates that perhaps the problem occurred because he is running Windows XP, which I do not run.
I played around briefly with the bookmarks functions in Mozilla, but didn't see any obvious way to import IE Favorites manually. Any suggestions appreciated.
It's always been clear that Jerry Pournelle is a very smart man, but his post yesterday [link here after 27 October] made it clear that he's a very brave man as well. In that post, Pournelle states outright that real differences in intelligence exist among different human populations. That's not news to anyone who understands anything about the topic, but it is a subject that has been made taboo by the forces of Political Correctness.
I can imagine the squeals of outrage now gathering in Pournelle's inbox. He'll be called a racist and accused of advocating "bad science." All of that is crap, of course. What he writes is truth based upon observable facts. The most intelligent subgroup in the human population is, on average, Ashkenazi Jews. Behind them are the Chinese, followed by whites of European origin, followed by the Japanese, and so on down the list to American blacks in next-to-last place, and African blacks in last place.
None of this says anything about the intelligence of any particular person, but merely about how the curves plot. There are moronic Ashkenazi Jews, just as there are brilliant African blacks. Neither is intelligence the sole determinant of human value, nor even the chief determinant. But the fact remains that these differences are important, with implications for education and other aspects of our civilization. The logical inference is that our advanced placement classes should be largely populated by Jews, Chinese, American whites and Japanese, with blacks proportionately underrepresented. Those more intelligent groups should be more heavily represented among our physicians, engineers, and scientists. To the extent that that's true, it's not racism operating, but reason and natural selection.
Similarly, the emphasis on state-to-state variations in the SAT and other standardized test scores should take population groups into account. North Carolina and other Southern states rank very low by these measures if one looks at only the raw data, and accordingly are damned for providing poor education to their students. But if you normalize those scores by taking into account the percentage of students who take the tests and the weighted numbers by population group, you find that in fact the scores are similar from state to state. That is, the scores of white North Carolina students are similar to those of white Minnesota students, as are those of Chinese North Carolina students versus Chinese California students, as are those of black North Carolina students versus black Iowa students. For obvious reasons, states in which a higher percentage of all students take the test have lower average scores than states in which a lower percentage of all students take the tests. States in which a high proportion of students who take the standardized tests are from more intelligent population sub-groups have higher average scores than states in which a relatively high number of students who take the standardized tests are from less intelligent population sub-groups. Well, duh.
Saturday, 26 October 2002
10:02 - The Russians pulled off a miracle, successfully storming the theatre with only 67 hostages killed. I use the word "only" advisedly. Obviously, 67 hostages dead is 67 too many, but the miracle is that ten times as many weren't killed. About 750 hostages are still alive, thanks to those Russian special forces troops. Imagine the difficulties involved. Hundreds of hostages intermixed with 40 terrorists, who were armed with automatic weapons. The whole place wired with explosives. It would have been difficult enough had the terrorists each been wearing a blaze orange shirt with bulls-eye targets front and rear. I can't imagine having to sort out terrorists from hostages when all were wearing similar clothes. But the Russian special forces troops managed somehow.
None of that is any consolation to the families and friends of the 67 dead hostages, of course. But the families and friends of the 750 rescued hostages owe a vote of thanks to the brave young soldiers who risked their own lives to rescue those hostages.
The second-guessing by the media is truly foul: "Were some of the hostages killed by rescuers' bullets?" "Did two of the hostages die from suffocating in their own vomit as a result of the gas used?" Even if those speculations turn out to be true, how are they relevant? The inference is that somehow the rescuers were at fault for some of the hostage deaths, and the implication is that the rescue was botched. Give me a break. All 800+ hostages might as well have been dead already. Those who did die were killed by the terrorists. The rescuers weren't responsible for any of the hostage deaths. They brought 750 hostages back from the dead.
Sunday, 27 October 2002
9:06 (EST) - All our clocks are changed, the computer clocks automatically, and all the others manually. What a stupid thing to have to go through twice a year. I keep the clock in my truck set to UCT (formerly GMT), and I'd just as soon everyone used that. But if we do have to have an offset, can't we at least have an unchanging one? This DST crap annoys me more as I get older.
There's another huge Leonids meteor storm this year, and it's the last that any of us will live to see. The Leonids peak for a few years on a 33-year cycle, but those peaks vary dramatically in meteor volumes. Last year's Leonids storm was the show of the century, both because of the high number of meteors visible and because it occurred during a new moon. This year's storm may be even heavier than last year's, but occurs during a full moon, which reduces the number of meteors visible. Even with that reduction, this is a show that you don't want to miss. The next Leonids storm of this magnitude won't occur until the 22nd century, so this is an event comparable to the return of Comet Halley.
There are actually twin peaks this year, with the first spanning 106 minutes centered on 0403 UT 19 November (2303 EST 18 November) and the second spanning 122 minutes centered on 1046 UT 19 November (0546 EST 19 November). The radiant (the point from which the meteors appear to originate) is in the scimitar of Leo. For us, that will be just at the Eastern horizon during the first peak, which means that the meteors arriving during this peak will be "skimmers", arriving nearly parallel to us and therefore going through a lot of atmosphere. Unfortunately, Luna will be high during that peak. During the second peak, the radiant will be nearly at zenith for us, and Luna will be very low on the Western horizon.
If you have kids, get them out to see it. It's something they'll remember all their lives. Seeing one "shooting star" is neat. Seeing them arriving at 1,000 per hour, with peak rates of one per second or more, is incredible. For more information about how and when to view the Leonids storm, see this Sky & Telescope article.
Copyright © 1998, 1999, 2000, 2001, 2002, 2003, 2004 by Robert Bruce Thompson. All Rights Reserved.