THE VIEW FROM CHAOS MANOR
View 195 March 4 - 10, 2002
Highlights this week:
This is a day book. It's not all that well edited. I try to keep this up daily, but sometimes I can't. I'll keep trying. See also the monthly COMPUTING AT CHAOS MANOR column, 4,000 - 7,000 words, depending. (Older columns here.) For more on what this page is about, please go to the VIEW PAGE. If you have never read the explanatory material on that page, please do so.
If you are not paying for this place, click here...
For Previous Weeks of the View, SEE VIEW HOME PAGE
Search: type in string and press return.
|For an index
of previous pages of view, see VIEWDEX.
See also the New Order page, which tries to make order of chaos. These will be useful.
For the rest, see What is this place? for some details on where you have got to.
If you subscribed:
If you didn't and haven't, why not?
For the BYTE story, click here.
The atomz Search returns:
Search: type in string and press return.
The freefind search remains:
March 4, 2002
I am back, the new column is up on BYTE and it's the last they have, so I have deadlines if there is to be anything new for next Monday...
There is a ton of outraged mail regarding my remarks on the VIA chipset and other chip support for AMD processors. I have taken the trouble to get some validating opinions and I stand by my view:
until recently AMD chips have been going into motherboards with predictable problems, largely based on the VIA chipsets; even boards that used the AMD 760 support chips were compromised by the inclusion of VIA chips in the southbridge.
What I said, and stand by, is that a new generation of support boards for AMD chips promises to provide real competition for Intel and that's good.
There is a fair amount of mail, which I was working on posting with answers, but some AMD enthusiasts though it would be amusing to add me to a bunch of spam lists, and encourage flames. This professional behavior on the part of the partisans of a major company really makes me sympathetic to the object of their enthusiasm. Unfortunately it leaves me without the time to continue the discussion; that time is being used to set my spam filters.
In another discussion group I got this, which causes a chuckle:
Rasputin was a Skoptsy, of the mild sort. I wonder if the campuses are ripe for the hard-core variety of Skoptsyism. It'd sure be fun to watch.
And then we have this:
Accrediting Body Approves Astrology School
An Arizona-based school of astrology appears to be the first to boast accreditation from an agency recognized by the U.S. Department of Education. In August, the Accrediting Commission of Career Schools and Colleges of Technology (ACCSCT) granted accreditation to the Astrological Institute, located in the Phoenix suburb of Scottsdale. The move paves the way for students at the school to seek federal educational assistance, such as student loans.
The Institute offers a certificate program in "astrology and psychology." Full-time students can complete the course of study in 12 months. The program, which is an outgrowth of classes offered by a bookstore, costs $5300.
The meaning of ACCSCT accreditation remains something of a mystery. According to the Associated Press, an official of the agency described the group's stamp of approval as little more than an affirmation that a school's teachers are qualified and its graduates able to qualify for employment. Yet, the ACCSCT website says its goal is "to help good schools become better schools through setting high standards of educational quality . . ." Similarly, it states that "the accrediting process also assures students and the public that high standards of career education are being met."
In a four-minute interview with National Public Radio, Joyce Jensen, president of the Institute portrayed the accreditation as a major step toward academic credibility--a move toward "changing the mindset" that has left astrology with little scientific respect. She compared the situation of astrology today to that of Gallileo centuries ago.
As for astrophysicist Neil deGrasse Tyson, who told AP that astrology had been discredited 600 years ago, Jensen commented, "maybe he still believes the Earth is flat."
Which should be a comfort to everyone. Now we won't have unaccredited astrologers giving false advice.
|This week:||Tuesday, March
I have deadlines and a problem, and if one of you knows what I should do, please tell me.
I have a 2 GHz system. With an el cheapo GeForce 2 video board it does just fine at Everquest, although the 2d graphics including text are not very good. I installed an ATI Radeon 8500 board. Everquest recognized the board, but advises drivers that do not exist on the ATI web board. I have downloaded and installed the latest ATI drivers for a W2K system.
When I go to Everquest the 3D models -- the characters -- look fine and the speed is good, but there is almost no background: it is as if they are all at midnight in fog even in bright daylight. Adjusting the gamma to make that bright doesn't make it useable and washes out the 3D images as well.
I am certain this is a setting of some kind, or a driver I need, but at the moment I'm stumped. Any help appreciated. Please do not speculate: I need help from someone who knows what the problem is. I've got a call in to David Em but I haven't been able to reach him, and Alex is in Detroit supporting an auto show, and the deadlines are upon me...
More. I went to www.rage3d.com and found the absolute latest drivers. That doesn't do any good either.
Everquest says when I log on that I have a Radeon 8500 board and I should download either driver version 3286 or 3286 before playing. I have no way of discovering where these 3286 reside; the ones on ATI sites are higher numbers. It may be that Everquest wants me to use obsolete drivers.
Well I know part of the story now. The rest goes in the column. Apparently you can do it with XP but not Win2K.
The following report says they have made neutrons in cold fusion:
Alas other neutron experts say they don't get neutrons. See mail.
I have tried the 3286 drivers: thanks to all those who sent links. They don't work either.
I am told that others have managed Everquest with ATI RADEON with XP. That's next. I do these silly things so you don't have to.
Deadlines are upon me and there is other stuff to work on.
The lead next Monday will be on the AMD flap.
And continuing to work, the Great Hall gets a bit chaotic. But it's ordered chaos: things are sorted and easily removed.
The computers are back, but they are on rolling stands as is the work bench. The new projector that will be in the column is on the table. Because of the flash you can't really see that the desktop is projected on the far wall, but in normal light it's quite readable: report on the Plus projector in the column. That's the output of an ATI All In Wonder RADEON 8500 you see up there. The computer is a Dual Processor AMD Athlon 1.2 GHz system. There are some boxes of software stacked out there, but they'll go away like the work benches when the column is done...
Now back to work.
March 7, 2002
I will make the deadline and the lead next Monday will be the great AMD flap.
But there's still a lot to do.
March 8, 2002
Well, the column is done and off. Now the cleaup. I am NOT going to let things go back to the chaotic mess I usually live in.
And another virus warning. Of course the Internet connections have stopped working when I want to upload this to my site. Very odd.
I have sent a mailing to subscribers. Many have been returned. I have listed all those on the updated today badmail page .
If you subscribe and didn't get the mailing, your mailbox may be full (a common cause of returned mail) or you have subscribed under an address that is no longer valid. If the latter please send me mail, subject SUBSCRIBER ADDRESS CHANGE and include
old address = your old email address
new address = new email address.
I need the old one to make the change. Thanks
Incidentally, excite @ home just says it can't deliver. It doesn't tell me to whom it couldn't deliver, so I can't ever list those in badmail. Excite @ home doesn't exist any longer...
I have 19 messages returned because they have excite @ home addresses. Next week I am going to eliminate all entries in the subscriber list that use that domain address.
Note that I intend to prune this list of those whose mail bounces and do not show renewals. If you think you subscribe and have not heard from me in a while, it may be time to renew, and it's almost certainly time to send me an update to your email address. I send these warnings out when I think appropriate.
Roland got this cookie today:
The difference between science and the fuzzy subjects is that science requires reasoning while those other subjects merely require scholarship. -- Robert Heinlein
It's a good aphorism, but it does itself require the use of reason: what is a "fuzzy subject"? Is law? History? Certainly good history requires both reasoning and scholarship.
The "social sciences" are perhaps the worst here, in that they do require something like scholarship, and generally nothing like reasoning.
Robert was a dear friend and a brilliant man, but I would love to debate this one with him. It certainly contains the germ of a truth, but it is far to easy to apply it far too broadly...
It is also certainly the case that what Edith Efrom called "Regulatory Science" which is what many "environmental scientists" approve requires scholarship. It also requires reasoning, but the reasoning is on the closed system that emerges from scholarship; and that, I think, is where the real truth lies.
I long ago did an essay on "The Voodoo Science" in which I pointed out that novelists require only plausibility, lawyers need evidence, but science requires data -- and also requires that ALL the data be taken into account, which means going outside closed systems to check with the real world. And that, I think, is the real difference between the fuzzy subjects and the sciences and the humanities: if you endlessly apply logic to a closed system with no chance of checking against reality, you had better have chosen the right closed system -- and we have no way of knowing what that one is.
Vanderbilt's restatement of Pournelle's critique of Heinlein's aphorism:
The difference between science and the fuzzy subjects is that science requires accounting for _all_ the data.
I suspect there is a close inverse correlation between the amount of relevant data either carelessly or carefully ignored and the uselessness of the resultant "science".
March 9, 2002
One would think there would be more out there about it. I haven't seen a fix yet, either.
Today is the aftermath of the column. Yesterday I'd actually intended to take a train down to the beach house where Roberta went a couple of days ago, but I didn't manage it. Probably not today, either.
More on the above story:
I have not seen any Microsoft patch or fix for this.
I am attempting a test. Don't follow this test link. It may do nothing and it may do something unexpected. And in fact it does. Harmless but a bit scary. You try this at your own risk. The fix is in the above URL. One would expect Microsoft to Do Something, and Fast.
You've earned your A+ cert in nasty writing, it didn't do a thing.
Norton 5.0 caught your little attempt, chewed it up, and spat it into quarantine so fast I never had a chance to sa "Well mazletov, it's Asimov".
Keep up the good work.
Now that's odd. It didn't for me. I had better see to my installation. Thanks. I put that on the site and went there and it sure popped up the calculator...
OK, my Norton was configured wrong. It catches it now. Without it, there's the calculator. Note that this only works in Internet Explorer 5.5 and later. And Roland says
Here's a registry workaround:
Live long and prosper...
March 10, 2002
The exploit above now works with XP as well as NT: the path name to the calculator file is different, so I have put them both in.
Meanwhile Norton has decided that I can't have access to the page even in FrontPage (I set it to scan files as they open) so I have to reset that. But oddly enough although it won't let me open them in here, it still pops up the calculator, which is very disturbing.
I am resetting Norton and fixing my registry before posting this. It looks like I will need to do this on all my systems.
Note that this won't work -- I think -- to run any program not already on your system, but it ought to run all those it can find. I'll try another test.
I think I have it fixed up now so that it opens Freecell on XP systems and the calculator on W2K systems.
Now something else I just learned: I have on an XP system tried to delete C:/windows/system32/freecell.exe (actually I first moved it, then renamed it, then deleted it, and in every case it was back again in seconds.) I was doing this to do a test of paths for this exploit, but ye flipping gods, you can't delete a game? It seems that way on W2K too: I can't delete freecell. If I do it is back again, very quickly.
This turns out to be a feature, and if you go to the DLL cache and try to get rid of freecell.exe for good and all, it warns you that you are about to get rid of a critical system file, etc. I hadn't noticed all this. It's a feature.
I was going to get rid of freecell to see what happens if this is told to run a file that isn't there, then one that is there. I can do that another way. My test shows that if you give this a lot of programs and paths, it will keep trying until it finds one; run that; and then it does no more. But that should be enough, no?
While you are playing, look at this one, which is courtesy of Dr. Ed Hume. If you don't see anything try moving the cursor around.