Monday, August 31, 1998

Yesterday morning, I was updating the web page, speaking of changing from work mode to play mode and spending Saturday evening fighting with a differential SCSI drive over at our friends, the Tuckers. As I was writing that, about 8:30 yesterday morning, my friend Robin Weiner called to say she was desperate. She'd upgraded her Win95 box to Win98. Everything went fine until she tried to install her HP scanner, which trashed the system. She had work for clients that she needed to get done, and no computer with which to do it.

She'd already formatted C: and tried unsuccessfully several times to reinstall Win95, which she assumed was necessary because the Win98 CD was an upgrade version. I told her to bring the system over here so we could work on it. I'd have gone over there, but there are two big advantages to working here. First, I have all kinds of components that we could swap in and out as needed to discover faulty hardware. Second, I can smoke my pipe here without feeling guilty.

She'd formatted her C: drive again, so we tried doing a clean Win95 install. No dice. It got to the final step, where it's supposed to configure the system and then restart, but it locked solid. At that point, I began to suspect that something was screwed up in the BIOS settings, so I started checking them. I discovered that the second physical hard disk in the system had somehow been disabled in BIOS, whether by the Win98 install or not, I don't know. I set both IDE channels to autodetect for both master and slave and restart the system.

Win95 came up fine and we were able to upgrade to Win98 successfully. At that point, we started adding peripherals back in one-by-one. Win98 detected the sound card and installed and configured the drivers properly. We shut down again and installed Robin's USR Sportster 56K Voice Modem. Strangely, although Win98 detected the new modem, it had no driver for it. Robin didn't have the driver CD with her, and we were unable to locate a Win98 driver on the 3Com web site, so I told her to wait until she got home. When she booted the system at home, it again detected the presence of the new modem. She fed it the USR driver CD, and it installed and configured the modem perfectly.

The final problem was getting her HP ScanJet installed. It uses a SCSI adapter, and the instructions tell you to install the HP scanner software before installing the card. That's what caused Win98 to crash and burn the first time, so I told Robin to go ahead and install the HP SCSI card without first installing the HP scanner software. Sure enough, Win98 detected the SCSI card and installed the proper drivers for it. Once that was done, Robin was able to install her scanning application and scan successfully. I suspect the problem was that the pre-Win98 HP software was installing real-mode SCSI drivers that conflicted with Win98.

At any rate, I spent the weekend doing pretty much the same type of things I do during the week. It's fortunate that I like what I do. Time to get back to work on the book...


Late afternoon: This has been one of those typical Mondays. I did get some work done on the book. I also got the towels folded, the bathroom door fixed so that it will stay on the latch, and some books I needed ordered from Amazon.com. Other than that, it's been a frustrating day. This all started when I wanted to do a simple thing - format a bootable floppy disk with Win95 so that I could boot Win95 on one of my NT systems and run a diagnostic utility. I have only one system with Windows 95 installed, and I'd pulled the monitor off of it to use while working on old kerby, which appears to have a dead hard drive.

That one system, which dual-boots Windows 95 as mandy and Red Hat Linux as freya, is screwed up. It wouldn't boot at all at first. The BIOS screen displayed and it hung. Since I didn't hear the hard disk spin up, I assumed that that was causing the problem. I gave the system a smack during boot and, sure enough, the hard disk spun up and the system booted. Originally, I thought this problem was due to a bad IDE cable, which I replaced a month or two ago, but it now appears that the hard disk itself is causing the problem. It's a Western Digital 1 GB IDE, so it's no great loss. Still, this is aggravating. It may be coincidence, but the 850 MB Western Digital IDE in old kerby also appears to have died.

Once I got mandy booted, I figured my problem was solved. That turned out not to be the case. The floppy disk drive is apparently hosed. Nothing I tried would let me even read a floppy, let alone format one. This machine is built on an antique AT desktop case with a no-name 486 DX/2-80 system board. I think I'm going to salvage what's usable - 32 MB of SIMMs, a decent IDE CD-ROM drive, and possibly a couple of other components - and use them to build a new low-end box to run Linux. I'll probably build it around a low-end Pentium or K6 system board. I have two 16 MB SIMMs in this one. Old kerby has two 16 MB SIMMs and two 32 MB SIMMs in it. I'll probably swap memory around and put four 16 MB SIMMs in one box and two 32 MB SIMMs in the other, giving each of them 64 MB of RAM. That should be plenty both for Linux and for Windows 98. Old kerby is a Gateway 486 DX/2-66, which should be adequate for a Win98 test bed, particularly with 64 MB.

That's three systems I need to build or upgrade, counting the Pentium II system I've promised to build for Barbara to replace her current machine, thoth, which is a Gateway Pentium/133 tower. She's been stuck with that slow old Gateway for too long now. It runs Windows NT 4 Server as the PDC for one of my domains, and is also the primary backup box. I'll probably leave it in her office and put a new 300 MHz Pentium II box along side it.

That also brings finding an SVGA switchbox to a higher point on the to-do list. I located a Belkin box that looks appropriate on the Insight web site. I'll probably get that ordered, along with the necessary cables, sometime later this week. I also need to make time to get a hair cut sometime soon, or Barbara will take her gasoline-powered hedge trimmer to me.

Tuesday, September 1, 1998

Nothing like the start of a new month to remind me that the countdown clock is ticking on the deadline for my new book. The first chapter is due by mid-month, but I hope to beat that. This project is different from the ones that I have been doing, and getting into the swing is going to take some adjustments in the way I work.

My wife asked me last night if I'd gotten my Trooper inspected yet. It was due by yesterday, but of course I hadn't gotten around to doing it. I'll be driving illegally the next time I pull out of the driveway. I use this truck 100% for business, and put less than 1,500 miles a year on it, so it's easy to let stuff like this slip. Barbara is off Friday, and is taking her Trooper over to Mock Tire for an oil change, so I'll probably follow her over and get the damn thing inspected. That does kind of enforce writing discipline for this week, though. I can't be driving around in an illegal truck all week to pick up parts for these new computers, so I'll spend this week writing instead of building computers.

Thursday is Barbara's last day at work for a while. She's taking Friday and the following week off to knock around home. On September 12, she leaves with her parents for a two-week bus tour of eastern Canada. I'll be on my own for those two weeks. I told Barbara I was going to invite all the loose women I know to parties every night while she's gone. Trouble is, I don't know any loose women. She didn't sound concerned. Come to think of it, September 10 is our 15th wedding anniversary, so I'd better come up with a good present idea fast.

I'd better get back to work on the chapter...

Late afternoon: I've spent the entire day working on the chapter, and have actually made some progress, although not as much as I'd like. I wasn't going to update this page for the second time in a day, but then I read something on Jerry Pournelle's web site that makes it clear that laxity in updating your web site can sometimes have fatal results.

There's been quite a bit of discussion on Pournelle's site about the fact that the US was tricked into targeting the pharmaceutical plant in Sudan and the camp in Afghanistan. Apparently, what's-his-name intentionally suckered US electronic intelligence gathering by using cell phones with the full knowledge that the US was listening in on the conversations and would act upon what it learned.

The pharmaceutical plant was apparently just that, and the terrorist camp didn't have any terrorists in residence, just ordinary Muslims there for a religious retreat. I haven't seen this mentioned in the newspaper or TV news, although I may have missed it. I don't pay much attention to the "news" that lands on my front walk every morning, and I haven't bothered to watch TV news much in years, so perhaps I'm doing them a disservice.

However, apparently the real reason that that pharmaceutical plant in Sudan got Tomahawk'd to rubble is that they failed to update their web site in a timely manner. I'm not making this up. Pournelle posted the following quote from the LA Times on his web site:

"US Officials said that at the time of the strike, they knew that the plant had had a 'grand opening' celebration to publicize it as a major new source of medicines for the Sudan, which has been rent by a lengthy civil war.

But the officials said they did not believe that the plant actually produced such medicines, because they saw no evidence of such an output when they accessed a Web site for it. Web sites for five other pharmaceutical plants in Sudan listed the medicines produced at those plants." (LA TIMES Sept 1 1998 page A4)

If true, and I have no reason to believe it is not, this means that the US attacked a sovereign foreign country, killed and injured innocent people, and destroyed a building, all because they didn't like the looks of the pharmaceutical company's web site. Can it really be possible that all of this would have been avoided if someone had gotten around to updating the web site in a timely manner? I find this whole mess incredible.

Wednesday, September 2, 1998

Back to the chapter-writing grind this morning. Barbara works noon until 9:00 p.m. on Wednesdays, and I usually head out to her library to have dinner with her. Not today, though. I think I'll keep a low profile until I get my Trooper inspected.

I watched some of the U. S. Open last night, and added a section to musings about The U.S. Open and Big Servers. You can't trust Cyclops. You can't trust the IBM speed gun. There's not a lot you can trust these days.

Thursday, September 3, 1998

Got a fair amount done yesterday, although it didn't show up in the word count. My stack of stuff to read is getting taller. Yesterday, PC Magazine showed up, along with review copies of a couple of O'Reilly books - Outlook Annoyances and PalmPilot: The Ultimate Guide. I notice by the UPS tracking system that the three books I just ordered from Amazon.com for research purposes should arrive today. I just started the latest Sue Grafton, N is for Noose, last night, and the latest Ridley Pearson is also queued up in the stack on my nightstand. Then there's a foot-high stack of new mysteries that Barbara brought me, along with an assorted bunch of donated paperbacks she brought for me to read first because she doesn't have time to catalog them right now. An embarrassment of riches. Back to work on the book...


Ugh. Just flushed the toilet downstairs, whose output is routed to an ejector pump. The ejector pump jammed, and I ended up with about 3.5 gallons of water all over the bathroom floor. Fortunately, it was clean water. Thank god for the the ShopVac.

Friday, September 4, 1998

Finished N is for Noose last night, and got about half way through the latest Ridley Pearson. UPS delivered the books I'd ordered for research, so I have a stack of stuff to get read, both for enjoyment and for work. Barbara and I are off to Mock Tire shortly for our 9:00 a.m. appointments, her truck for an oil change and mine for an inspection. Yesterday was Barbara's last work day before vacation. She'll be off through the end of the month, her longest vacation in years. She'll stick around home through the end of next week, when she leaves with her parents for a two week bus tour of Canada. I'll stick around here to take care of my mother and write.


Word 97 is about to drive me mad. It apparently takes a dislike to certain documents. The document that contains the chapter I'm working on right now is one of those. Originally, Word would GPF reproducibly each time I exited it after using that document. I did some reformatting of the document, and the problem went away for a while. Now it's back, but in a different form. Now, everything appears normal when I exit Word, but the system slows to a crawl. Calling up Task Manager shows that Word, although it's not showing up in the list of running applications, is still resident and using 99% of the processor. I have to kill the process manually. Still, it's not losing any data. But it's a pain in the butt. This is almost enough to make me learn FrameMaker.

Saturday, September 5, 1998

I have, I think, discovered a secondary sex characteristic never before reported in the literature. It has to do with a person's attitude to what should be put in the dishwasher (or the washing machine, come to that). My attitude, which I think I share with most men, is Darwinian. If it can't survive the dishwasher, better we find out now, before it has a chance to pass on its genes. Women always have pity for the weak, and so sort things that should be washed by hand. As I was loading the dishwasher, the dialog went something like this:

Barbara: "Are you insane? You can't put 18th century crystal in the dishwasher!"

Robert: "Why not? It needs washed."

Barbara: "It's too delicate. You have to hand wash it."

Robert: "Whadya mean, delicate? It's glass, for god's sake. A little hot water and soap shouldn't hurt it. Besides which, I'm more likely to drop it than the dishwasher is to damage it."

Barbara: "It's not dishwasher-safe."

Robert: "Sure it is. It says so right here on the stem - 'Dyshe-washere saefe.'"

Barbara: "Don't be ridiculous. There were no dishwashers in 1780."

Robert: "Hah. Shows how much you know. Leonardo Da Vinci invented one in 1483. Ben Franklin's improved model sold in the millions. Well, in the dozens, anyway."

Barbara: "You're impossible. If you don't want to wash it, just say so."

Well, perhaps this exchange is slightly exaggerated, but that was the essence of it. So, because the discoverer of a phenomenon gets to name it, I hereby dub this Dishwasher Darwinism. A quick search of AltaVista and Northern Light for +"dishwasher darwinism" didn't yield any hits, so perhaps I'll trademark the term.

I ended up hand-washing the crystal, of course.

Sunday, September 5, 1998

As usual on Sunday mornings, Barbara is off playing golf with her father. I'll take this chance to do a little work on the web site. Tomorrow, I'll be back in heads-down mode working on the book.

Updated: 31 May 2002 14:22

Copyright © 1998, 1999, 2000, 2001, 2002, 2003, 2004 by Robert Bruce Thompson. All Rights Reserved.