photo-rbt.jpg (2942 bytes)

Email Robert

Daynotes Journal

Week of 6 December 1999

Sunday, 12 December 1999 10:17

A (mostly) daily journal of the trials, tribulations, and random observations of Robert Bruce Thompson, a writer of computer books.


wpoison

 

 

 

 

Search [tips]

TTG Home

Robert Home

Daynotes Home

Links

Special Reports

Current Topics


Jump to most recent update

 Add Daynotes Journal to Favorites (IE4/5 only)


Monday, 6 December 1999

[Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week]


Back up at last. My site was not responding when I first checked it at about 07:00 today, and has been down throughout the day. When I checked about 16:00, it was still down, but by 16:15 it had come back up.

Barbara ran some errands this morning. Among them was dropping by Billings Electronics with the JVC CD changer to see if he could get it to let go of the CDs that were stuck in there. He pressed a spring, the cartridge popped out, and he informed Barbara that there was nothing wrong with the hardware. The cartridge contains five or six plastic CD carriers. When Barbara loaded the cartridge this time, for some reason she loaded fewer CDs than it can hold, and failed to put the empty carriers back into the cartridge. Apparently, this caused the system to lock up tight, which I consider poor design. At any rate, Barbara now has her CDs back and we again have a functioning CD changer. 

Billings Electronics was first recommended to me by my friend David Rowe. David was impressed because when he took something in to be repaired, Mr. Billings told him that it wasn't worth repairing and that he should buy a new one instead. We've had the same experience several times now. I took in an old VCR. Mr. Billings told me that he could repair it, but that the parts and labor would total about $100. He said that although the VCR had originally been a $400 unit, I could buy a better one now for $150. If I fixed the old one, I'd have a repaired five year old VCR that something else might fail on shortly. I took in an older TV. Same deal. And he never charges me for his advice, although I always try to get him to accept at least a small fee. He says just to keep bringing in stuff and recommending him to my friends because he makes his money fixing things that it makes sense to fix. How refreshing it is to encounter someone like Mr. Billings in this day and age.

I need to get this published up to the server to see how the weekly changeover works. My mail is still floating around in the ether, but it should be coming in over the next few hours as the sending SMTP servers retry. More later, probably tomorrow.

 


 

 

 

Search [tips]

TTG Home

Robert Home

Daynotes Home

Links

Special Reports

Current Topics


Tuesday, 7 December 1999

[Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week]


I was curious to see how badly pair Network's move had impacted my traffic, so I downloaded and processed my web stats for yesterday. They were, as expected, off badly. I got a total of only about 700 page reads yesterday, and nearly 200 of those were from the Atomz spider indexing my site, leaving only 500 or so "real" page reads. On a normal Monday, I'd expect something like 1,200 to 1,400 "real" page reads, mostly of this week's and last week's journal pages.

It didn't help that my site was inaccessible for much of the day, including the entire periods from 10:00 to 16:00 and from 17:00 to 19:00. I suspect it was also inaccessible for most of the 9:00 to 10:00 hour, and I know that it was dead from about 16:30 through 19:45. Although some hits show up from 05:00 through 10:00, I know that I couldn't get to the site from 06:45 onward, and I suspect the same is true for many others.

We had some excitement with Malcolm yesterday. I was sitting in my office working when I heard Malcolm screaming. I jumped up and ran down the hall looking for him. He'd stopped screaming by the time I got down the hall, but I knew from the sound that Malcolm was truly terrified or injured. He wasn't in our bedroom. He wasn't in Barbara's office. He finally screamed again, and I found him in the master bathroom. He'd curled up around the back of the toilet to take a nap and woken up wedged. I finally had to grab both his back legs and pull him out backwards to extract him.

The excitement continued last night. I heard Barbara shouting at him in the library, where she is installing and configuring our Saturnalia tree. She was chasing Malcolm around trying to prevent him from eating something. He ate it before she could get to him. She thought it was one of those white Styrofoam packing peanuts. That scared the hell out of me, because I was afraid it would plug up his tiny little alimentary canal, requiring surgery. We immediately called our vet, Sue Stephens at home. Barbara always feels guilty about taking advantage of our friendship like that, but I tell her that I fully expect Sue and Robin to call when they need computer help, so what's the difference? What are friends for, if not to get advice from?

At any rate, Sue said to give him one tablespoon of hydrogen peroxide to induce vomiting, and to repeat after 20 minutes if he hadn't thrown up by then. That raised the first problem: how to get a tablespoon of nasty-tasting liquid down the throat of a squirm-puppy. We settled on Barbara holding him down whilst I sucked up the proper amount into a turkey baster and shot it down his throat. We then got him out to the front yard, watched and waited. Sure enough, after ten minutes or so, he threw up. Lots of partially digested puppy chow was visible in the mess, but no Styrofoam peanut. So we went in and got a bottle of water and poured it over the mess while shining a flashlight on it and trying to keep Malcolm and Duncan away from it. No joy.

Back into the house for a second dose. This one was worse, because Malcolm knew what was about to happen. But we got the hydrogen peroxide into him, and it was my turn to go stand outside freezing in the dark. This dose worked even faster, and he threw up a white, slimy thing that I can only assume was the partially digested remains of the peanut. He's fine this morning, so we assume that we got to him in time.

What never ceases to amaze me about dogs is that they don't hold any of this stuff against you. After being held down and having us squirt obnoxious stuff down his throat, Malcolm comes right back to us for attention and curls up in our laps. I guess that by dog rules, a senior pack member can really do anything he wants without having it held against him. Speaking of which, Barbara was sitting reading on the sofa last night while Duncan and Malcolm tussled on the floor. They make distinct "tussle noises" which sound like ferocious growling to outsiders, but which we can easily identify as playing sounds. In mid-tussle, Barbara heard Malcolm yipping in pain. She looked down and saw Malcolm was dangling from Duncan's fangs. Duncan had picked him up by the ear, and Malcolm was not amused. Life's tough when you're a little guy.

I've changed my search engine from Thunderstone to Atomz. Thunderstone originally said that they'd index my site automatically once a week. That quickly turned into once every two weeks, and at one point they went two full months between index passes. So I switched to using www.atomz.com, which allows you to specify when and how often your site is parsed.

Welcome to the latest on-line daily journal keeper. Dan Bowman now has The Timesink up and running. Check it out.

* * * * *

-----Original Message-----
From: Sjon Svenson [mailto:sjon@svenson.com]
Sent: Sunday, December 05, 1999 5:33 PM
To: thompson@ttgnet.com
Subject: Re: Hurt feelings

That PNG problem is a bit strange. When I click on an thumbnail in Netscape it pops up (well, popping in the sense of snail) normally. In Opera and MSIE4 it doesn't show but I get an empty page. The backward-forward trick works. This is in Win 95 and Win 98. I don't have NT to test it. Under OS/2 I don't get pictures, only GIF is supported in the old IBM browser (a V1 product) and I don't mind that because I use that browser only for reading the IBM online manuals.

Well the cheaper motherboards you find here are not jumpered (yet). Of course the non-Intel chips are quite likely to work, if they adhere to the old specifications, as the Celeron does. what I meant to point out is that Intel is keeping its concurrents hopping around and testing. No one is allowed to complacently collect money while wearing out their buns :-) .

Yes, the behavior of PNGs is odd to say the least. It appears that even the same browser may treat them differently, depending on factors that I can't get a grip on. I have, for example, one copy of IE5.0 that renders PNGs correctly whether they're accessed directly or via an in-line image tag, and another copy of IE5.0 on a different machine that displays PNGs properly only if they're referenced by an in-line image tag. I have no idea what determines how IE handles them. I wish I did.

As far as the Socket 370 stuff, I still think it's a tempest in a teapot. But we'll see what the market thinks.

* * * * *

-----Original Message-----
From: Rick Boatright [mailto:boatright@cjnetworks.com]
Sent: Sunday, December 05, 1999 9:28 PM
To: webmaster@ttgnet.com
Cc: sjon@svenson.com
Subject: purse science

Oh. I can _not_ let that one go unanswered...

>Ph.D's all over the place and can any one of them explain how finding a few grams of water ice on Mars for billions of $ will change anything here for us?

I started to answer this letter in detail point by point. I withdrew that, and try again with a simple answer that I HOPE the author can understand.

Mars science, and other science has payoffs. Pure research has always shown a profit. It always will, but regarding MARS missions, let's ask one simple question. Where does the questioner think that the millions where SPENT? Did we send the MONEY to mars? No.

The research is good and useful, and the money circulated in OUR economy. Where's the beef?

* * * * *

-----Original Message-----
From: Roger G. Smith [mailto:rgsmith@c-gate.net]
Sent: Monday, December 06, 1999 7:54 AM
To: webmaster@ttgnet.com
Subject: hardware books

You wrote: "I'm greatly looking forward to getting these hardware books done and starting in on Windows 2000. " ...at which point it will be time to revise the hardware books -- <grin>

At least you seem to be making the effort to do it right, which should establish the title and lead to continuous sales. There are many hardware titles now, but precious few *useful* ones.

Good luck.

-Roger

Thanks. Yes, it is very slow work. When I was writing software books, I was accustomed to cranking out a rough-draft chapter in about a week. It's more like a month on a hardware book. Oh, I could knock out hardware chapters one a week if I wanted to. The problem is that in order to make the chapters actually useful, I have to do all this stuff. I may spend one day actually writing and four days building and configuring hardware, experimenting with various permutations, and so on. I'm beginning to suspect that many hardware books merely copy information from other books, errors and all. I know that I've found some identical statements in several books that turn out not to be the case. I wonder how much of the other stuff is simply copied also.

 


 

 

 

Search [tips]

TTG Home

Robert Home

Daynotes Home

Links

Special Reports

Current Topics


Wednesday, 8 December 1999

[Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week]


My mother told me yesterday that someone had called her from the airport (she's extension 1 on the automated attendant, so she gets a lot of calls intended for me from people who are too lazy to listen to the message) to say that a truck would be delivering something for me in the afternoon. My Hitachi 19" monitor, no doubt. The truck showed up yesterday afternoon, and the guy hand-trucked the monitor into the house. It weighs 64 pounds (29 Kg). I remember a time when I wouldn't have thought twice about picking up that much in each hand and carrying it a long distance. No more, though. I let him use the hand truck.

Now the problem is what to do with it. It ultimately gets connected to kiwi (my new main workstation), of course, but right now there's no room for it on my desk. Kiwi sits on the floor at the bottom right. The machine at the far right of the desk is kerby, my current main workstation. On the desk in front of kerby is an L. L. Bean catalog, which reminds me that I'd better order Barbara's Saturnalia presents. That switchbox on top of kerby switches the Dell 17" monitor between kiwi and kerby. The keyboard and mouse I'm temporarily using on kiwi are sitting on top of the monitor. Just visible on the floor to the left of the filing cabinet is the APC Smart-UPS/1100 that powers kiwi and kerby. That little box in the far corner is a 675 VA Tripp Lite OmniSmart UPS that powers odin (and usually another tower system that normally sits under the left side of the desk but is presently MIA.)

Immediately to the left of the 17" monitor is odin, a Dell Dimension XPSM200s Pentium/200 system that I run Windows 98 on to use for screen shots and so on. Those 5.25" floppy cases were salvaged from the basement recently (I never throw anything away), and are now used to store CDs. There are a couple of CD wallets on top that contain CDs I use regularly. To the left of odin is an old Mag Innovision 15" monitor that's good enough for what I use it for, which is mainly capturing screen shots from odin and doing maintenance and configuration tasks on the tower server that ordinarily sits on the floor beneath the scanner.

There's just not room for a 19" monitor on that desk without some serious re-arranging. I think what I'm going to do is remove the Mag 15" monitor, shift everything that currently sits to the right of it over to the left, and put the Hitachi 19" in the space thus freed. I'll connect the Dell 17" monitor to the switchbox and connect the switch to odin and kerby.

Actually, that's only a stop-gap, because there's still not really room for everything. I think I'll move the scanner back to Barbara's office. One problem with that scheme. The scanner has USB and SCSI interfaces. It's currently connected to odin via USB. But Barbara is running NT4 Server, which means no USB support, which means I need to connect it to her box using SCSI, which means I need the appropriate external SCSI cable.

"A thing for everyplace, and everyplace in its thing."

* * * * *

-----Original Message-----
From: Chuck Waggoner [waggoner at gis dot net]
Sent: Tuesday, December 07, 1999 5:10 PM
To: 'Robert Bruce Thompson'
Subject: Lighted Brumalia Tree

Dear Mr. Electricity,

We have one of the most extravagant wastes that I've ever seen.

Across the street from us is a small park. Behind that is a modest 50 car parking lot for the commuter rail stop. Beside that, at the end of the parking lot that borders the street, is a seven-foot Brumalia tree, decorated with a couple or three strands of lights.

And right next to that is a BIG Onan diesel generator built on its own integral trailer, running 24 hours a day, powering lights on the tree. The generator has an AC ammeter which top scales at 300 amps. The lowest marked reading is 50 amps. When I observed that meter, it looked like the needle was in minus territory, so that gives some idea of the draw.

That generator is every bit as large as the ones we use for big TV remote broadcasts where we can't get a power drop, and which powers equipment truck, sound truck, and all lighting for the program. Of course, the generator is owned by the big state park agency, so one must assume tax dollars are therefore involved in this genius scheme to light that modest, lone Brumalia tree.

They sunk a brand-new 20 foot pole next to the generator, and a teeny tiny extension cord runs from the generator to the top of that pole, which then runs wiring in heavy gauge metal tubing, down the pole to a very industrial grade-looking weather covered outdoor power outlet--probably inspected and approved by our city for heavy duty use.

Now directly across the street from the tree is a little grocery store/deli. Why in Brumalia's name didn't they just run a cord over there to the big outdoor sign they have, which is also about 20 feet in the air, drop it down to the heavy industrial grade-looking weather covered outdoor outlet that they have over there, and pay the owner 20 bucks for his electricity and trouble?

--Puzzled in Boston 

I long ago ceased to be surprised at anything the government did, but this does seem a bit extreme.

* * * * *

-----Original Message-----
From: Dave Farquhar [mailto:farquhar@lcms.org]
Sent: Tuesday, December 07, 1999 3:45 PM
To: thompson@ttgnet.com
Subject: IE PNG handling

Bob,

I notice on one of my boxes with IE4 installed, IE4 uses Apple's QuickTime (which I also have installed) to handle PNG files. That may be part of the mystery; I know QuickTime is common, but far from ubiquitous. But the IE4/QuickTime combo handles PNGs about as well as Netscape 4.x does.

I knew that Navigator and IE4 required external help, but I'd prefer to have IE5 display PNGs natively. I know it will do so, both because Microsoft says so, and because I've seen it do it. I've about concluded that IE5 by default displays PNGs correctly (well, more or less), but that installing IrfanView somehow damaged that ability. On a clean install, IE5 displays PNGs whether they are embedded in an HTML page or linked to directly. On the machines that I've installed IrfanView on, IE5 displays PNGs only if they're embedded in an HTML page with an in-line image tag. Clicking on a linked PNG simply displays the dialog that prompts you to download the file or run it directly. I can't swear that it's IrfanView, but that's the way it looks at present. It makes no difference how file associations are set, and I can't find anything in the registry that might be causing the problem. Oh, well.

* * * * *

-----Original Message-----
From: Jan Swijsen [mailto:qjsw@oce.nl]
Sent: Wednesday, December 08, 1999 11:09 AM
To: Robert Bruce Thomson
Subject: hardware books

>...suspect that many hardware books merely copy information from other books, errors and all.

Yep, I think you can take that suspicion for fact. Something in the style of an author starting on a hardware book because he knows a lot about it and does his work correct, until he hits an area that he doesn't know a thing of and that still needs to go in the book.

This doesn't only apply to hardware books. I remember my father laughing terribly at some geography textbooks. In one of his own books he had a picture of the earth, the sun and the moon to illustrate the various phases of the moon. That picture was reversed in the layout process (separate layers were printed on transparent film before making the print plates, the films containing the picture were reversed). This had slipped the reviews but in the actual book there was an errata page notifying this error, and the picture was corrected in the second edition. Some years later two other textbooks came out with the same, wrong, picture. And of course no reference to an original source.

And I sympathize with those authors. No one can be an expert on everything, and producing a hardware book almost requires that you depend on other people's expertise. I guess that's my problem. I don't take people's word for stuff. I actually do it. For example, with CD-R, I actually installed both ATAPI and SCSI CD-R drives under both Windows 98 and Windows NT4 and burned a bunch of CDs. I tried different master/slave arrangements. I tried putting the CD-R and hard drive on the same SCSI channel versus separate channels. I tried different brands of blank discs in different environments. All told, I put literally a month's worth of work into verifying stuff that turned into a few pages of text. 

Doing things this way obviously takes time, and a lot of it. But even after doing this, there are two problems: (a) someone reading the material has no way to judge how reliable it is vis-a-vis material written by someone else who just copied others' mistakes, and (b) although the information is valid for the hardware I used, it may not be for other hardware. If I find that a Smart & Friendly SAF-798 ATAPI CD burner works reliably under the conditions I test, that doesn't necessarily mean that another brand of ATAPI CD burner will also be reliable under those conditions.  I wish there were some means of addressing those issues, but I don't know of any.

 


 

 

 

Search [tips]

TTG Home

Robert Home

Daynotes Home

Links

Special Reports

Current Topics


Thursday, 9 December 1999

[Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week]


It has come to my attention that several members of the Daynotes group are filthy swine when it comes to maintaining their workspaces in a neat, clean, and uncluttered fashion. As an example to encourage those filthy swine (you know who you are) to clean up their acts, I'm posting a picture of my main desk below. This is actually the typical appearance of my desk (at least so far). I didn't clean up specifically for the purpose of taking this picture.

Kiwi sits on the floor at the bottom right. Sitting on top of kiwi is kerby, my current main workstation. The monitor on the right is the new Hitachi 751 19" display, with the keyboard for kerby on top of it. Kerby's mouse sits on the blue mouse pad. The mouse immediately to its left, sitting directly on the desk, is one of the new Microsoft IntelliMouse with IntelliEye models. They are purely optical, eliminating both the roller ball and the need for a mouse pad. That's a blessing for me, because I have to clean my old mechanical mice about once a week. Unlike old-style optical mice, the IntelliMouse with IntelliEye doesn't require a special surface. It works fine on most surfaces, excepting reflective ones and ones with no pattern at all. It works fine, for example, on the bare wood of my desk, picking up the underlying grain pattern as a reference. Just as an experiment, I tried using the top of kerby as a mousing surface. That worked fine, too, even though it's a featureless beige. The IntelliEye mouse is pretty impressive. The keyboard directly in front of the 19" monitor is one of the new Microsoft Natural Keyboard Pro models. It has a bunch of dedicated (and reprogrammable) keys that provide one-touch access to various browser functions, email, and so on, although I haven't installed the drivers to enable those functions yet. 

Between the monitors sits a KVM switch, which currently switches the 17" Dell monitor on the left between kerby and odin (to the monitor's left). At the top rear of the scanner, you can barely see two video cables that will soon connect the KVM switch to the two tower boxes on the far left floor. On top of the KVM switch and the 17" monitor are a set of Altec Lansing speakers with dual inputs, which are currently connected to odin and kerby. I haven't gotten around to installing speakers for kiwi or the two tower boxes. The tower to the far left is an old Gateway Pentium/133 box that will soon run Linux. The tower immediately to its right is Barbara's old main workstation, a Pentium II/300 with 128 MB, which will soon be a Windows 2000 Server. The Tripp Lite 675 UPS still sits behind them, but I'm afraid it's a bit light to run both towers, odin, the scanner, the 17" monitor, and so on. I have a 2 KVA Smart Power Systems SineSmart SS2000 UPS that I'll probably swap in there. With the American Power Conversion Smart-UPS SU1000NET that sits immediately to the left rear of the filing cabinet, that gives me 3 KVA of UPS under my desk, which should be more than sufficient. I'll relocate the 675 VA Tripp Lite to my test-bed bench.

In the interests of journalistic integrity, Barbara insisted that I turn around after taking the picture above and shoot in the opposite direction. Oh, well.

Several people have requested that I post a picture of Malcolm modeling his new snout-wear. Several people have also suggested that perhaps I was overstating the case a bit when I called Malcolm "demonic." So, to kill two birds with one stone, here's a picture of demonic Malcolm wearing his new muzzle. It's actually on upside-down. When we install it properly, the part of the muzzle against his throat in this picture actually covers his snout from the tip of his nose to about half way between his eyes and ears. When we put it on that way, he ran around bumping into things. As far as the demonic thing, you can see it in his eyes. In this case, he looks like a traffic signal. If he had three eyes, I'm sure the third one would have been yellow. I've heard of redeye before, but greeneye is a new one on me. I also have flash photos of Malcolm showing his eyes reflecting as bright blue, yellow, orange, pink, and a rather pretty violet color. This characteristic is called "chatoyance" and is particularly common in demons.

 


 

 

 

Search [tips]

TTG Home

Robert Home

Daynotes Home

Links

Special Reports

Current Topics


Friday, 10 December 1999

[Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week]


I really must do something about the mail forwarding at pair Networks. It's completely unreliable in several different respects, some of which are related to the way it parses the autoforwards and others to the way pair Networks has chosen to implement autoforwarding. As an example of the latter, I get some (but not all) return receipts addressed to barbara@ttgnet.com. According to pair Networks, they do this intentionally to prevent mail loops. Well, if it's intentional it should be consistent, and it's not. More to the point, it shouldn't be configured that way at all. No other service provider I know of does that. As an example of the former, I have a message sitting in my inbox which Steve Tucker forwarded to me. The message was from Tom Syroid, and was addressed

To: Steve DeLassus; Robert Thompson; Moshe Bar; Matt Beland; Brian Bilbrey

I have a line in my autoforward configuration file on the pair server that reads:

FORWARD steve steve@wakeolda.com

In theory, that means that any message addressed to steve@ttgnet.com is automatically forwarded to steve@wakeolda.com, and that indeed happens. The problem is that on Tom's message (or any message that includes "steve" anywhere in the To: or CC: line), the copy intended for me is instead forwarded to steve@wakeolda.com. Fortunately, Steve is very good at catching those and forwarding them to me. But Steve is on the road a lot, and I could end up not seeing a critical message for hours or days if he's unable to check his mail. That's unacceptable. When I contacted pair tech support about the problem, they suggested making the forward explicit, e.g.

FORWARD steve@ttgnet.com steve@wakeolda.com

I tried that method some time ago, but it introduced problems of its own, the details of which I forget. I think I'll try that again now, come to think of it. Perhaps that will also solve the other bizarre problem. If someone sends a message to both Barbara (barbara@ttgnet.com) and me (thompson@ttgnet.com), I never get a copy of the message. Perhaps using fully-qualified local address names will also solve that. We'll see. I probably should just have gotten Barbara her own POP mailbox, but the autoforward would be perfectly adequate if it worked as it should. I've been told that pair Networks will start allowing additional POP mailboxes without charge after the first of the year, so perhaps I'll look into that.

During a long telephone conversation the other day with Pournelle, we started talking about vitamins, dietary supplements and so on. Barbara and I are a bit concerned about her dad, who's now in his late 70's, and I wanted to get Pournelle's advice. Pournelle's theory is that nature builds us to have children, rear them, and then die once they start having children, thus preserving the food supply for the grandkids. And that actually makes sense when you think it through. Assuming puberty at 11 to 13, that means that biologically we should be parents at 12 or so and grandparents in our mid-twenties. And it's right around the mid-twenties when we start going downhill fast. Our physical powers begin to decline, as do our mental abilities. In the absence of intervention, our teeth begin to fall out. Pournelle convinced me: humans have a design life of 25 to 30 years, rather than the biblical three score and ten.

Thinking all this through, I began to be a bit concerned about myself too. Knowing Jerry, I knew he would have researched all this stuff thoroughly, so I sought his advice with the following message:

I'm 46 years old. What concerns me is that my memory is no longer what it once was. I'm sure that I still fall in the 99.99th percentile in terms of my ability to remember, but my memory used to be truly stellar. Until I was about 35, for example, I literally never wrote down a phone number. I could remember any phone number I'd ever called even once. Twenty-five years ago, I played duplicate bridge seriously, and I could name literally every card in all four hands of every deal in a tournament I'd played six months or a year previously. 

What's rather strange is that I still remember, for example, that Connie Stewart's phone number, which I dialed once 35 years ago, was 654-0263, but I can't remember your number at the Beach house, which you gave to me only a few months ago. I've noticed lately that I occasionally can't retrieve a word I want. I know that the word exists, and I know that it has a subtly different meaning from the word I want to replace with it, but it just won't come to me. I've never used a thesaurus before, but I may start.

I'm sure that this is due at least in part to Conan Doyle's "full attic" issue. I suspect that my attic is now full, and for each new thing I remember, I forget an old thing. But I'm concerned that age is also taking its toll, so I want to do whatever is necessary to eliminate or at least reduce the downward slide. Perhaps I should test my memory now by memorizing long strings of random numbers and then compare my current results with a similar series of tests after taking these drugs for several weeks.

Any advice you have will be genuinely appreciated.

Jerry recommended I start with SAM-e and CoQ10. I forwarded Barbara that message, along with the one from Dr. Jim listing the supplements he takes. Barbara returned from her drugstore run Wednesday with a whole collection of bottles for me--1,000 mg Vitamin C, 1,200 mg garlic, 400 IU Vitamin E, 150 mg CoQ10, and 200 mg SAM-e. I took the first batch of everything but the SAM-e on Wednesday evening, and the first dose of SAM-e yesterday morning. We'll see how things go.

 

 


 

 

 

Search [tips]

TTG Home

Robert Home

Daynotes Home

Links

Special Reports

Current Topics


Saturday, 11 December 1999

[Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week]


First impressions of kiwi: it's fast, as you might expect a dual Pentium III/550 system to be. Overall, kiwi "feels" perhaps four times faster than kerby, the Pentium II/300 that it replaces as my main workstation. About half of that, of course, is due to the relative processor speeds. Probably as important as the faster CPU is that kiwi uses U2W SCSI throughout and a 10,000 RPM Seagate Cheetah as the system drive. Disk access is much faster than with the EIDE drive in kerby. It also can't hurt that kiwi has 256 MB of memory, versus 128 MB on kerby

Dual processors have little to do with how fast a system "feels" when running typical applications, since most mainstream programs run little faster with dual processors than they do with one. But the dual processors really come into play when the system is running a lot of stuff. Nothing I try bogs this system down. Regardless of how many windows I have open and how many background processes running, it just keeps chugging along at 550 MHz. With a single processor, I used to run out of steam when I had a lot of things going on at once. Not with this system. I can, for example, burn a CD while listening to an audio CD (kiwi has a CD burner, a CD-ROM drive, and a DVD-ROM drive) while publishing my web site while downloading my mail while surfing the web. Everything proceeds at full speed. Basically, there's always a Pentium III/550 available to dedicate to the foreground task, so no matter what's going on in the background, the foreground task seems as snappy as ever.

The Hitachi 751 19" monitor is very nice, as expected. I ran it for a while at 1600 X 1200 at 75 Hz refresh. I expected to see some flicker or image instability at only 75 Hz at that resolution, but the combination of the Matrox Millennium G400 and the Hitachi 751 looked perfect to me. No flicker. Rock solid. I cut it back to 1280 X 1024 at 85 Hz, though, because things were just too small to see at 1600 X 1200. Even at 1280 X 1024, I ended up changing to the Windows Standard (Large) scheme. With the Windows Standard scheme, menu items and toolbar icons were too small to be usable. Although 1280 X 1024 doesn't sound much larger than 1024 X 768, nor 19" much larger than 17", the difference is really surprising. I have a lot more screen real estate now. I can see where it will be useful

I really must buy some Saturnalia presents for Barbara. I've actually made a start of sorts. So far, I've gotten her a pretty blue, 14 foot, Category 5 Ethernet drop cable. That, and some Cat 5 modular snap-in jacks to go with it. Just what every woman dreams of finding under the Saturnalia tree. That even beats the 3/8" drive 12-point metric socket set I got her one year. And they say men aren't creative when it comes to choosing gifts.

My thanks to Gary Berg, who jumped in to help with my forwarding problems at pair Networks. I haven't quite gotten up the nerve to start editing my .procmailrc file manually, but here's the exchange of messages relating to that:

-----Original Message-----
From: Gary M. Berg [mailto:Gary_Berg@attglobal.net]
Sent: Friday, December 10, 1999 10:11 AM
To: Robert Bruce Thompson
Subject: Pair network and forwards

Bob,

Been there, fixed it. This is the .mailproto file I started with:

# Main forwards
#
# The .procmailrc file will need to be edited for each of these to
# change the ":0" in front of each of these to "0:c".
#
FORWARD To jbunke@myowndomain.com jbunke@ispdomain.com
FORWARD To gary_berg@myowndomain.com gary_berg@ispdomain.com
#
# These will discard everything already forwarded above.These are needed
# to keep all of the modified ":0c" messages which were forwarded from being
# sent to the catchall forward at the end.
#
DISCARD To jbunke@myowndomain.com
DISCARD To gary_berg@myowndomain.com
#
# Catchall - everything not trashed above comes to my CompuServe account
# Also go in and remove the entry in the .procmailrc for this which
# conditions to not forward from a DAEMON.
#
FORWARD To . gary_berg@secondispdomain.com

Note the comments. By editing the :0 to a :0c a copy of the message is edited and processing continues. Otherwise, on the first match the message is forwarded and nothing else happens to it. Also, by removing the generated line for the last forward which mentions a DAEMON anything which falls through goes into my second outside mailbox. Until I killed the DAEMON mention I had a handful of newsletters emails which shot straight on into my POP mailbox at Pair. 

I puzzled this all out from the procmail information. Using the .mailproto file and the pmproto program made it easy to get the general outline in place. Pair support didn't seem to be able to help with this, but eventually I got it to work correctly. I guess I spent 2-3 hours on testing it, and then another hour or two tweaking it over time. You can, of course, use more FORWARD commands, but be sure to pair them with DISCARD commands. I assume there is some way in a FORWARD command to forward it to your normal POP mailbox on Pair, but I haven't checked that out yet. I hope that Pair does start supporting free POP mailboxes, especially with more than 1Mb of space. I don't use more than a few hundred K of my allotment at Pair because what I really want is a permanent mailbox for myself and my wife. Because I need to control the forwarding to send her stuff somewhere else I had to get a Basic account instead of an FTP account.

Thanks. I was in playing around with my mail forwarding, and now have things screwed up. Barbara is getting mail addressed directly to her by individuals just fine, but she belongs to a couple of mailing lists, and every message from those listserves is coming to the default box and thence to me. I even tried putting the .mailproto file back exactly as it was and then re-running pmproto, but it's still doing it. I've mailed urgent@pair.com to ask them if they can fix it.

That was all before I got your message. I'll give the method you suggest a try, although the "discard" lines scare me. I assume that you have a typo in your comments. Early on, you refer to changing ":0" to "0:c", but later you have it as ":0c". I assume that the latter is correct (and that that's a "zero" rather than a "cap O")

Yes, mailing list messages are a real PITA with procmail. Most such messages come through with the destination mailbox as part of the SMTP header only, not in the body of the mail message. The only way I can suggest dealing with that is more FORWARD commands based on the source of the email or something in the title line. Pair has a pointer to the docs for Procmail on their site. One of the sections explains why mailing list messages can't be matched in Procmail.

I hope that sometime in the future Pair provides a better tool for mail handling. Just having separate POP boxes which can have their own forwards would be great. Ideally you'd be able to specify forwarding for each box separately as well as for the "catch-all".

>> although the "discard" lines scare me. <<

Yes, I had the advantage of being able to play with the messages when I had nothing external pointed to my domain. There is some mechanism you can use to file a safety copy of all incoming mail, if I recall correctly.

Also, I believe you can give procmail the name of it's resource file on the command line somehow, and then turn it loose on an input file. So you can have your current .procmailrc file in effect while testing a new one against sample email files.

>> I assume that you have a typo in your comments. <<

Probably. It is indeed ":0c" - the colon is a flag that this is the start of a new matching command. It is indeed a "zero".

Yes, but the mailing list traffic *was* working this morning. I made one simple change, altering "barbara" to "barbara@ttgnet.com" and then I started getting all her listserve mail. Even when I changed it back to exactly how it was originally, I still get her listserve mail.

The truth is, I don't want to learn the nuances of procmail. I just want messages addressed to "barbara@ttgnet.com" to be automatically forwarded to "thompsrb@bellsouth.net". That shouldn't be too hard, should it? pair support told me they were going to replace procmail with something else, but that's been months ago.

Thanks for all your help. I may give editing my .procmailrc file manually a shot, or I may just tell pair to set up a POP mailbox for barbara@ttgnet.com.

>> I made one simple change, altering "barbara" to "barbara@ttgnet.com" and then <<

You did re-run pmproto after you put back the original configuration file (.mailproto) that pmproto reads? If you didn't re-run pmproto, you'd still be running with the modified profile for procmail (.procmailrc).

>> The truth is, I don't want to learn the nuances of procmail. <<

Oh, sure, I agree with you. Pair certainly needs to upgrade their mail forwarding capability. Among other things it needs to deal with the delivery of a message to all mailboxes or forwards on the domain, even if more than one is in the "To:" header. And it needs to be able to process the SMTP header so that mail sent to a mailing list can be routed properly.

Actually, "all" that Pair would have to do is to modify their email system to write the SMTP routing header into the text headers of the email message so that procmail could see it. It could be an additional "To:" entry, or an "X-SMTPTo:" or something. Ideally they'd modify procmail so it included this header when you told it to match on a "To:".

I object to the current system of additional mailboxes on Pair in two ways:

1) I shouldn't have to pay extra for another mailbox.

2) My overall usage of space, web+email, should go against my account balance instead of $1/month/MB of email storage. If I have 30Mb on their server it shouldn't matter if it's all in email or all in web pages.

Hmmm, when you get an additional POP mailbox at Pair, how is it associated with the proper incoming name? I'll bet it's a second user ID on their server that shares the same home directory with your regular domain. I wonder how you tell it that barbara@ttgnet.com goes into that mailbox? I hope you don't have to use Procmail!

I was trying to look at Pair's "pair2000" site today and it appears to no longer be functional. I don't know if it was a casualty of the move or if that server is down. I was hoping to find a description of the changes they hoped to make for mailboxes and forwarding support. I did find some of those pages cached on Google.

What docs I could find indicated that they hoped to roll Pair 2000 out later this year after they recover from their data center move. Those cached pages had the move in mid-November, so perhaps it will be delayed until January.

I did re-run pmproto each time, which is why it's so weird that it's not working as it was this morning.

Good points about pair. They told me long ago that they were going to stop using procmail and switch to something like qmail, but I haven't heard any more about it.

It appears that probably 50-60% of the mailing list mail I get contains a line like:

"X-Envelope-To: <me@mydomain.com>"

which is read by procmail in processing. However, I got an email this morning from a friend who used Bcc to hide from the recipients who she sent the message to. Probably because she sent it to 10-15 people, and didn't want them all to show in the "To:" section. That would have only landed in the catch-all forward on my system (the mail was sent from AOL). So you'll have the same problem on anything sent with a BCC to Barbara - it'll land in your catchall mailbox because there will be no indication in the body of the message that it was sent to you.

So if Pair added writing in an "X-Envelope-To:" entry (when there wasn't one) based on the SMTP envelope procmail would work fine. Except for the hassle of handling the multiple receivers in the same domain. A little work on pmproto to do what I indicated you could do would solve that problem too.

I hope you can get her mailing list subscription working again the way it used to!

Thanks. I think what I'll ultimately do is just let pair dump all *.ttgnet.com mail into the main mailbox. I'll then have my Linux system POP all the mail from pair and deliver it to local mailboxes on the Linux system. Neither Time-Warner Cable nor BellSouth plan to make any provision for issuing static IP addresses to their cable modem and xDSL users, or I'd just bag pair entirely and run Apache locally. I don't really mind not having my web server local, but I'd *really* like to have my mail server under my direct control.

* * * * *

-----Original Message-----
From: ROBERT RUDZKI [mailto:rasterho@pacbell.net]
Sent: Friday, December 10, 1999 10:05 PM
To: RBT
Subject: Vitamins are ok but the other stuff I have to wonder about.

Sorry to hear the years are taking their toll of your mind and body, the part of human aging that surprises me is that I have known people in their 90's sharp as a tack, never sick a day in their lives, and they die peacefully in their sleep. Others lose their minds in a slow and painful process through Alzheimer's starting in their 60's and are drooling idiots in wheel chairs in a nursing home 10 years later and live for 15 more years after that putting the families and caregivers through hell until they finally die. What scares me bad is you can't tell which way you will end up...

While the majority effects of vitamin deficiencies are well known and easily fixed by diet and eating vitamin supplements, all that new age herbal health stuff makes me suspicious. Has any of this really been double-blind tested on people? How do we know what these herbs do? Folklore? Nearly every human culture has alternative remedies that predate modern medicine and diet knowledge, of course they are all different. Kind of like the Bible being the True Word of God, so why are there different versions and which ones are wrong?

We can point to the various medical practices Western society believed in, that we now know to be dangerous, unnecessary and just plain wrong, ie, bleeding, transfusing newborns, circumcising male babies, pre-frontal lobotomies, intact D&E of a healthy 29 week fetus, and a host of others your doctor readers could add in, I am sure.

I take 2 generic multi-vitamins, 800 units of Vitamin E, and a 500mg Vitamin C daily. The Vitamin E makes my chest and facial hair a bit thicker, but since I have worn a beard since the day after I retired from the USAF back in 1991, I can't see much other effect if I stop taking it. Not taking the other stuff has less effect, although getting enough sleep and keeping hydrated properly affects my feeling good or bad much more depending if I get enough.

I don't know how much non-caffeine beverage you drink in a day, but most Americans go around chronically de-hydrated. [According to most flight surgeons at USAF survival school and all the hot country initial briefings I sat through in 20 years!] As they say when you feel thirsty you are at least a quart behind on your body's need for water. I like drinking club soda or sparkling waters for the taste, our tap water is quite good but very hard and has to run for a minute or so until the mains water reaches the tap from the street. Sitting in the small pipes overnight, it is just awful. [Our coldwater pipes are a mixture of new PVC and 50 year old galvanized and cast iron mains in the street]

I remember you said you like Coke in 3 liter bottles, just be aware the more Coke you drink, the more fluid you lose since caffeine is a diuretic...

The Pournelle '3.5" Magnum shotgun' approach to herbals and supplements scares me, what potential interactions exist among the 27 pills he takes daily? Once again who can vouch for what this stuff really does, if anything? What human research has been done on any of this stuff? I have heard a rumor that Stabilant-22 is actually re-badged DMSO for a higher price...

I saw a survey taken of heath food store shoppers, they all were there spending quite a bit of money for their health they said, but half of them admitted having driven to the store not wearing a seat belt...

For the benefit of your cat-owning readers, can you post some pictures of 'expressing anal glands on dogs'? This could help remind us of why we own cats instead of dogs... =8^-)

Robert Rudzki
rasterho@pacbell.net
www.home.pacbell.net/rasterho

I'm taking the same vitamins that Dr. Jim takes himself, and I don't believe there's any downside to them. As to the CoQ10 and SAM-e, I don't know if they're effective, but I doubt they're harmful. As far as my daily fluid intake, it's typically about 2 litres of Coke Classic, a pot or two of decaf coffee, perhaps 1.5 litres of tap water, and miscellaneous glasses of iced tea, orange juice, etc. All told, perhaps two gallons per day. I don't think I need worry about dehydration. Expressing anal glands is a negligible price to pay for the benefits of having a dog rather than a cat.

 


 

 

 

Search [tips]

TTG Home

Robert Home

Daynotes Home

Links

Special Reports

Current Topics


Sunday, 12 December 1999

[Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week]


Duncan and Malcolm are starting to get along pretty well. As I walked through the den yesterday with my camera, I noticed them sleeping on the sofa. Malcolm woke up before the flash finished charging.

duncan-malcolm-couch.jpg (47750 bytes)

I installed Red Hat Linux 6.1 yesterday. Short take: Linux is making great strides, but it's still not usable by ordinary mortals. I decided to install Linux on an old Gateway Pentium/133 with 64 MB and a 3 GB hard drive. I'm not a complete UNIX novice, having installed and configured XENIX more than a few times fifteen years ago, and having worked with BSDI, Linux and other variants infrequently since. I can usually get UNIX installed without problems. Once it's up and running, I know the basic commands for navigating directories; copying, moving, deleting, viewing, changing ownership and permissions, and otherwise manipulating files; shutting down the system, adding users, creating links, and so on. I'm certainly not an expert, but I do know more than the average Windows user is likely to.

At any rate, the first problem came when I attempted to boot Linux from diskette. As it turned out, the boot diskette supplied by Red Hat was defective. It took only a couple of minutes to move the RH 6.1 CD over to kiwi, format a floppy, and run rawrite to copy the Intel boot image file to the new floppy. That floppy booted fine, but this problem would have been a showstopper for many novice users.

Once that was done, Linux setup proceeded without problems. It prompted me to choose the type of mouse, emphasizing that choosing the correct type was very important. This system uses an AT motherboard, with an old-style AT keyboard connector, but a PS/2 mouse connector. I choose the Bus Mouse option, although the mouse is actually labeled "Mouse Port Compatible 2.0A". The mouse seemed to work normally during subsequent GUI portions of setup, so I thought I'd chosen rightly.

Setup prompts you to enter and confirm the password for root, and also allows you to create new users. I created user thompson and continued setup. After setup completed and the system rebooted, I entered thompson and my password at the login prompt, and Linux displayed a $ prompt. Showstopper number two for the average user. The intuitive thing to do was to type startx and press Enter, so that's what I did. Well, perhaps not intuitive for the average Windows user. Given that I'd installed RH6.1 in the default KDE workstation configuration, it would have been nice if X had started automatically. 

But that's neither here nor there, because X didn't start at all. It blew up with an error message that it couldn't open the mouse. Showstopper number three. Given that this is the Intel version of UNIX, and given that about 99% of Intel systems upon which Linux is likely to be installed have Microsoft mice, how hard would it have been to detect and configure the mouse type automatically? All other operating systems I've installed do so, so Linux clearly could also.

Oh, well. I'll keep working with Linux. It's certainly a viable server platform right now, but I suspect it's going to be at least another year or two before it has what it needs to become a mainstream operating system for ordinary users. But I don't doubt that that will happen. When it does, Microsoft is in seriously bad trouble.

 


[Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week]

 

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