photo-rbt.jpg (2942 bytes)

Email Robert

Daynotes Journal

Week of 14 February 2000

Friday, 05 July 2002 08:22

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


Monday, 14 February 2000

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


I'm now officially famous, having been (partially) quoted in The Times. Chris Ward-Johnson from Chateau Keyboard asked me last week to say something quotable about W2K. Here's what Chris submitted:

Robert Bruce Thompson, President of network consulting company Triad Technology Group, Inc. (http://www.ttgnet.com) and author of many computer networking and PC hardware books for O'Reilly & Associates, is scathing in his criticism of Windows 2000.

"Windows 2000 Professional is mostly hype. It’s essentially NT4 Workstation with Device Manager, Plug-'N-Play, DVD, USB, and DirectX, all of which could and should have been incorporated in NT4. We see no compelling reason to upgrade existing business desktops, and regard W2KP as a poor choice for home systems. W2KP is slower and no more stable than NT4 in our testing, lacks adequate drivers for most video and many sound cards, and has some compatibility problems with legacy hardware. We recommend installing W2KP only on new business desktops, or as an upgrade for very recent notebooks, for which it offers some real advantages."

And here's what was actually printed:

Robert Bruce Thompson, the president of network consulting company Triad Technology Group and the author of many computer networking and PC hardware books, doubts that Windows 2000 Professional (W2KP) offers any real advantages to business users.

"We see no compelling reason to upgrade existing business desktops, and regard W2KP as a poor choice for home systems," he says.

"W2KP is slower and no more stable than NT4 in our testing, lacks adequate drivers for most video and many sound cards, and has some compatibility problems with legacy hardware."

Using the middle part of the quote and cutting both ends rather changed the flavor of what I said.

Oh, well. It's back to work for me. Input devices this time.

* * * * *

-----Original Message-----
From: Thomas Gatermann [mailto:tgatermann@access2k1.net]
Sent: Sunday, February 13, 2000 2:02 PM
To: thompson@ttgnet.com
Subject: UPS power cords

Hello. I'm a friend of Dave Farquhar, and he suggested that I write to you with a UPS power question, that he and I could not come to a final conclusion on.

I've recently upgraded my second computer and am wanting to buy a second UPS to use on it. But I have come upon a problem dealing with the length of the cords on them. I have not found one that has a power cord longer than six feet. This is just about three feet short of what I need to be able to reach my wall outlet. My first thought was to buy a heavy gauge extension cord to take of care of the situation. But I have read in the instructions with my current UPS (Which is an APC) that it is not wise to do that. My question is, is this because most people typically buy a cord that is underrated and can't handle the power load put on them? Dave and I can't see why if I buy a heavy duty enough gauge cord that I couldn't do it. But neither of us really have any experience with this.

I also was wondering how much is to much on a single outlet. I have my main machine plugged in the above APC UPS running a 19" monitor, a 266Mhz k6-2 with two hard drives, two CD-ROMs in it, and a scanner. A friend suggested that I just plug the second UPS into the empty socket on the outlet. But it would seem to me that having two UPS' running off one outlet might be dangerous.

Any thoughts or suggestions would be greatly appreciated.

Sincerely,

Thomas Gatermann
tgatermann@access2k1.net

I suspect the reason that APC recommends not using an extension cord is that doing so is a violation of fire codes in most places. You're supposed to disconnect extension cords (including corded surge protectors) from the receptacle when whatever is connected to them is turned off. Technically, if you do something that violates the fire code and it causes a fire, your homeowner's insurance may refuse to pay. In your position, though, I wouldn't hesitate to use a short, heavy-duty extension cord, although you do so at your own risk.

How much is too much on one receptacle depends on how many amps that receptacle is rated for and how many other things are connected to the same circuit. Very few receptacles are on a dedicated circuit. Instead, the same circuit that powers that receptacle may also power the receptacle in the kitchen where you have your coffee maker plugged in, the overhead light in your foyer, and the receptacle in the den where your TV is connected. Figuring out which receptacles (and other things like ceiling lights) are connected to a given circuit requires some experimentation. I've often gone around with a collection of small lamps, table radios, and so on, throwing breakers to figure out what connects where. Unless you do that, you'll never know until a breaker blows. Even at that, there may be "hidden" stuff on a circuit that you'll never think to look for.

Residential circuits are almost always 15- or 20-Amp. At 120VAC nominal, that means a 15A circuit can support up to an 1,800W load, and a 20A circuit up to a 2,400W load. If your UPSs are typical 280VA to 650VA units, chances are you'll be able to connect the second UPS without problems. If there are a lot of other things on the circuit, the worst that's going to happen is that the main breaker will blow. That just happened here when Barbara plugged in her new 12A vacuum cleaner. That doesn't leave a lot of slack on a 15A circuit, and my UPSs went berserk.

* * * * *

-----Original Message-----
From: TTG [mailto:ttgnet@operamail.com]
Sent: Sunday, February 13, 2000 2:35 PM
To: Robert Bruce Thompson
Subject: RE: An unsolvable instance of Freecell: a game from Windows'95

Yes. From the postscript:

The starting position here actually does not occur as one of the positions generated by the Freecell program. The position 11982 actually is proved to be unsolvable, by exhaustive computer search. Follow the link above, and links reachable from there. In addition, the newer versions of the program have positions numbered -1 and -2, which are also unsolvable.

Thanks. I just took a look at that one, and it does indeed seem insolvable at first glance. I'll take your work for it that computers say the same, presumably after using a brute-force attempt.

 


 

 

 

Search [tips]

TTG Home

Robert Home

Daynotes Home

Links

Special Reports

Current Topics


Tuesday, 15 February 2000

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


Barbara is off to play golf with her father, leaving me with the kids. Malcolm is teething right now, losing baby teeth and growing his big-boy teeth, which makes him even more demonic than usual. When Barbara leaves, he turns really demonic, running around, whining constantly, and stealing anything he can get his fangs into. I chanced upon a partial solution yesterday that seems to help. Putting him in his crate while Barbara is gone is a non-starter. He whines, yips, and barks constantly, which makes it hard to get any work done. But if I (a) toss in one of Barbara's old leather slippers, and (b) play some Bach, that seems to calm him down a lot. Interestingly, it seems to be Bach-specific. I tried Vivaldi's The Four Seasons the other day, but it had no beneficial effect whatever. I may try to fool him today with some Telemann.

Back to work on Input Devices.

* * * * *

-----Original Message-----
From: Chuck Waggoner [waggoner at gis dot net]
Sent: Monday, February 14, 2000 12:26 PM
To: webmaster@ttgnet.com
Subject: UPS'es

The subject appears regularly in your pages, but I'm baffled why computing continues to require them--at least for workstations. It occurs to me, that if Quicken can figure out how to make absolutely nothing volatile (after it was very much so in its early days) in only a few years development, why can't workstation OS'es be developed so that only the operation in progress is volatile? The current situation doesn't seem good, or really necessary, to me, but is a boon to the UPS industry.

Probably because there is too much going on. I just used Task Manager to check processes. My system, which is relatively lightly loaded at the moment, currently has more than 50 processes running. I assume that Quicken is using a Transaction Processing model, but not everything a computer does is amenable to that. There are certainly things, like a journaling filesystem, that can be done to improve robustness, but in general anything that improves robustness reduces performance. With adequate UPSs available for $100 or so, I don't see that it's much of an issue.

 


 

 

 

Search [tips]

TTG Home

Robert Home

Daynotes Home

Links

Special Reports

Current Topics


Wednesday, 16 February 2000

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


Here's what disturbs me about Windows 98. Last week, Barbara brought home her new Ferrari® canister vacuum cleaner, which draws 12 amps. She plugged it in, turned it on, and promptly blew the breaker on the circuit that several of my computers use. After calming her down, I simply shut down all non-essential PCs, along with lights and other stuff that I didn't need to work. One of those PCs was the Windows 98 box I keep for taking screen shots and so on. That box also has the scanner attached to it. Yesterday afternoon, Barbara stopped in with a book from which she wanted to photocopy a page. The Windows 98 box was still turned off, so I powered it up. Windows 98 displayed a "new hardware found" message and told me that it was installing drivers for a Dell 1025HX or something like that. Now, nothing about this box has changed for months. It's probably been rebooted at least 10 times in that period, and never brought up that dialog until just now. I find that a cause for concern. Software should be, if nothing else, consistent. Windows 98 is anything but.

Evidence continues to pile up that it was not my imagination when I reported some time ago that Windows 2000 Professional is noticeably slower than Windows NT Workstation. In the latest news, CNet reports that "Dell is leading the movement among PC makers to two processor systems in part due to Windows 2000 Professional's need for more robust hardware to achieve the same performance as Windows NT." Read the entire article here. Microsoft responds that Windows 2000 Professional is plenty fast, and that it runs just as fast as NT4 Workstation on a "PII 200" with 64 MB of RAM. That's fine if you're using a Pentium II/200, although Intel never produced such a processor as far as I know.

I do know that I'm in no hurry to install the shipping version of Windows 2000 Professional, not least because it will run only fifty times before it requires that you register it. I don't like that requirement even one little bit. Perhaps Microsoft should GPL Windows 2000. As Open Source it would at least have some prayer of competing with Linux. That may sound ridiculous now, but Linux is beginning to flex its muscles as a client operating system. In a couple of years, Windows may have a minority market share in the workstation OS segment.

There's an interesting article in The Register this morning about the results of some benchmark testing of AD versus NDS that one Utah company did for Microsoft and Novell. Both companies attempted to skew the results in their own favor, of course, but any neutral observer has to be impressed with just how slow and unreliable AD is. Microsoft is in big trouble. As Microsoft and others have said repeatedly, Microsoft is betting the company on Windows 2000, and it's beginning to look as though they've made a very bad bet. 

* * * * *

-----Original Message-----
From: Chris Ward-Johnson [mailto:chriswj@mostxlnt.co.uk]
Sent: Tuesday, February 15, 2000 10:55 AM
To: 'Robert Bruce Thompson'
Subject: UPSes

Chuck Waggoner misses the whole point of UPSes as far as I'm concerned - downtime. We get at least one or two power cuts a month here in rural France, and at least one or two glitches per week which dim the lights and make the APC UPSes sing. Without them, the machines would be re-booting each and every time. With them I can carry on regardless and don't have to wait for my machine to start up again. I take the regular precautions against losing information and, with Word, this means I can only lose one minute's worth of work at a time, but that's irrelevant - it's the irritation of having to wait for the damned machines to reboot. And, being a child of the noughties, I CANNOT WAIT. Well, not more than a couple of seconds anyway.

Regards

Chris Ward-Johnson
Chateau Keyboard - Computing at the eating edge
http://www.chateaukeyboard.com

There is that...

* * * * *

-----Original Message-----
From: Dan Bowman [mailto:DanBowman@worldnet.att.net]
Sent: Wednesday, February 16, 2000 2:17 AM
To: 'Dave Farquhar'; tgatermann@access2k1.net
Cc: 'RBT'
Subject: Extension cords (long piece)

Hi Thomas,

I offered Dave some additional information on the Great Extension Cord Question and he bit; I'm copying to Bob as a cross check in case I get anything too far out of line. Background: Ten years in the fire service (A.S.); ten years in plant management; unlimited opinions on anything I think I might know something about (factual information optional). Reference: 1996 National Electrical Code (...and what else do I have on my bookshelf?)

One of APC's (and many other manufacturer's) issues is the question of liability: the longer and more exposed a cord is the more susceptible it is to damage from abrasion and weight/object generated cuts. (Offhand, there are two surge suppressor manufacturers that limit their cord lengths to three feet.) Another issue with cord length can be overheating due to the overload of the wires. As the cord length increases, the wire size has to increase (or the cord has to be de-rated) to compensate for the resistance of the wire. My best example of walking the edge: a modern hair dryer will pull every bit of fifteen amps (the maximum available in many household circuits) through 15 gauge wire (yeah, really; I just checked and both dryers here say 15 gauge--to handle 1875 watts) on the dryer itself. The National Electrical Code specs 14 gauge (bigger the number, the smaller the size) as the minimum wire for a fifteen amp circuit. If you've ever felt the cord after drying the hair of a wiggly kid (or dog), you'll recall it as warm. The math: P=I*E; E=I*R => P=I^2*R. I=15A R=.0155 (3.1 ohms/1000ft of 14 ga / 1000 (to get /ft) * 5 (for cord length) gives you 3.5 VA(~watts) dissipated through the cord (conservative, as the resistance of 15 gauge will be higher). That's all well and good in the bathroom with a limited run time and plenty of air flow; put this type of borderline numbers under boxes and such and you can end up with a problem. That's why you de-rate wire if you run multiple conductors in conduit and raceways: heat generation.

Solution: Current limit the cord with a circuit breaker to match the length and wire gauge. That limits the current to something that will not generate significant heat through your cord's insulation (minimize your fire hazard and slow the deterioration of the cord insulation as a bonus). Require all extension cords in commercial buildings have an integral breaker and your fire safety is handled by paranoid manufacturers who de-rate their equipment. This also prevents the use of two wire "Zip cord" extensions. ...and that's how it is handled locally.

Okay, what about home use? Be just as paranoid and use commercial equipment. Don't buy a standard extension cord; use an outlet extension cord or a surge suppressor extension with that integral breaker. Both of these are available with ten foot cords at the various office supply and computer stores (heck, W.W.Grainger sells one with a fifteen foot cord). If you overload the cord, the breaker on the cord will pop before your main breaker. Don't run the cords under furniture or boxes; keep them where you can see them (handy if you have pets that like to chew things; sometimes self-limiting in this case but a very real hazard).

Money where my mouth is: I use an IsoBar eight outlet block to the main desk area. The UPSs for Wolf and Athena are plugged into it (...and I'm not commenting about surge suppressors in line with UPSs at this point) as are other sundry pieces of equipment and another block that holds six 'bricks' for various things. That same wall outlet feeds one of the fifteen footers to the bench area. If I add anything else in this room, I will be running a new drop to the bench and another to the desk. I second Bob's comments about household wiring having lots of surprises (my garage door opener is on the same circuit as the TV in the living room); and please note, apartment wiring is notoriously under-wired for outlets.

Think of these cords as distribution panels: you can plug in several things to one; but let's not put too many in a chain. Another nice touch if you have outlets behind furniture is to use a outlet adapter that allows your plugs to exit parallel to the wall; this really minimizes cord crimps and pressure tears as furniture is pushed back against the wall (available at most hardware stores).

More than you wanted to know,

Dan Bowman mailto:danbowman@worldnet.att.net

Good points. One thing you don't mention is the vicious circle effect of temperature on the ability of a wire to carry current:

  1. As current increases, temperature increases.
  2. As temperature increases, resistance increases.
  3. As resistance increases, voltage drops.
  4. As voltage drops, current increases.
  5. Goto step #1.

As far as your reference to connecting UPSs and suppressors, as you know but some may not, some fires have been attributed to connecting a surge suppressor in-line between a UPS and the equipment. As far as I know, the problem occurs only with inexpensive UPSs whose output waveform is square wave. But it's bad practice to use a surge suppressor between the UPS and the equipment. As far as I know, there's no problem using a surge suppressor between the UPS and the wall receptacle. In fact, I've been doing it that way for years.

 


 

 

 

Search [tips]

TTG Home

Robert Home

Daynotes Home

Links

Special Reports

Current Topics


Thursday, 17 February 2000

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


Today's was going to be a short update anyway, because I have a lot of things to do and I'm not feeling all that great. It'll be even shorter than planned, however. When I checked my mail this morning, I found a 72 page chapter draft from Tom Syroid waiting in my Inbox. He's on a Death March to finish Outlook 2000 in a Nutshell entirely by tomorrow (!), so I have to get a quick tech review done on that and send it back to him today. 

More tomorrow. Sorry.


 

 

 

Search [tips]

TTG Home

Robert Home

Daynotes Home

Links

Special Reports

Current Topics


Friday, 18 February 2000

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


I got through the tech review pass on Tom Syroid's chapter for Outlook in a Nutshell yesterday morning, and then turned to working on my own chapter on Input Devices. I immediately ran into a problem. Keyboards, no problem. Mice and trackballs, no problem. Game controllers, big problem. 

This sounds arrogant, but I'm used to knowing more than 99% of my likely readers about 99% of the stuff I'm writing about. Naturally so, because I have a great deal of experience with most of this stuff, and the time to spend uncovering the details that I don't know about, whether by research or by actually working with the stuff. For example, I wanted to find out the truth about RDRAM and SDRAM and the relative performance of the 440BX and 820 chipsets, so I called up Intel. They sent me two motherboards. One, the CC820, uses SDRAM. The other, the VC820, is identical except that it uses RDRAM. So I can experiment and find out for myself. Most people have neither the time to do that nor the easy access to components that I do. So I can usually speak authoritatively based on actual experience.

But with game controllers, the situation is different. Oh, I have a stack of game controllers sitting here, and I do intend to work with them. But I'm not a gamer, and am not likely to become one, so whatever I conclude is likely to be less generally useful than usual. I'm not sure that there's really a good solution to the problem, but I have to write something. I just hope I don't make a complete idiot of myself in print with regard to game controllers. It's not so much that I mind appearing an idiot as that errors in my coverage of game controllers may call into question the validity of my conclusions on other matters. The easy thing to do would be to wimp out and just not cover game controllers. But I've never been one to take the easy way out. I guess I'll just depend more heavily than usual on my tech reviewers.

* * * * *

-----Original Message-----
From: Dan Bowman [mailto:dbowman@americanambulance.net]
Sent: Wednesday, February 16, 2000 1:35 PM
To: 'Robert Bruce Thompson'; 'Dave Farquhar'; tgatermann@access2k1.net
Cc: 'Dan@home'
Subject: Follow-up: Extension cords (long piece)

Yep, and power dissipation through the wiring goes up as the square of the current which leads to more trouble very quickly.

Thanks for the detail on the surge suppression/UPS issue; I hadn't nailed it down as I didn't have the exact story. I'd heard not to use them together, but that didn't make sense as most decent UPSs have surge suppression built in (and it's okay to build layers on to layers as the IsoBar line does). What I missed was that they were putting the surge units downstream from squarewave outputs.

For the uninitiated, power (VAs or Watts, and they are not the same) is an 'area under the curve' function; a square wave has much more area under it's curve than a sine wave. The RMS voltage of a wall outlet may be 120 volts, but its ability to deliver current (and thus power) is limited by the curve of the sine wave; a square wave can deliver much more power for a given voltage since it has no 'curve' to it. This results in a recipe of destruction for sensitive devices downstream, especially those with inductors in them (such as surge suppressors). With proper engineering, you can limit the 'area under the curve' electronically to power-limit a device; this is how some motor controls and many lamp dimmers work.

Good point. In addition to the under-the-curve power issue you mention, square wave AC is essentially bi-polar DC with a near-zero rise time. That non-existent rise time means that components are rapidly and repeatedly subjected to changes from -120V to +120V, a deltaV of 240V in microseconds which occurs every few milliseconds.

* * * * *

-----Original Message-----
From: Dan Bowman [danbowman@worldnet.att.net]
Sent: Wednesday, February 16, 2000 2:15 PM
To: swallbridge@home.com
Cc: 'Gang'
Subject: UPS batteries

ref: Shawn's current at Tuesday, February 15, 2000, 11:08:38 PM

Shawn,

If you don't mind working with enough raw power to make neat sparks and start a fire, you might take a look inside that UPS. If it's an APC (or any of several others), the batteries are user replaceable. When I ran the plant and MIS here, I'd get the number from the battery itself and check with my local electronic parts house; I'd find batteries that APC wanted US$65 for selling at US$20+/-. Triplite units worked the same way, but were a little more of a hazard to work on as their electronics was exposed (APC offers a battery compartment that isolates you from the line voltages). If the potential for high-amperage discharges to the case doesn't excite you, the same guy that sells you the battery (or the kid behind the counter) can likely replace it for you for a buck or two.

YMMV

Another good point, and one I'd never thought of. Of course, what you pay APC for a replacement battery also includes return shipping for the dead battery and proper disposal.

* * * * *

-----Original Message-----
From: J.H. Ricketson [mailto:culam@micron.net]
Sent: Thursday, February 17, 2000 2:46 AM
To: thompson@ttgnet.com
Subject: Why I'll probably never buy W2K

Dear Bob,

In your 16 Feb Daynotes you said, among other things,

"I do know that I'm in no hurry to install the shipping version of Windows 2000 Professional, not least because it will run only fifty times before it requires that you register it. I don't like that requirement even one little bit. Perhaps Microsoft should GPL Windows 2000. As Open Source it would at least have some prayer of competing with Linux. That may sound ridiculous now, but Linux is beginning to flex its muscles as a client operating system. In a couple of years, Windows may have a minority market share in the workstation OS segment."

That about sums up why I am less and less inclined to buy W2K - or any other SW that, in effect, has a SW dongle. I lost a dongle once - and with it US$80 (in 1974$) of SW and hours of work. No more dongles. Not to mention the requirement to submit my security & privacy to the tender mercies of Microsoft. No Way, Jose! The only hope I can see is that some bright young cracker will soon defeat the copy protection scheme, and that this latest effort will go the historical path of all such schemes. I think that this scheme, if not retracted, may break W2K - the OS that MS "bet the farm on." I hope so.

Meanwhile, Linux looks better & better to me, now that I can use VMWare to run the Windows apps that I need to communicate to the Rest of the World. Never thought I'd see the day that I would be a Linux fan!

Regards,

JHR 
--
[J.H. Ricketson in San Pablo]
culam@micron.net

I predict that Windows 2000 will get off to a very slow start. Millions of copies of Windows 2000 Professional will be installed on new PCs, and Microsoft will attempt to make much of that. Very few organizations will do large-scale upgrades of existing Windows NT 4 Workstation systems to Windows 2000 Professional, although a few will do so. Again, Microsoft will attempt to make much of that. Most organizations that run NT4 Server will buy a test copy of W2KS, just as NetWare 3.1X shops all bought one copy of NetWare 4.0. As with NetWare 4.0, very few organizations will deploy W2KS in production until at least a couple of service packs have been released. Some will, and again Microsoft will attempt to make much of that. 

Microsoft should have released W2K 18 months ago. If they had, it would have been widely deployed by now. Instead, they've given Linux 18 months to gain mind- and market-share in server space. The natural caution of administrators about deploying a new server operating system means that widespread deployment of W2K probably won't start for a year or so, during which interval Linux will make additional advances. My guess is that 2001 will be a death struggle between W2K and Linux for server space. 

Novell hates Microsoft, and has always wanted to do something to damage them. If Novell is smart--alas, they have never shown any evidence of that, at least in marketing--they will release a free (not Open Source) version of NDS for Linux. In one stroke, that would give Linux a robust directory service, which is really the last remaining thing it needs to compete on an equal footing with W2K in server space. Novell can derive revenue either by limiting the size of the NDS database in the free Linux release (probably best) or by later releasing a non-free NDS for Linux 2.0 that is a must-have upgrade for commercial shops.

If I were Microsoft, I'd be running scared right now.

* * * * *

-----Original Message-----
From: J.H. Ricketson [mailto:culam@micron.net]
Sent: Thursday, February 17, 2000 8:05 PM
To: thompson@ttgnet.com
Subject: Are We Real People? (rant)

Dear Bob,

I really did a burn on this.

According to Stanford Professor Norman Nie, director of the SIQSS and co-investigator of the study along with Professor Lutz Erbing of the Free University of Berlin, a key finding is that "the more hours people use the Internet, the less time they spend with real human beings."

It would seem that the professor is still back in the 1850s - or at best, the 1950s. I submit that the people I correspond with on the 'Net are real people. Very real. Not answer bots. I love the freedom the 'Net gives me. It keeps me in touch with my granddaughter in Seoul, people all over the world. People who have become my good friends, and have been of invaluable help to me.

I am hearing-impaired to the extent that normal face-to-face social interaction is a waste of time. The 'Net allows me to interact on a level field with anyone in the whole world, and gives me freedom that would otherwise be impossible.

We ARE real people. And I say - A plague on the professor's house, and all who think they can generate another The-sky-is-falling scenario such as global warming, tobacco threats, etc.

Thanks for hearing my rant.

Regards,

JHR 
--
[J.H. Ricketson in San Pablo] 
culam@micron.net

Probably just an unfortunate choice of phrasing, although I agree with you that the concept is contemptible. He probably thinks nothing of jumping in his car and driving 15 or 20 miles across town to have dinner with friends, or calling them on the phone to keep in touch. But those technologies have been around a long time, so he probably regards them as natural extensions of the village life he unknowingly advocates. Luddites have always been with us, and they've always had a point, although never the point they thought they had. Friends are where you find them, and the Net simply allows you to choose from a larger universe. Sociologists are idiots anyway. I don't let them upset me.

 


 

 

 

Search [tips]

TTG Home

Robert Home

Daynotes Home

Links

Special Reports

Current Topics


Saturday, 19 February 2000

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


Barbara is off this morning to a Carolina Border Collie Rescue board meeting, where they will presumably decide weighty issues before having pizza for lunch. I tried to convince her to take Malcolm along--a Border Collie pup should be a big hit at a Border Collie Rescue group meeting--but she says no dogs are allowed. Seems like a strange rule, but perhaps it's just me.

With the W2K rollout, Microsoft is apparently belatedly admitting what a POS* Windows 98 is. They're also saying nasty things about NT4. Apparently, W2K runs for 90 days without crashing, while Windows 98 crashes every two days, and Windows NT4 every five days, or something like that. I don't know what planet they're on. My Windows NT systems commonly run for 90 days or longer--sometimes much longer--between reboots. I just checked the System Idle Process on kerby and found that it's been up for 1736+ hours, or about 72 days. Kerby does a lot of stuff, including managing our shared Internet connection and publishing this web site.. Old thoth has been up for 5189+ hours, or something over 216 days, although admittedly it's not doing much. My main system, kiwi, has been up for 502+ hours, or about 21 days. And as I recall, the only reason for that relatively short up-time is that I restarted it three weeks ago to run W2KP. Otherwise, it runs for months on end.

* POS = "problematic operating system" (or something like that)

If you need memory, now's the time to grab some. Spot prices have taken a nose-dive, with 64 Mbit SDRAMs as cheap as they've been in many months. You should be able to pick up SDRAM for under $1/MB now.

 


 

 

 

Search [tips]

TTG Home

Robert Home

Daynotes Home

Links

Special Reports

Current Topics


Sunday, 20 February 2000

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


Well, FrontPage 2000 has screwed me AGAIN. As I called up this page to edit it, I happened to notice that my copyright notice was missing from the bottom. Instead, there was a FrontPage comment, "Comment: This border appears in all pages in your Web. Replace this comment with your own content." Well, hell, I had replaced it with my own comment. A long time ago, in fact. Sometime in 1998. 

The copyright notice is (was) in what FrontPage calls a Shared Border. A Shared Border is an area of the page layout in which you can enter text that will subsequently appear on every page in the web that has visibility of that border enabled. There are four such Shared Borders, top, bottom, left, and right. They're ideal for something like a copyright notice that you want to appear on every single page in the web. I'd enabled the bottom Shared Border and put my copyright notice in it. Now it's gone. Well, it's back. But that means, of course, that every page in the web has now changed. That has two implications. 

First, every page in the web that has the Date/Timestamp Bot in it will now show today's date and time. That's a useful Bot. I use it at the top of this page, for example, to change the last modified date/time automatically. But it's less useful than it should be. It should have an option to change the date only when the page is manually edited, as opposed to being changed by an automatic update like the one that changing the bottom Shared Border causes. In fact, it does have such an option. Under Date and Time Properties for the Bot, there are two option buttons, Date this page was last edited and Date this page was last automatically updated. Sound like just the thing, you might think. The trouble is, it doesn't work. I have and have always had the first option button marked for every occurrence of the Date/Tiime Bot. FrontPage doesn't care. It happily updates all the date/timestamps on my old pages every time it does an automatic updated. Weekly journal pages from a year ago and more will have today's date on them. 

The second problem with the automatic update is just a minor aggravation. Because every single page in the web has changed, the next time I publish FrontPage will have to upload every single page in the web. This gets old fast. Oh, well. I suppose I'll take this opportunity to do a global link check and some other stuff that'll change a lot of pages. If I have to publish the whole web again, I might just as well get that out of the way. I hate Microsoft.

 


[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.