jp.jpg (13389 bytes)

CHAOS MANOR MAIL

Mail 140 February 12 - 18, 2001

read book now

HOME

VIEW

MAIL

Columns

BOOK Reviews

emailblimp.gif (23130 bytes)mailto:jerryp@jerrypournelle.com

CLICK ON THE BLIMP TO SEND MAIL TO ME

The current page will always have the name currentmail.html and may be bookmarked. For previous weeks, go to the MAIL HOME PAGE.

FOR THE CURRENT VIEW PAGE CLICK HERE

If you are not paying for this place, click here...

IF YOU SEND MAIL it may be published; if you want it private SAY SO AT THE TOP of the mail. I try to respect confidences, but there is only me, and this is Chaos Manor. If you want a mail address other than the one from which you sent the mail to appear, PUT THAT AT THE END OF THE LETTER as a signature.

I try to answer mail, but mostly I can't get to all of it. I read it all, although not always the instant it comes in. I do have books to write too...  I am reminded of H. P. Lovecraft who slowly starved to death while answering fan mail. 

Day-by-day...
Monday -- Tuesday -- Wednesday -- Thursday -- Friday -- Saturday -- Sunday
 
atomz search

Search: type in string and press return.

 

or the freefind search

 
   Search this site or the web        powered by FreeFind
 
  Site search Web search


Boiler Plate:

If you want to PAY FOR THIS there are problems, but I keep the latest HERE. I'm trying. MY THANKS to all of you who sent money.  Some of you went to a lot of trouble to send money from overseas. Thank you! There are also some new payment methods. I am preparing a special (electronic) mailing to all those who paid: there will be a couple of these. I am also toying with the notion of a subscriber section of the page. LET ME KNOW your thoughts.
.

If you subscribed:

atom.gif (1053 bytes) CLICK HERE for a Special Request.

If you didn't and haven't, why not?

If this seems a lot about paying think of it as the Subscription Drive Nag. You'll see more.

Highlights this week:

  •  
  •  
  •  
  •  

Search: type in string and press return.

 

LAST WEEK                                                               NEXT WEEK

line6.gif (917 bytes)

 
This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

TOP

Monday, February 12, 2001  

Roland says finally a use for spam: http://www.spammimic.com/explain.shtml 


I was a Titan crew dog in the early 80's at Davis Monthan Air Force Base and pulled my share of alerts as a deputy commander. While I can't speak for the early Titan days, the mods that were in place while I was in the hole allowed us to launch before we would get hit. As I recall, the missile would be on the way well within 3-5 minutes and given the early warning capabilities we had, that was enough time for the bird to clear the silo and be downrange before the incoming stuff arrived.

To tell the truth, I was _way_ more scared of the fuel/oxidizer than I was of Russian missiles.

"Titan II, A History of a Cold War Missile Program," by David K. Stumpf (ISBN 1-55728-601-9) is excellent book covering the development, deployment, and deactivation of the system.

Karl Sandwell-Weiss

Interesting. I was in charge of future basing studies at Aerospace in 1964 and in those times R/V sink was the standard term for Titan...


In your recent column on Byte, you mention the possibility of IBM dropping AIX in favor of Linux. It is also my belief that IBM wants to tap the resources of the open software community in order to reduce costs in OS development. They don't actually charge you for the OS when you buy an RS6K machine, you just have to pay for the media (currently something like US$ 50). Given that, outsourcing OS development makes sense (at least economically). OTOH, though, AIX, as one of my fellow AIX Admins (who has a lot more insight than I have) stated, is still lightyears away from Linux in some aspects of kernel design and structure. It especially shines when handling extreme amounts of data, producing I/O bandwidth usually associated with mainframe type computers (this is very propably because some of the AIX developers MUST come from the OS390/MVS department). Linux has advantages when it comes to pure number crunching, Torvalds and his gang seem to make better use of advanced processor features and floating-point processing. There have been several discussions on comp.unix.aix about this; if your time allows, I would suggest you spend some time there, you'll find quite a lot of sane UNIX aficionados very different from the flame warriors who use to plague you about your position concerning non-MS Systems.

Once again, thanks for the good (and sometimes hilarious) writing over all those years.

K.B.

One day I will have time to do everything I ought to do...

Thanks


Mr. Pournelle:

Besides being a great fan of your speculative fiction (I have read every fiction book you have published, first in Spanish translation and lately in English, since I now reside in Canada), also, I really enjoy your writings for the great, lamented magazine Byte.

Reading your column dated February 12 you mention that the Mac version of MS Office used to be better than the one for Wintel systems, until Office 2000 came out.

Well... Guess what?

I am using MS Office 2001 for the Mac (released last year in November I think), and for my money is the best version yet, in any platform.

With kind regards

Carlo Jaunez-Silva West End Offset Plate Service, Toronto, Ontario, Canada Senior Prepress Manager &; All-Around Goof -- Un Chileno, afuera.

Intriguing. Thanks

 

TOP

 

This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

TOP

Tuesday, February 13, 2001

We can open with a love letter:

I have been reading your columns off and on for years. Your writing is generally convoluted and illogical at best. Most of the time your facts are wrong and your discussions of win2k server are just plain stupid. You have no idea what you are talking about with this software. You have this superior attitude that most phd's have but you fool very few. You should stick to your fiction writing and leave server topics to people who have done the months of study who don't rely upon their constantly advertised phd to give them credibility about all knowledge in the universe like you do.

Sherosky, Rick (R.) [rsherosk@ford.com]

Well. Let us take this seriously.

 I will not plead guilty to convoluted: whatever faults I have, I write clearly, and you can tell what I mean. Many years ago I was on the David Susskind show with Isaac Asimov. I had known Isaac for years, agreed with him sometimes and argued about other matters. In the course of the show Susskind asked Isaac what was the secret of his popularity. "I'm clear," Isaac said. "I may be wrong but I'm clear about it."

I never forgot that, and haven't yet. Convoluted? No. Wordy, perhaps, particularly on the web site where I don't edit to sweat out needless words; and sometimes that happens in columns.

Illogical? I do not think so, but few people do think they are illogical. Logic is the act of drawing conclusions from assumptions. The assumptions need not be true, but in the real world we hope to deduce assumptions (principles) from observation, and build on that in the hopes of building a picture of the real world. Of course that implies a number of assumptions about the nature of the universe and the relationship of our models to it, but this isn't an essay on epistemology. I do note that  'Illogical at best' implies that I'm never logical at all, and that is demonstrably untrue. Q.E.D.

As to all knowledge in the universe, I have often said, if you pretend to know everything you should be right most of the time, and that implies paying attention when people correct you. Come to that I got that phrase from Isaac too, probably about the same time, 1974 or so.

I will agree with one of the points: possession of credentials doesn't prove much about what you know. A Ph.D. degree demonstrates one certain characteristic: stamina. It's an obstacle course designed to get you to quit; the minimum intelligence isn't all that high, but the minimum determination to finish and get the darned degree is. Ph.D.'s may or may not be smart, but every darned one of us is stubborn.

As to my knowledge about servers, I cheerfully admit I am not up to the level of knowledge of some of my readers.  Fortunately, I have access to people who have put in the months of study -- only months? Really? That's not long at all! I presume Mr. Sherosky meant months of study to learn the intricacies of servers; he is not, I fear, too clear on that. But I have readers and editors who know this stuff, and I generally send my columns to them for sanity checking before the rest of the world sees them. In the old days of BYTE we had some 30 real experts in Peterborough who went over everything I wrote. Alas, those glory days are no more, and Rick and Phil and Tommy and Ken aren't up there as a backup system for me. (Why, by the way, was I the lead columnist when I cheerfully admit most of the BYTE technical editors knew more than I did? Because, of course, whatever faults I have, I am clear. But you knew that.)

I don't have the big BYTE staff any longer but I do have Bob Thompson and Roland Dobbins and my son Alex and David Em and Eric Pobirs and a bunch of others available, and beyond them are thousands of readers who don't hesitate to tell me, with specifics, when I get something wrong. So I fear I have to reject Mr. Sherosky's charges as at best unproven. Of course I used his letter largely as a reason to write this bit, but surely I can be excused for that.

Note that most of my column is built around what I DO, and that "I do all these silly things so you don't have to." I may do the wrong silly things but at least I report that.

When someone shows me where I was wrong, I pay attention. Fortunately I don't get many letters like Mr. Sherosky's, which, I fear, isn't very helpful because it's not very specific.


"Remember to put these big rocks in first or you'll never get them in at all." I can't add much to that.

 Sure you can.

F'example, I'll bet your kids turned out to be bigger rocks than you expected (And even if they didn't, you have readers for whom they were unexpected big rocks); thus, 1. Save, or be prepared to make, room for unexpected big rocks; and, 2. Make sure unexpected big rocks displace gravel, sand, and water, not other big rocks!

Very Respectfullly, Rod McFadden Centreville, VA

Well said. Thanks.


Dr. Jerry:

One of my co-workers came up with an interesting comment.

One of these days somebody's going to make an email virus with the title "Virus alert!!" - and get everybody!

Steven Dunn

Yea, verily. I never open attachments.  Well, hardly ever, and then only not only from people I know, but when an attachment is expected or explained. Random messages, even those that purport to come from people I regularly work with, do not get opened nor do attachments; I look at them with an ASCII viewer first. I was bit by Melissa, but not since.


And a voice from the past:

You might enjoy this one. We seem to have similar taste games: Free demo at www.battlefront.com  Like chess it is fairly easy to learn but a lifetime to master.

I still remember meeting you at my booth at San Francisco Computer Faire many eons ago!

Happy Adventuring!

------------<begin signature block>------------ 

Say Yoho - Everything spins around and suddenly I'm elsewhere... M. Scott Adams (Not Dilbert, Adventure!)

http://www.msadams.com  Msadams@msadams.com Scott.Adams@Avistainc.com

Microsoft Games Beta Tester ID: 155880

When your life expectancy is near zero everything is meaningful - Dr Stephen Hawking When your life expectancy is near full any little thing can depress you - Dr James Dobson jr When you stop learning you start dying - M. Scott Adams ------------<end signature block>------------

Which does in fact look like interesting game. Thanks.


And in a more serious vein:

Jerry, You were having trouble downloading a copy of IE that didn't require you to go to the web site each time. Regarding downloading anything from Microsoft, here's the Corporate Update Site. It let's you download files and post them to a server for deployment throughout your Enterprise, instead of each individual desktop having to download from Microsoft's website. It's especially good if you implement a policy of limited web access.

http://corporate.windowsupdate.microsoft.com/en/default.asp 

Tracy Walters Networking Rocky Mountain Technology Group "Most computer problems can be solved by a suitable charge of high explosive."

I have had trouble navigating that site which seems to change often, but thanks..

Jerry

In "Are We Wearing Out the Guard and the Reserve?" in the 2/2001 issue of Air Force Magazine, there was a comment on page 39 that caught my eye: "[T]he Reserve Component Employment 2005 study . . . suggested that reserve forces were a particularly good fit for the emerging mission of homeland defense. The Pentagon has already established 10 Rapid Assessment and Initial Detection teams-consisting of Army National Guard and ANG [Air National Guard] personnel-to assist civil authorities in responding to a terrorist attack involving weapons of mass destruction. The Fiscal 2000 defense bill authorized the creation of 17 more RAID teams."

I suppose I should be glad they are thinking about all of this, but the reference to "the emerging mission of homeland defense" is chilling. I remember the Nike domes around the southern California of my boyhood. I guess it's coming back to that.

Ed Hume

The Nike bases don't scare me. It's that defending the homeland is an "emerging mission", while sticking our noses in everyone else's business all over the world is natural. Actually, the simplest way to reduce the threat of terrorist attacks at home is to stop beating on people abroad. I don't think Switzerland has to greatly expand anti-terrorist forces. There is no power on earth that can take a drink from the Mississippi without our let and leave. The United States is under no threat of invasion by anyone.

So who wants us harmed? Perhaps Sudan, whose pharmaceutical plant was destroyed by US Missiles fired because -- because -- well, just because. The Democratic Party and former President Clinton assure us it was not wagging the dog to distract the country from revelations about White House interns, and sure they are honorable men. There are those in Pakistan who hate us because US missiles rained down on a convention of religious Moslem physicians at a meeting in Afghanistan. The Taliban hates us because the missiles fell on their country, and many Pakistani's are unhappy because we killed some of their civilian physicians.  But again we are assured it was all necessary, and all, all are honorable men.

But perhaps, now, such actions won't be so necessary? In which case we might not need so many preparations to combat terrorism?

As a maritime power we need to keep open the sea lanes. We need a Navy. But keeping the seas free of pirates and enforcing freedom of the seas for all has seldom generated hatreds sufficient to generate terrorist activities.

If we are a Republic, we have one set of defense requirements. If we are to impose a New World Order -- which is Imperialism under another name -- we need another. Rome began with a mission "to protect the weak and make humble the proud," and that was the beginning of Empire. The end of the Roman Empire came with the sons of Aurelius; no Emperor after that was Roman at all. And the Legions soon discovered the dread secret, that Emperors could be made in places other than Rome. But of course none of that will ever happen here.

The case for Empire was made by Allbright: what is the good of having a great and powerful army if we do not use it to right the wrongs of the world? It is a powerful case: if we can right wrongs, should we not do so?  But in the Sudan they have been butchering and enslaving Christians and racial minorities (actually religion is closely correlated with tribe there, so it's hard to tell the difference between religious and racial persecutions) -- they have done this for decades, and we seem not eager to do anything about it. Only to bombard their pharmaceutical plant. 

In the Balkans we have managed to exchange one set of persecuting thugs for another. In Kuwait we have restored a Royal Family who fled to the fleshpots of Europe while we took their country back for them. In Haiti we have quietly allowed the restoration of the Tonton Macoute under another name and with different membership, but little has changed.

The US had two highly spectacularly successful imperialist adventures: the reconstruction of Japan and Germany under proconsuls Douglas Macarthur and Lucius Clay. Those ventures worked and well. But those days are over.

Still, Allbright had a point: is it moral to stand by and do nothing about the ills of the world when we obviously have the power to right all wrongs? And we are then left to argue whether we really have the power that a large and powerful army seems to imply.  Talleyrand told Bonapart that "you can do anything with a bayonet but sit upon it," a lesson that few seem to have learned; I would argue that a large and powerful army is not in fact the means for doing good that Allbright thought it was. 

Going about righting the world's wrongs by direct action is one possible mission for America. The other is the policy put forth by John Adams and his son John Quincy Adams: "We are the friends of liberty everywhere, but we are the guardians only of our own." The purpose of America is to clean up its own act, be the shining example, the beacon of freedom, and if need be the arsenal of democracy. That is the other policy; and under that one we will not need the oppressive measures that a vigorous anti-terrorism program requires.

When we have a literacy rate to be proud of; when the city of Washington is the envy of the world as an example of splendid schools, ordered liberty, clean streets, arts, culture, beauty, no crime and no gangs, tranquil and productive -- when we have achieved all that, perhaps we can begin to think of exporting our ordered liberty by force, as we did with Germany and Japan. But perhaps we will not need to?


In your column, you said...

IBM and Compaq have both made major investments in ensuring their systems are compatible with Linux; IBM is putting a lot of effort into developing a Journaling File System, or JFS, for Linux, and even have it running on their big S/390 and above mainframes (see IBM and Compaq.)! My guess is that IBM will probably phase out their proprietary UNIX, called AIX, in favor of Linux across their entire line of hardware. That will give Linux a real boost.

JFS is the file system used on an AS/400 (or whatever they are calling it these days, IBM renamed their whole lineup ;-) ). It was ported to Warp Server about two years ago along with something called LVM or Logical Volume Manager. Then they did something that, at the time, was shocking. Both are now Open Source. IBM putting their own technology in the Open Source arena, very shocking.

With LVM and JFS on os/2, you have the ability to create a single drive letter that spans drives. You can make drives have any letter you want as well. You can have drive Q, for example, be part of IDE drive 0, all of IDE drive 1, part of three different scsi drives, and all of a fourth scsi drive, all in the same system. Another feature is that if you have a raid for example, and add a drive to it, you can expand drive Q into that new space..... ON THE FLY, with NO rebooting required. You will still want to do this in the wee hours as it does slow down the server while it is doing this (duh). If you somehow crash the server (rare with OS/2 servers), a check disk of a JFS drive often goes so fast that if you blink, you will miss it... even for big drives. All it does is replay the Journal. With HPFS (you still have to boot using it), it can take an hour for a large drives check disk.

How much of this will wind up on Linux remains to be seen. Since the source code is avail, someone SHOULD port both to windows as it too is in bad need of a new file system. The one used on the consumer OS is nearly as old as MS itself. I hear that NTFS is somewhat older than HPFS in technology.

David Eckard 

------ It's not hard to meet expenses... they're everywhere. Webmaster http://www.billswoodshed.com Webmaster http://www.ntrnet.net/~swordedg 

Fascinating. Thanks.


Jerry:

I've read and enjoyed your columns over the years.

I have been trying to understand the open source/free software model.

I have read The Cathedral and the Bazaar, by Eric S. Raymond.

I understand how many developers can advance a project.

What I do not understand is the ECONOMIC MODEL.

Do the Linux contributors advocate that all software be free??

Should PhotoShop, SAP, AutoCad, etc. be open source/free????

This seems so socialist to me.

Why don't other professionals offer their services for free.....doctors, lawyers, CPA's, etc.???

Can you point me in a direction where I can become more informed on this subject????

Gregg Gilbert

Well, it's complicated. There are also many shades of opinion here. And the revenue model varies from one to another, and is always the central problem: why invest in open source software companies? How do we get return on investment? What is the revenue stream?

For the views of the various Linux contributors you will have to go read their own words. I make no doubt that it won't be long before you see some of them here. As to "socialist", some perhaps would agree; others would be shocked since they are not trying to force anyone else to their model. It's they way they want to operate. And many open source advocates have day jobs, as professionals, and collect money for consulting.

I'll leave it at that for the moment. As I say, I make no doubt we will hear more.

From: Steve Setzer <setzer@backfence.net> Subject:Open Source Economic Models Jerry,

Mr. Gregg Gilbert, and anyone else interested in an extensive analysis of potential economic models for open source companies, should read another essay by Eric S. Raymond (ESR) entitled "The Magic Cauldron."

http://www.tuxedo.org/~esr/writings/magic-cauldron/magic-cauldron.html 

Having read it a few times, I think ESR's basic analyses of individual economic models is correct, but he is overly optimistic in thinking that most of them will pan out for a large number of projects and companies. I think he does not give enough credence to the unique personalities driving the projects he holds up as examples. It is nice, clear writing however, with footnotes.

Steve S.

Jerry,

Gregg Gilbert wrote:

> I have read The Cathedral and the Bazaar, by Eric S. Raymond. > > I understand how many developers can advance a project. > > What I do not understand is the ECONOMIC MODEL.

There's more than one economic model. Eric Raymond mentions some of them in "The Magic Cauldron" (which someone else mentioned to you as I was writing this).

Linux-related companies are experimenting with many different economic models - ranging from pure open source (selling CDs (for the convenience of those without fat pipes) and/or selling services - like Red Hat and Ximian) to pure proprietary (expensive licenses for little shrinkwrapped boxes - like Computer Associates) . Some will succeed, and some will fail.

IBM appears to going the "services" route (mostly). That's not too surprising, since they seem to make most of their profits on services in general (while losing a boatload of cash on hardware sales).

> Do the Linux contributors advocate that all software be free??

Some do (Richard Stallman), and some don't. We don't all speak with one voice, ya know. And even Eric Raymond has told some companies that it didn't make sense to open source their software (like log-cutting programs at a lumber mill).

BTW - the issue is free as in "free speech", not "free beer" (though low/no cost certainly has its attractions).

My own preference is for open source software - both for its price AND its fixability.

On the other hand, when Mom wanted a better way to maintain the Dodge Family website (www.dodgefamily.org), I suggested IBM's Top Page (now Home Page) for Linux (yep, commercial software). (Front Page had been choking often - I know you're familiar with THAT, Jerry)

On the third hand, it turns out Home Page wasn't really designed to cope with 6000+ files (lots of people in the family tree) totalling 130 megabytes. IBM suggests that we shell out hundreds of dollars for their commercial-grade stuff (not to mention a server to run it on).

THAT'S the sort of artificial limitation which would be frowned on in the open source world, so I'm back to preferring open source software.

> Should PhotoShop, SAP, AutoCad, etc. be open source/free????

That's up to their respective owners - I wouldn't insist on them taking that route.

But I've had enough trouble with proprietary software (my day job is occupied almost entirely by Microsoft Access) that I'll always lean toward the open alternatives.

> This seems so socialist to me.

Errr ... no. Socialism is when someone from the government points a gun at you and says "hand it over, or ELSE".

> Why don't other professionals offer their services for free.....doctors, lawyers, CPA's, etc.???

Actually, some do - after their day jobs are done, or during vacations or sabbaticals.

Calvin Dodge

Gregg Gilbert wonders whether the Linux contributors advocate that all software should be free.

Some of them do; most do not. They are a diverse group, and not easily lumped together.

I think most of the Linux contributors would agree that people can charge money for software. However, as you noted, for a variety of reasons they are willing to contribute code that will be given away.

Companies like IBM are willing to contribute to Linux because Linux is making them money, and they forsee even more money if Linux gets even better.

Some people add features they, themselves, wish for. That's why you sometimes see odd little features added before big, important ones; you get the features that people write and donate, not necessarily the ones you wish for. -- Steve R. Hastings "Vita est" steve@hastings.org http://www.blarg.net/~steveha

 

 

TOP

 

This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

TOP

Wednesday,

In San Francisco for AAAS

 

 

TOP

 

 

 

This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

read book now

TOP

Thursday,

AAAS meeting

 

 

 

TOP

 

 

 

This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

TOP

Friday, February 16, 2001

 

Hello Dr. Pournelle,

Here's a silly thing that I did that might be useful for your other readers.

I have an external (parallel port) Backpack CD Rewriter on my win 98 system, and on the printer out line I have a PriMax autoswitch that sends the printer output to either A or B printers. I've used this rewriter extensively and without trouble for several months, even when printing from a CD in the writer.

The other day I pulled one of the printers off for a trip to another office, but left the Primax switch in place.

I don't know a thing about autoswitches, but I can select A or B with a little button, and it electronically does the switching. It is supposed to default to "Printer A" but there are times during the day that I will notice that "Printer B" is selected. It seems almost random, but I suppose it is a stray signal of some sort that is perceived as the software command to switch printers.

Anyway, since I removed the printer on the B side of the switch, if the switch happens to be selected to B (remember that it seems to randomly go there) and I try to access the Microsolutions Backpack CD Rewriter, with Adaptec DirectCD running, my computer REBOOTS immediately if I try to access the drive, or even use My Computer or Windows Explorer to look at the different drives on my system. I worked most of the day yesterday with no trouble, but when I tried to save a file in Word, and went to Save As and clicked on my computer to select a different hard drive, the machine did reboot right there.

I first noticed this when I returned from the other office and turned my computer on. It booted up, put up the Windows splash, then rebooted soon after (when the Adaptec DirectCD software tried to load) and did it again and again and again, until I could figure out what was causing it (at least the DirectCD software put up it's own splash when it loads).

I remember a column a long time ago when you were talking about SCSI devices needed a terminator, but I never dreamed that would also happen with a parallel port device. I can (and usually do) leave the B printer off, and the drive works fine. I have neither the time nor the interest to worry about what pins are making a circuit when a printer is plugged in but not on, and that's not really the point of the story.

Who knows whose fault this is, Microsolution's Backpack, Windows 98, or Direct CD (I'll put my money on it). But if you or your readers have weird stuff happening with this sort of setup with a electronic switch, take the switch out and see if it doesn't cure it.

Occam's razor, in other words.

I love your site, check it daily.

Charles Duell

cduell@rmi.net 

I never had that problem, but I don't print through parallel ports any longer. My printer is on the Ethernet. Sometimes I will put a color printer on a parallel port, but not often. Thanks for the experiment. Of course you didn't intend to be an experimental subject...

Jerry,

Have you seen http://news.cnet.com/investor/news/newsitem/0-9900-1028-4825719-RHAT.html?tag=ltnc  ?

If so, what's your take on it?

I'd love to hear exactly what Jim means by "stifle innovation" and "legislators need to understand the threat".

What precisely does he want legislators to "do" about the "threat"?

In spite of his comments at the end of the article, I suspect his company is running scared of Linux.

your obedient servant,

Calvin Dodge

I have a lot of mail on this astounding -- and pretty silly -- statement. I don't have time to do adequate commentary. But it is a statement by one person, who seems to be a bit demented... I'll have more on this next week.

Jerry,

One of your readers wrote:

With LVM and JFS on os/2, you have the ability to create a single drive letter that spans drives. You can make drives have any letter you want as well. You can have drive Q, for example, be part of IDE drive 0, all of IDE drive 1, part of three different scsi drives, and all of a fourth scsi drive, all in the same system. Another feature is that if you have a raid for example, and add a drive to it, you can expand drive Q into that new space..... ON THE FLY, with NO rebooting required. You will still want to do this in the wee hours as it does slow down the server while it is doing this (duh). If you somehow crash the server (rare with OS/2 servers), a check disk of a JFS drive often goes so fast that if you blink, you will miss it... even for big drives. All it does is replay the Journal. With HPFS (you still have to boot using it), it can take an hour for a large drives check disk. How much of this will wind up on Linux remains to be seen. Since the source code is avail, someone SHOULD port both to windows as it too is in bad need of a new file system. The one used on the consumer OS is nearly as old as MS itself. I hear that NTFS is somewhat older than HPFS in technology.

-------

NTFS Version 5 under Windows 2000 is hardly behind the times. Because there is a tendency to attack what folks who use MS products say, here's an excerpt from the Windows 2000 Server Administrator's Companion, Chapter 14, Disk Management, Pages 484 and 485:

Remote Management

The new Disk Management snap-in in Windows 2000 Server lets you manage not only the local hard disks but also drives on other computers running Windows 2000, enabling the administrator to manage disk tasks and space allocations from a workstation without having to sit at the machine that is being administered. This capability is a boon for remote site management and - using the MMC - makes it easy to delegate authority and administrative responsibilities for a group of computers to others without having to give them full administrative privileges.

Dynamic Disks

The other major feature that Disk Management adds in Windows 2000 is the concept of dynamic disks. By converting a disk to a dynamic disk, you give Disk Management the ability to manage it in new ways, without requiring a reboot in most cases. You can extend a disk volume, span a volume across multiple physical disks, stripe the volume for improved performance, mirror it, or add it to a RAID-5 array - all from the MMC and all without a reboot, once the disk is converted to a dynamic disk. The initial creation or conversion of the first of your basic disks to a dynamic disk will require a reboot, unfortunately, but once you're gotten over that hurdle, you'll breeze through the remaining tasks. When combined with the new remote management functionality, dynamic disks give the system administrator powerful tools for managing the type and configuration of hard disk storage across the enterprise.

Page 451 -Mounted Volumes:

Windows 2000 borrows a concept a from the UNIX world by adding the ability to mount a volume on a subfolder of an existing drive letter. A mounted volume may also have a drive letter associated with it, although it does not need to, and it can be mounted at more than one point, giving it multiple entry points into the same storage.

There are other features...but these seem to push the buttons of most folks.

Tracy Walters Networking Coordinator Rocky Mountain Technology Group

 

This from a tech support supervisor (at another company):

Check out http://www.flex.com/sign_up/  for the most BOFHish support policy in the world.

And wow. He's right....


Dr. Pournelle,

Regarding the parallel port switch that reboots a computer...

A few friends of mine put together a little game LAN party and one of the new toys we were trying out was a programmable game pad designed for first-person shooter games. I don't even remember the name of the pad. In any case, it plugged in to the joystick port, keyboard port, and if I recall correctly, a serial port. We plugged it into a computer and tried to reboot, and the computer was DEAD. No POST, no video, no nothing. Hmmm. Unplug gamepad, plug it into another computer, reboot. No POST, no video, no nothing.

Since both of those computers had identical motherboards, we figured it was a mobo issue, and in a fit of idiocy we plugged the gamepad into yet another computer and rebooted. No POST, no video, no nothing. This gamepad from hell had managed to completely scramble the BIOS on 3 computers via a gameport, keyboard port, and serial port. We still don't know how this happened, and we immediately gift wrapped the device up and sent it to our worst enemy for Christmas.

Fortunately, we managed to figure out the power up procedure that reset the BIOS settings to defaults and all three computers were back up and running within an hour or two, but even though in our group we had a dozen people who are HIGHLY skilled computer types (networking, programming, you name it we had it), none of us could figure out how this little gamepad had done what it did.

Getting back to the point, it doesn't surprise me that a seemingly innocent device hooked up to an interface port could cause spontaneous hardware faults because I've seen something even more weird.

Sean Long seanlong@micron.net

Wow. I never had anything like that happen. Thanks!

 

 

Subject: Basic physics laws need not apply

Dr. Pournelle,

<P>You might be interested in knowing that basic physics laws have been repelled in California.

<P>EETimes has a story titled "California mandates electric-vehicle sales by 2003" (see <A HREF="http://www.eetimes.com/story/OEG20010202S0024"> http://www.eetimes.com/story/OEG20010202S0024</A>. In a nutshell, the California Air Resource Board (CARB) has refused to reconsider its previous ukase which mandates that 10% of the cars sold in California by 2003 would have to be zero-emission. And CARB is thereby proving it is either hoping the laws of physics will change, or that it is as totally, raving *mad* as a British cow!

<P> The problem is that fuel cell technologies, which might be the only solution to viable ZE cars, is neither reliable nor cheap enough yet to be fielded under millions of hoods. Fuel cell companies are making huge progress and are even proposing home generators, but no commercial fuel cell car engine is envisionnable within five years, let alone two.

<P>Which means these ZE cars will have to be electric. And that's where any engineer worth his salt is seriously doubting the CARB bureaucrats have any sanity. Let's see. A current lightweight city car with an internal combustion engine must have about 70 HP in order to safely merge into Californian traffic. At 736 W per HP, make that 50 kW. Retain this figure for an electric car. Now, assume a usable 4-hour battery charge, which is 200 kW-hour. If you assume an order-of-magnitude battery price drop and retain the futuristic $150/kW-hour quoted in the article, that's $30,000 for the battery alone (but let's still assume it is lightweight!). That's an expensive car.

<P>But wait, the best part is yet to come. See, the good, happy, environment-conscious Californian driver must recharge his ZE car overnight. Assume a 10-hour recharge time and 100% efficiency (ha!) for your 200-kWhour battery pack. That's a 20 kW load. Now, suppose that just a meager 100,000 good, happy etc. drivers are recharging their car. That's a whooping 2-gigawatt load. Or, in layman terms, the whole output of a nuclear power reactor. For 800,000 ZE car drivers, you'd need 16 GW, or two whole nuclear power plants with 4 cores per plant. Power plants which California refuses to build.

<P> How does that fit into the current Californian energy shortage, pray tell? The CARB has the answer: To quote the article, "The board's members insisted, however, that the current shortages would have no effect on the viability of their plan."

<P> So there. No effect. Behold, unimaginative laws of physics; behold, petty arithmetic, the Board Has Spoken!

<P> California might need to import power, but it could obviously export pompous fools by the truckload.

-- Fred Mora 

No argument ghefe

 

 

 

 

 

 

 

TOP

 

 

This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

TOP

Saturday,

 

 

 

TOP

 

 

This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

read book now

TOP

Sunday, February 18, 2001

From Trent Telenko:

There is a great article on our failures in Kosovo in using "Information Superiority" despite having great intelligence gathering at the link below.

http://call.army.mil/call/fmso/fmsopubs/issues/kosovo.htm 

The following conclusions are clipped from the article with a few comments (in these) from me:

o Total information superiority did not allow us to achieve a political or diplomatic victory. Like Saddam Hussein, Milosevic is still in power, and the Belgrade Agreement was a far cry from what was sought at Rambouillet.

(The article is dated from before Milosevic realizing he should have stolen the election before the voting and not after. The article is still spot on this point.)

o Total information superiority did not enable NATO to locate the Serbian armed forces' center of gravity, the police and paramilitaries doing the killing.

(I disagree with the article here. Killing security units made up of thugs and press ganged political opponents would not shake Milosevic's hold on power. Destroying Serbian civilian infrastructure, which won the war for us, did.)

o Total information superiority did not counter rumor nor prejudiced reporting. For example, to cite an instance not covered in this analysis, information superiority did not allow NATO to know, even approximately, how many Kosovo civilians were killed before the bombing started. Instead of 100,000 Kosovo victims, as rumors suggested, 10,000 now appears to be closer to the truth. Would NATO have gone to war over 10,000 people? To date, only some 2,500 bodies have been discovered.

(The Serbs took the bodies to heavy metal processing plant in Kosovo and fed them into the ore processing furnaces to vaporize them, according to NPR reports on the findings of the Kosovo war crimes tribunal. This was a Serb "lesson learned" from Bosnia.)

o Total information superiority was affected by politicians, who demanded that pilots fly above a certain height to minimize casualties, thereby degrading the effectiveness of information systems.

(A military coalition is only as strong as its weakest members. NATO had, and has, a bunch of them.)

o Total information superiority was manipulated, if the debate over the total number of tanks destroyed is any indicator, by asymmetric offsets (e.g., fake tanks, other decoys) and by a study of NATO air operation templates.

(Based on some of the things I have read on the Gulf War, we may not have hit 10% of Iraqi tanks prior to the Desert Saber ground fighting. The Iraqis used the same tricks the Egyptians did with the Israelis. They set burning tires on the rear decks of live tanks and put gasoline containers in camouflaged decoy positions so they would burn convincingly when hit. Not for nothing did the USAF avoid a BDA investigation of destroyed tanks when we owned the battlefields of southern Iraq! The attacks did condition the Iraqis to camp at a distance from their vehicles. This, more than anything else, is why the "Battle of 73 Easting" was so one sided. Most of the Iraqi tanks that were destroyed were not crewed when the troop from the 2nd ACR came over the rise.)

o Total information superiority did not result in NATO communications working without serious problems, even after years of practi

ce and in the face of no radio-electronic counterattacks.

(This is still a real problem. Recent operations with German Tornados showed that they had the same spread spectrum digital radios as USAF planes but lacked common cryptographic set up so they could not talk to each other. The USAF just recently informed allied air forces that either they met USAF communications standards or they will not fly with USAF strike packages in the future. I suspect the politicians will have a say in that one. If the recent Iraqi bombings are any indication.)

Interesting. I'll have something to say another time. Thanks.

 

On Insurance and genetic defects

I'm really looking forward to mail you get on this topic. It's one that I find very interesting, because I suspect the marching tide of technology is about to radically change the nature of "health insurance".

I say this because it seems to me that insurance is based on the concept of unpredictable risk (I realize that "unpredictable risk" is sort of an oxymoron, since risk by definition should be unpredictable, but that's not how that word is often bandied about, where it usually just means "something bad").

When unpredictable risks become predictable... I think you ultimately eliminate insurance for those former risks. If you don't, you'll have not only the insurance companies, but the insured individuals who aren't "at risk" legitimately up in arms. Why in Ghu's name should someone who can, say, eat high cholesterol foods without running a risk of heart attacks be willing to pay the same "insurance" premium as me, who has to watch what he eats? (BTW, I've read that there are such people -- apparently there was a mutation in Italy about a 1,000 years ago that allows certain individuals to tolerate cholesterol levels that would have "normal" folks giving our physicians conniption fits).

Of course, predictability isn't absolute; I'm assuming that here just to keep things simple. But I think the same logic applies when predictability is "good", but not "perfect".

It'll be interesting to see how this all works out. Will we abandon health insurance for medical savings accounts? Dragoon everyone into a "health insurance" system to share the cost of our genetic diversity (which I think would require government control of the program, since what other institution can require participation of everyone)? Or something else?

It's interesting how progress messes up pre-existing conditions. Seems to me I recall a guy writing a series of scifi stories with that as one of the premises. But I'm still waiting for the invention your writing buddy Niven hypothesized which would "force us to legalize murder" <grin>.

- Mark mark@arcabama.com

Good summary of the problem. 

 

 

 

 

 

 

  TOP

 

 

birdline.gif (1428 bytes)