| Friday, December
22, 2000
We can open with
Christmas Cheer mail:
To quote you from December 21, 2000: "Mark and
Eric are crawling around in the attic pulling Ethernet Level 5 cable ...
"
A brilliant expert such as yourself should already
know that there is no such thing as Ethernet Level 5 cable. I could leave
it at that but I won't since you seem intent on the idea that if I am the
one telling you I must be lying.
CATEGORY 5, or "Cat 5", (not 'Level' as
you stated) is not just for ethernet [sic] data communications. It was
originally made for TELEPHONE communications, but it quickly found its way
into the world of ethernet [sic] because it was more convenient to use
cable that had already be strung than restringing cable. What is the
difference between 'Cat 5' and conventional telephone wire? Basically the
number of twisted-pairs. Cat-5 has 4 pairs and conventional telephone has
2 pairs (incidentally, the 2 inner wires are the primary pair, or line 1,
and the outer pair is the secondary, or line 2). With Cat 5, you use RJ-45
connectors and with conventional phone you use RJ-11. Most digital phones
use RJ-45 with either Cat 5 or Cat 3 - but Cat 5 is more prevalent since
the increased cost is miniscule compared to the increased quality of
transmission.
Ethernet Level 5 cables... do you put this quality
of detail into all of your writing?
DarrenR114@aol.com
Darren; on behalf of; Darren Remington [DarrenR114@netroenterprises.com]
Let's take this in order, because it does give me
a chance to make clear the difference between a professional publication
and a daybook; some people evidently don't know, and thank you for the
opportunity to tell all.
My professional publications get examined by
others before they go to the editors, who themselves have a look. In the
old days of BYTE in Peterborough we had one of the best groups of experts
in the world, Tommy Thompson on Macs, Jon Udall on Internet, Halfhill,
Russell, Rick, and a dozen other really top people. I had a technical editor whose
job it was to ask me about details, both trivial and important, to be sure
I meant to say what I had said, and to tell me if I was just dead wrong.
In those days, too, it was my habit to send copy to people being written
about with the generic statement that "I will correct errors of fact,
and I will listen to arguments about errors of judgment, and I reserve the
right to determine which is which." I often got replies. Once in a
great while I would correct an error, and I think once I corrected a
judgment (had to do with a message in an early DOS, that said "now
trashing your hard disk." Bill Gates himself called on that one: a
summer hire had put the message in, and in fact it did nothing; my problem
had to do with a bug Microsoft had NOT known about, about a log file that
didn't properly close; they fixed that before we went to press, and I was
able to tell the whole story including the coincidence that I got the
awful error message about "illegal copy of DOS now trashing your hard
disk" and my getting the disk problem, AND sbout how by the die you
read that they fix would be available although this being pre-Internet you
had to download it tediously from a bulletin board. Ah well.)
But all that happened over a period of a month or
more, and the column did not hit the stands until about 3 months after it
was submitted. This meant that I had to predict what would be interesting
after 12 weeks. I got things RIGHT, but it was a bit slow getting to the
world. I am reminded of a journalistic meeting on accuracy; the Editor of
the Christian Science Monitor said "Everything in my paper is
true," to which some wag replied "Yes, Peter, and it has been
for a long, long time."
Now I turn in the column on the 7th, and it
appears on line about 2 weeks later. I turn in a full 8,000 words or so,
and they break it into weekly parts, supply headers. and such like, and do
some sanity checking, but nothing like what the old BYTE did; I don't have
a single Technical Editor and another (Walt Williams for the entire
history of my column in BYTE) style editor (he didn't have much to do
except that McGraw Hill had some editorial diktats that had to be
followed, and he did tend to break my overly long sentences into shorter
ones, which in fact was a good thing to do, my non-fiction writing habits
being perhaps overly influenced by Macauley). However, I do send it out
for sanity checks to appropriate people including Bob Thompson and Roland
Dobbins and David Em and Eric Pobirs and my son Alex before I send it in
to the editors. We do try to get things right, and certainly we would have
caught the "Level 5" vs. "Category 5" error.
This, however, is a day book. When I wrote that I
wondered "It's Category 5 or Level 5, I got it wrong once, I wonder
which it is?" but frankly I didn't want to disturb Mark who was
crawling around between floors drilling holes, or Eric who was feeding
pulling wire (I think that's the proper term, they aren't here now, I mean
the thin wire they pass along a channel to serve as the means of pulling
the Level 5 Ethernet cable and the whatever level telephone lines they are
pulling for my DSL or T1 connection depending on what the phone company
thinks it can do; either way I wanted new wire from the telephone closet
to the Cable/Server room). It didn't seem worth the effort. Mea
culpa, mea culpa, mea maxima culpa... I was quite
certain that one or another reader would inform me if I had it wrong, and
I cannot think that it makes very much difference to be wrong for a day or
so in a day book. I make no doubt that Bob Thompson spotted the error
within an hour and didn't bother to tell me because it didn't seem
terribly important. Still, best to get things right, and I will, and
thanks for telling me.
As to the two addresses above and Mr. Remington's
other remarks, he sent me several notes in his amicable style about Pat Buchanan
[that fascist, says Remington, and I didn't really want to write an essay
on what fascism is and was, and the whole question of class warfare,
fascism being a Marxist doctrine that accepts much of Marx's analysis but
proposes entirely different solutions: rather than elimination of all
classes but the proletariat, fascism proposes that the State require the
classes to work together and to be represented in the parliament as class
representatives rather than geographical districts or other means] and the
Palm Beach votes. In one case I asked for sources because he said some
things I had not heard, and given his casual use of "that fascist
Buchanan" I did rather want a source other than him.
Whereupon he sent another of his polite notes
saying that by asking for sources I had accused him of lying. At that
point I made a rule that put further communications from him into the JUNK
folder which I look at in the leisure I generally don't have. So he took
the trouble to find a new address in order to send me this note, and my
thanks; I don't like to have errors of fact. I can live with errors of
judgment.
I also don't do a lot of rewriting of day book
entries and replies to mail. I can't. I have two choices: go the way I go
now, with much of this top of my head and first draft, or NOT TO DO IT AT
ALL. Given that a fair number of you have paid
subscriptions (thanks!) I presume that most want me to keep on as I
have. Those who can't stand it have the option of not reading; apparently
there are some who can't stand it, but continue to read anyway. I haven't
been a practicing psychologist for 40 years (and my title when I was was
"Human Factors Engineer and Aviation Psychologist") and I know
no more about masochism than most well-read people.
It turns out that much of what Mr. Remington says
is wrong; for the definitive story on Level 5, Category 5, and that I
should have Level 5e at least see below.
And another letter which is the final
word (so far as
I am concerned) in a very odd exchange. I warn you it is long. In any
event you
are warned not to be deceived by the obviously -- to me obvious
anyway -- composite photo of the Earth at night. The Earth usually has
CLOUDS. Satellites do not see through solid objects and therefore no
single picture could possibly have encompassed the entire Earth, so
therefore the picture HAD to be assembled from pictures taken at different
times. I wonder if there is any single person reading this to whom that
was NOT obvious??? Anyway, here we go. If you want to skip this, click
here.
My apologies in advance for posting this
exchange, which was polite if, to me at least, longer than it needed to
be, just after the one above; I do not intend to belittle Mr. Sloan, I just
find it hard to imagine that anyone able to find this web site would be
deceived. We begin with the final exchange:
Sir,
I suppose I'm not being clear: I'll try again.
You say "We have had this one before but it is
worth mentioning again. And look at Korea...". Clearly you are
pointing out that ther lights in South Korea are far brighter and
widespread than in North Korea, indicating more industrialization and by
inference a stronger economy.
If this picture is unretouched, with the pictures of
North and South Korea taken at the same time and on the same night, then
this would be good evidence. However, this picture is art, not science and
we have no information about the provenance of the multiple images used to
make this picture. I certainly would not know who to ask for that
information, and it is not provided. We have no idea of when and how the
pictures of North and South Korea were taken, just that they were glued
together. So, we can come to no objective conclusions about the lights
visible in each country.
To ask us to "And look at Korea..." in
this case is to ask us to examine a piece of artwork and come to some kind
of conclusion about geopolitical reality. This is disingenuous.
On a different topic, I was appalled by the article
about the "copy-protecting of hard drives". I suspect this is a
trial balloon sent up by Big Media to see if it'll fly. However, I suspect
that this will fade away, since as the "Register" article points
out: "The Register understands there is fierce opposition to the plan
from Microsoft and its OEM customers.". Since Microsoft still wields
the big stick in the industry, that should be the kiss of death (I
hope!!!).
I hope you're feeling better!
John Sloan
-----Original Message----- From: Jerry Pournelle [mailto:jerryp@jerrypournelle.com]
Sent: December 22, 2000 12:03 AM To: john.sloan@trader.com Subject: RE: I
wonder what this picture really means...
I cannot imagine what you are talking about, or why. There was no more
secret about this picture than there was about the cloud free orbital
picture of the United States. Those who made it are well known. Any
reporter can talk to them. And what in the world importance is it what I
think the implications are to begin with?
If you think North Korea is normally more brightly lit, and the other
patterns you can see, that is your privilege, but I cannot imagine anyone
going to the trouble to make such a picture for that reason.
-----Original Message----- From: John N. Sloan [mailto:john.sloan@trader.com]
Sent: Thursday, December 21, 2000 5:53 PM To: Jerry Pournelle Subject: RE:
I wonder what this picture really means...
Sir,
You say: "O course it is a composite." I
can't agree. This picture could easily be entirely fake, created in
software. You have no proof that it is, or is not, a composite picture.
My point was that even if it is a composite and it
has been made by skillful technicians who made their best effort to make
it representative of the night sky across the world, it still should not
be used for any photo interpretation. There are too many variables in what
composition does.
The picture cannot be used to determine that the
lights in North Korea were usually less bright than in South Korea, since
the photos of those two countries could have been taken at different times
and under different circumstances. So, any judgements we may come to about
the difference in their economies from this photograph would be
unjustified.
Don't get me wrong. I am aware the South Korean
economy is far stronger than the North Korean, along with the level of
industrialization and the standard of living. The political systems cannot
be compared, with much greater freedom in the South. And of course, this
freedom has been assisted by the courageous efforts of the USA for many
years.
However, the picture in question should not be used
as evidence for this fact! It is a pretty picture and nothing more.
Essentially, the picture doesn't "mean" anything since we can't
use it to learn anything.
Am I splitting hairs? Perhaps. However, any photo
interpreter worth their salt knows to learn anything from a picture they
need to start with the ORIGINAL photograph, not the photo that has been
composited, airbrushed and tweaked until it might show something that
isn't there and hides what is really there.
Respectfully yours,
John Sloan
-----Original Message----- From: Jerry Pournelle [mailto:jerryp@jerrypournelle.com]
Sent: December 21, 2000 6:43 PM To: john.sloan@trader.com Subject: RE: I
wonder what this picture really means...
? Of course it is a composite. Who in his right mind would suppose
otherwise?
-----Original Message----- From: John Sloan [mailto:john.sloan@trader.com]
Sent: Thursday, December 21, 2000 12:58 PM To: jerryp@jerrypournelle.com
Subject: I wonder what this picture really means...
Sir,
While looking at the picture at http://antwrp.gsfc.nasa.gov/apod/image/0011/earthlights_dmsp_big.jpg
, I was struck by the following thoughts: 1) There are no clouds. 2) There
is no sunshine anywhere. 3) The camera that took this "picture"
seems to be able to see every point on the globe, and put it all into one
picture. This picture may be a composite, with different sections taken at
different times and pasted together. When this happens, you usually get
artifacts where you can see the edges of the individual photographs that
were pasted together, but there are none that are apparent here. Looks
like someone has been busy with the digital equivalent of an airbrush! At
best, this is a cleverly massaged composite of carefully selected images
chosen to portray the world at night, that has been adjusted according to
whatever the technician "thought" it should look like. At worst,
it is an "artist's conception" of what the world would look like
from space at night. Either way, it is not an "image" of
anything any more than a painting is a photograph. It would be naive to
try to conclude anything about the relative industrialization of North and
South Korea from something like this that has spent a lot more time under
the airbrush than anywhere else! Please don't get me wrong: This is a
pretty and interesting picture, but it simply has no objective meaning. It
just isn't real.
Respectfully yours,
John Sloan
As I said in the beginning I thought it obvious
to anyone. Clearly I assume an intelligent readership. If anyone was led
astray by my remarks, I would suppose that error corrected now. I would be
astonished if there were one single reader to whom that applies. If there
were, the deception is clearly ended.
Regarding the IDE copy protection (and Eric tells
me that SCSI won't be far behind) I will have much to say, but I think in
the column.
It turns out this is NOT the end of the
discussion. See below.
For those who skipped the
above we resume our previously scheduled letter column:
Dr. Pournelle,
A personal note - I have been an avid reader of your
science fiction and various columns for over twenty years. Even when I'm
not in complete agreement you, I find your writing compelling and thought
provoking. It is a rare occasion when I don't agree with you, probably
because your writing has been so influential to me.
I have noticed that you haven't addressed many
comments to the growing discussion regarding the Digital Millennium
Copywrite Act (DMCA) and the repercussions from it. I subscribe to a mail
group <dvd-discuss@eon.law.harvard.edu> that has been following
various issues such as the DECSS and CueCat. I'm making the assumption
that you are aware of the DECSS issue while CueCat is a little more
obscure.
In a nutshell - A company gave away millions of cat
shaped bar code scanners through Radio Shack stores and magazine
subscriptions. They encrypted the output; however, not surprisingly, many
curious people reversed engineered the hardware and discovered that it
could be used as regular bar code scanner with a simple software solution.
The company began to send out letters threatening to have these people
prosecuted under the DMCA stating that reverse engineering is illegal
under the Act.
Now there is this:
http://www.theregister.co.uk/content/2/15620.html
"The Register has broken a story of the latest
tragedy of copyright mania in the computer industry. Intel and IBM have
invented and are pushing a change to the standard spec for PC hard drives
that would make each one enforce "copy protection" on the data
stored on the hard drive. You wouldn't be able to copy data from your own
hard drive to another drive, or back it up, without permission from some
third party. Every drive would have a unique ID and unique keys, and would
encrypt the data it stores -- not to protect YOU, the drive's owner, but
to protect unnamed third parties AGAINST you." (Quoted from John Tyre
at the dvd-discussion site).
It's easy to envision a day when many actions we now
take for granted are illegal. Backing up an image of your hard drive or a
making a copy of your software in the event of media failure or software
glitch. Or taping a show or game on television to watch at a later time.
The law states that you have a right to do so; however, now it also states
that it is illegal for companies to make or sell a device to allow you to
do so.
How do you as an author feel about copyright?
In Section 8 of the Constitution http://www.constitution.org/constit_.htm
, one of the powers given to Congress is this "To promote the
Progress of Science and useful Arts, by securing for limited Times to
Authors and Inventors the exclusive Right to their respective Writings and
Discoveries;" . Originally, the time was 14 years renewable for
another 14 years. In the 1800's, it was increased to 40 years. Even
allowing for the longer life spans of today, the period of life plus 90
years seem excessive. The writings of the Framers seem to indicate that
the limited time referred to a time span within the lifetime of the Author
or Inventor, not the lifetime of the Universe. Jack Valenti (President of
the Motion Picture Association of American) feels otherwise. He is on
record as stating that it should be for eternity less a day.
Many works of film are deteriorating and being lost.
Great (and mundane) writings are no longer available to the public because
the copyright holder (corporations in many cases) are unwilling to spend
the money to reprint them. Where is the promotion of Science and Art - the
main reason for the copyright?
In the same vein - the gold rush of gene patenting.
Companies are patenting genes. Not designer genes that they may have
created, but existing genes. It appears that they could hold the human
race hostage.
The Great Gene Grab http://www.techreview.com/articles/oct00/regalado.htm
Toward Sharing the Genome http://www.techreview.com/articles/oct00/shulman.htm
The Case for Gene Patents http://www.techreview.com/articles/oct00/haseltine.htm
What happens to the rule of law when the majority of
the population feel the law is wrong and ignore it?
Sorry to be so verbose, but I value your opinions
and wonder how your readers feel about these issues.
Cordially,
Dean Sanchez
First, I was happy enough with the 26 years
renewable Copyrights; I believe "Life plus fifty years" or even
longer is unconstitutional and clearly well beyond the intentions of the
Framers.
When a law is silly and unacceptable to a
population, the results are never good. Prohibition is the most obvious.
The current Drug Wars are I fear another.
There is a vast conflict between Internet
Procedures and Intellectual Property; between "information wants to
be free" and the rights of creators. On the other hand, my publishers
just paid a substantial advance to renew their rights to publish Lucifer's
Hammer, which was first published in the 70's; we still make good money on
that book, and I am grateful. And people do PAY
for this place even though they don't have to. So there is life after
Gnapster...
But we have seen in Microsoft Reader format most
of my books now offered on the Internet by small pirates for free. MOTE,
HAMMER, FOOTFALL, they are all up there and more; Eric showed me several
just yesterday. I need to think on all this.
And BYTE pays me for my columns. There are no
residual rights, but then there seldom are for journalism. The bottom line
is that it's possible to make a living as a writer, but the competition is
stiffer. I can't think that is all bad.
The disk copyright stuff is another story and
deserves longer treatment and I have to go out.
Now for the definitive word on Category and
Level 5:
Okay, here's a detailed response to Mr. Remington.
Unlike Mr. Remington (who claims to understand LAN cabling but does not)
and you (who have never claimed to be a cabling expert), I actually do
understand LAN cabling, having installed or been responsible for
installing literally hundreds of miles of the stuff at more than 100
sites, and having been certified by AT&;T on their structured cabling
system. So, point-by-point:
> A brilliant expert such as yourself
should already know that there is no such thing as Ethernet Level 5 cable.
I could leave it at that but I won't since you seem intent on the idea
that if I am the one telling you I must be lying.
Not lying, but mistaken. What Mr. Remington refers
to is "Category 5" cable. The specifications for Category 5
cabling and connectors were originally published in the TIA/EIA 568
commercial building telecommunications wiring standards document with
supporting Technical Supplement Bulletins TSB-36 and TSB-40/40A. That
document and the supporting TSBs were incorporated as TIA/EIA 586A in 1995
and adopted as ANSI standard ANSI/TIA/EIA-568-A. In the years before the
formal adoption of the standard, various vendors supplied cabling and
connectors which were not formally standards-certified, but which complied
with the draft standard. One of those, as I recall, was Anixter, which in
1991 began selling cable that supported 100 Mb/s connections. And they did
in fact label that cable "Level 5". In short,
"Category" numbers refer to the formal TIA/EIA-568
classifications, while "Level" numbers refer to Anixter's
products and classifications. Anixter's "Level" classifications
are a de facto standard because of their size and influence in the cabling
industry. You are certainly not wrong to refer to cable to be used for
100BaseT Ethernet as "Level 5". If Mr. Remington had much
experience with LAN cabling, he'd be aware that the folks who work with
the stuff for a living use the terms "Cat 5" and "Level
5" pretty much interchangeably.
> CATEGORY 5, or "Cat 5", (not
'Level' as you stated) is not just for ethernet [sic] data communications.
It was originally made for TELEPHONE communications, but it quickly found
its way into the world of ethernet [sic] because it was more convenient to
use cable that had already be strung than restringing cable.
Wrong again, Mr. Remington. Category 5 (Level 5)
cable was not originally made for telephone communications. It was
originally designed for 100 Mb/s data communications, and it was
originally extraordinarily expensive. Back in the early 90's, we used to
installed Cat 3 cable (about $40/box of 1,000 feet) for both telephone and
10BaseT. When Cat/Level 5 cable first became available in volume, it cost
literally 10 times what Cat 3 did. We originally paid about $400/box for
Cat/Level 5 cable, and that was the stuff with a PVC sheath. The
plenum-rated stuff was ungodly expensive. During that period, most new
buildings and those being rewired were cabled with a mix--Cat 3 for voice
and Cat 5 for data. As Cat 5 production ramped up, the cost differential
between Cat 5 and Cat 3 got smaller and smaller, and smart people began
spec'ing all Cat 5.
> What is the difference between 'Cat 5'
and conventional telephone wire? Basically the number of twisted-pairs.
Cat-5 has 4 pairs and conventional telephone has 2 pairs (incidentally,
the 2 inner wires are the primary pair, or line 1, and the outer pair is
the secondary, or line 2).
Wrong again, Mr. Remington. The number of pairs has
nothing to do with the Category rating, except as I'll explain below.
Nearly all horizontal (station) cabling runs are done with 4-pair cable,
whatever its Category rating. No installer I know (and I know a lot of
them) would even consider using 2-pair cable. It's not enough cheaper to
make a difference (most of the cost is in the labor in pulling runs) and
you always want to have the spare pairs available. What differentiates Cat
5 from Cat 3 cable is a bunch of technical electrical/electronic
characteristics (such as NeXT or Near-End-Crosstalk), which we don't need
to get into, but which have nothing to do with the number of pairs other
than peripherally. By that, I mean that the larger the number of pairs in
a cable, the more difficult it is to meet higher Category requirements.
For example, when I was wiring a new 10-story jail building, I would have
loved to have had 100-pair (or even 25-pair) Category 5 riser cable
available. But it wasn't, so we had to run a bunch of 4-pair Cat 5 cables
instead. (larger multi-pair Cat 5 cables became available subsequently.)
Mr. Remington is also confused about pairs. He
mentions outer and inner, by which I assume he refers to the pinout of
connectors, which has nothing to do with the cable. A standard 4-pair
cable uses the color coding Blue, Orange, Green, and Brown for pairs 1
through 4 respectively. The first (blue) pair is the "best" pair
in terms of electrical/electronic characteristics. Orange is next best,
and so on. There are various connection pinout schemes, the most common of
which are 568A and 568B, which specify where each wire of each pair
connects to the plug. The best I can get from Mr. Remington's confused
comments about "primary" and "secondary" and
"inner" and "outer" is that he has, to put it
charitably, an imperfect understanding of connector pinouts.
> With Cat 5, you use RJ-45 connectors and
with conventional phone you use RJ-11. Most digital phones use RJ-45 with
either Cat 5 or Cat 3 - but Cat 5 is more prevalent since the increased
cost is miniscule compared to the increased quality of transmission.
Wrong again, Mr. Remington. The "RJ"
(registered jack) terminology is obsolete USOC (Universal Service Ordering
Code) stuff. 10BaseT and 100BaseT Ethernet does not use "RJ-45"
connectors (although admittedly many knowledgeable people use the term
RJ-45 loosely). It uses 8P8C (8-position, 8-connector) plugs and jacks.
But then I suppose that mistake is no more serious than not capitalizing
the first letter of Ethernet.
> Ethernet Level 5 cables... do you put
this quality of detail into all of your writing?
Do you put this quality of detail into all of your
writing, Mr. Remington?
-- Robert Bruce Thompson thompson@ttgnet.com http://www.ttgnet.com/rbt/thisweek.html
Actually, I wouldn't have put this up were it not
that we may as well get all the details straight; I would not want someone
to repeat Mr. Remington's errors (or mine) thinking them authoritative.
Mr. Thompson on the other hand really is an authority. And now we know
more than most of us really wanted to... My apologies to Mr. Remington,
who I suspect meant well and didn't express himself as well as he would
have liked.
And there is more, also worth posting:
Dr. Pournelle -
I too saw your use of the term "Level 5"
when you meant to say "Category 5" and decided to ignore it
until I saw the letter from Darren Remington. He said:
"A brilliant expert such as yourself
should already know that there is no such thing as Ethernet Level 5
cable."
Wrong! The Anixter Levels Program existed before
Category 5 was established, and in fact Category 5 was based largely on
Anixter's Level 5. See:
http://www.anixter.com/techlib/standard/cabling/d0502p34.htm
"It was originally made for TELEPHONE
communications, but it quickly found its way into the world of ethernet
[sic] because it was more convenient to use cable that had already be
strung than restringing cable."
I wonder where he came up with this myth. Telephone
cabling was around long before the Category system was established, and
much of the installed base of telephone wiring only meets Category 1 (no
performance criteria). In fact the quality of cable many companies have
been installing is so bad that the FCC recently mandated that the minimum
quality cable that could be used in premises telephone wiring is Category
3 (in other words cable suitable for at best 10BaseT Ethernet). See:
http://www.bicsi.org/FCC.htm
"What is the difference between 'Cat 5'
and conventional telephone wire? Basically the number of twisted-pairs.
Cat-5 has 4 pairs and conventional telephone has 2 pairs (incidentally,
the 2 inner wires are the primary pair, or line 1, and the outer pair is
the secondary, or line 2)."
Wrong on almost all counts. Category 5 cable is most
commonly sold as 4 pairs, but 2 pair Category 5 cable is available. Common
100BaseT and the older 10BaseT Ethernet only use 2 pairs anyway. The major
difference between Category 5 and telephone cable is the twisting. Quality
telephone wire has always been twisted, but huge quantities of untwisted
"quad" have been used in premise wiring. Category 5 has a much
higher twist rate than telephone wire, and each of the pairs is twisted at
a slightly different rate to reduce crosstalk between pairs. Flat
telephone cord where there are parallel conductors, and you can therefore
speak of "inner wires" versus outer, is only suitable for
connecting a telephone to the wall plate and should never be used for
computer networking. It is true that on a two line telephone the inner
pair of contacts on the plug is for line 1 and the outer is for line two.
"With Cat 5, you use RJ-45 connectors
and with conventional phone you use RJ-11."
I would have let this go except for his attitude.
There is no such thing as an RJ-45 or RJ-11 connector. The proper
terminology is an 8 position 8 contact non-keyed modular connector or a 6
position 2 contact non-keyed modular connector. RJ-45 and RJ-11 refer to
specific wirings of those connectors according to the telephone industry's
Universal Service Ordering Code (USOC). A two line phone such as he spoke
of uses a 6 position 4 contact non-keyed modular connector wired per
RJ-14.
The pair assignment on the "RJ-45"
connector as used for Ethernet is very different from any of the USOC
wirings. Miswiring the pairs onto the connectors by those who only know
telephone usage is a very common cause of poor network operation and high
data error rates.
I am afraid that it is too late to change the
popular usage away from RJ-45 for the common connector used in Ethernet
wiring, but that does not make it correct.
With respect to the cabling you have just had
installed, I hope you did not put in Category 5. In today's world I would
suggest a minimum of Category 5E be used. Category 5E is what Category 5
should have been but was not. Basically the Category 5 Standard was too
loose, and could not assure reliable operation under all conditions.
Category 5E significantly tightened the Standard and should be the minimum
quality of data wiring installed today.
Category 6 is being worked on, but has not been
released. Meanwhile Anixter has come out with a Level 6 and Level 7 which
assure higher levels of interconnect system performance. To achieve Level
7 performance requires precise matching of the wire and connectors used in
a link. Some vendors such as Krone make all the components, while others
such as Belden's wonderful MediaTwist cable must be matched up with
suitable connectors from Panduit.
If you are considering future Gigabit Ethernet
operation, it might pay to install Level 7 now.
Ray A. Rayburn Audio (at) Technologist (dot) com
Former Telecom engineer and current Audio over Ethernet engineer. www
(dot) SoundFirst (dot) com
I believe that it was 5E that they put in; it was
all under the supervision of my son Alex and his experts, and they know
FAR more than I do about it. AND IT WORKS. This is going out over the new
Ethernet lines, and the old Blue Wire running across the floor is GONE.
And Roland adds:
*Everyone* calls the connectors RJ-45s, even though
they really aren't. And most people who've been in networking for a while
(15 years in my case) use 'Level-5' and 'Cat-5' to mean 'Cat-5E'.
As for Gigabit Ethernet, I've never done it over
copper - I've always used either MultiMode Fibre (MMF, for runs of up to
500m or so), or Single Mode Fibre (SMF, for longer runs), and fibre of
differing core diameters - measured in microns - for both types. For
Gigabit Ethernet and for SANs (see below), the fibre I've used has had
what are called 'SC-type' connectors on the ends.
Also, the Gigabit Interface Converters (I think
that's what the acronym means), or GBICs, which plug into Gigabit Ethernet
LAN cards and switches and so forth for Gigabit Ethernet over fibre, come
in different flavors which must be matched to the type of fibre used. One
network I designed made use of 'Z'-type GBICs plugged into Cisco Catalyst
5500-series switches, with a run distance of some 2 miles between them.
If I remember correctly, by using the right GBICs
and fibre and blades in one's switches, it's possible to run Gigabit
Ethernet (and Gigabit EtherChannel, which can consist of up to 8 Gigabit
Ethernet links aggregated to form one big, fat virtual path) to distances
of around 10 miles or so. For organizations which are in large campus
environments or can secure right-of-way and permits to dig ditches for the
fibre, the benefits are obvious.
This is what George Gilder was talking about years
ago when he was telling us that rather than complicated telco-sponsored
initiatives like ATM, what we really needed was lots of 'dark fibre' which
could be provisioned as we pleased. He was right.
Fibre-Channel Aribtrated Loop (FC-AL) storage
subsystems also use GBICs to connect high-end servers to Storage Area
Network (SAN) switches, and thence to large disk units. I recently
configured a couple of Compaq DL580 servers (4 PIII/700 Xeons &; 4GB
RAM each) in such a fashion, linking them through meshed SAN switches to
two separate Compaq RA4100 RAID enclosures containing 256GB each of
RAID-5+1 SCSI-3 storage using 10K RPM drives. The GBICs present in the
server Host Bus Adaptors (HBAs, or disk controllers), the SAN switches,
and the RAID boxes were all the same type as those used for the Gigabit
Ethernet NICs in each of the servers, and the fibre cables I used for the
storage were MMF cables identical to that used for the Gigabit Ethernet
links.
Sun have come out with a purely FC-AL storage
subsystem, which they call a T3. It consists of FC-AL HBAs for the
server(s), and 36GB 10K FC-AL *hard drives* in the T3 RAID enclosure. They
even support internal pure FC-AL drives inside their new 280R UltraSPARC
III-based server. No copper cabling involved at all. I'm sure Compaq and
the rest won't be far behind with this purely FC-AL drive subsystems.
I'd be very interested to learn whether anyone's
actually doing Gigabit Ethernet over copper; I've never seen it,
personally.
--- Roland Dobbins <roland_dobbins@yahoo.com>
Gilder and I have been chasing each other for
years; it was pretty clear to me that he read my old BYTE column in paper
days since once I listened to him give a keynote speech at a CDROM
conference that was pretty close to a column I had published several
months earlier. In any event, he is a good evangelist for our
technologies, and if I sound a bit jealous, well, I did got top billing
over him in the Millennial edition of the American Enterprise Institute
issue on assessing future technologies... And that will have to do.
I had thought there was
little else to say about the NASA "Earth at
night" photo, but in fact there is:
Dear Dr. Pournelle:
After reading John Sloan's emails regarding the
Earth at night photo, I was overcome by curiosity about its origin
(perhaps I have too much free time?). I dug around a bit on the website
where the photo is hosted, and found some information. Here is the link:
http://antwrp.gsfc.nasa.gov/apod/ap001127.html
This web site features a different astronomy photo
each day, along with information about the photo and relevant links. Very
cool!
I did some further surfing and found my way to the
NASA "earth observatory" website's "Lights Study",
which explains in detail how scientists use city light data to map
urbanization:
http://earthobservatory.nasa.gov/Study/Lights/
Fascinating stuff!
As to the origin of the photo, according to the NASA
web site:
"This image of Earth’s city lights was
created with data from the Defense Meteorological Satellite Program (DMSP)
Operational Linescan System (OLS). Originally designed to view clouds by
moonlight, the OLS is also used to map the locations of permanent lights
on the Earth’s surface."
I was surprised to learn that the images used to
build the composite were actually taken over a period of several new
moons, when the satellites were sensitive enough to record the
illumination from city lights.
Regarding the economic/industrial implications of
the picture, to quote the web site:
"Urbanization in any country generally
begins when large-scale commerce takes root and most new jobs are to be
found in the factories and financial centers in cities."
This would seem to be a fairly obvious observation,
and I'm not at all sure why Mr. Sloan has such difficulty with it. In
point of fact, this photograph does clearly illustrate that North Korea is
less urbanized (industrialized) than South Korea.
According to information I found on the NASA
website, almost every assertion Mr. Sloan makes is simply wrong. He is
remarkably consistent that way. However, in one small detail he was
correct. The Photo is indeed "art" in the sense that it is a
composite picture superimposed upon a map (the night time geography would
not be nearly so discernable otherwise). However, the photo is also
clearly "science", presenting factual data. Enhancements like
the map only serve to increase the usefulness of the photo. Here is a USA
at night photo without a map: http://antwrp.gsfc.nasa.gov/apod/image/usanight_dmsp_big.gif
Not nearly as easy to interpret!
Mr. Sloan attempts to make the point that even if
the photo is a composite (it certainly is!), then "There
are too many variables in what composition does."
The folks over at NASA would seem to disagree. Perhaps Mr. Sloan can fire
off a hastily written and poorly researched email setting them straight on
this matter. One should always play to one's strengths.
In closing, thanks for linking to the photo in the
first place, and for your thought-provoking comments about it. I learned a
lot as a result, and Mr. Sloan's notes to you were nothing if not
entertaining.
Warmest Regards, Joe Shockley
I had not thought about the Moon, but once I did
it was obvious that those had to be taken during a New Moon or the Dark of
the Moon, else they would never have shown up so well.
And from Clark Meyers
He begins with quotation from the site where the photo is to be found:
This image of Earth’s city lights was
created with data from the Defense Meteorological Satellite Program (DMSP)
Operational Linescan System (OLS). Originally designed to view clouds by
moonlight, the OLS is also used to map the locations of permanent lights
on the Earth’s surface.....
The above image is actually a composite of
hundreds of pictures made by the orbiting DMSP satellites.
Quoting first from the FAQ's then from the further
information on the site -
Q13: What if I used to be a millionaire but
then I believed something I read on APOD and now own only a single dented
bucket? A13: There are no guarantees. Use APOD information at your own
risk.
Earth at Night Credit: C. Mayhew &; R.
Simmon (NASA/GSFC), NOAA/ NGDC, DMSP Digital Archive Explanation: This is
what the Earth looks like at night. Can you find your favorite country or
city? Surprisingly, city lights make this task quite possible. Human-made
lights highlight particularly developed or populated areas of the Earth's
surface, including the seaboards of Europe, the eastern United States, and
Japan. Many large cities are located near rivers or oceans so that they
can exchange goods cheaply by boat. Particularly dark areas include the
central parts of South America, Africa, Asia, and Australia. The above
image is actually a composite of hundreds of pictures made by the orbiting
DMSP satellites.
This image of Earth’s city lights was
created with data from the Defense Meteorological Satellite Program (DMSP)
Operational Linescan System (OLS). Originally designed to view clouds by
moonlight, the OLS is also used to map the locations of permanent lights
on the Earth’s surface.
The brightest areas of the Earth are the most
urbanized, but not necessarily the most populated. (Compare western Europe
with China and India.) Cities tend to grow along coastlines and
transportation networks. Even without the underlying map, the outlines of
many continents would still be visible. The United States interstate
highway system appears as a lattice connecting the brighter dots of city
centers. In Russia, the Trans-Siberian railroad is a thin line stretching
from Moscow through the center of Asia to Vladivostok. The Nile River,
from the Aswan Dam to the Mediterranean Sea, is another bright thread
through an otherwise dark region.
Even more than 100 years after the invention
of the electric light, some regions remain thinly populated and unlit.
Antarctica is entirely dark. The interior jungles of Africa and South
America are mostly dark, but lights are beginning to appear there. Deserts
in Africa, Arabia, Australia, Mongolia, and the United States are poorly
lit as well (except along the coast), along with the boreal forests of
Canada and Russia, and the great mountains of the Himalaya.
The Earth Observatory article Bright Lights,
Big City describes how NASA scientists use city light data to map
urbanization.
Image by Craig Mayhew and Robert Simmon, NASA
GSFC, based on DMSP data
My own perspective - Granted that in order to be
considered evidence in an American court of law the preparation of the
image would have to be explained by people who could testify to all the
facts of preparation I personally credit it. The information and
methodology underlying this image is printed in respected journals and of
course comes from reputable sources. I have looked at satellite
photographs and photo-recon strips of terrain I have walked over and lived
on and I am inclined to believe the pictures of land I have never seen
bear the same relationship to that land as the pictures of land I have
seen. Calibration for satellite photographs I have kibitzed uses well
defined and instrumented artificial ground targets laid out and then
simultaneously viewed by the ground crew with eyeball and instruments, low
flying aircraft, high flying aircraft and satellites - timed to suit the
satellite of course.
I worked in the preparation of film images for
admission as evidence in Court in which even before the original image was
made, camera and film format was considered, then lens and film itself,
then darkroom techniques including print size and paper and so on - all
the facts about final decisions were available to both sides in Court of
course. With Digital Photography all the same considerations apply with
different answers. The general rule for admitting an image of say a
traffic accident is to have someone, ideally the photographer, say in
effect "I took the picture and it looks about like what I saw with my
eyes". Original Image is a nebulous term.
Call this image a visual aid rather than evidence if
you will.
Clark
So some good has come from it all, and I was
incorrect about it being uninteresting. My thanks to Mr. Sloan and to both
Mr. Meyers and Mr. Shockley. And the final word is below.
And here is a good statement of the encryption
problem:
The main thing that bothers me about hard drive
encryption is that, at some point, you have to get access to the
unencrypted data.
How would Quicken handle your finances if it didn't
get access to clean data?
And once you have access to the unencrypted data,
what prevents you from doing whatever you want with it?
So for this hard drive encryption to work, you would
have to have support from the OS vendors. And if it is under the control
of the OS, it should allow you to copy data from drive to drive.
DVD's work because a typical DVD drive is an
embedded system, and the fact we can now break DVD encryption arose
because of a mistake in one software program that was used by the DeCSS
system. Though it would have been broken eventually by someone willing to
spend enough computer time at it.
While following some of your links, it appears that
the encryption will be an optional feature. So, if you take advantage of
the fact that Windows 2000 already supports encryption in their filesystem,
I could see where they could use this to make any filesystem safe. But
probably for special applications. Such as military use, laptop systems,
etc. With a laptop, you could probably do some of this unencryption in the
BIOS, so if you don't login correctly, you can't get at the data on the
hard drive even if you remove the hard drive. You couldn't even get at it
with block level access.
But for normal use, people would not put up with a
system where the loss of a password can totally prevent access to the
data.
Kevin Kreiser
And I see little to object to. I note that the
SETI experiment has shown how to get massive amounts of computing power
working in parallel on any problem that interests the computer community.
The implications ought to be obvious to us although they will not be to
the Congress and the bureaucracy who think legislation is omnipotent.
I think the last word on this is what Mr. Thompson
said on his web site: this plan will either become mandatory, or it will
fail. Literally no rational person will buy a hard disk that has, as a
*feature*, the ability to completely deny access to your data.
My father is fond of quoting Wyland's Law of
Automation: "Anything that can be automatically done *for* you can be
automatically done *to* you. Some day the drive may malfunction and simply
decide not to let you have your data, period. It's just another
catastrophic failure mode. Who wants to buy equipment with extra failure
modes built in?
As Thompson points out, if any one drive in your
system is one of the new ones, all of them must be, because it is a
feature of the drive that it won't let you copy data to unregistered
drives. You must buy nothing but registered drives, and you must register
them to the Authorities and receive the codes that will allow them to talk
to each other.
If we follow that line of reasoning, it follows that
you also need a registered Zip drive, a registered tape backup drive, and
so on. Actually, it is more likely that each Zip disk and each tape, not
the Zip drive and tape drive, will be registered. There are already
removable flash memory devices with anti-copying serial numbers, so that
you can download music that will only play on one or two flash chips.
It then follows that the Authorities will need a
very high-bandwidth connection to the Internet and lots of fast servers to
provide an unending stream of authorization codes: "Hey, Steve, I
can't copy any data on this new Zip disk!" "Oh, you just need to
register it with the Authorities. This time of day, it usually takes four
or five minutes..."
Who are the people who think they can sell this to
us? IBM couldn't even convince the market to adopt MicroChannel, and that
wasn't a tenth as odious as this idea.
Hey, when this system is in place, do you suppose
they will propose version 2.0 of the system, where all your data gets sent
to the Authorities so they can decide (on a byte-by-byte basis) whether
your drives are allowed to copy it among themselves? After all, what good
will this system do if it always grants every request to register a disk?
You might possibly be engaged in copying stolen music or ThoughtCrime
essays.
I can just picture a future Byte review of hard
drives: "And for the performance benchmarks, we connected our test
computer directly to a T1 line and ran the tests at 2:00 A.M. to minimize
the effect of Copying Authorizations on the tests..."
Actually, this scheme is to hard drives exactly what
Newspeak is to language: an attempt to make it impossible to copy
forbidden data (versus an attempt to make it impossible to think forbidden
ideas). Therefore I propose we call the new drives "Newspeak hard
drives". -- Steve R. Hastings "Vita est" steve@hastings.org
http://www.blarg.net/~steveha
Well said.
TOP
|
CLICK HERE for
a Special Request. 
