>About 2004... the professor brought us a bunch of those old purple sheets for a handout
Impressive! The stencil had to be typed on a typewriter or other impact printer, as well as finding the mimeo stencils, and a supply of alcohol, and the mimeo drum!
It would be harder to use the old dictaphone (the acetate belt type, with acetate old and brittle since it'd have to be pre-(1970?) manufacture).
The idea of analog computers was to make a box with operational amplifiers to generate and manipulate signals. This is still alive today, in all circuits that use operational ampllifiers, but it isn't surrounded by a box and labeled 'computer'. An online electronics parts supplier lists, under 'operational amplifiers', over thirty thousand items.
Rent _Mutiny on the Bounty_ (1962), and look at the DVD extras menu for a nifty video of the ship they reproduced for the film. Wood cost makes it infeasible to do that nowadays, and skills are probably scarce, too.
The 'pen knife' is just a whittling tool; modern ones
would carve a point in a quill just fine, if you can
find someone who's mastered the technique.
There are quill pens available in modern times, as museum-reproduction-quality decorative items. Calligraphy aficionados have the skills and tools to sharpen 'em, too. 'head clamp' is now more used for brain surgery
than for photography. As for 'paper tape readers', a recent one was made to transcribe player-piano tapes to MIDI format. If you mean the 1" tape for ASR-33 teletypes, that's more easily read with a scanner and character recognition software... but it could be argued that such a software solution is still a 'paper tape reader'.
Are you referring to "point-contact devices" as in working on similar principles to cat's whisker or crystal-crystal rectifiers?
No.
Actually, I'd say yes here. The previous technology was a gizmo called a 'coherer' which was basically a bottle of metal granules. If it stopped rectifying, ya had to shake the bottle. That technology goes back nearly a century.
Copyright is only 200 years old and it was the big publisher who invented the copyright, not the content creators.
The truth is, the publishers weren't paying the content creators anything. The original intent was to aid the author, so the effect of copyright was to add a cost burden to the publishers. It didn't HURT the publishers (after all, they pass those costs to the book buyers), but it was long after the birth of copyright that the speculative buyers of rights got our legislatures to extend the period. And to impose confiscatory and punitive unwarranted judgments on 'violators'.
That's rawther difficult without abandoning computers altogether. Are there any new monitors that don't support HDCP? I thought it was a requirement for the HDMI license, and every TV monitor has an HDCP-capable HDMI port nowadays.
There's a lot of DVI monitors that plug into my HDTV component box, that aren't HDCP equipped. The 'every TV monitor' comment omits these as an option. That is a definite drawback to the consumer.
National Aeronautics and Space Administration, NASA, has always had an eye on weather, because it's a key item in aeronautics.
I don't want them to drop this particular one of their duties. It could be argued that G. W. Bush DID want them to drop the ball on weather. It's a crazy-eddie thing.
I somehow get the feeling we are missing something here." agreement that could allow Verizon to speed some online content to Internet users".
What we're missing, is this is a no-news item.
Google isn't a single server on the internet, it's multi-homed; the agreement allows Google and their affiliates to use some Verizon sites as though they're another point-of-presence like a Google home node. It means using some extra-cost long lines from Verizon's collection to speed the traffic from a Google server through a Verizon link, so the next step (from the Verizon long line to your home) is a short hop.
The Verizon priority lines wouldn't be carrying all internet traffic, just an agreed-upon subset. So, it's not a 'net neutrality' issue at all, because those Verizon links aren't open to the internet at large.
Other 'net neutrality' issues relate to an ISP that goes behind the customer's back and, by reading his messages, figures out that they can cut their costs by sabotaging the communication. It was a case like the mail carrier burning your mail messages that spurred the FCC to the controversial action; because this kind of fraud is interstate commerce, and because it is hard to detect, some kind of regulatory action is warranted (even if it's not currently supported by FCC's legal powers).
Otherwise, what has happened once, can happen again.
You could throw something together, or just make a nice organized list in a spreadsheet. With appropriate filters it can be quite useable. For only a few hundred/thousand items spreadsheets make nice databases.
I'll second this. There's lots of gobbledygook numbering schemes on transistors and ICs, and having a spreadsheet that reminds you that the TL494 was used in the Casio power supply (or whatever) is very useful. Knowing that it was $0.83 from Mouser in 2003 is also useful. You can't easily replace that info from an internet search, so KEEP IT in the database.
If you buy a bag of 100 house-labeled 7805 voltage regulators, add that house number to your database.
Searching for all diodes, and examining the voltage and current ratings, also fits a spreadsheet listing model. Having lots of parts on one screen helps here
My spreadsheet has columns for part number, known uses, functional type (NPN or NPN power or NMOS...) supplier and price, and of course you can add any kind of info that suits your mental model of importance.
For resistors and capacitors, which DON'T have long part descriptions/part numbers, I generally like the big-array-of-drawers parts organizers. Low-value resistors (under ten ohms) on the top row, 100 ohm to 1000 ohm on the second row, etc.
There's three things you can do: you can donate/sell/discard the item, you can keep it against future need, you can offer it to friends/neighbors etc.
The donate/sell path is useful to the next owner (don't knock WEP, for a lot of folk it's quite sufficient), provided he/she can figure out the configuration procedure. Scribble the configuration address on the case, and fasten the AC adapter securely to the router, if you go this path.
Keeping it, you can turn off the transmitter and DHCP functions and it's a switch (and if your wilderness cabin needs connectivity, it can come out of mothballs with a simple push of the RESET button to be a full router). In case your 802.11N goes down, configuring it to only talk to your three MAC addresses gives you a backup router that isn't likely to be hacked into.
I'd keep it, myself. All the 'extra' functions of newer items are minor frippery compared with the core value, fast-enough wireless connection.
Actually, the specifics that were mentioned in the article indicated that lots of the money was spent to make computer-readable data of the old records (Arlington has over two centuries of records). That might actually be worth what they paid.
We all think of 'making a database' as the center of this kind of problem, but IT ISN'T. Pretty certainly, the monies spent weren't spent on building softwares... and the employees didn't, in their everyday work, feel the need to get every shovel-pusher a computer terminal, they didn't think they WANTED software. They just wanted to 'stay organized'. In the absence of a computerized system, and in comparison with other cemeteries that DID get their records onto database computers, that isn't working. Something has to change.
Army oversight of the cemetery operations was limited, they didn't worry as long as the visitor experience was good. So, naturally, the administrators ignored everything other than the visible tokens of the cemetery operation.
It's interesting how this will increase the adoption of iSCSI storage, yet the original reason to go to iSCSI will be lost since fiber cables will have to be laid.
That seems a tad disingenuous. The real reason for iSCSI was a Microsoft price structure that made a network file service very expensive unless it went in through the 'disk-on-SCSI-bus' back door.
Linux and iSCSI was a way around the high cost of a MS server/client system. None of the Linux-only or Macintosh network systems were so encumbered, and worked quite well without any iSCSI.
I'm not sure about all those busses, but NuBus is an IEEE standard, and all that happened is that some major users dropped it (Apple, Next). It isn't really dead.
MicroChannel had notoriously 'trade secret' specifications, when IBM stopped supporting it it really DID die.
"Triangles" are obviously defined as existing in a plane, not mapped onto the surface of a sphere. However, you do hint at a valid point: can triangles with angles that don't add up to 180 degrees be derived from non-Euclidean geometry? Any Math majors out there?
Recovering ex-math major here; yes, in Lobachevskiian (negative curvature space) geometry, the angle defect (difference from 180 degrees) of a triangle is a measure of its area. Actually, that angle defect IS the area.
The parallel postulate, for Euclidean geometry, is: in a plane, given a line, and a point not on that line, there is one and only one line through the point which is parallel to the given line. For Lobachevskiian space, it ends "...there is more than one line through the point which is parallel to the given line".
Chemically, that "equation" just doesn't balance without an input of energy.
The article DOES explain this, the salt imbalance makes a kind of battery.
It's brilliant! Solar energy concentrates a brine, which then (just as dissimilar metals make a thermocouple) causes current and builds an electric potential when connected via a membrane (impermeable except to Na+ ions) to a less-concentrated brine.
So, the difference in concentration of ions between two channels results in a diffusion from more-concentrated to less-concentrated, OF A CHARGED ION. That means electric current flows, until the charge buildup raises the electric potential enough to stop the diffusion.
The solar input concentrates the brine, the resulting (small) voltage then is electrically applied to the to-be-desalinated channel, and (in the absence of a concentration difference) the electric field causes the ions to leave the to-be-desalinated stream.
Thus, it's a solar-concentration-of-salt that makes the desalination occur. The electricity caused by the diffusion is active ALL NIGHT until the concentration of salt goes down, so the concentrated brine is an effective load-leveling device for the whole plant.
The 'electric input' part of the process is entirely for pumping the brines around, so it can be a small fraction of the brute-force desalination energy requirement. Heck, you could use wave or wind power for that.
Solar collectors for this kind of gizmo are just open-air trays of brine. Can't get any more cost-effective than THAT.
You need to keep in mind that Japan doesn't have all the land needed to deploy hundred of wind turbines.
Thus, it is oddly appropriate that Japan is at the right latitude to send machinery into the jet stream. There, the wind is strong enough, it might not take hundreds of turbines. One needs only enough land area for a good anchor (come to that, sea area would do as well), and a good-sized kite to maneuver into the strongest flow.
First off NO ONE accused him of being a "thief" or a "criminal", that's just his idea of drama.
It might be that the real issue is just the lookup of the correct part number. The only way the dealer has of getting the right part (it might have to be special ordered, and returns aren't profitable) is to type in the serial number or registered owner name in the warranty database application.
There is a lot of stupidity in the world, and treating all repair/accessory/upgrade transactions as though it were a warranty parts issue is a very common kind of stupidity. A bad manufacturer will require jumps through warranty hoops for every transaction. It sounds like Alienware doesn't have a catalog of parts that can be read without lots of form-fill-out including info (a serial number?) that this particular user cannot/will not disclose.
Dude, you are SO far off the mark. There are two incompatible 1G ethernet standards using fiber, one at 1300 um, and one at 850 um. I don't think either uses the connectors that my older 100baseFX converters use.
Fiber is nice. Fiber has its uses. Compatibility isn't its forte.
Don't think it's just copper! Yes, the wire is copper, and that's nice and durable (there's centuries-old roofs in Denmark still in use), BUT the insulation matters. It matters a LOT, actually.
The dissipation and attenuation of the signal is dominated by losses in the polymer used for insulation, and that polymer will age/shrink/get brittle. The particular insulation used is a crosslinked thermoplastic (it's hard to dent with a fingernail), which has very good aging characteristic, but time alone will tell.
Expect 30 years or more from the insulation. If the run is short, attenuation won't be an issue. That leaves only the issue of embrittlement, so always leave LOTS OF SLACK, or the first touch on your aged wiring will pulverize the polymer at the nearest stress point. Protected wire (inside walls or conduit) can probably last centuries.
Any degradation of the copper (green sulfides, black oxides) would surprise me. I've pulled out 70-year installations of copper wire, the metal under the cotton wrap was in very good condition. The varnish insulation was NOT in very good condition.
I've always found crimping solid core to be troublesome.
The RJ45 terminals for solid core are designed differently from the RJ45 terminals for stranded core wire, and the forces for barrel-crimping to solid core wire are also higher than for stranded wire. So, it's a matter of different materials required, or maybe different tooling.
It should be noted that lots of telco punchdown equipment (like the whole 66 series) isn't right for data wiring. The 110 series is the one that comes in CAT5 compliant equipment.
This is VERY interesting work: it has the effect of replacing an old theory (original sin) with a known phenomenon, susceptible to experiment, as the probable cause for Adam and Eve tasting the fruit and getting expelled from Eden.
By Occam's razor, there's no reason to hold on to the "original sin" hypothesis, since the alternative is simpler. I wonder when we will hear from the Vatican on this development?
That is correct, but it overstates the magnitude of the problem. If the PF is 50%, then the utility has to supply (transmit) twice the current. That extra current is not used up by the load, but it does translate into additional transmission losses.
Actually, it DRASTICALLY overstates the problem, in another respect. If there were only a generator, lossy transmission line, and load, the power factor would be important in transmission losses (power not billed to customer). That is true of some large industrial plants, and it's why power factor billing is used.
It's completely unrealistic for a residential community running motors (refrigerator, air conditioning, blowers, etc.) which act as motor-generator-flywheel systems, and which are on the load side of the long transmission line. The excess current due to power factor needn't traverse any longer wire than the space between houses on your block.
The article is deliberately slanted to be provocative.
I love the part about Apple using better SATA drives than normal SATA drives. I bet if you pull the drive it is just a standard OEM drive.
No, it isn't a standard drive; the OEM branch of (Seagate or Hitachi or whoever) makes a new part number for Apple. That isn't the real point, anyhow: many (or most) highend server applications are covered by service contracts (Applecare in the Xserve case) that require the official drives. If you want the service, this is part of how you pay for it. There are benefits, at the organization level, that are worth the cost.
The (very real) differences in server-qualified drives aren't visible, but they are known to the Applecare providers, and that makes the service contract less expensive. Apple, Dell, HP are competitive in the server market, it just may be that customers are getting what they pay for.
These servers are $5k boxes, roughly, and disks don't fail often; the "extra" cost is just noise.
It's intended for satellite communication (so the reflection issue is unimportant). The difference from circular polarization is that the transmission has variable-repeat-time of the polarization cycle, The classic circular polarizing systems use AM or FM modulation, on a fixed rotation rate.
Here, the rotation rate is the modulated signal element.
Slashdot Mirror
>About 2004 ... the professor brought us a bunch of those old purple sheets for a handout
Impressive! The stencil had to be typed on a typewriter or other impact printer,
as well as finding the mimeo stencils, and a supply of alcohol, and the mimeo drum!
It would be harder to use the old dictaphone (the acetate
belt type, with acetate old and brittle since it'd have to be pre-(1970?) manufacture).
The idea of analog computers was to make a box
with operational amplifiers to generate and manipulate signals. This is still alive today, in all circuits that
use operational ampllifiers, but it isn't surrounded by a box and labeled 'computer'.
An online electronics parts supplier lists, under
'operational amplifiers', over thirty thousand items.
Rent _Mutiny on the Bounty_ (1962), and look at the DVD extras menu for a nifty video of the ship they
reproduced for the film. Wood cost makes it
infeasible to do that nowadays, and skills are
probably scarce, too.
The 'pen knife' is just a whittling tool; modern ones
would carve a point in a quill just fine, if you can
find someone who's mastered the technique.
There are quill pens available in modern times,
as museum-reproduction-quality decorative items.
Calligraphy aficionados have the skills and tools to
sharpen 'em, too.
'head clamp' is now more used for brain surgery
than for photography.
As for 'paper tape readers', a recent one was
made to transcribe player-piano tapes to MIDI
format. If you mean the 1" tape for ASR-33
teletypes, that's more easily read with a
scanner and character recognition software...
but it could be argued that such a software
solution is still a 'paper tape reader'.
No.
Actually, I'd say yes here. The previous technology was
a gizmo called a 'coherer' which was basically a bottle of metal
granules. If it stopped rectifying, ya had to shake the bottle.
That technology goes back nearly a century.
>
Copyright is only 200 years old and it was the big publisher who invented the copyright, not the content creators.
The truth is, the publishers weren't paying the content creators anything.
The original intent was to aid the author, so the effect of copyright was to
add a cost burden to the publishers. It didn't HURT the publishers (after
all, they pass those costs to the book buyers), but it was long after the
birth of copyright that the speculative buyers of rights got our legislatures
to extend the period. And to impose confiscatory and punitive unwarranted
judgments on 'violators'.
That's rawther difficult without abandoning computers altogether. Are there any new monitors that don't support HDCP? I thought it was a requirement for the HDMI license, and every TV monitor has an HDCP-capable HDMI port nowadays.
There's a lot of DVI monitors that plug into my HDTV component box,
that aren't HDCP equipped. The 'every TV monitor' comment omits
these as an option. That is a definite drawback to the consumer.
National Aeronautics and Space Administration, NASA,
has always had an eye on weather, because it's a key
item in aeronautics.
I don't want them to drop this particular one of their
duties.
It could be argued that G. W. Bush DID want them to drop
the ball on weather. It's a crazy-eddie thing.
I somehow get the feeling we are missing something here." agreement that could allow Verizon to speed some online content to Internet users".
What we're missing, is this is a no-news item.
Google isn't a single server on the internet, it's multi-homed; the agreement
allows Google and their affiliates to use some Verizon sites as though
they're another point-of-presence like a Google home node. It means
using some extra-cost long lines from Verizon's collection to speed the
traffic from a Google server through a Verizon link, so the next step
(from the Verizon long line to your home) is a short hop.
The Verizon priority lines wouldn't be carrying all internet traffic, just
an agreed-upon subset. So, it's not a 'net neutrality' issue
at all, because those Verizon links aren't open to the internet
at large.
Other 'net neutrality' issues relate to an ISP that goes behind
the customer's back and, by reading his messages, figures
out that they can cut their costs by sabotaging the communication.
It was a case like the mail carrier burning your mail messages
that spurred the FCC to the controversial action; because this
kind of fraud is interstate commerce, and because it is hard
to detect, some kind of regulatory action is warranted (even
if it's not currently supported by FCC's legal powers).
Otherwise, what has happened once, can happen again.
You could throw something together, or just make a nice organized list in a spreadsheet. With appropriate filters it can be quite useable.
For only a few hundred/thousand items spreadsheets make nice databases.
I'll second this. There's lots of gobbledygook numbering schemes on
transistors and ICs, and having a spreadsheet that reminds you
that the TL494 was used in the Casio power supply (or whatever)
is very useful. Knowing that it was $0.83 from Mouser in 2003
is also useful. You can't easily replace that info from an
internet search, so KEEP IT in the database.
If you buy a bag of 100 house-labeled 7805 voltage regulators,
add that house number to your database.
Searching for all diodes, and examining the voltage and current ratings,
also fits a spreadsheet listing model. Having lots of parts on one
screen helps here
My spreadsheet has columns for part number, known uses,
functional type (NPN or NPN power or NMOS...) supplier and price,
and of course you can add any kind of info that suits your mental
model of importance.
For resistors and capacitors, which DON'T have long part
descriptions/part numbers, I generally like the big-array-of-drawers
parts organizers. Low-value resistors (under ten ohms) on
the top row, 100 ohm to 1000 ohm on the second row,
etc.
There's three things you can do: you can donate/sell/discard the
item, you can keep it against future need, you can offer it
to friends/neighbors etc.
The donate/sell path is useful to the next owner (don't knock
WEP, for a lot of folk it's quite sufficient), provided he/she
can figure out the configuration procedure. Scribble the
configuration address on the case, and fasten the AC adapter
securely to the router, if you go this path.
Keeping it, you can turn off the transmitter and DHCP functions
and it's a switch (and if your wilderness cabin needs connectivity,
it can come out of mothballs with a simple push of the RESET
button to be a full router). In case your 802.11N goes down,
configuring it to only talk to your three MAC addresses gives you
a backup router that isn't likely to be hacked into.
I'd keep it, myself. All the 'extra' functions of newer
items are minor frippery compared with the core value, fast-enough
wireless connection.
Actually, the specifics that were mentioned in the article indicated
that lots of the money was spent to make computer-readable data
of the old records (Arlington has over two centuries of records). That
might actually be worth what they paid.
We all think of 'making a database' as the center of this kind
of problem, but IT ISN'T. Pretty certainly, the monies
spent weren't spent on building softwares... and the
employees didn't, in their everyday work, feel the need to
get every shovel-pusher a computer terminal, they didn't
think they WANTED software. They just wanted to 'stay
organized'. In the absence of a computerized system,
and in comparison with other cemeteries that DID get
their records onto database computers, that isn't
working. Something has to change.
Army oversight of the cemetery operations was limited, they
didn't worry as long as the visitor experience was good. So,
naturally, the administrators ignored everything other than
the visible tokens of the cemetery operation.
It's interesting how this will increase the adoption of iSCSI storage, yet the original reason to go to iSCSI will be lost since fiber cables will have to be laid.
That seems a tad disingenuous. The real reason for iSCSI was a
Microsoft price structure that made a network file service very
expensive unless it went in through the 'disk-on-SCSI-bus'
back door.
Linux and iSCSI was a way around the high cost of
a MS server/client system. None of the Linux-only or Macintosh
network systems were so encumbered, and worked
quite well without any iSCSI.
I'm not sure about all those busses, but NuBus is
an IEEE standard, and all that happened is that some
major users dropped it (Apple, Next). It isn't
really dead.
MicroChannel had notoriously 'trade secret'
specifications, when IBM stopped supporting it
it really DID die.
"Triangles" are obviously defined as existing in a plane, not mapped onto the surface of a sphere. However, you do hint at a valid point: can triangles with angles that don't add up to 180 degrees be derived from non-Euclidean geometry? Any Math majors out there?
Recovering ex-math major here; yes, in Lobachevskiian (negative curvature space) geometry, the angle defect (difference from 180 degrees) of a triangle is a measure of its area. Actually, that angle defect IS the area.
The parallel postulate, for Euclidean geometry, is: in a plane, given a line, and a point not on that line, there is one and only one line through the point which is parallel to the given line. For Lobachevskiian
space, it ends "...there is more than one line through the point which is parallel to the given line".
More important than the cost is the question of effectiveness.
In their diagram, they have this schematic in the critical location:
[Salt water]<----(+)----[Brine]----(-)----->[Salt water]
Chemically, that "equation" just doesn't balance without an input of energy.
The article DOES explain this, the salt imbalance makes
a kind of battery.
It's brilliant! Solar energy concentrates a brine, which
then (just as dissimilar metals make a thermocouple)
causes current and builds an electric potential
when connected via a membrane (impermeable
except to Na+ ions) to a less-concentrated brine.
So, the difference in concentration of ions between two
channels results in a diffusion from more-concentrated to
less-concentrated, OF A CHARGED ION. That means
electric current flows, until the charge buildup raises
the electric potential enough to stop the diffusion.
The solar input concentrates the brine, the resulting
(small) voltage then is electrically applied to the to-be-desalinated
channel, and (in the absence of a concentration difference)
the electric field causes the ions to leave the
to-be-desalinated stream.
Thus, it's a solar-concentration-of-salt that makes
the desalination occur. The electricity caused by the
diffusion is active ALL NIGHT until the concentration
of salt goes down, so the concentrated brine is
an effective load-leveling device for the whole plant.
The 'electric input' part of the process is entirely for
pumping the brines around, so it can be a small fraction
of the brute-force desalination energy requirement.
Heck, you could use wave or wind power for that.
Solar collectors for this kind of gizmo are just open-air
trays of brine. Can't get any more cost-effective than
THAT.
You need to keep in mind that Japan doesn't have all the land needed to deploy hundred of wind turbines.
Thus, it is oddly appropriate that Japan is at the right latitude to send
machinery into the jet stream. There, the wind is strong enough, it might not
take hundreds of turbines. One needs only enough land area for a
good anchor (come to that, sea area would do as well), and a good-sized
kite to maneuver into the strongest flow.
This guy is an idiot.
First off NO ONE accused him of being a "thief" or a "criminal", that's just his idea of drama.
It might be that the real issue is just the lookup of the
correct part number. The only way the dealer has
of getting the right part (it might have to be special ordered,
and returns aren't profitable) is to type in the serial
number or registered owner name in the
warranty database application.
There is a lot of stupidity in the world, and treating all
repair/accessory/upgrade transactions as though
it were a warranty parts issue is a very common
kind of stupidity. A bad manufacturer will
require jumps through warranty hoops for every
transaction. It sounds like Alienware doesn't have
a catalog of parts that can be read without lots
of form-fill-out including info (a serial number?)
that this particular user cannot/will not disclose.
Single mode fiber? For compatibility?
Dude, you are SO far off the mark. There are two
incompatible 1G ethernet standards using fiber,
one at 1300 um, and one at 850 um. I don't think
either uses the connectors that my older 100baseFX
converters use.
Fiber is nice. Fiber has its uses. Compatibility isn't
its forte.
Don't think it's just copper! Yes, the wire is copper, and
that's nice and durable (there's centuries-old roofs
in Denmark still in use), BUT the insulation matters.
It matters a LOT, actually.
The dissipation and attenuation of the signal is
dominated by losses in the polymer used for
insulation, and that polymer will age/shrink/get
brittle. The particular insulation used is a
crosslinked thermoplastic (it's hard to dent with
a fingernail), which has very good aging characteristic,
but time alone will tell.
Expect 30 years or more from the insulation. If
the run is short, attenuation won't be an issue.
That leaves only the issue of embrittlement, so
always leave LOTS OF SLACK, or the first touch on
your aged wiring will pulverize the polymer at the
nearest stress point. Protected wire (inside walls
or conduit) can probably last centuries.
Any degradation of the copper (green sulfides, black
oxides) would surprise me. I've pulled out 70-year
installations of copper wire, the metal under the cotton
wrap was in very good condition. The varnish insulation
was NOT in very good condition.
I've always found crimping solid core to be troublesome.
The RJ45 terminals for solid core are designed differently
from the RJ45 terminals for stranded core wire,
and the forces for barrel-crimping to solid core wire
are also higher than for stranded wire. So, it's a
matter of different materials required, or maybe different
tooling.
It should be noted that lots of telco punchdown equipment
(like the whole 66 series) isn't right for data wiring. The
110 series is the one that comes in CAT5 compliant
equipment.
This is VERY interesting work: it has the effect
of replacing an old theory (original sin) with a
known phenomenon, susceptible to experiment,
as the probable cause for Adam and Eve tasting
the fruit and getting expelled from Eden.
By Occam's razor, there's no reason to hold
on to the "original sin" hypothesis, since the
alternative is simpler. I wonder when we will
hear from the Vatican on this development?
That is correct, but it overstates the magnitude of the problem. If the PF is 50%, then the utility has to supply (transmit) twice the current. That extra current is not used up by the load, but it does translate into additional transmission losses.
Actually, it DRASTICALLY overstates the problem, in another
respect. If there were only a generator, lossy transmission
line, and load, the power factor would be important in
transmission losses (power not billed to customer).
That is true of some large industrial plants, and it's why
power factor billing is used.
It's completely unrealistic for a residential community
running motors (refrigerator, air conditioning, blowers, etc.)
which act as motor-generator-flywheel systems, and
which are on the load side of the long transmission line.
The excess current due to power factor needn't traverse
any longer wire than the space between houses
on your block.
The article is deliberately slanted to be provocative.
I love the part about Apple using better SATA drives than normal SATA drives. I bet if you pull the drive it is just a standard OEM drive.
No, it isn't a standard drive; the OEM branch of (Seagate
or Hitachi or whoever) makes a new part number for
Apple. That isn't the real point, anyhow: many
(or most) highend server applications are covered
by service contracts (Applecare in the Xserve case)
that require the official drives. If you want the service,
this is part of how you pay for it. There are benefits,
at the organization level, that are worth the cost.
The (very real) differences in server-qualified drives
aren't visible, but they are known to the Applecare
providers, and that makes the service contract
less expensive. Apple, Dell, HP are competitive
in the server market, it just may be that customers
are getting what they pay for.
These servers are $5k boxes, roughly, and disks don't
fail often; the "extra" cost is just noise.
It's intended for satellite communication (so the reflection
issue is unimportant). The difference from circular polarization
is that the transmission has variable-repeat-time
of the polarization cycle, The classic circular polarizing
systems use AM or FM modulation, on a fixed rotation rate.
Here, the rotation rate is the modulated signal element.