High electrical conductivity and high thermal conductivity do tend to run together, but exceptions are numerous, including sapphire which (in some range of temperatures) is actually a better heat conductor than copper, yet is an hard insulator. A slightly worse heat conductor, alumina, is also a hard insulator.
There should be two standards: one for binary packaging RPM style the other for compile-script based packaging ebuild style. These two serve different purposes and should not be lumped together.
Do you know anything about TERENA finances? Basically, I wonder if this sourceforge is a more reliable long-term site than sf.net hosted by a commercial company with moderately shaky finances.
This is what he means. On the best mills, if you are careful about temperature of the surface being worked and if you machine slowly with sharp bits you can get 0.0001" precision, i.e. 2.5 micron. Going beyond that requires a polishing jig. It is not uncommon for people to give specs to the machinist with the precision of a few micron but that HAS to be dimensions at some given temperature. At those precisions, thermal expansion matters.
Kleinbauer claims 3 mil accuracy whereas pcb lines can be 5 mil wide or even smaller. I am saying this because I had a design which needed to have ultra thin traces and have them straight (low capacitance and inductance were key). This method just isn't accurate enough for the most demanding PCBs.
176thousand * 25 dollars a month = 4.4Mil _a month_ So 2003 loss is circa $50 Mil. That's beyond most stock option even today. It's getting close to serious money.
I do not see any reference to any material actually being mentioned in the APL article. In fact the APL article is merely a simulation on a computer for some idealized transmission line. Does anyone have a reference to actual experimental evidence, assuming it exists?
1. The little girl merely does not eat with her parents. She never abandons them. 2. It seemed to me that Haku was assisting Yubaba because she took his name from him. Certainly he was released by remembering who he was. 3. Repents? Did we watch the same movie? She lost a wager/was outsmarted and decided to lose gracefully. So she is a classy villain, bite me. 4. Methinks this is just an insult, not an argument. 5.
As for Disney movies, I didn't grow up in the USA so I was certainly not brainwashed by them. In fact I have not watched a complete Disney movie in my life because they either put me to sleep or cause revulsion. This movie by contrast was just bland, not entirely without merit but also nothing to write home about. After having this discussion I think that my standards are just MUCH higher than yours.
Don't remember character names but here goes: 1. The little girl (main character) is clearly overall good, as are here parents (minor characters though). 2. The guy who helps her (river spirit or some such) is clearly good, despite a bad spell on him. Neither of the two main characters deals with any substantial moral issues or questions. 3. The old broad running the bathhouse is clearly bad. 4. The many-armed geezer operating bath machinery is your typical rough on the outside good on the inside character. Nothing new, nothing to think about here. 5. The youngish lady helping and overseeing the girl is good. No wrenching moral dilemmas here either. Shall I continue? Name one character (better yet describe one character) who is not somehow cliche.
No juicy details (IMHO). Simple predicatble plot, one dimensional characters which are either good or bad but not much in between, and somehow I found it really hard to care about the characters. They are somehow very artificial. I suppose if you can't live a day without anime, then this is OK but it is IMHO sad that this won best animated picture.
The major reason CPUs keep getting hotter as they go faster is because as feature size shrinks, the leakage currents go up. However there is a lot of research on controlling leakage. It's just that up to a point so long as people don't complain too much CPU makers can ignore leakage and require ever bigger heatsinks. But now at around 90 nm, leakage is becoming the bottleneck not only for heat production but even for signal propagation, so I expect to see tons of research turn into tons of practical solutions, like double gated FETs and the like. So in fact heat dissipation will probably saturate long before IC industry starts to stagnate.
Apodizing is lossy, so it remains to be seen if this technique will satisfy the purists in the photography as art community. I do see how it could be useful in many other places, especially in research microscopy. I wonder if electron optics (SEM, TEM etc) can make use of these ideas.
Thanks for the links. It seems this system has a downside, namely it introduces its own artefacts, similar to ghosting. Look at http://www.colorado.edu/isl/intimages/focusinv.htm l and this will become clear. I wonder if this is inherent in their technique or just the imperfections of "1.0 release" of their tech.
BeFs did this and AFAIK Reiser is working on this for his FS. And besides, MS has promised this for their future Windows, so you know it will be copied and made mainstream eventually.
This is useful in many ways. If you know how to configure windows devices thru control panel then you can do the same in Linux without learning about and mucking with other options. You need never know that there is such a thing as/dev but you can still be productive. Many ISPs give directions on how to configure internet settings using windows front-end, down to which button to click. Having the exact same buttons means someone can get on the net from Linux without learning about IP. In short, if the goal is to avoid learning while staying productive this is useful.
Your post got me to wondering: with dynamic binary recompilation you are taking code written in machine language and compiling it into another machine's language. This is not much different from taking code in a human-readable language like C and compiling that. So why is there any speed loss with this form of emulation, assuming you don't do JIT compile things but precompile all your binaries? And why do current emulators still slow down with disk access?
You are arguing my points. If you move jobs to third world countries you can expect lower ethics standards throughout. So yes, low level workers will also try to steal. However, this is easier to eversee and control and also each worker has access to less and can steal less. The higher up the food chain you go the more one can steal. Indeed, the CEO is just the highest of the food chain and is not the "owner" and the shareholders will pick the CEO who is not likely to steal. However, if the latest scandals in the US are any indication, CEO ethics is questionable here, so this might precipitate management outsourcing if it continues.
Generally outsourcing positions higher on the management scale is harder. This is because the higher ups have more access to resources and can steal/mismanage more. So for instance in Russia their business elite has not reinvested in factories for a while and they now face crumbling infrastructure. The few top managers are responsible. Basically the fear is that otusourcing top management to someone from a banana republic you turn your company into a banana republic. The other aspect of it is that top management is a small group of people often acquainted with each other, so they don't need a labor union to protect their inetrests in an organized fashion. That said, as production and support shift to thrid world countries, management will follow but much slower. This will accelerate drastically if the US ceases to be the largest consumer market. Then companies will shift headquarters en mass.
What we need is a law making sending spam a felony then establish a separate anti-spam police force to hunt down spammers. I think a mandatory 1 years jail sentence (max security) for 500+ spam emails would quickly teach spammers what the deal is. And frankly, we would also need to actively seek extradition of foreign spammers, although I would also support the use of remote controlled weapons to kill spammers where extradition is impossible. Lastly, the law should provide for heavy fines for corporations which send unsolicited e-mail, with something like a three strikes law: the first time you pay a fine, the next time you pay THE FINE, the third time you are out of business and all your assets go to the state/get sold at an auction. If the spamming by a corporation happens more than once under the same management then the management should go to jail for 1 year (mandatory minimum).
Re:How does a website spend $80mln?
on
Salon Asks for Help
·
· Score: 3, Interesting
Why should I support Salon? If the poster is right and they have figured out a half-promising ad model then they should declare Chapter 11. Not Chapter 7, just Chapter 11. Get out of onerous contracts like that lease that suffocates them now and reorganize, maybe even move to a cheaper city or burb. Why should I pay for their lack of finacial advice?
The blame rests squarely on the labels. IMHO. IMNSHO. These guys take acts they think will make it big, then take their cut for the business/legal/sales stuff. And as a customer, I would appreciate if someone picked the best there is and filtered out crap. I would be willing to pay for that service. That's the fair business model for music labels. But they use their monopoly (ok collusion) to set their "cut" to be nearly everything. So for artists the failure is indeed the label monopoly. For the consumers, the failure is that the labels don't filter out crap. They seem to be willing to pick any sh*t that's similar to what sells already and promote that. Commercial noise goes up, quality and innovation goes down. This lack of taste and artistic appreciation fails the artists too, since the labels _have_ to get an even bigger cut because so many of their acts fail miserably. And this artistic ineptness comes from the fact that the same guys run the show year in, year out. Any industry will run out of ideas without fresh blood. So the failure is strictly that of the labels, no matter how you look at it.
Slashdot Mirror
High electrical conductivity and high thermal
conductivity do tend to run together, but exceptions
are numerous, including sapphire which (in some
range of temperatures) is actually a better heat
conductor than copper, yet is an hard insulator.
A slightly worse heat conductor, alumina, is
also a hard insulator.
There should be two standards:
one for binary packaging RPM style
the other for compile-script based packaging ebuild style.
These two serve different purposes and should
not be lumped together.
Do you know anything about TERENA finances?
Basically, I wonder if this sourceforge is a
more reliable long-term site than sf.net hosted
by a commercial company with moderately shaky
finances.
Just idle wondering [see subject].
This is what he means. On the best mills, if you
are careful about temperature of the surface being
worked and if you machine slowly with sharp bits
you can get 0.0001" precision, i.e. 2.5 micron.
Going beyond that requires a polishing jig.
It is not uncommon for people to give specs to the
machinist with the precision of a few micron but
that HAS to be dimensions at some given temperature.
At those precisions, thermal expansion matters.
Kleinbauer claims 3 mil accuracy whereas pcb lines
can be 5 mil wide or even smaller. I am saying this
because I had a design which needed to have ultra
thin traces and have them straight (low capacitance
and inductance were key). This method just isn't
accurate enough for the most demanding PCBs.
176thousand * 25 dollars a month = 4.4Mil _a month_
So 2003 loss is circa $50 Mil. That's beyond most
stock option even today. It's getting close to
serious money.
I do not see any reference to any material
actually being mentioned in the APL article.
In fact the APL article is merely a simulation
on a computer for some idealized transmission
line. Does anyone have a reference to actual
experimental evidence, assuming it exists?
1. The little girl merely does not eat with her
parents. She never abandons them.
2. It seemed to me that Haku was assisting Yubaba
because she took his name from him. Certainly
he was released by remembering who he was.
3. Repents? Did we watch the same movie? She lost
a wager/was outsmarted and decided to lose
gracefully. So she is a classy villain, bite me.
4. Methinks this is just an insult, not an argument.
5.
As for Disney movies, I didn't grow up in the USA so
I was certainly not brainwashed by them. In fact
I have not watched a complete Disney movie in my
life because they either put me to sleep or cause
revulsion. This movie by contrast was just bland,
not entirely without merit but also nothing to
write home about. After having this discussion
I think that my standards are just MUCH higher
than yours.
Two words: YURI NORSTEIN.
Look it up.
Don't remember character names but here goes:
1. The little girl (main character) is clearly
overall good, as are here parents (minor
characters though).
2. The guy who helps her (river spirit or some
such) is clearly good, despite a bad spell on
him.
Neither of the two main characters deals with any
substantial moral issues or questions.
3. The old broad running the bathhouse is clearly
bad.
4. The many-armed geezer operating bath machinery
is your typical rough on the outside good on the
inside character. Nothing new, nothing to think
about here.
5. The youngish lady helping and overseeing the
girl is good. No wrenching moral dilemmas here either.
Shall I continue? Name one character (better yet
describe one character) who is not somehow
cliche.
No juicy details (IMHO). Simple predicatble plot,
one dimensional characters which are either good
or bad but not much in between, and somehow I found
it really hard to care about the characters. They
are somehow very artificial. I suppose if you can't
live a day without anime, then this is OK but it
is IMHO sad that this won best animated picture.
The major reason CPUs keep getting hotter as they
go faster is because as feature size shrinks, the
leakage currents go up. However there is a lot of
research on controlling leakage. It's just that
up to a point so long as people don't complain
too much CPU makers can ignore leakage and require
ever bigger heatsinks. But now at around 90 nm,
leakage is becoming the bottleneck not only for
heat production but even for signal propagation,
so I expect to see tons of research turn into
tons of practical solutions, like double gated
FETs and the like. So in fact heat dissipation
will probably saturate long before IC industry
starts to stagnate.
Apodizing is lossy, so it remains to be seen if
this technique will satisfy the purists in the
photography as art community. I do see how it
could be useful in many other places, especially
in research microscopy. I wonder if electron
optics (SEM, TEM etc) can make use of these ideas.
Thanks for the links. It seems this system has am l
downside, namely it introduces its own artefacts,
similar to ghosting. Look at http://www.colorado.edu/isl/intimages/focusinv.ht
and this will become clear. I wonder if this is
inherent in their technique or just the imperfections
of "1.0 release" of their tech.
BeFs did this and AFAIK Reiser is working on this
for his FS. And besides, MS has promised this for
their future Windows, so you know it will be
copied and made mainstream eventually.
This is useful in many ways. If you know how to /dev
configure windows devices thru control panel
then you can do the same in Linux without learning
about and mucking with other options. You need
never know that there is such a thing as
but you can still be productive. Many ISPs give
directions on how to configure internet settings
using windows front-end, down to which button to
click. Having the exact same buttons means someone
can get on the net from Linux without learning
about IP. In short, if the goal is to avoid
learning while staying productive this is useful.
His link actually has the meaning as "ho-hum"
which is pretty close to your first meaning.
Your post got me to wondering: with dynamic binary
recompilation you are taking code written in
machine language and compiling it into another
machine's language. This is not much different
from taking code in a human-readable language
like C and compiling that. So why is there any
speed loss with this form of emulation, assuming
you don't do JIT compile things but precompile
all your binaries? And why do current emulators
still slow down with disk access?
You are arguing my points. If you move jobs to
third world countries you can expect lower
ethics standards throughout. So yes, low level
workers will also try to steal. However, this
is easier to eversee and control and also each
worker has access to less and can steal less.
The higher up the food chain you go the more
one can steal. Indeed, the CEO is just the highest
of the food chain and is not the "owner" and
the shareholders will pick the CEO who is not
likely to steal. However, if the latest scandals
in the US are any indication, CEO ethics is
questionable here, so this might precipitate
management outsourcing if it continues.
Generally outsourcing positions higher on
the management scale is harder. This is because
the higher ups have more access to resources
and can steal/mismanage more. So for instance
in Russia their business elite has not
reinvested in factories for a while and they now
face crumbling infrastructure. The few top
managers are responsible. Basically the fear is
that otusourcing top management to someone from
a banana republic you turn your company into a
banana republic.
The other aspect of it is that top management is
a small group of people often acquainted with
each other, so they don't need a labor union to
protect their inetrests in an organized fashion.
That said, as production and support shift to
thrid world countries, management will follow but
much slower. This will accelerate drastically if
the US ceases to be the largest consumer market.
Then companies will shift headquarters en mass.
What we need is a law making sending spam a felony
then establish a separate anti-spam police force
to hunt down spammers. I think a mandatory
1 years jail sentence (max security) for 500+
spam emails would quickly teach spammers what
the deal is. And frankly, we would also need
to actively seek extradition of foreign spammers,
although I would also support the use of remote
controlled weapons to kill spammers where
extradition is impossible. Lastly, the law should
provide for heavy fines for corporations which
send unsolicited e-mail, with something like a
three strikes law: the first time you pay a fine,
the next time you pay THE FINE, the third time
you are out of business and all your assets go
to the state/get sold at an auction. If the spamming
by a corporation happens more than once under
the same management then the management should
go to jail for 1 year (mandatory minimum).
Why should I support Salon? If the poster is right
and they have figured out a half-promising ad model
then they should declare Chapter 11. Not Chapter 7,
just Chapter 11. Get out of onerous contracts like
that lease that suffocates them now and reorganize,
maybe even move to a cheaper city or burb. Why
should I pay for their lack of finacial advice?
The blame rests squarely on the labels. IMHO. IMNSHO.
These guys take acts they think will make it big,
then take their cut for the business/legal/sales
stuff. And as a customer, I would appreciate if
someone picked the best there is and filtered out
crap. I would be willing to pay for that service.
That's the fair business model for music labels.
But they use their monopoly (ok collusion) to set
their "cut" to be nearly everything. So for artists
the failure is indeed the label monopoly. For the
consumers, the failure is that the labels don't
filter out crap. They seem to be willing to pick
any sh*t that's similar to what sells already and
promote that. Commercial noise goes up, quality
and innovation goes down. This lack of taste and
artistic appreciation fails the artists too, since
the labels _have_ to get an even bigger cut
because so many of their acts fail miserably.
And this artistic ineptness comes from the fact
that the same guys run the show year in, year out.
Any industry will run out of ideas without fresh blood.
So the failure is strictly that of the labels, no
matter how you look at it.
Ok, fine, but then why are their stills so
crappy? I mean look at those color blotches.
Is there any reason for that or is that the
best HDTV can get?