..., unlike electrons. There goes your dream of an all-optical transparent switch. You can still control light with electrical signals, or reflect it with a MEMS micromirror, and I guess you can achieve total bandwidth in 100 terabit range right now -- what is problematic (especially with MEMS approach) is packets becoming too long: terabit with 1ms MEMS switch time -> gigabit packet.
Anyway, someone who really needs small packet switching at fiber speed in 100 terabits/sec might as well go with superconductor approaches (you do not expect such a router to fit into a single pizza box, right? You can integrate cooler in a rack easily) -- see, e.g., http://www.google.com/patents?vid=USPAT6917537&id= kkIVAAAAEBAJ&dq=bunyk (correlation between inventor's name and my/.nickname is not entirely accidental:) ).
There is no such thing as "speed of light inside a metal", because, you know, metals are pretty much non-transparent, as in, light does not propagate through them! If in doubt, find metal object and try to see through it!:)
"Speed of electricity" that GP was referring to was a cute (if not entirely scientfic) way to trick the reader into thinking that "speed of ligth" and "speed of EM wave" in given medium are somehow different -- nope, they are not.
Speed of E/M wave in SiO2 insulator between two sides of copper microstrip line and of light in fiber, made of the same SiO2 are exactly the same.
As to making fibers out of something with \epsilon about 1 (close to that of vacuum/air) -- you can not! Or, at least, it will stop working like a fiber you can route around, read some day about why optical fiber works.
Scientists are no less susceptible to bribes and threats, and no less prone to intellectual whoredom than regular people. should start with "modern publically/govt. funded scientists". Modern "science", since just before WWII (if you have to put a threshold somewhere) was too dependant on government grants, which (surprize!) were funnelled to things having military/national pride/ national "happiness" applications (in that order).
Before that time the great minds who called themselves "scientists" were mostly financially independent, if not outright wealthy -- thus, much more independent of public opinion and public funding.
Scientists working for private corporations (old Bell Labs or IBM T.J.. Watson center, anyone?) tend to have a bit less of this whoredom, I hope.
... there is a fine difference between heavily promoted (at the time) product from a mojor supplier, which might have got some custom applications happily running on it -- and almost totally forgotten and obscure product (from the same supplier, who, it seems, tried to downplay it as much as it could, for somewhat obvious reasons). I was just surprized when I read the list, I almost mis-read it as Minix!:-)
And I did like your tagline too... "Shine on, you, crazy diamond!"
... Thanks for clarifying the distinction between non-Windows and non-Microsoft! But please tell me, which of a very large, high volume (1500+ stores) retail chain, an international, premier law firm, and one of the largest US based banks actually uses Xenix now? Or was it way back then?
Let's say the machines are designed to operate at a 0.5% error rate and that we all agree that's a tolerable number. ... on an airplane with avionics good to 0.5%?
(As in, like dozens planes a day just dropping dead -- I bet there are more than 4800 commercial airplanes flying on every given day).
Hmm, I was always (back in Soviet Russia, really!) taught that these 6 billion people look forward every day towards some bread... Or rice... (In rare years we were not at war with -- equally "Communist" -- China).
What makes you think all of The World's Underprivileged People are going to go after the bluefish tuna tomorrow, if I may ask?
you do sound as if you think the guy spent five years of his life to "make a tool" -- as far as I can read into the story, it is a bit of regex hacking in URL name to replace one name with another! If this is going to escalate any further, expect t-shirts with 3-line Perl scripts and taglines "this helps terrorists!?"
"Tools" (in your definition) are too easy to make...
... and of course everyone is thinking about sending frozen bodies of austronauts to some remote star, but in reality it should be useful when a soldier is hit in, say, some remote desert and it takes 15 hours to airlift him to more advanced places where an operation can be performed. Source of HS4 can be portable and if his metabolism is three times slower it would feel like an 5 hour flight.
Also, what happened to "development workstation runs Fedora, production server runs RHEL" mindset? One of most basic ideas of OSS in business environment is that you do not have to pay for _per_seat_ licenses, only _per_deployed_server_ (and even that only if you want 24/7 support and brand name to rely upon)!
And, of course, it is going to cost you somewhat more per server license if vendor in question gives you unlimited workstation licenses "for free" and still somehow manages to compete with fully closed source shops.
If that guy's programmers are so stupid^H^H^H^H^H^H "detached from reality";-) that they need RHEL level of support (or, RHEL stability instead of FC's cutting edge, "what gonna make it to RHEL when app is deployed" attitudes) on their development boxes, I guess that the original decision to go OSS was made by some managers mostly because they think it is "cool"...
Paul B.
P.S. FC vs. RHEL comparison above is just an example; YMMV; I do work for a start-up where some very smart people have chosen gentoo and slack on their main workstations, maybe because availability of that choice motivates them -- but if your app heavily depends on some feature which is present in RHEL but not available on, say, Gentoo I think you have bigger problems that license cost!;-)
IANAL, never played one on TV, nor even dated one, but I think that what you are describing is a cartel, "prohibited by antitrust laws in most countries"... Not that I would not suggest to you, nor to the original poster, organizing in this particular way with your competitors, but maybe it is not entirely wise to post your intentions on/. ?:)
Maple is good, Mathematica is good, what the heck, good old Common LISP is good (and knows about ratios) -- and Forth did not even have floating point numbers, proclaiming in its main defining book something along the lines of "real men do not use floating point" -- it was before political correctness times, when Forth was actually used to control robotics/avionics, including some Space Shuttle programs (see http://forth.gsfc.nasa.gov/ for mission list), and, of course, before it became PostScript!:-)
"Pluton" IS Pluto, transliterated... Uran, Neptun, Pluton are three last planets in Russian, or whatever they are called now. I had to stop reading and give myself some time to parse (in lexical, not synctactic way!:-) ) the announcement to realize that what they are talking about is just a "pluto-ish" object!
Congratulations! You've just invented the earliest attempt to build superconductive electronics (SCE), dating back to late 50s, or something -- quantrons these gates were called (I think, way before my time...), controllable by neighboring magnetic fields and using SC/non-SC to distinguish between ``1'' and ``0''.
In early 60s Brian Josephson discovered interesting properties of what became to be known a Josephson junction, a tunnel barrier between two SC which would be superconducting or normal depending on the current one attempts to pass through it. The next SCE family was called "latching logic" -- switch some on, some off, and they remain in this state for as long as you are providing the bias current. Basis for 80s IBM SCE supercomputer project.
(R)SFQ is based on a similar kind of sufficiently shunted Josephson junction, which would not stay in resistive state, but resulting currents (or absence of them) in SC loops can be interpreted as ones or zeros. According to some accounts, ``R'' stands for "Russian", though originally it was "resistive", then "rapid".
Finally, when you use JJs in a SC loop and use the fact that you (nor Nature) have no clue which direction the current flows, you get superconducting quantum computing...
Yes, I think that 110GHz was Ft (everyone would be reporting THAT for a transistor, right?) -- I am wondering what a ring oscillator frequency would be. And I did spend a year of my life not too long ago messing with top-of-the-line InP HBTs, your 500GHz figure for experimental stuff is rather accurate (one can get above 300GHz for "production" devices).
Hey, it is "standard" "silicon" process, but they compete with other GaAs/InP/SiGe bipolar transistors, not yoru garden variety CMOS FETs -- and for other technologies Ft of more than 100GHz is not unheard of. Neat trick, and you will see them in your cellphone front-end, maybe soon, but do not hold your breath for 20GHz processors (and if someone makes 'em, please *do not hold them with your bare hands*! -- they gonna be HOT!);-)
... with a nick of cookiepuss must be the hight of my/. experience, but still -- you get more than one try to answer the "security" question, and if for all of them the secret answer is "Red" you have advantage over the bad guy who might try to work on actually guessing the real answer.
Slashdot Mirror
..., unlike electrons. There goes your dream of an all-optical transparent switch. You can still control light with electrical signals, or reflect it with a MEMS micromirror, and I guess you can achieve total bandwidth in 100 terabit range right now -- what is problematic (especially with MEMS approach) is packets becoming too long: terabit with 1ms MEMS switch time -> gigabit packet.
= kkIVAAAAEBAJ&dq=bunyk /.nickname is not entirely accidental :) ).
Anyway, someone who really needs small packet switching at fiber speed in 100 terabits/sec might as well go with superconductor approaches (you do not expect such a router to fit into a single pizza box, right? You can integrate cooler in a rack easily) -- see, e.g., http://www.google.com/patents?vid=USPAT6917537&id
(correlation between inventor's name and my
Well, maybe one of these days...
Paul B.
There is no such thing as "speed of light inside a metal", because, you know, metals are pretty much non-transparent, as in, light does not propagate through them! If in doubt, find metal object and try to see through it! :)
"Speed of electricity" that GP was referring to was a cute (if not entirely scientfic) way to trick the reader into thinking that "speed of ligth" and "speed of EM wave" in given medium are somehow different -- nope, they are not.
Speed of E/M wave in SiO2 insulator between two sides of copper microstrip line and of light in fiber, made of the same SiO2 are exactly the same.
As to making fibers out of something with \epsilon about 1 (close to that of vacuum/air) -- you can not! Or, at least, it will stop working like a fiber you can route around, read some day about why optical fiber works.
Paul B.
Scientists are no less susceptible to bribes and threats, and no less prone to intellectual whoredom than regular people. should start with "modern publically/govt. funded scientists". Modern "science", since just before WWII (if you have to put a threshold somewhere) was too dependant on government grants, which (surprize!) were funnelled to things having military/national pride/ national "happiness" applications (in that order).
:)
Before that time the great minds who called themselves "scientists" were mostly financially independent, if not outright wealthy -- thus, much more independent of public opinion and public funding.
Scientists working for private corporations (old Bell Labs or IBM T.J.. Watson center, anyone?) tend to have a bit less of this whoredom, I hope.
But whining academics do get on my nerves!
Paul B.
... there is a fine difference between heavily promoted (at the time) product from a mojor supplier, which might have got some custom applications happily running on it -- and almost totally forgotten and obscure product (from the same supplier, who, it seems, tried to downplay it as much as it could, for somewhat obvious reasons). I was just surprized when I read the list, I almost mis-read it as Minix! :-)
And I did like your tagline too... "Shine on, you, crazy diamond!"
Paul B.
... Thanks for clarifying the distinction between non-Windows and non-Microsoft! But please tell me, which of a very large, high volume (1500+ stores) retail chain, an international, premier law firm, and one of the largest US based banks actually uses Xenix now? Or was it way back then?
Paul B.
Let's say the machines are designed to operate at a 0.5% error rate and that we all agree that's a tolerable number. ... on an airplane with avionics good to 0.5%?
(As in, like dozens planes a day just dropping dead -- I bet there are more than 4800 commercial airplanes flying on every given day).
Paul B.
Hey dude!
/.tters from Vancouver, of all places??? :)
Are there really any
Just kidding,
Paul B.
%^#%$@&($&@%^!!!
Paul B.
... the appetites of all 6 billion people.
Hmm, I was always (back in Soviet Russia, really!) taught that these 6 billion people look forward every day towards some bread... Or rice... (In rare years we were not at war with -- equally "Communist" -- China).
What makes you think all of The World's Underprivileged People are going to go after the bluefish tuna tomorrow, if I may ask?
Paul B.
don't make a tool to DO it
you do sound as if you think the guy spent five years of his life to "make a tool" -- as far as I can read into the story, it is a bit of regex hacking in URL name to replace one name with another!
If this is going to escalate any further, expect t-shirts with 3-line Perl scripts and taglines "this helps terrorists!?"
"Tools" (in your definition) are too easy to make...
Paul
... and of course everyone is thinking about sending frozen bodies of austronauts to some remote star, but in reality it should be useful when a soldier is hit in, say, some remote desert and it takes 15 hours to airlift him to more advanced places where an operation can be performed. Source of HS4 can be portable and if his metabolism is three times slower it would feel like an 5 hour flight.
Paul B.
http://www.reason.com/rb/rb100606.shtml
Paul B.
Agreed!
;-) that they need RHEL level of support (or, RHEL stability instead of FC's cutting edge, "what gonna make it to RHEL when app is deployed" attitudes) on their development boxes, I guess that the original decision to go OSS was made by some managers mostly because they think it is "cool"...
;-)
Also, what happened to "development workstation runs Fedora, production server runs RHEL" mindset? One of most basic ideas of OSS in business environment is that you do not have to pay for _per_seat_ licenses, only _per_deployed_server_ (and even that only if you want 24/7 support and brand name to rely upon)!
And, of course, it is going to cost you somewhat more per server license if vendor in question gives you unlimited workstation licenses "for free" and still somehow manages to compete with fully closed source shops.
If that guy's programmers are so stupid^H^H^H^H^H^H "detached from reality"
Paul B.
P.S. FC vs. RHEL comparison above is just an example; YMMV; I do work for a start-up where some very smart people have chosen gentoo and slack on their main workstations, maybe because availability of that choice motivates them -- but if your app heavily depends on some feature which is present in RHEL but not available on, say, Gentoo I think you have bigger problems that license cost!
IANAL, never played one on TV, nor even dated one, but I think that what you are describing is a cartel, "prohibited by antitrust laws in most countries"... Not that I would not suggest to you, nor to the original poster, organizing in this particular way with your competitors, but maybe it is not entirely wise to post your intentions on /. ? :)
Paul B.
Maple is pretty good for HUGE numbers.
:-)
Maple is good, Mathematica is good, what the heck, good old Common LISP is good (and knows about ratios) -- and Forth did not even have floating point numbers, proclaiming in its main defining book something along the lines of "real men do not use floating point" -- it was before political correctness times, when Forth was actually used to control robotics/avionics, including some Space Shuttle programs (see http://forth.gsfc.nasa.gov/ for mission list), and, of course, before it became PostScript!
Paul B.
"Pluton" IS Pluto, transliterated... Uran, Neptun, Pluton are three last planets in Russian, or whatever they are called now. I had to stop reading and give myself some time to parse (in lexical, not synctactic way! :-) ) the announcement to realize that what they are talking about is just a "pluto-ish" object!
Paul B.
Congratulations! You've just invented the earliest attempt to build superconductive electronics (SCE), dating back to late 50s, or something -- quantrons these gates were called (I think, way before my time...), controllable by neighboring magnetic fields and using SC/non-SC to distinguish between ``1'' and ``0''.
In early 60s Brian Josephson discovered interesting properties of what became to be known a Josephson junction, a tunnel barrier between two SC which would be superconducting or normal depending on the current one attempts to pass through it. The next SCE family was called "latching logic" -- switch some on, some off, and they remain in this state for as long as you are providing the bias current. Basis for 80s IBM SCE supercomputer project.
(R)SFQ is based on a similar kind of sufficiently shunted Josephson junction, which would not stay in resistive state, but resulting currents (or absence of them) in SC loops can be interpreted as ones or zeros. According to some accounts, ``R'' stands for "Russian", though originally it was "resistive", then "rapid".
Finally, when you use JJs in a SC loop and use the fact that you (nor Nature) have no clue which direction the current flows, you get superconducting quantum computing...
Paul B.
Check out the sidebar under "Published Stuff", especially this link... Next objection, please...
Paul B.
... I can say definitely, YES!
Paul B.
Yes, I think that 110GHz was Ft (everyone would be reporting THAT for a transistor, right?) -- I am wondering what a ring oscillator frequency would be. And I did spend a year of my life not too long ago messing with top-of-the-line InP HBTs, your 500GHz figure for experimental stuff is rather accurate (one can get above 300GHz for "production" devices).
Paul B.
\
Hey, it is "standard" "silicon" process, but they compete with other GaAs/InP/SiGe bipolar transistors, not yoru garden variety CMOS FETs -- and for other technologies Ft of more than 100GHz is not unheard of. Neat trick, and you will see them in your cellphone front-end, maybe soon, but do not hold your breath for 20GHz processors (and if someone makes 'em, please *do not hold them with your bare hands*! -- they gonna be HOT!) ;-)
Paul B.
"NOBODY expects the Spanish Inquisition!" ;-)
Paul B.
... of all! -- National Security Agency, or NSA (for short) -- really, the largest employer of mathematicians of all...
Paul B.
... with a nick of cookiepuss must be the hight of my /. experience, but still -- you get more than one try to answer the "security" question, and if for all of them the secret answer is "Red" you have advantage over the bad guy who might try to work on actually guessing the real answer.
Paul B.
Your mother maiden name? / your city of birth,
Your pet's name? / your GF nickname,
Your pet? / Ultraviolet
And so on...
Paul B.