IANAL, and it has been four years since I took a "Space Treaties and Legislation" class... Under The Outer Space Treaty (short title), no country can stake claim on an extraterrestial body (including the Moon) -- says nothing about private organizations. The Moon Treaty (short title), which only a handful of countries have signed (none of the major space powers have signed it), is based on The Law of the Sea Convention and is more muddy regarding private ownership rights. As to whether the U.S. could currently claim ownership, territorial waters traditionally had been defined by the range of the most advanced artillary of the time (weapons of the last few decades has made this absurd, which is why there is now an arbitrary distance used). Even if the U.S. claimed ownership of the Moon, it is currently in no position to defend that claim. Christopher A. Bohn
I, for one, am glad to see that civilized discourse resolved the matter, rather than this becoming the (apparently) much-anticipated courtroom test of the GPL. Christopher A. Bohn
They made the announcement back in March. I remember talking to one of my professors about the research & academic opportunities it presented. http://slashdot.org/articles/99/03/02/1133216.shtml Christopher A. Bohn
Perhaps that's why the investigation was launched. But I assure you that not only do I remember the events (well, at least the tripping), I had never forgotten them, so they are not "recovered memories." Christopher A. Bohn
Early on, I displayed some symptoms. Like my father, I didn't speak until I was 3 yrs old (and even then, I developed my own language before I switched over to English -- e.g., my parents tell me part of the vocabulary was "ruck-row" for truck. Later, I tended to trip frequently over nothing (or, as my parents put it, over black scuff marks on the floor), that the number of bruises I had actually led to a brief abused-child investigation. Since then, I've adjusted pretty well. Still irritate my mother when I don't telephone her frequently (Q: nothing new to report, so what's the point in calling her? A: to make her happy. That'll be $2.35 long-distance charge, please). Christopher A. Bohn
Wouldn't it be like suing people for having bibliographies in their book/articles
The difference is that a printed bibliography does not give you access to the material (of course, this is one of the benefits of an electronic bibliography). As an example, suppose "Huckleberry Finn" (or "Origin of Species") is outlawed in your state. Including "Huckleberry Finn" in your bibliography is fine, but including contact information for the underground library is another matter.
But this brings up an interesting thought. If it's illegal to include a hyperlink to (hypothetically) http://joes_music.com/like_a_virgin.mp3, then can you get around this problem by printing the URL as plain text?
You're right. A simple loop with no interloop dependences is something a parallelizing compiler should be able to handle on its own. The problem is such loops are often only a small part of the application, and Amdahl's Law tells us that the speedup we get by parallelizing an application is a function of the fraction of the code that gets parallelized. It would be a waste to through 100 processors at this application if your parallelizing compiler could only parallelize 2% of the code. IMO, to get near-optimal parallelization, you need the insight only available to a human programmer. Christopher A. Bohn
What you're describing is still explicitly identifiying available parallelization. That it's part of the language doesn't alter the fact that you're telling the compiler where it can parallelize the code. Christopher A. Bohn
Oh, yeah -- nearly forgot. Another issue is that software that provides good performance on SMP can actually run slower on a distributed system than it would on a uniprocessor, due to the communication overhead introduced by the network. Christopher A. Bohn
Unless you identify the parallel nature of your software, then this won't be able to help an individual application (though it might help with distributing multiple apps similar to what you can see with SMP). No OS can be expected to identify the parallel nature on its own. The programmer must do it, either with threading or with message passing. You might try a parallelizing compiler, but they have only limited insight into possible parallel nature, and cannot be expected to provide near-optimal parallelization. Christopher A. Bohn
Plans for shuttle operations out of Vandenberg were canned when the DoD decided it wasn't going to rely on the shuttle for its launches. SLC-6 has since been turned into a commercial launch facility. Christopher A. Bohn
The problem is it's not that simple. Add in orbital inclinations and eccentricities, and we no longer have a collection of bodies that pose no possible threat to each other. Granted, in most conceivable scenarios, there won't be a "head-on" collision (the only conceivable scenario would be two bodies in near-polar orbit), but you don't need to have a "head-on" collision to cause damage. Christopher A. Bohn
The problem is not with the material large enough for this to work with. The problem is with the material that is too small to even track, such as paint chips and naturally-occuring micrometeorites. Christopher A. Bohn
!? Given how much the FSF has given away, please explain to the rest of us how RMS is ungrateful. I don't agree with all he says, but I'd say he's been rather generous. Christopher A. Bohn
Two possible outcomes: - Nothing breaks it, and this becomes a marketing high-point for Microsoft - It gets broken, and Microsoft engineers now have solid data (vice anecdotal) as to where the problems are. Especially if this was compiled with the debug option switched on. Christopher A. Bohn
"We don't do background checks, and we have no way of knowing if you answer the questionaire truthfully." Truly fascinating that they would state this so openly. Christopher A. Bohn
This guy's info is, at best, dated. "Tempest" hasn't been the approved term since 1996 -- "EMSEC" (Emissions Security) is now the official term. And last I checked, the EMSEC instructions don't provide allowances for all this shielding (whether they should is a different discussion) -- the "minimum spacing" requirements to separate classified & unclassified systems (and even classified systems rated for different levels of classification) would still hold. Christopher A. Bohn
The technology formerly known as "Wolfpack" (now NT Enterprise Cluster Server, or something like that) is intended for high availability (failover, etc). NT high performance clustering has been done, however. See, especially the "NT SuperCluster" at UIUC. There are a couple versions of MPI available for NT. I know of four, two no-longer-supported-but-still-available academic versions and two commercial versions derived from the academic versions. MPI-Pro (from MPI Software Technology, Inc.) and PateNT (from Genias) are the two commercial verions. Don't recall URLs, but MPI-Pro is advertised in Linux Journal, and if you're really curious, we all know how to use a search engine. Christopher A. Bohn
Technically, you could not build a Beowulf from BeOS -- the definition that Don Becker, Tom Sterling, et.al., provide for "What is a Beowulf?" is that it must have a free OS (Linux, *BSD, other) so that, if need be, optimizations to the OS can be made. That said, there's no reason you couldn't set up a "Cluster of Workstations" (COW) which is basically what you're thinking of -- it's been done with Solaris, NT,... The limiting factor is getting interprocess communication libraries. Christopher A. Bohn
Think of it as more of a "trickle-charge" -- not enough to fully power your laptop, but enough to delay changing batteries. By how much, I dunno -- Using the POOMA method, I'd guess on the order of 15-30 minutes from a few hours of typing. Christopher A. Bohn
About a forth of the way down this page (sorry, no cid link)...
Keyboard Electric Generators (Score:1)
by EngrBohn (cbohn@ieee.org) on Thursday April 01, @09:25AM EDT (User Info) http://members.aol.com/EngrBohn/ Actually, the technology to do this exists now. Not enough to fully charge your batteries, but enough to slow down the discharge (how much? dunno, this is a back-of-envelope analysis). By attaching small magnets to the underside of the keys, and wires near the keys, then when you depress a key, a small electric current will be induced in the nearby wire. Of course, when you release the key, a reverse current is then induced -- we'd have to use itsy-bitsy rectifier bridges to maintain a positive voltage across the battery terminals, but with IC technology, that shouldn't be hard. Of course, you'll have to keep your floppies (and maybe hard drive) away from your keyboard now... cb Christopher A. Bohn Oooh! What does this button do!?
Keyboard Electric Generators (Score:1) by Metiu on Thursday April 01, @04:13PM EDT (User Info) http:// If you really want energy, why not use piezoelectric crystals? They are very sensitive to knocks, and they would work perfectly with my IBM "heavy -foot- touch" keyboard... (BTW very good design for the site, it took a long time to build it I guess... looks like there are some very good unemployed geeks out there... hire 'em!)
Slashdot Mirror
IANAL, and it has been four years since I took a "Space Treaties and Legislation" class...
Under The Outer Space Treaty (short title), no country can stake claim on an extraterrestial body (including the Moon) -- says nothing about private organizations. The Moon Treaty (short title), which only a handful of countries have signed (none of the major space powers have signed it), is based on The Law of the Sea Convention and is more muddy regarding private ownership rights.
As to whether the U.S. could currently claim ownership, territorial waters traditionally had been defined by the range of the most advanced artillary of the time (weapons of the last few decades has made this absurd, which is why there is now an arbitrary distance used). Even if the U.S. claimed ownership of the Moon, it is currently in no position to defend that claim.
Christopher A. Bohn
- MS WinCE team has concluded that open source development tools are superior to MS' WinCE development tools
- MS is secretly funding ports of these tools to CE
The author then asks whether MS will take the next step and open-source WinCE, and answers his own question as "unlikely".Christopher A. Bohn
I, for one, am glad to see that civilized discourse resolved the matter, rather than this becoming the (apparently) much-anticipated courtroom test of the GPL.
Christopher A. Bohn
Mailing lists are not being mirrored.
Christopher A. Bohn
They made the announcement back in March. I remember talking to one of my professors about the research & academic opportunities it presented. /03/02/1133216.shtml
http://slashdot.org/articles/99
Christopher A. Bohn
Perhaps that's why the investigation was launched. But I assure you that not only do I remember the events (well, at least the tripping), I had never forgotten them, so they are not "recovered memories."
Christopher A. Bohn
Early on, I displayed some symptoms. Like my father, I didn't speak until I was 3 yrs old (and even then, I developed my own language before I switched over to English -- e.g., my parents tell me part of the vocabulary was "ruck-row" for truck. Later, I tended to trip frequently over nothing (or, as my parents put it, over black scuff marks on the floor), that the number of bruises I had actually led to a brief abused-child investigation.
Since then, I've adjusted pretty well. Still irritate my mother when I don't telephone her frequently (Q: nothing new to report, so what's the point in calling her? A: to make her happy. That'll be $2.35 long-distance charge, please).
Christopher A. Bohn
The difference is that a printed bibliography does not give you access to the material (of course, this is one of the benefits of an electronic bibliography). As an example, suppose "Huckleberry Finn" (or "Origin of Species") is outlawed in your state. Including "Huckleberry Finn" in your bibliography is fine, but including contact information for the underground library is another matter.
But this brings up an interesting thought. If it's illegal to include a hyperlink to (hypothetically) http://joes_music.com/like_a_virgin.mp3, then can you get around this problem by printing the URL as plain text?
Christopher A. Bohn
You're right. A simple loop with no interloop dependences is something a parallelizing compiler should be able to handle on its own. The problem is such loops are often only a small part of the application, and Amdahl's Law tells us that the speedup we get by parallelizing an application is a function of the fraction of the code that gets parallelized. It would be a waste to through 100 processors at this application if your parallelizing compiler could only parallelize 2% of the code.
IMO, to get near-optimal parallelization, you need the insight only available to a human programmer.
Christopher A. Bohn
What you're describing is still explicitly identifiying available parallelization. That it's part of the language doesn't alter the fact that you're telling the compiler where it can parallelize the code.
Christopher A. Bohn
Oh, yeah -- nearly forgot. Another issue is that software that provides good performance on SMP can actually run slower on a distributed system than it would on a uniprocessor, due to the communication overhead introduced by the network.
Christopher A. Bohn
Unless you identify the parallel nature of your software, then this won't be able to help an individual application (though it might help with distributing multiple apps similar to what you can see with SMP). No OS can be expected to identify the parallel nature on its own. The programmer must do it, either with threading or with message passing. You might try a parallelizing compiler, but they have only limited insight into possible parallel nature, and cannot be expected to provide near-optimal parallelization.
Christopher A. Bohn
Plans for shuttle operations out of Vandenberg were canned when the DoD decided it wasn't going to rely on the shuttle for its launches. SLC-6 has since been turned into a commercial launch facility.
Christopher A. Bohn
The problem is it's not that simple. Add in orbital inclinations and eccentricities, and we no longer have a collection of bodies that pose no possible threat to each other.
Granted, in most conceivable scenarios, there won't be a "head-on" collision (the only conceivable scenario would be two bodies in near-polar orbit), but you don't need to have a "head-on" collision to cause damage.
Christopher A. Bohn
The problem is not with the material large enough for this to work with. The problem is with the material that is too small to even track, such as paint chips and naturally-occuring micrometeorites.
Christopher A. Bohn
!?
Given how much the FSF has given away, please explain to the rest of us how RMS is ungrateful. I don't agree with all he says, but I'd say he's been rather generous.
Christopher A. Bohn
Frankly, I'd rather not see GPL tested in court, since that would require first that somebody attempt to violate the GPL.
Christopher A. Bohn
Two possible outcomes:
- Nothing breaks it, and this becomes a marketing high-point for Microsoft - It gets broken, and Microsoft engineers now have solid data (vice anecdotal) as to where the problems are. Especially if this was compiled with the debug option switched on.
Christopher A. Bohn
"We don't do background checks, and we have no way of knowing if you answer the questionaire truthfully."
Truly fascinating that they would state this so openly.
Christopher A. Bohn
Just passed this to my wife, and she says 8-10 hours is about right for her, while she does other things around the house.
Christopher A. Bohn
This guy's info is, at best, dated. "Tempest" hasn't been the approved term since 1996 -- "EMSEC" (Emissions Security) is now the official term. And last I checked, the EMSEC instructions don't provide allowances for all this shielding (whether they should is a different discussion) -- the "minimum spacing" requirements to separate classified & unclassified systems (and even classified systems rated for different levels of classification) would still hold.
Christopher A. Bohn
The technology formerly known as "Wolfpack" (now NT Enterprise Cluster Server, or something like that) is intended for high availability (failover, etc). NT high performance clustering has been done, however. See, especially the "NT SuperCluster" at UIUC. There are a couple versions of MPI available for NT. I know of four, two no-longer-supported-but-still-available academic versions and two commercial versions derived from the academic versions. MPI-Pro (from MPI Software Technology, Inc.) and PateNT (from Genias) are the two commercial verions. Don't recall URLs, but MPI-Pro is advertised in Linux Journal, and if you're really curious, we all know how to use a search engine.
Christopher A. Bohn
Technically, you could not build a Beowulf from BeOS -- the definition that Don Becker, Tom Sterling, et.al., provide for "What is a Beowulf?" is that it must have a free OS (Linux, *BSD, other) so that, if need be, optimizations to the OS can be made. ... The limiting factor is getting interprocess communication libraries.
That said, there's no reason you couldn't set up a "Cluster of Workstations" (COW) which is basically what you're thinking of -- it's been done with Solaris, NT,
Christopher A. Bohn
Think of it as more of a "trickle-charge" -- not enough to fully power your laptop, but enough to delay changing batteries. By how much, I dunno -- Using the POOMA method, I'd guess on the order of 15-30 minutes from a few hours of typing.
Christopher A. Bohn
About a forth of the way down this page (sorry, no cid link)...
Christopher A. Bohn