What is bad is the additional restrictions. And if you cannot freely reditribute your modifications to others, I -for one- question how "open" such source is.
What's to prevent you from distributing a set of patches? After all, that's basically how BSD got started - before 4.4BSD, you needed an AT&T license to legally run BSD.
Sun's move follows what RMS wanted from "free software" before the advent of the GPL - the ability to go in and fix broken code.
OTOH, it probably would be a bad idea for Linux or *BSD kernel developers to peruse through the code unless they thoroughly document what code they've seen (and even then...).
When you neglect security to a point where accidents are bound to happen sooner or later, do you still not think we should hold the responsible accountable?
Question is - who was responsible?
When the plant was built, UC had an agreement with the local government that no housing be built near the plant - at the time of the accident, housing had been built right up to the plant boundaries. The plant was supposed to have a Chemical Engineer on-site at all times - at the time of the accident, the fellow in charge was at home having tea.
Along that line, did UC have total operational control of the plant? That is, was the accident a result of something that UC had control over, or was due to circumstances under local control?
So, the technology may be new to the USA, but there's are wealth of knowledge on designing and running these reactors elsewhere in the world.
Lessee, Peach Bottom unit 1 went on-line in the early 1960's and Fort St. Vrain was running from the mid-70's to the mid-80's. Both were HTGR's and not the lower temp Magnox or AGR designs.
The big problem with Fort St. Vrain was that someone decided that it would be slightly more efficient using steam turbines to run the He circulators rather than electric motors - GA and the plant operators never did get the seal problem fixed.
Oh yes, they're arguably quite a bit safer than PWRs as well!
Yes and no. The metal water reactions that occur even in a moderate melt-down (e.g. Three Mile Island) release enough hydrogen to combine with the iodine - and that works as a wonderful containment. OTOH, the fuel design for HTGR's have a much higher marging against melting than a typical PWR.
I was referring to putting the Lithium targets in a natural uranium fueled, heavy water moderated fission reactor - e.g. the tritium production reactors (now shut down) at Savanah River.
Third, Tritium is a little annoying to get, but heavy water moderated fission reactors produce the stuff as a waste product, and those aren't going away anytime soon even if we get fusion working.
Huh?
The traditional reason for using heavy water reactors is to allow use of natural uranium - the Tritium production comes from putting lithium targets in the reactor for an (n,T) reaction. You will get some tritium from deuterium absorbing neutrons, but that is a real PITA to extract.
PhD thesis or not, it's just her opinion, and there seem to be plenty that contradict hers.
It certainly is her opinion, but I'd trust her opinion on this subject. One of the issues will be separating the He3 from all of the other volatiles in lunar rock - probably the biggest pain will be He4 (natural choice would be gaseous diffusion).
The D-He3 reaction does have the advantage of producing a lot less neutrons than the "standard" D-T reaction. The fact that most of the energy is being carried away by a charged particle is also a potential big plus.
On the gripping hand, I do have a friend whose PhD thesis was the chemistry of moon rocks - and her opinion was that mining He3 would be impractical.
It can't all be bad. Electronic miniturisation seems to be mostly okay so why not mechanical?
While I'd quibble with the OP's numbers, the general thrust is correct - efficiency does take a big dive when a combustion engine is scaled down. Gas turbines are more affected by scaling than reciprocating engines - the smallest model aircraft engines are on the order of 0.2cc and the largest diesel engines have several cubic meter displacement per cylinder.
The OP had mixed up flow regimes - the microturbines would have problems with the laminar flow boundary layer being larger than the spacing between blades (i.e. viscous flow). In larger turbines, the boundary layer is small compared to blade size and can be treated more as a potential flow than viscous flow.
As for electronic miniturization - looks like we are approaching the limits of silicon based technology. Back about 15 years ago, the majority of the delays in logic design was due to gate switching (interconnect delays could be ignored). Now the interconnect delays are the big problem and will be getting even worse as feature size shrinks (scaling laws again).
Linus has finally gotten the hint about documenting where the Linux code came from. One of the reasons that the SCO lawsuits were a threat was that there was no consistent documentation of the Linux source tree. Clear and consistent documentation would have made it easy to say "Oh you think our foo.h was copied from SCO's foo.h? Well, our foo.h was provided by Arthur Dent and Edmund Blackadder and they had not see any of the SCO source."
For example if SUN releases solaris under GPL no problem. Linux will take from it and improve itself. If it's under the BSD-like the same thing will occur.
Which is precisely the reason that Solaris will probably not be released under the GPL or a Berkeley license. This doesn't prevent features from being implemented in Linux, but that code would have to be written from scratch - and the process in which the code was written would have to be documented (which is something that Linus has very belatedly started to do).
On the gripping hand, having the source allows you a chance to fix a problem - either by having a better idea of what the software actually does or by modifying the source and re-compiling. IIRC, one of RMS's beefs against closed source software was the inability to fix problems - the GPL came about later because of the propietary fork of emacs.
Licensing is not a religious issue for me. I have no qualms about people releasing code under GPL (and wrote a dissenting e-mail to BizWeek's Wildstrom when he suggested that Linux be licensed by something other than the GPL), but I also object to demands that all code be released under GPL or Berkeley licenses.
P.S. I use the term "Berkeley license" instead of BSD since SPICE was released with a very similar license and SPICE predates BSD.
Your rant touched on three points: 1) where's the source?; 2) Where's ZFS?; 3) Where's the Linux compatibility?
For the first point, I'd be surprised if they released the code by FCS in January, but even more surprised if it isn't released by Dec '05.
The ZFS filesystem will probably not make it to the FCS release in January, but the Sun folks on Usenet seem confident that it will be in the first MU in the May-July timeframe.
Sun already has the lxrun libraries and is working on "Janus" - which should be in by the firt MU release. On this last point, Sun is compensating for Linux's deviation from the SysV standard for the x86 ABI.
W-e-l-l, just take that magnet and vacuum chamber you bought for the cyclotron, whip up an ion gun and a couple of collection buckets- and voila, you have yourself a calutron. Get soem Uranium and you too can produce highly enriched uranium.
You obviously have never designed or shimmed a magnet before - makes perfect sense to me (BTDT) - then again I get paid to work on magnets.
Re:pollution in cities is much better now
on
Killer Ozone?
·
· Score: 1
Uhh.. I see that people have failed to mention that pollution in cities has really gone down in the past 50 years.
That is if you are talking about air pollution. For other forms of pollution, the story is a little different.
Go back towards the end of the 19th century and one of the big urban pollution problems was horse manure (I'd rather deal with ozone than finely ground horseshit). One of the big selling points of the electric streetcar was that it didn't have the pollution problems associated with the horsecar.
Then there was the problem of smoke from coal burning furnaces (natural gas is a lot cleaner burning). A related issue now is that many areas are prohibiting installation of wood burning fireplaces - the bumper sticker "Split wood - not atoms" is really bad advice.
A datum in support of your original point - the air in San Diego County is much cleaner (i.e. fewer smoggy days) than it was in the late 1970's - with the exception of cold nights when people light off their fireplaces (and of course the first few days of the Cedar Fire).
The power transistors in my Prius handle that kind of power, and the ones that drive locomotives handle quite a bit more.
The IGBT's that you'ld find on a late model GE (EMD's have been using GTO thyristors) typically are good for a few kHz. The GE's use an inverter per axle, so a 6,000 hp unit will have 1,000 hp or 750 kW per inverter (which is understating the VA capability by a bit). A megawatt class tube from Eimac will go up to 30 MHz or so (and require about 30 kW just to light up the cathode). The USAF BMEWS stations had some microwave tubes good for a megawatt or two.
The ultimate tubes were the thyratrons built for the Pacific Intertie back in 1968 could carry a couple thousand amps and hold off 150 kV.
I worked on a 4CX35000 amplifier that would put out 800KW (albeit in short pulses). It used a 4CX3000 running class-C as a driver and that stage needed about 1KW input just to get the tube out of cut-off.
One interesting lesson from this beast is that any kind of film resistor (e.g. carbon, metal or metal-oxide) is not in the same league as good old carbon composition resistors (kind of knew that before hand, but carbon comps became hard to get after about 1992).
...but far less likely to survive a fall of 6 inches.
Obviously you haven't read up on the history of the proximity fuze. Deak Parsons and team found out how to make a vacuum tube survive in the nose of a 5 inch naval shell - with initial acceleration of several thousands g's.
Can you name a more widely used application of tubes now days?
High power RF amplifiers. Tubes have several advantages here, better high frequency response, can run a LOT hotter and are typically more electrically rugged (i.e. a tube can recover from an arc).
You're not getting the point - in order to go to direct election of the president, the constitution needs to be amended - which requires ratification of 3/4 of the states - and a lot of the states that would be needed for ratification would stand to lose if the Electoral College goes bye-bye.
I don't know if you're a USian or not, but you obviously don't grok basic US civics.
I'll see your free speech suppression and raids on ______ists via the Sedition Act of 1918 and raise you secret searches and the elimination of habeas corpus via the "War on Terror".
How about the 20 to 100 million death from influenza that were a pretty much direct consequence of the US involvement in WW1? (See "The Great Influenza" by Barry). After reading Barry's book, I really wonder if a lot of the inspiration for 1984 came from the US in WW1.
So constitution[al] democracy includes locking people away without charges, trials, or lawyers?
Ever been to Manzanar in California's Owens Valley???? 110,000 people locked away for simply being of Japanese descent by order of FDR (with considerable prodding from Earl Warren). I don't recall any court even contemplating granting writ of habeus corpus to these detainees. Many of these people got cheated out of some very valuable property when they were forced to move. Then again, a lot of people got screwed over in WW2 on the pretense that there was a war going on.
Anyway, back to WW1 - several members of my mom's family grew up speaking German at home, but refused to admit that they knew the language after the propaganda campaign of WW1.
TV station KTLA edited the tape before broadcasting it and passing it along to other stations. What most people didn't see until the first trial was Rodney leaping out of the car before he was pounced upon - which led to the acquittal of all four of the officers involved in the first trial.
Microsoft said it lost the source code to MS-DOS. Pretty much the same way that they can't find the e-mails for the Burst case.
One question about source code for OS's - if a company can't find the source code for a 5 year old release of its software - do I really want to trust their software to handle my data??
Time 6AM on the day of a full moon - wondered about the connections between tides and earthquakes since then.
What most people don't know is that if the shaking lasted maybe 20 seconds longer it could have been the worst natural disaster in US history. The near failure of the Van Norman dam scared the bejeebers out of the Cal Department of Water Resources and they called for a lot of earth fill dams to be rebuilt.
Slashdot Mirror
What's to prevent you from distributing a set of patches? After all, that's basically how BSD got started - before 4.4BSD, you needed an AT&T license to legally run BSD.
Sun's move follows what RMS wanted from "free software" before the advent of the GPL - the ability to go in and fix broken code.
OTOH, it probably would be a bad idea for Linux or *BSD kernel developers to peruse through the code unless they thoroughly document what code they've seen (and even then...).
Question is - who was responsible?
When the plant was built, UC had an agreement with the local government that no housing be built near the plant - at the time of the accident, housing had been built right up to the plant boundaries. The plant was supposed to have a Chemical Engineer on-site at all times - at the time of the accident, the fellow in charge was at home having tea.
Along that line, did UC have total operational control of the plant? That is, was the accident a result of something that UC had control over, or was due to circumstances under local control?
That's pretty much my experience as well. The new fomat is much less useful than the old format (give me plain text).
Lessee, Peach Bottom unit 1 went on-line in the early 1960's and Fort St. Vrain was running from the mid-70's to the mid-80's. Both were HTGR's and not the lower temp Magnox or AGR designs.
The big problem with Fort St. Vrain was that someone decided that it would be slightly more efficient using steam turbines to run the He circulators rather than electric motors - GA and the plant operators never did get the seal problem fixed.
Oh yes, they're arguably quite a bit safer than PWRs as well!
Yes and no. The metal water reactions that occur even in a moderate melt-down (e.g. Three Mile Island) release enough hydrogen to combine with the iodine - and that works as a wonderful containment. OTOH, the fuel design for HTGR's have a much higher marging against melting than a typical PWR.
I was referring to putting the Lithium targets in a natural uranium fueled, heavy water moderated fission reactor - e.g. the tritium production reactors (now shut down) at Savanah River.
Huh?
The traditional reason for using heavy water reactors is to allow use of natural uranium - the Tritium production comes from putting lithium targets in the reactor for an (n,T) reaction. You will get some tritium from deuterium absorbing neutrons, but that is a real PITA to extract.
It certainly is her opinion, but I'd trust her opinion on this subject. One of the issues will be separating the He3 from all of the other volatiles in lunar rock - probably the biggest pain will be He4 (natural choice would be gaseous diffusion).
On the gripping hand, I do have a friend whose PhD thesis was the chemistry of moon rocks - and her opinion was that mining He3 would be impractical.
While I'd quibble with the OP's numbers, the general thrust is correct - efficiency does take a big dive when a combustion engine is scaled down. Gas turbines are more affected by scaling than reciprocating engines - the smallest model aircraft engines are on the order of 0.2cc and the largest diesel engines have several cubic meter displacement per cylinder.
The OP had mixed up flow regimes - the microturbines would have problems with the laminar flow boundary layer being larger than the spacing between blades (i.e. viscous flow). In larger turbines, the boundary layer is small compared to blade size and can be treated more as a potential flow than viscous flow.
As for electronic miniturization - looks like we are approaching the limits of silicon based technology. Back about 15 years ago, the majority of the delays in logic design was due to gate switching (interconnect delays could be ignored). Now the interconnect delays are the big problem and will be getting even worse as feature size shrinks (scaling laws again).
Linus has finally gotten the hint about documenting where the Linux code came from. One of the reasons that the SCO lawsuits were a threat was that there was no consistent documentation of the Linux source tree. Clear and consistent documentation would have made it easy to say "Oh you think our foo.h was copied from SCO's foo.h? Well, our foo.h was provided by Arthur Dent and Edmund Blackadder and they had not see any of the SCO source."
Which is precisely the reason that Solaris will probably not be released under the GPL or a Berkeley license. This doesn't prevent features from being implemented in Linux, but that code would have to be written from scratch - and the process in which the code was written would have to be documented (which is something that Linus has very belatedly started to do).
On the gripping hand, having the source allows you a chance to fix a problem - either by having a better idea of what the software actually does or by modifying the source and re-compiling. IIRC, one of RMS's beefs against closed source software was the inability to fix problems - the GPL came about later because of the propietary fork of emacs.
Licensing is not a religious issue for me. I have no qualms about people releasing code under GPL (and wrote a dissenting e-mail to BizWeek's Wildstrom when he suggested that Linux be licensed by something other than the GPL), but I also object to demands that all code be released under GPL or Berkeley licenses.
P.S. I use the term "Berkeley license" instead of BSD since SPICE was released with a very similar license and SPICE predates BSD.
For the first point, I'd be surprised if they released the code by FCS in January, but even more surprised if it isn't released by Dec '05.
The ZFS filesystem will probably not make it to the FCS release in January, but the Sun folks on Usenet seem confident that it will be in the first MU in the May-July timeframe.
Sun already has the lxrun libraries and is working on "Janus" - which should be in by the firt MU release. On this last point, Sun is compensating for Linux's deviation from the SysV standard for the x86 ABI.
W-e-l-l, just take that magnet and vacuum chamber you bought for the cyclotron, whip up an ion gun and a couple of collection buckets- and voila, you have yourself a calutron. Get soem Uranium and you too can produce highly enriched uranium.
*rolls eyes*
You obviously have never designed or shimmed a magnet before - makes perfect sense to me (BTDT) - then again I get paid to work on magnets.
That is if you are talking about air pollution. For other forms of pollution, the story is a little different.
Go back towards the end of the 19th century and one of the big urban pollution problems was horse manure (I'd rather deal with ozone than finely ground horseshit). One of the big selling points of the electric streetcar was that it didn't have the pollution problems associated with the horsecar.
Then there was the problem of smoke from coal burning furnaces (natural gas is a lot cleaner burning). A related issue now is that many areas are prohibiting installation of wood burning fireplaces - the bumper sticker "Split wood - not atoms" is really bad advice.
A datum in support of your original point - the air in San Diego County is much cleaner (i.e. fewer smoggy days) than it was in the late 1970's - with the exception of cold nights when people light off their fireplaces (and of course the first few days of the Cedar Fire).
The IGBT's that you'ld find on a late model GE (EMD's have been using GTO thyristors) typically are good for a few kHz. The GE's use an inverter per axle, so a 6,000 hp unit will have 1,000 hp or 750 kW per inverter (which is understating the VA capability by a bit). A megawatt class tube from Eimac will go up to 30 MHz or so (and require about 30 kW just to light up the cathode). The USAF BMEWS stations had some microwave tubes good for a megawatt or two.
The ultimate tubes were the thyratrons built for the Pacific Intertie back in 1968 could carry a couple thousand amps and hold off 150 kV.
I worked on a 4CX35000 amplifier that would put out 800KW (albeit in short pulses). It used a 4CX3000 running class-C as a driver and that stage needed about 1KW input just to get the tube out of cut-off.
One interesting lesson from this beast is that any kind of film resistor (e.g. carbon, metal or metal-oxide) is not in the same league as good old carbon composition resistors (kind of knew that before hand, but carbon comps became hard to get after about 1992).
Obviously you haven't read up on the history of the proximity fuze. Deak Parsons and team found out how to make a vacuum tube survive in the nose of a 5 inch naval shell - with initial acceleration of several thousands g's.
High power RF amplifiers. Tubes have several advantages here, better high frequency response, can run a LOT hotter and are typically more electrically rugged (i.e. a tube can recover from an arc).
I don't know if you're a USian or not, but you obviously don't grok basic US civics.
It ain't gonna happen.
How about the 20 to 100 million death from influenza that were a pretty much direct consequence of the US involvement in WW1? (See "The Great Influenza" by Barry). After reading Barry's book, I really wonder if a lot of the inspiration for 1984 came from the US in WW1.
So constitution[al] democracy includes locking people away without charges, trials, or lawyers?
Ever been to Manzanar in California's Owens Valley???? 110,000 people locked away for simply being of Japanese descent by order of FDR (with considerable prodding from Earl Warren). I don't recall any court even contemplating granting writ of habeus corpus to these detainees. Many of these people got cheated out of some very valuable property when they were forced to move. Then again, a lot of people got screwed over in WW2 on the pretense that there was a war going on.
Anyway, back to WW1 - several members of my mom's family grew up speaking German at home, but refused to admit that they knew the language after the propaganda campaign of WW1.
TV station KTLA edited the tape before broadcasting it and passing it along to other stations. What most people didn't see until the first trial was Rodney leaping out of the car before he was pounced upon - which led to the acquittal of all four of the officers involved in the first trial.
One question about source code for OS's - if a company can't find the source code for a 5 year old release of its software - do I really want to trust their software to handle my data??
What most people don't know is that if the shaking lasted maybe 20 seconds longer it could have been the worst natural disaster in US history. The near failure of the Van Norman dam scared the bejeebers out of the Cal Department of Water Resources and they called for a lot of earth fill dams to be rebuilt.