Which reminds me of something... Isn't it time for C and its likes to let us specify explicitly how many bits we want for a variable? I would like to tell the compiler that a variable should be _exactly_ 32 bits and another one _at least_ 64 bits.
Sounds like you want the "kind" functionality of Fortran 90.
Back in the bad old days, a REAL in Fortran could be anywhere from 32 to 64 bits - a program that ran fine using REAL on a CDC-6600 (60 bits) might die horribly using REAL on an IBM 360 (32 bits but using hexadecimal arithemtic).
The joke in the early days of my employer was that prospective employees were given a psychological screening - if they passed, they weren't hired. After getting to know some of my co-workers, I started to wonder is there was some truth to that. That's probably why I'm still there...
Anyone remember the article from a few months back about many CEO's being sociopaths?
The Cairo prject was M$'s attempt to finsih killing off OS/2 and kill off the various desktop UNIX distro's (HP had a nice candidate with the 900/712 with Lotus 123 and Ami-Pro running natively on HP-UX). Kind of thinking that the WinFS idea is like speech recognition (or Duke Nukem Forever) - remember reading Jerry Pournelle quoting Bill Godbout about the 80286 will be powerfull enough for speech recognition, this was ca 1982.
My take on the Wall Street crowd is that they have a poor understanding of what makes the industry tick and what may be good for Wall Street on the short term is not necessarily good for the customers. Sun's stock performance since the bottom of the market (post bubble) has been better than the blue chips. The stock price is still w-a-y below the peak, but the peak stock was pure bubble.
I'm not familiar with the details of modern civilian breeder technology, but current weapons reactors run on a three month fuel cycle, and this is likely to be the case for civilian reactors too (the 238U jacket will have to be cycled roughly every 90 days.) This is because 240Pu starts to build up significantly after that time, making the fuel difficult to handle.
240Pu is a problem for making weapons, but not that big a deal for reactors.
Normal light-water plants have a 3 year fuel cyle, where 1/3rd of the fuel gets swapped out each year. The lifetime limits for breeder reactor fuel is based on the material damage from fast neutrons.
At best, fission is still a stop-gap: supplies of fissionables are limited, on the order of a century or two at most, perhaps much less.
We've got plenty of fissionables (which include U-238 and Th-232), but the supply of fissiles (e.g U-235) is much more limited. Uranium is actually quite common, typical granite has about 1 gram per tonne. Anyway, the whole issue of limited supply of fissile material versus fissionable was what was behind the development of breeder reactors - with the integral fast reactor having some intriguing attributes (and significant engineering challenges).
Also TFA is inaccurate in talking about nuclear waste; the problem is not the U and Pu in spent fuel, which can be processed and reused, but thorium, radium, radon, and radioactive lead isotopes.
WTF are you talking about??? Thorium is naturally occuring, radium and radon are part of the natural decay chain of U-238, and the only way lead becomes radioactive is by activation (typically by neutrons). The radioactive waste from reactors consists of fission products and transuranics (plutonium, americium, curium, etc.) - fisson products typically have short half lifes, while the transuranics often have very long half-lifes.
Argh! I can't stand it anymore... Your point is dead on, but then you went and polluted it with a really inappropriate analogy.
Maybe a more appropriate analogy would be that of electric service in that utilities are rarely sized to handle all customers drawing their maximum rated power simultaneously - which would cause a meltdown of the transmission and distribution networks.
The telcos were having problems when dial-up service was becoming popular since the networks were designed for relatively brief calls.
One of the worst AUP's for broadband was Adelphia, that limited total bandwidth usage to 2.5GB a month - which could easily be reached with a dial-up connection with nightly 100MB downloads.
I did see a mention of some small outfit that was making noises of doing a single core version of the T1 - presumably small and low power. I was not able to find out anything more of the company mentioned.
Would be nice to have a low power (say 5 to 10 watts) Sparc board to run Solaris - something like a Sparc equivalent to the Soekris boards.
I do have memories of MS being a halfways decent company at the same time that DRI was having some issues with arrogance - but that's when Allen was still pretty much in charge. During the DOS 2.0 days, MS was willing to allow customers other than IBM to package their own version of MS-DOS - Compaq's DOS v3.31 was a notable example. It may have been CPQ's DOS v3.31 embarrassed MS so much, that they decided put a stop to vendor specific versions of DOS.
After Ballmer stepped in, support for fringe platforms (i.e. not strictly PC-compatible) was pretty much dropped, up through 2.0, MS-DOS ran on quite a variety of 8086/8 boxes.
Now to think of it, MS dropping Xenix happened about this same time frame.
Stopping Nazis and dirty sneak attacking Japs wasn't too hard a sell
Actually, stopping Nazi's was a hard sell, the only reason the US declared war on Germany in WW2 was that Germany declared war first - and only after Roosevelt had goaded Hitler into declaring war first. The US public was in no mood to get involved with another war in Europe after the mess of our involvement in WW1 (which was probably a much larger mistake than getting involved in Iraq).
To get back on topic, the main reason the US was able to defeat Japan was that it started a major Naval construction program 1939-40, the Essex didn't come off the ways until 1943.
Really? I remember scramjets being discussed in the 1960's
I've got an encyclopedia set titled "Above and Beyond" published in the late 60's showing the difference between a ramjet and scramjet. Work was being done by the Marquardt corp. The Nation Aerospace Plane (NASP, ca 1990) was supposed to be using an external scramjet.
It was Bill Stout, not Ed Heineman who coined the phrase "Simplicate and add lightness".
Nevertheless, Heineman was still one heck of an airplane designer.
As far as the 8" floppy, ISTR that they were intended to replace punched cards, 77 tracks with 26 sectors (hard coded) came out to be pretty close to a box of 2000 hollerith cards (80 columns with 12 bits per column). 8" drives were available before the end of 1975, and the VAX came out in 1977(?). One of the uses for the flopies was loading the microprogram store on the VAX and IBM machines of the same era.
At the same time, 86-DOS (the original name for what is now MS-DOS) could also handle varying sized sectors - generally used 128 byte sectors for SSSD 8" disks and 1024 byte sectors for DSDD 8" disks. Tim Paterson recommened using a cluster size that was the square root of the disk capacity in bytes, so the SSSD disks used four sectors per cluster (512 bytes) and the DSDD used 1 sector per cluster (1024 bytes).
What was done with CP/M and 86-DOS, but not PC/MS-DOS, was distributing source code for the BIOS (called io.sys under 86-DOS). With that source code, it was relatively easy to support whatever sector size you wanted.
Glad to know that I'm not the only one who liked Word for DOS better than WfW. One of the nice things was that I could set up different 'normal.sty' in different subdirectories - have one for letters, one for other stuff, etc. Making a global change in formatting was simply a matter of attaching a different style sheet.
One thing that drives me up the friggin wall with Word and related word processors (Star/OpenOffice) is the way they handle graphics. I find it a lot easier to define a frame and then put in the graphic (where the frame defines whether the graphic will be cropped or re-sized to fit), instead of putting in a graphic, finding that it wildly out of scale with the rest of the document and then try to wrestle it into place without the @$%*&#!!! word processor deciding that it is going to re-arrange text to suit the eff'ing picture. What makes this worse is that 14 years ago, I was using a package that did it the right way and that package could run on a 25MHz 68040 with 16MB of memory.
On a similar vein, a former co-worker says he was much more productive typing out long documents on Lotus Manuscript than any currently available word processor - mainly because of the text interface.
It seems to me that the vision of the future we all have today is nowhere near as optimistic as the vision of the future they had in the '50's.
I've got a copy of the "Tomorrowland" DVD released by Disney a couple of years ago - mainly stuff about space made in the 1950's. Especially enjoyed the one about Mars - since I first saw it in 1962 or 63. What was a bit weird was seeing Ward Kimball as he looked in the 1950's as compared to what he was like in person in the 1990's (and he was still quite a character).
One thing a lot of people forget is that the 1950's were a time of tremendous changes in day to day technology. In 1950, rockets just being sent to the lower reaches of space, A-bombs were to be delivered by long range bombers, electronics meant vacuum tubes, TV's were rare, computers were largely experimental, most trains were hauled by steam locomotives, airliners were piston powered, cars typically had flathead engines, 6V electrical systems and air-conditioning was only found on Cadillacs, and many cities still had streetcars. In 1960, satellites had been to the moon, H-bombs were to be delivered by ICBM's, electronics meant transistors, TV's were common (color TV's were rare but not unheard of), computers were comon for large businesses, the steam locomotive was essentially extinct, airliners were jets, flathead engines had disappeared (with the exception of the Continentals used in the Checkers), cars used 12V electrical systems, air-conditioning was a common option and streetcars were almost gone.
We may finally get the videophone that was originally envisioned in the 50's.
One of the things that kept it from happening earlier (that is ca 2000) was that most "broadband" connections had their upstream speeds capped to 128kbps - though many of the newer offerings have a much higher upstream speeds.
Reading your post got me to thinking about a couple of memories from the 1970's. The first was reading Mac's Service Shop column in the April (?) 1972 issue of PopTronics where they were talking about the brand new HP-35 and knowing the world was going to be a different place. The next was seeing the first West Coast Computer Faire in 1977 - the equipment was pretty primitive by todays standards, but there was an optimism of what the future might bring that has been gone for at least the last decade if not the last two decades - a slightly faster intel box running a slightly improved version of windoze isn't much to look forward to.
One of Sun's goals for Java was to have a write once, run anywhere environment. MS wanted to embrace and extend Java so that it would be very easy to write a Java program that ran only on Windows. The project files that VC spits out are "an abomination in the eyes of the Lord".
Hard to believe that UNIX came out of Berkeley in 1960's since the first tape from Bell Labs arrived in December 1973. I'm guessing the CS department wanted to have a replacement for Kronos that went away when the CDC-6400 'B' machine got shipped off in January 1973.
2 gig ECC dimms and I'm seeing a 40% failure rate from multiple manufacturers. They stacked the damn chips one on top of another. There's no where for the heat to go. Of course they're going to fail.
Cpnversely, one of the things that impressed me about the physical design of Sun's SunBlade-1500 was a shroud over the DIMM's to promote cooling of said modules - the other end of the shroud went to the chassis exhaust fan. Weird that we're having to worry about thermal management in memory - though the the intel 2167's on the Seattle Computer's first 64K board were getting a bit toasty.
"If you tell a lie big enough and keep repeating it, people will eventually come to believe it. The lie can be maintained only for such time as the State can shield the people from the political, economic and/or military consequences of the lie. It thus becomes vitally important for the State to use all of its powers to repress dissent, for the truth is the mortal enemy of the lie, and thus by extension, the truth is the greatest enemy of the State."
-- Joseph Goebbels, German Minister of Propaganda, 1933-1945
And what most people don't realize is that Goebbels was describing George Creel's activities in the U.S. during WW1 - and dissent was heavily repressed in the US during WW1.
Why would MS try to appease ODF loving governments with an XML format and then piss them off again by embedding binary data for everything?
For the same friggin' reason they put the brain dead Posix inteface in Windows NT - so they could claim to meet the requirements of Posix compatibiliy without intending any serious use of it. Remember the fuss with Kerberos??
Also consider: If your datacenter is already built for DC, throw some solar photovoltaic panels on the roof. Inverters are a large part of most PV systems' expense, and you can skip that part. Why not start offsetting your grid demand now?
Glad to know that I'm not the only one thinking this would be a good idea. I would avoid the "Home Power" magazine approach of connecting the panels directly to the batteries - bad idea to have the batteries supplying short circuit current - and you should get significantly more power out of the panels by operating them at peak power operating point rather than forcing them to run at battery voltage.
LoL It would be fun to get the DC from old rotary converters for a data center.
Rotary converters??? The hell. How about the M-G sets from the old Milwaukee Electrification?? Especially the Janney substation with the 3MW M-G set from C.U.T. along with the three 1.5MW M-G sets. 'course most data centers would have a wee bit of trouble digesting 3.3kV...
You are on the right path in that it is easier to make a large AC/DC converter run efficiently than a bunch of small ones. In many ways you are descibing the old DC substations used in some older downtowns - where the rotary converters were feeding +/- 120V battery banks - talk about reliable power.
umm NO, despite British Government propaganda, it was POLES who cracked enigma, who captured first Enigmas, and who deciphered Enigma codes by HAND on a daily basis until Germany invaded us:(
And the Poles did a darn good job of it. One minor correction - Poland was invaded by both Germany and the Soviet Union.
Perhaps the best work of pre-war cryptanalysis was Friedman's breaking of the Japanese diplomatic cipher (AKA the Purple Code) - without any idea of what the machine looked like.
Like that one time, Satan decided that all railroad tracks should be the same distance apart, so that every train could work on every track, so people would ride around on the trains, which sucked out their immortal souls.
We all owe a debt of gratitude to the USSR for fighting Satan on that one...
Actually owe the debt to the Czars - Russia's 5 foot gauge predated the Soviets by decades. The US had quite a bit of 5 foot guage track until the 1880's - mainly in the old South, a little bit of 6 foot (mainly the Erie RR) and quite a bit of 3 foot in the west (especially Colorado) with the last real 3 footer being the White Pass and Yukon (the D&RGW stopped everything but tourist ops in 1968).
Satan must stick to the USA, since most of the rest of the world goes for 240V.
That's my favorite, of course (good old European politics)
Especially since the 120V standard dates back to Edison's Pearl Street installation of 1882. FWIW, 32V DC appliances were common in rural areas of the US (Windchargers), the last 50 Hz distribution in the US was phased out in 1948 (Southern California Edison), the last DC distribution was phased out very recently.
Slashdot Mirror
Sounds like you want the "kind" functionality of Fortran 90.
Back in the bad old days, a REAL in Fortran could be anywhere from 32 to 64 bits - a program that ran fine using REAL on a CDC-6600 (60 bits) might die horribly using REAL on an IBM 360 (32 bits but using hexadecimal arithemtic).
Anyone remember the article from a few months back about many CEO's being sociopaths?
The Cairo prject was M$'s attempt to finsih killing off OS/2 and kill off the various desktop UNIX distro's (HP had a nice candidate with the 900/712 with Lotus 123 and Ami-Pro running natively on HP-UX). Kind of thinking that the WinFS idea is like speech recognition (or Duke Nukem Forever) - remember reading Jerry Pournelle quoting Bill Godbout about the 80286 will be powerfull enough for speech recognition, this was ca 1982.
240Pu is a problem for making weapons, but not that big a deal for reactors.
Normal light-water plants have a 3 year fuel cyle, where 1/3rd of the fuel gets swapped out each year. The lifetime limits for breeder reactor fuel is based on the material damage from fast neutrons.
We've got plenty of fissionables (which include U-238 and Th-232), but the supply of fissiles (e.g U-235) is much more limited. Uranium is actually quite common, typical granite has about 1 gram per tonne. Anyway, the whole issue of limited supply of fissile material versus fissionable was what was behind the development of breeder reactors - with the integral fast reactor having some intriguing attributes (and significant engineering challenges).
Also TFA is inaccurate in talking about nuclear waste; the problem is not the U and Pu in spent fuel, which can be processed and reused, but thorium, radium, radon, and radioactive lead isotopes.
WTF are you talking about??? Thorium is naturally occuring, radium and radon are part of the natural decay chain of U-238, and the only way lead becomes radioactive is by activation (typically by neutrons). The radioactive waste from reactors consists of fission products and transuranics (plutonium, americium, curium, etc.) - fisson products typically have short half lifes, while the transuranics often have very long half-lifes.
Maybe a more appropriate analogy would be that of electric service in that utilities are rarely sized to handle all customers drawing their maximum rated power simultaneously - which would cause a meltdown of the transmission and distribution networks.
The telcos were having problems when dial-up service was becoming popular since the networks were designed for relatively brief calls.
One of the worst AUP's for broadband was Adelphia, that limited total bandwidth usage to 2.5GB a month - which could easily be reached with a dial-up connection with nightly 100MB downloads.
Would be nice to have a low power (say 5 to 10 watts) Sparc board to run Solaris - something like a Sparc equivalent to the Soekris boards.
After Ballmer stepped in, support for fringe platforms (i.e. not strictly PC-compatible) was pretty much dropped, up through 2.0, MS-DOS ran on quite a variety of 8086/8 boxes.
Now to think of it, MS dropping Xenix happened about this same time frame.
Actually, stopping Nazi's was a hard sell, the only reason the US declared war on Germany in WW2 was that Germany declared war first - and only after Roosevelt had goaded Hitler into declaring war first. The US public was in no mood to get involved with another war in Europe after the mess of our involvement in WW1 (which was probably a much larger mistake than getting involved in Iraq).
To get back on topic, the main reason the US was able to defeat Japan was that it started a major Naval construction program 1939-40, the Essex didn't come off the ways until 1943.
I've got an encyclopedia set titled "Above and Beyond" published in the late 60's showing the difference between a ramjet and scramjet. Work was being done by the Marquardt corp. The Nation Aerospace Plane (NASP, ca 1990) was supposed to be using an external scramjet.
Nevertheless, Heineman was still one heck of an airplane designer.
As far as the 8" floppy, ISTR that they were intended to replace punched cards, 77 tracks with 26 sectors (hard coded) came out to be pretty close to a box of 2000 hollerith cards (80 columns with 12 bits per column). 8" drives were available before the end of 1975, and the VAX came out in 1977(?). One of the uses for the flopies was loading the microprogram store on the VAX and IBM machines of the same era.
What was done with CP/M and 86-DOS, but not PC/MS-DOS, was distributing source code for the BIOS (called io.sys under 86-DOS). With that source code, it was relatively easy to support whatever sector size you wanted.
One thing that drives me up the friggin wall with Word and related word processors (Star/OpenOffice) is the way they handle graphics. I find it a lot easier to define a frame and then put in the graphic (where the frame defines whether the graphic will be cropped or re-sized to fit), instead of putting in a graphic, finding that it wildly out of scale with the rest of the document and then try to wrestle it into place without the @$%*&#!!! word processor deciding that it is going to re-arrange text to suit the eff'ing picture. What makes this worse is that 14 years ago, I was using a package that did it the right way and that package could run on a 25MHz 68040 with 16MB of memory.
On a similar vein, a former co-worker says he was much more productive typing out long documents on Lotus Manuscript than any currently available word processor - mainly because of the text interface.
I've got a copy of the "Tomorrowland" DVD released by Disney a couple of years ago - mainly stuff about space made in the 1950's. Especially enjoyed the one about Mars - since I first saw it in 1962 or 63. What was a bit weird was seeing Ward Kimball as he looked in the 1950's as compared to what he was like in person in the 1990's (and he was still quite a character).
One thing a lot of people forget is that the 1950's were a time of tremendous changes in day to day technology. In 1950, rockets just being sent to the lower reaches of space, A-bombs were to be delivered by long range bombers, electronics meant vacuum tubes, TV's were rare, computers were largely experimental, most trains were hauled by steam locomotives, airliners were piston powered, cars typically had flathead engines, 6V electrical systems and air-conditioning was only found on Cadillacs, and many cities still had streetcars. In 1960, satellites had been to the moon, H-bombs were to be delivered by ICBM's, electronics meant transistors, TV's were common (color TV's were rare but not unheard of), computers were comon for large businesses, the steam locomotive was essentially extinct, airliners were jets, flathead engines had disappeared (with the exception of the Continentals used in the Checkers), cars used 12V electrical systems, air-conditioning was a common option and streetcars were almost gone.
We may finally get the videophone that was originally envisioned in the 50's.
One of the things that kept it from happening earlier (that is ca 2000) was that most "broadband" connections had their upstream speeds capped to 128kbps - though many of the newer offerings have a much higher upstream speeds.
Reading your post got me to thinking about a couple of memories from the 1970's. The first was reading Mac's Service Shop column in the April (?) 1972 issue of PopTronics where they were talking about the brand new HP-35 and knowing the world was going to be a different place. The next was seeing the first West Coast Computer Faire in 1977 - the equipment was pretty primitive by todays standards, but there was an optimism of what the future might bring that has been gone for at least the last decade if not the last two decades - a slightly faster intel box running a slightly improved version of windoze isn't much to look forward to.
One of Sun's goals for Java was to have a write once, run anywhere environment. MS wanted to embrace and extend Java so that it would be very easy to write a Java program that ran only on Windows. The project files that VC spits out are "an abomination in the eyes of the Lord".
Hard to believe that UNIX came out of Berkeley in 1960's since the first tape from Bell Labs arrived in December 1973. I'm guessing the CS department wanted to have a replacement for Kronos that went away when the CDC-6400 'B' machine got shipped off in January 1973.
D Day One??? As in Duane Allen Day by any chance? (Apologies if I messed up the spelling.)
For the same friggin' reason they put the brain dead Posix inteface in Windows NT - so they could claim to meet the requirements of Posix compatibiliy without intending any serious use of it. Remember the fuss with Kerberos??
Glad to know that I'm not the only one thinking this would be a good idea. I would avoid the "Home Power" magazine approach of connecting the panels directly to the batteries - bad idea to have the batteries supplying short circuit current - and you should get significantly more power out of the panels by operating them at peak power operating point rather than forcing them to run at battery voltage.
The +5 mod points are well deserved.
Rotary converters??? The hell. How about the M-G sets from the old Milwaukee Electrification?? Especially the Janney substation with the 3MW M-G set from C.U.T. along with the three 1.5MW M-G sets. 'course most data centers would have a wee bit of trouble digesting 3.3kV...
You are on the right path in that it is easier to make a large AC/DC converter run efficiently than a bunch of small ones. In many ways you are descibing the old DC substations used in some older downtowns - where the rotary converters were feeding +/- 120V battery banks - talk about reliable power.
Neat link on the NYC subway system.
And the Poles did a darn good job of it. One minor correction - Poland was invaded by both Germany and the Soviet Union.
Perhaps the best work of pre-war cryptanalysis was Friedman's breaking of the Japanese diplomatic cipher (AKA the Purple Code) - without any idea of what the machine looked like.