The point was that the Ace was potentially a much more advanced design than the Ferranti/Manchester machine that gave rise to Leo. Although people like Newman and Blackett achieved quite a lot, I still think they were greatly handicapped by the failure of the Government and the Civil Service to make the most of the Turing design.
Let's not forget that Turing was unusual for his time in that he had practical skills as well as theoretical. He could actually machine the parts for relays and wire up electronics, at a time when mathematicians never got their hands dirty. (His time in the US, I am sure, contributed a lot to this.) His claim to be the father of software is based on his papers which actually discussed the whole organisation of a data processing center as well as the design of software itself, (before such things existed) and his early work with the Manchester computer, which involved advanced work into biological patterns. Since he had also been a lead consultant to the British Government in codebreaking in WW2 - not limited to Enigma by any means, but extending to voice encryption - he covered almost as many bases as John von Neumann.
It's a bit sickening that already posts on this thread are making gay-bashing remarks about him. The history of how he was discarded by the British Government, believed to be partly at the instigation of the US government, is a sad story of how intolerance helped the British lose their early lead in computing. If he had been born forty years later, he'd probably be running an equivalent of Apple,Oracle, Sun or Microsoft, and no-one would care about what he did in his spare time.
I don't have time to investigate this stuff, but I wonder how many of these patents are so-called design patents? A design patent is basically a way of restricting the use of a feature in a particular context. It enables someone to prevent you from making something which is basically a knock-off, in ways which are outside the realm of ordinary copyright.
It's a sad fact that the US patent system, which began as a way of protecting the small inventor from greedy corporations, is now a way for the greedy corporations to suppress the small inventor.
Correct me if I am wrong, but I thought the difference between US and EU patents was that US patents are backdated to the claimed date of invention while EU patents are based on date of filing. In the EU it should not be possible to patent any existing technology that is in the public domain - and that means all of OSS, by definition.
In the US it is all too possible for something to be well established prior art, but an inventor claims to have made the invention prior to the first date of open publication. Having been involved with both US and European patents until about 1995, I considered the US system to be deeply screwed - the opportunity for fraud is immense. (though yes, that didn't stop me from filing US patent applications...)The EU system should not be so bad.
If this still applies, the important thing is for all ideas and concepts being brought to the OSS table to be published as soon as possible after they arise, thus creating prior art even if it is only in a very buggy bit of code.
Of course, if the US gets the entire IP world to rely on "date of invention", we're all screwed, and I'm going to buy a farm and retire.
Is that actually SCO seems to have been of some use, on the basis of what doesn't kill you makes you stronger. They have:
Caused major corporates like IBM and HP to assess how important FOSS is to them, and act accordingly
Probably influenced the acquisition of SuSE by Novell, and helped commercialise Linux
Raised the profile of Linux in the world at large, by making business analysts realise there is something here that people will fight over
Probably speeded up improvements in the FOSS creation process, helping to ensure that its IP status is robust
Made a heroine of PJ
Given the entire IT industry a new hate figure to mold in wax and stick pins in, doubtless helping Bill G sleep better at nights of the full moon
Enriched several poor and deserving attorneys and helped to ensure that neither DaimlerChrysler nor Porsche have too many layoffs
Further educated some of the people who thought they understood the stock market
Darl McB deserves some sort of award for all this. With his remarkable combination of tact, diplomacy and tireless negotation, he should at least be made Ambassador to Iraq.
Vexatious litigants keep courts in business and contribute mightily to the revenue streams of lawyers. It is very hard indeed to have a litigant declared vexatious while he still has money (note this isn't a legal inclusive pronoun: it's usually a he). This isn't sarcasm, there are lawyers in my family.
I didn't say Shaw was right, I was making a comment about arrogant academics and how they are perceived by non-academics.
Dijkstra reminds me of my thermodynamics teacher, who used to make remarks like "you can't understand thermodynamics unless you begin with astrophysics." He was a Nobel prizewinner, but he was also a {censored because he's still alive and might conceivably read this}.
Edsger Dijkstra is all too typical of the arrogant academics who gave rise to Shaw's comment "Those who can, do, those who can't, teach. He's like the academic fanboys who argue that PostgreSQL is a real RDBMS, MySQL can't really be used for anything serious.
Don't believe it, kids. If your brain hasn't been ruined by age 7, you can unlearn any bad habits you pick up. His remark is of a stupidity level equal to "if you learn French at school, you won't be able to learn German."
As a matter of fact, not only did I once inherit a program that someone had written - well - on a BBC micro that was a pleasure to maintain, I once myself had to write a quick and dirty assembler for an obscure microprocessor in HP Basic, having no other resources available in a crisis. Despite which I have never once had the urge to use labels in C.
Well, that seems to be the general approach here...it's not happening, or it's natural, perhaps the poles are getting colder...nyaah,nyaah, I can't hear you.
The point is surely not that the Earth gets hotter and colder. I accept that (where I live I can look out the window and see some leftovers from the last glaciation or so.) Rather, it is that the heating up is very, very rapid in geological terms. During the 19th Century when the age of the Earth was realised, it was understood that natural processes were very slow. Now they are happening really rather fast, and the satellite data suggests it is faster than previously believed. There has been, in geological terms, a step change in atmospheric carbon dioxide, and a lagging step change in temperature. (as an aside why can't a geek site manage subscript and superscript? Step changes are usually bad news. I have just become a grandfather and I can't help contrasting when I was born into a post-WW2 world rather full of optimism despite McCarthy et al, and my granddaughter being born into a world where accelerating climate change, population migration, hydraulic, food and energy wars may be the norm. A load of/.ers announcing that everything is just fine does nothing for my peace of mind. You are the intelligent people, for the most part. If you aren't taking it seriously, what are the morons doing?
Is an Apple G5 GHz the same as an AMD Opteron or Athlong 64 GHz, assuming similar cache sizes? And is a G4 GHz the same as a Mobile Pentium GHz?
This seems closer to me to an apples to...well, not apples obviously...comparison.
I have this feeling that inside the P4 is really a small Maxwell's Demon, the whole GHz thing is irrelevant, it's how hard the electromagnet can be made to poke it to keep it working.
Seriously. The world seems to be going more and more down the route of the Roman Empire, in which a small class became fabulously rich and the vast majority were poor and acted as a reservoir to supply cheap labor and the Army. It's perhaps worth remembering that Julius Caesar was the chosen figurehead of the capitalist party (the industrialists or the Plebians) against the traditional agricultural/religious aristocracy, the patricians. The difference between Caesar and Bush was that Caesar was a man of enormous ability who dealt effectively with internal problems and foreign threats, but that's a side issue.
The problem for the Empire was that it gradually outsourced everything to the provinces - the grain supply (Egypt), mining, other agriculture. Like the US it imported the most able provincials and gave them citizenship to encourage them to support the system. But eventually the focus of power moved to the provinces, Rome itself became decadent (who needed to earn a proper living?) Even most of the army was recruited abroad. And the Empire collapsed. The remains of the Empire that survived - in Byzantium - was a statist civilisation in which capitalism was rigorously controlled, based around many small artisans and companies of very limited size, in which the Government interfered in production, distribution and exchange. Sound like anywhere?
Endless outsourcing may be capitalism, but what happens on the day when R&D is carried out abroad, manufacturing is all done abroad, the Internet, cheap broadcasting and cheaper film making has destroyed the US dominance in media, the US army is too small to control even a small dissident country (look at the problems posed by Iraq...we could kill everybody, but imagine the backlash), the rest of the world sees that the Emperor has no clothes, and the dollar collapses? Live off intellectual property? Can you imagine the rest of the world agreeing to observe US patents which frequently would not get through the assessment stage in European countries? At that point the super-rich will be sitting on piles of worthless dollars, and farming may look like the smart option again.
OK, it probably won't be that bad. But too much policy at the moment seems to be predicated on the idea that the US can control the rest of the world financially or militarily, and the example of Rome shows that unrestricted capitalism is likely to destroy the very factors that make that possible.
I believe that a lot of research is ongoing into this. As well as gold, there is the possibility of cleaning up heavy metal residues from industrial processes, and I believe that some nickel contaminated lakes have resulted in shellfish that can survive high nickel concentrations and accumulate nickel, providing a possible cleanup mechanism. I would have liked to have something like this at the back end of a nickel plating plant, instead of producing loads of contaminated, hard to treat sludge.
But then we (homo sap. sap.)are good at this: we can accumulate lead in our bones from drinking water or contaminated air, and I believe that mercury too can get collected by the body (gets resorbed in the lower intestine, I think.)
If true, this is more durable than tape or dvd, it's bootable, and I'm going to be able to Ghost my entire laptop to it. It's cheaper than the Snap server I've used up to now for backup. Backing up to 7 DVD+RW every week simply is not an efficient use of my time.
As for reliablility, we have an accounting system for which we still do a daily backup on 6 Zip discs (one per day). It works, why change it? After 6 years the original disks are still running, though the original SCSI ZIP has been replaced with USB. That original SCSI zip is still in use elsewhere for non-critical backup. I think that counts as a long term test.
My perception of Iomega drives is that a lot of people don't need them, but for some of us there is just no other solution that is near as good.
It was obviously written for settlers in the early 20th century and had all kinds of stuff on the different types of paddlewheels for different applications. By the sound of it you would need an undershot wheel with large buckets, unfortunately far from unobtrusive. Noise could be a major problem unless you used sucessive belt step-up drives rather than gearing, but the basic setup would need to resemble an automotive alternator system, which can produce a fairly constant output power despite fluctuations in rpm.
However, there would be many potential problems, especially the difficulty and cost of fixing a large overhung wheel with an asymmetric load over a river with fluctuating height (the wheel axis is going to need to rise and fall) and the regulatory problems: I guess you would need a license and it might be hard to obtain.
Another solution might be a hydraulic ram. There is the remains of one near where I live, that could raise water nearly 200ft. without an external power source, and was very simple and reliable. I guess some sort of license would be needed, but they are unobtrusive- there is nothing to see above water level but the exit pipe and the compression tank. Once the water is in a storage tank at high level, it can power a conventional turbine or an overshot wheel (more efficient than undershot), and the output can be adjusted to give fairly constant generator rpm regardless of load. Hydraulic rams can be noisy.
However, I wouldn't recommend going down either of these routes unless you are a qualified mechanical or civil (structural) engineer or both, and have good contacts in other disciplines.
The smallest hydro generator I have seen working, by the way, is at the end of the River Lyn in England. It's way bigger than you are likely to want ( I think I recall it's about 100KW) but when I was there in the early 90s it was still working. It attracts a lot of visitors from the US, and the whole place (including the water powered gravity railway) is a wonderful example of English quaintness.
The rover would have to move continuously to stay in the correct temperature zone. So you would need to know in advance that you could travel round a significant part of the circumference without encountering obstacles.
It would obviously have to stay in the dark because any level of sunshine would overheat it. So it would never see a sunrise or sunset. It would just crawl sadly around staying in the zone that current electronics and motors can handle (say -25 to 70C) until the batteries ran down (no solar power, you see.)
How much power are you using? I have a highly energy efficient house and the payback on photovoltaics is never - they would fail long before payback. If your payback is 6 years you are presumably paying approx. $5000 a year for electricity, assuming your 120% figure is correct. And how much have you allowed for maintenance? Your name isn't Charles Windsor by any chance?
However, I agree with you in part. Unfortunately it is hard to store electricity in any quantity and we therefore need sources which are not daylight dependent. A mixed approach of lots of small windfarms, photovoltaics, hydro, coppicing and steam generation, bioDiesel, enhanced insulation, and (heresy this) a few nuclear plants to provide a base load for trains and essential services could be a lot more robust overall than the present system.
Yes, a sufficiently powerful radar can have interesting effects especially if wires have become heated enough for the insulation to start to carbonise.
I would also suspect, knowing a bit about electrical grids, that there may be unexpected connections around and some of those wires may still be live.
I know of at least one case where a factory had an unusually high power load and the new manager determined to investigate. One evening after work he went round turning off the main switches one after another. One switch had no apparent effect, so he left it turned off. He then later heard that the golf club half a mile down the road had suddenly been totally deprived of power.
The Mac has traded on three things: Ease of use, looks, and low power consumption. In recent years, the low power consumption has come to the fore because of the benefits to portables.
The Powerbook is the VW Golf DTi (up to 150BHP, but at over 40mpg), the iBook is the Polo equivalent.
There is no BMW equivalent in the computer world (Compaq were the nearest thing till Ms. Fiorino came - perhaps she should be put in charge of Iraq?), but I'd suggest that Sony is roughly equivalent to Honda, and Toshiba to Toyota.
All analogies are suspect, but if you're at all interested in marketing you need to think about them sooner or later.
Actually, I've had a thought. The new generation of AMD64 portables should eventually have 64 bit Linux drivers for all the important stuff, including the gigabit Ethernet. Perhaps someone should start planning a walk in supercomputer creation based around machines like the eMachines 6805/6807 and the Acer 1501/1502. As they both have easily interchangeable hard disks, it should even be possible to release a suitable disk image in advance. And when the G5 Powerbook or whatever emerges, the opportunity for the ultimate archiecture head to head would present itself.
That's interesting and supports my suspicions. I've always assumed that, when push comes to shove, IBM has more and better resources to throw at architectures for solving serious problems than Intel. Intel now seems to be backtracking away from the P4 towards the derivatives of the PIII architecture despite its two competitors clocking around 2GHz (I'm counting the G5 and AMD64 as the competitors).
Although a G5,P4 etc. can theoretically do several GFlops, AFAIUI they cannot reach a sustained GFlop on a large scale problem because there is insufficient cache memory. The peak speed across the bus is only a few Gbytes/sec, and to sustain a GFlop on a large problem could mean a maximum data rate of 30Gbytes/sec (2 80-bit reads and 1 80-bit write per flop)and a minimum in excess of 10.
And that's before taking network and hard disk throttling into account.
Has anyone done any work on the actual sustainable processing rate for large data sets using currently available operating systems and hardware? Is Linpack actually representative? Forget all those graphics-bound melons-to-potatoes "benchmarks".
Slashdot Mirror
The point was that the Ace was potentially a much more advanced design than the Ferranti/Manchester machine that gave rise to Leo. Although people like Newman and Blackett achieved quite a lot, I still think they were greatly handicapped by the failure of the Government and the Civil Service to make the most of the Turing design.
It's a bit sickening that already posts on this thread are making gay-bashing remarks about him. The history of how he was discarded by the British Government, believed to be partly at the instigation of the US government, is a sad story of how intolerance helped the British lose their early lead in computing. If he had been born forty years later, he'd probably be running an equivalent of Apple,Oracle, Sun or Microsoft, and no-one would care about what he did in his spare time.
It's a sad fact that the US patent system, which began as a way of protecting the small inventor from greedy corporations, is now a way for the greedy corporations to suppress the small inventor.
In the US it is all too possible for something to be well established prior art, but an inventor claims to have made the invention prior to the first date of open publication. Having been involved with both US and European patents until about 1995, I considered the US system to be deeply screwed - the opportunity for fraud is immense. (though yes, that didn't stop me from filing US patent applications...)The EU system should not be so bad.
If this still applies, the important thing is for all ideas and concepts being brought to the OSS table to be published as soon as possible after they arise, thus creating prior art even if it is only in a very buggy bit of code.
Of course, if the US gets the entire IP world to rely on "date of invention", we're all screwed, and I'm going to buy a farm and retire.
- Caused major corporates like IBM and HP to assess how important FOSS is to them, and act accordingly
- Probably influenced the acquisition of SuSE by Novell, and helped commercialise Linux
- Raised the profile of Linux in the world at large, by making business analysts realise there is something here that people will fight over
- Probably speeded up improvements in the FOSS creation process, helping to ensure that its IP status is robust
- Made a heroine of PJ
- Given the entire IT industry a new hate figure to mold in wax and stick pins in, doubtless helping Bill G sleep better at nights of the full moon
- Enriched several poor and deserving attorneys and helped to ensure that neither DaimlerChrysler nor Porsche have too many layoffs
- Further educated some of the people who thought they understood the stock market
Darl McB deserves some sort of award for all this. With his remarkable combination of tact, diplomacy and tireless negotation, he should at least be made Ambassador to Iraq.Vexatious litigants keep courts in business and contribute mightily to the revenue streams of lawyers. It is very hard indeed to have a litigant declared vexatious while he still has money (note this isn't a legal inclusive pronoun: it's usually a he).
This isn't sarcasm, there are lawyers in my family.
Dijkstra reminds me of my thermodynamics teacher, who used to make remarks like "you can't understand thermodynamics unless you begin with astrophysics." He was a Nobel prizewinner, but he was also a {censored because he's still alive and might conceivably read this}.
Don't believe it, kids. If your brain hasn't been ruined by age 7, you can unlearn any bad habits you pick up. His remark is of a stupidity level equal to "if you learn French at school, you won't be able to learn German."
As a matter of fact, not only did I once inherit a program that someone had written - well - on a BBC micro that was a pleasure to maintain, I once myself had to write a quick and dirty assembler for an obscure microprocessor in HP Basic, having no other resources available in a crisis. Despite which I have never once had the urge to use labels in C.
The point is surely not that the Earth gets hotter and colder. I accept that (where I live I can look out the window and see some leftovers from the last glaciation or so.) /.ers announcing that everything is just fine does nothing for my peace of mind. You are the intelligent people, for the most part. If you aren't taking it seriously, what are the morons doing?
Rather, it is that the heating up is very, very rapid in geological terms. During the 19th Century when the age of the Earth was realised, it was understood that natural processes were very slow. Now they are happening really rather fast, and the satellite data suggests it is faster than previously believed. There has been, in geological terms, a step change in atmospheric carbon dioxide, and a lagging step change in temperature. (as an aside why can't a geek site manage subscript and superscript? Step changes are usually bad news. I have just become a grandfather and I can't help contrasting when I was born into a post-WW2 world rather full of optimism despite McCarthy et al, and my granddaughter being born into a world where accelerating climate change, population migration, hydraulic, food and energy wars may be the norm. A load of
This seems closer to me to an apples to ...well, not apples obviously...comparison.
I have this feeling that inside the P4 is really a small Maxwell's Demon, the whole GHz thing is irrelevant, it's how hard the electromagnet can be made to poke it to keep it working.
The problem for the Empire was that it gradually outsourced everything to the provinces - the grain supply (Egypt), mining, other agriculture. Like the US it imported the most able provincials and gave them citizenship to encourage them to support the system. But eventually the focus of power moved to the provinces, Rome itself became decadent (who needed to earn a proper living?) Even most of the army was recruited abroad. And the Empire collapsed. The remains of the Empire that survived - in Byzantium - was a statist civilisation in which capitalism was rigorously controlled, based around many small artisans and companies of very limited size, in which the Government interfered in production, distribution and exchange. Sound like anywhere?
Endless outsourcing may be capitalism, but what happens on the day when R&D is carried out abroad, manufacturing is all done abroad, the Internet, cheap broadcasting and cheaper film making has destroyed the US dominance in media, the US army is too small to control even a small dissident country (look at the problems posed by Iraq...we could kill everybody, but imagine the backlash), the rest of the world sees that the Emperor has no clothes, and the dollar collapses?
Live off intellectual property? Can you imagine the rest of the world agreeing to observe US patents which frequently would not get through the assessment stage in European countries?
At that point the super-rich will be sitting on piles of worthless dollars, and farming may look like the smart option again.
OK, it probably won't be that bad. But too much policy at the moment seems to be predicated on the idea that the US can control the rest of the world financially or militarily, and the example of Rome shows that unrestricted capitalism is likely to destroy the very factors that make that possible.
But then we (homo sap. sap.)are good at this: we can accumulate lead in our bones from drinking water or contaminated air, and I believe that mercury too can get collected by the body (gets resorbed in the lower intestine, I think.)
As for reliablility, we have an accounting system for which we still do a daily backup on 6 Zip discs (one per day). It works, why change it? After 6 years the original disks are still running, though the original SCSI ZIP has been replaced with USB. That original SCSI zip is still in use elsewhere for non-critical backup. I think that counts as a long term test.
My perception of Iomega drives is that a lot of people don't need them, but for some of us there is just no other solution that is near as good.
However, there would be many potential problems, especially the difficulty and cost of fixing a large overhung wheel with an asymmetric load over a river with fluctuating height (the wheel axis is going to need to rise and fall) and the regulatory problems: I guess you would need a license and it might be hard to obtain.
Another solution might be a hydraulic ram. There is the remains of one near where I live, that could raise water nearly 200ft. without an external power source, and was very simple and reliable. I guess some sort of license would be needed, but they are unobtrusive- there is nothing to see above water level but the exit pipe and the compression tank. Once the water is in a storage tank at high level, it can power a conventional turbine or an overshot wheel (more efficient than undershot), and the output can be adjusted to give fairly constant generator rpm regardless of load. Hydraulic rams can be noisy.
However, I wouldn't recommend going down either of these routes unless you are a qualified mechanical or civil (structural) engineer or both, and have good contacts in other disciplines.
The smallest hydro generator I have seen working, by the way, is at the end of the River Lyn in England. It's way bigger than you are likely to want ( I think I recall it's about 100KW) but when I was there in the early 90s it was still working. It attracts a lot of visitors from the US, and the whole place (including the water powered gravity railway) is a wonderful example of English quaintness.
The rover would have to move continuously to stay in the correct temperature zone. So you would need to know in advance that you could travel round a significant part of the circumference without encountering obstacles.
It would obviously have to stay in the dark because any level of sunshine would overheat it. So it would never see a sunrise or sunset. It would just crawl sadly around staying in the zone that current electronics and motors can handle (say -25 to 70C) until the batteries ran down (no solar power, you see.)
Who modded this interesting?
All I need now is the 500KW carbon dioxide laser that will allow me to take appropriate action.
However, I agree with you in part. Unfortunately it is hard to store electricity in any quantity and we therefore need sources which are not daylight dependent. A mixed approach of lots of small windfarms, photovoltaics, hydro, coppicing and steam generation, bioDiesel, enhanced insulation, and (heresy this) a few nuclear plants to provide a base load for trains and essential services could be a lot more robust overall than the present system.
No energy technology supported by a UK government and reported on the internet will ever produce more power than was consumed in publicising it.
Corollary: No energy technology will be supported by a US government unless it can (a)power an SUV and (b) create explosions.
I would also suspect, knowing a bit about electrical grids, that there may be unexpected connections around and some of those wires may still be live.
I know of at least one case where a factory had an unusually high power load and the new manager determined to investigate. One evening after work he went round turning off the main switches one after another. One switch had no apparent effect, so he left it turned off. He then later heard that the golf club half a mile down the road had suddenly been totally deprived of power.
The Powerbook is the VW Golf DTi (up to 150BHP, but at over 40mpg), the iBook is the Polo equivalent.
There is no BMW equivalent in the computer world (Compaq were the nearest thing till Ms. Fiorino came - perhaps she should be put in charge of Iraq?), but I'd suggest that Sony is roughly equivalent to Honda, and Toshiba to Toyota.
All analogies are suspect, but if you're at all interested in marketing you need to think about them sooner or later.
And yes, 64 bit FLOPS. I think I had a senior moment there.
Actually, I've had a thought. The new generation of AMD64 portables should eventually have 64 bit Linux drivers for all the important stuff, including the gigabit Ethernet. Perhaps someone should start planning a walk in supercomputer creation based around machines like the eMachines 6805/6807 and the Acer 1501/1502.
As they both have easily interchangeable hard disks, it should even be possible to release a suitable disk image in advance.
And when the G5 Powerbook or whatever emerges, the opportunity for the ultimate archiecture head to head would present itself.
Anyway, thanks for the info.
And that's before taking network and hard disk throttling into account.
Has anyone done any work on the actual sustainable processing rate for large data sets using currently available operating systems and hardware?
Is Linpack actually representative? Forget all those graphics-bound melons-to-potatoes "benchmarks".