You're defining a PM machine by excluding the 'machine' bit. Part of the definition of a machine is that it has to do Work (technical definition - The transfer of energy from one physical system to another).
A body set spinning on a (somewhat miraculous) journey along an isopotential of gravitational force in the universe will continue spinning for eternity (or thereabouts. The universe might collapse...)
The spinning body's still not a perpertual motion machine because it doesn't interact, and should it ever interact, it'll be subject to the laws of motion and thermodynamics and still not be a perpetual motion machine.
I wonder which way (physics or perpetual motion) your interest swings. The Carnot cycle places a limit on even ideal circumstances for the thermodynamic production of Work from Energy.
For a typical steam power plant, (800K hot, 300K cold), the maximum theoretically possible efficiency is ~60% for a 100% reversible reaction (hint: these don't exist in power plants). I seriously doubt it is possible get anywhere near 98% efficient without some new ground-breaking physics in the same vein as Newton -> Einstein.
My hardware is a G400 (I like the dual display) and Athlon 1800+. I get virtually identical results. At first I assumed (because a G400 shouldn't be anywhere near a Radeon!) that it was CPU-bound, but then I tried a remote-invocation as well, and got virtually the same results (see below). There's a 100mbit link between the two machines, if that matters...
This implies to me that there is a limitation on the number of requests per second that the X-server (irrespective of driver) can do, and that perhaps should be addressed. Either that, or a G400 really is the same speed as radeon 8500... The link to 'tanelorn' is via ssh as well (so it's encrypting and decrypting everything in the protocol stream...)
[simon@atlantis ~]$ x11perf -eschertilerect500 x11perf - X11 performance program, version 1.5 The XFree86 Project, Inc server version 40200000 on:0.0 from atlantis.mythology.gornall.net Sat Mar 22 11:03:29 2003
[root@tanelorn denyaccess]# x11perf -eschertilerect500 x11perf - X11 performance program, version 1.5 The XFree86 Project, Inc server version 40200000 on localhost:11.0 from tanelorn.mythology.gornall.net Sun Mar 23 11:10:28 2003
One other thing. I *like* X. I really couldn't live without the network transparency - editing files in co-located facilities via ssh with X-forwarding is just *so* much nicer than using 'vi'...
We host a fair few (60 or so) financial-orientated websites, with an average query-level of some 10-20 queries per second on the database supporting those sites.
We have an 8-way cluster of machines to support this (way-overkill for the most part) but recently, we've been (almost) hitting limits... The apache service has logged peaks of 1000 connections/second, with the DB query-level going as high as 70,000/second....
I'm actually fairly happy that the system can more-or-less cope with the load, but nonetheless, I want to make sure (or at least as-sure-as-possible) that we can't be easily DOS'd, so this weekend I'll be writing an Apache module to monitor the number-of-connections-per-second on an IP-by-IP basis, and take a decision to run a script depending on thresholds....
I think stateful firewalls could probably manage it but for historical reasons we're stuck with what we have, and having apache call a bandwidth-limiting script on an IP address that's registered 5000 hits in the last minute (for example) seems reasonable:-)
If there's something that can do this already, I'd like to know - I've found (ntal), but running a script per packet doesn't appeal:-( I prefer the idea of hitting a limit in Apache that triggers a script that limits access (dynamic firewalls)
Herein lies the difference. My browser window always pops up with (I think it's about:blank or something similar) as its' initial URL - anyway the URL area is empty.
If I want to overwrite existing text, I double-click the offending URL and hit delete first, but in general I don't need to do this.
I use linux as my desktop at home and at work, I'm a programmer by trade. Since we all check in and out via CVS, it's up to me which OS I use. Linux makes my life a lot easier:-)
I have a double-headed machine, normally with (say) 5 or 6 editor windows open - cutting and pasting is a common task, and it's just so much easier to do it all with the mouse, for me, at least. I guess everyone has their own way....
Maybe it's just me, but I *like* the X selection system. A windows-bound colleague of mine the other day exclaimed in jealousy when she saw me cutting and pasting between editors... Windows is just so damn clumsy.
To answer your question, I
1) Click on the browser desktop-icon to open a browser window 2) Highlight my URL with the 1st mouse button 3) Position my mouse over the URL bar, and click the middle button 4) There is no (4)
Well, yes, the data has to be presented, but not necessarily sent out of the FPGA - this is the basis of the 'Platform FPGA' architecture. As long as you can realise your processing within the FPGA, your output may be of significantly less bandwidth.
Pins aren't that limited either - you can get an FPGA with well over 1000 i/o pins, 168 1-clock-cycle 18-bit multipliers in hardware, and as many as will fit in firmware - with 8M "gates", that's a pretty nifty piece of hardware.
Not that you would, but clocking 512 pins at 200MHz gives you 12.8 GByte/sec both in and out...
The way you'd more likely use it though is to embed some of the processing subsequent to the FFT onboard as well, to reduce the output overhead - say you're looking for zero-crossing, or correlation, or anything where the output bandwidth is significantly less than the input (most signal processing tasks are...)
I thought about doing this for my own radio telescope. One day I might just get around to it:-)
The interesting thing about using an FPGA would be the speed of the FFT. With FFTW (the Fastest Fourier Transform in the West), I get approx 230 us (microseconds) per fft. Using an el-cheapo FFT, best efforts would be on the order of 15us, or approximately 15x faster... Bung several on a few PCBs, and you're talking super-computer speeds:-)
Say you use 16 FPGA implementations, that'd be the equivalent of 240 Athlon 1800XP's... With those sorts of speeds, you could do realtime chirp analysis for doppler effects on an incoming signal. That *would be cool*:-)
You're missing the point. You don't try to rewrite a P4 in an FPGA.
You can (for example) do a 4096 point FFT in a few milliseconds on an el-cheapo FPGA (20$ for 1-off pricing) whereas to get almost 1/10th the performance, you could buy a $400 P4/Athlon. Hmmm $20 vs $4000...
Horses for courses - the FPGA is a very low-barrier-to-entry solution, but it's not a general-purpose device - the speed tops out around about 400MHz anyway (and that's not a $20 FPGA!)
I find it interesting that you refer to the EU taking a stand against American protectionism. As an example, check out French farm subsidies. Pretty interesting comparison.
I think there's a world (no pun intended:-) of difference between an internal market and an external market. The EU is doing more or less what EU-people expect if they "protect" EU-people from external states.
The (horrendously large) french farm subsidies are an internal matter for the EU to sort out, and mainly came from the way in which the EU was set up, with Britain excluded from the EU until "appropriate" safeguards had been made for French farmers.
[History, as far as I recall]
Britain had a far more efficient farming style, wanted to join the EU => France was scared, so as anEU member France lobbied for EU subsidies as a condition for UK entrance. The UK eventually agreed that there were still sufficiently large advantages to be had by membership, and reduced its' original proposed EU payments as well. Britain entered the EU, and France kept their farmers employed.
[/History]
The ideal would be to wean people off subsidies, but I still see the above as the EU "protecting" member states (in this case, France) from external interests (in this case, the UK) . Whether I agree with the subsidies or not isn't really relevant...
As for poverty being the root cause of India & China's resurgence, I don't doubt it's an economic argument that's the cause of the dilemma. I was trying to point out that a practice is being established... Any innate industry feeds from its market, and if the market disappears, so does the industry...
The EU-15 zone is currently some 378.5 million people. There are another 10 countries currently under consideration for membership which will add a further 74.5 million people. This would make the US approximately 60% of the size.
Even Britain (approx the size of an averaged USA state) has some 59 million people...
Seriously, this will not sit well with American companies. It will not be allowed.
Actually, if the USA take that attitude with the EU, they're likely to be sent home with a flea in their ear. The EU doesn't have much of a history of bowing to American protectionism, witness the impending steel trade-war...
There are more people under EU law than there are under US law, and the EU is just starting to flex its' muscles a bit more. Negotiation is the key for getting your own way, either for member states or those outside the boundaries. Trying to impose a solution (by anyone, even founder member-states) is becoming more and more difficult.
Has anyone else noticed that plain 'ole numbers are becoming more important over time ? China and India are being cited as the future powerhouses of global commerce; the US and (to a lesser extent) the EU are outsourcing huge chunks of what would have been bread-and-butter work to external countries, etc. Maybe EU expansion isn't such a bad idea after all... Perhaps it'll be Russia next:-)
They do not (and won't for some time) run anywhere near current CPU speeds. 200MHz in the sort of FPGA he's talking about takes very good knowledge of the internal FPGA architecture, and excellent HDL skills. There's a top end limit (not sure where it is, but it's about 3-400MHz in an FPGA that cost $2k)
There is no way you'll be reconfiguring your CPU any time soon, but having an FPGA resource on-hand would be useful. Reconfiguring for sorenson/MPEG/M-JPEG would be a neat trick.
FPGA's get their speed not from the clock as much as from their inherent parallelism - you can run each of the units slower, but have multiple units where it may not make sense in the general case.
Simon
Re:It can be slowed down... perhaps
on
Cringely on P2P
·
· Score: 5, Insightful
At which point, you adopt a "spread-spectrum" approach to the data transmission. Chop each item up algorithmically into N blocks (so the split points can be determinable and reproducible across multiple servers), append metadata to the end of each block saying how to get the next from this, and encrypt each block with a key from the previous one. Use changing ports and servers (if it's a true P2P system) for access to each block.
The ISP filtering s/w would have to be *damn* good:-)
This doesn't cope with the blocking issue, so the "obvious" thing to do is to coerce the great unwashed into an involuntary P2P network using virus technology to steal bandwidth (disk & net).
There'd be no nasty virus payload (the authors would want the machines to be operating smoothly). The virus might even patch and protect against other virii just to keep it only infected with the P2P s/w!
If the virus can infect (ooh, say, IIS) then it could use HTTP as a transport without affecting normal behaviour.
It's coming, or something like it. It's just a matter of time before the arms race really kicks in.
Or then again, perhaps I've missed something obvious - it's very late over here in the UK:-)
In many respects, a simpler culture is far more resilient than a complex one. Increasing the complexity is a bit like walking up a mountain - the safe area to walk on gets smaller the higher you go up. Consider the magnetic flip to be a huge hand reaching down and translocating you a distance horizontally... would you prefer to be higher up the mountain (wheeeeee.....ouch) or farther down ?
In (slightly) more scientific terms, the advances we've made since those cavemen times are built on the premise of incremental change - we talk of "advances", ie: building on the past to get farther. Take away the foundations (communications is the major one, I guess, direction finding, etc.) and see how well everything that depends on them copes. Consider how an economy might react to (for example: the collapse of air traffic), and the subsequent secondary effects. None of this was even slightly worrying to the caveman, but our world is immensely dependent on excellent long-distance communications.
Yes, we have a far and away more complex civilisation than a caveman ever dreamed of. This is a weakness, not a strength. The payoff comes from what we can do with that technology, but if you remove that, you end up with a lot of hungry people in a small space...
I concur with the physics, btw, but you're really overestimating the resilience of our civilisation.
"There's been too much theology and not enough economic analysis in the debate so far," said Bradford L. Smith, Microsoft's general counsel, who oversees the company's global lobbying tea
I always thought you Americans took that Boston thing far too seriously:-)
Let's see, they've already transformed 10,000 machines which were previously enslaved to the windows drum over to the free'n'easy linux beat:-)
They've got another 100,000 scheduled for next year. That drum's just going to get louder, and louder, and louder. Can you hear it yet ? What you are hearing, ladies and gentlemen, is the hammering-in of the thin end of the wedge, and I for one can't wait for that wedge to grow.
Windows is the him-use-deep-magick-solve-problem approach, an oligarchy of high priests results with the local priests doling out consolences (note: not solutions...) from above in return for bloody coin.
Linux is a meritocracy, where librarians are shown their due worth, knowledge is open to all, and the only currency you need spend is time, the only fear you need have is looking stupid when asking beginner questions. Even then, you are mostly treated well because of the "There but for the grace of [insert deity] go I" mentality.
No, I'm not a librarian, but I much prefer the latter over the former:-)
You might be able to do that as well - depends whether you can get Xinerama to work. I'd guess you currently have a:0 and:1 display, thus you can't drag apps. If you run xinerama, you get a single merged:0 display, and can drag to your heart's content...
[simon@atlantis com.saltsw.pto]$ xdpyinfo | grep dimension
dimensions: 2560x1024 pixels (644x241 millimeters)... which is 2x 1280x1024 displays, next to each other. Makes a great development platform:-)
Slashdot Mirror
You're defining a PM machine by excluding the 'machine' bit. Part of the definition of a machine is that it has to do Work (technical definition - The transfer of energy from one physical system to another).
A body set spinning on a (somewhat miraculous) journey along an isopotential of gravitational force in the universe will continue spinning for eternity (or thereabouts. The universe might collapse...)
The spinning body's still not a perpertual motion machine because it doesn't interact, and should it ever interact, it'll be subject to the laws of motion and thermodynamics and still not be a perpetual motion machine.
Simon.
I wonder which way (physics or perpetual motion) your interest swings. The Carnot cycle places a limit on even ideal circumstances for the thermodynamic production of Work from Energy.
For a typical steam power plant, (800K hot, 300K cold), the maximum theoretically possible efficiency is ~60% for a 100% reversible reaction (hint: these don't exist in power plants). I seriously doubt it is possible get anywhere near 98% efficient without some new ground-breaking physics in the same vein as Newton -> Einstein.
Simon.
What ?
For the last year or so, the US has been averaging about 5.7% unemployment. The UK has been averaging about 5% for the last 2 years.
ATB,
Simon
My hardware is a G400 (I like the dual display) and Athlon 1800+. I get virtually identical results. At first I assumed (because a G400 shouldn't be anywhere near a Radeon!) that it was CPU-bound, but then I tried a remote-invocation as well, and got virtually the same results (see below). There's a 100mbit link between the two machines, if that matters...
:0.0
This implies to me that there is a limitation on the number of requests per second that the X-server (irrespective of driver) can do, and that perhaps should be addressed. Either that, or a G400 really is the same speed as radeon 8500... The link to 'tanelorn' is via ssh as well (so it's encrypting and decrypting everything in the protocol stream...)
[simon@atlantis ~]$ x11perf -eschertilerect500
x11perf - X11 performance program, version 1.5
The XFree86 Project, Inc server version 40200000 on
from atlantis.mythology.gornall.net
Sat Mar 22 11:03:29 2003
Sync time adjustment is 0.0537 msecs.
8000 reps @ 0.6795 msec ( 1470.0/sec): 500x500 tiled rectangle (216x208 tile)
8000 reps @ 0.6850 msec ( 1460.0/sec): 500x500 tiled rectangle (216x208 tile)
8000 reps @ 0.6794 msec ( 1470.0/sec): 500x500 tiled rectangle (216x208 tile)
8000 reps @ 0.7947 msec ( 1260.0/sec): 500x500 tiled rectangle (216x208 tile)
8000 reps @ 0.6966 msec ( 1440.0/sec): 500x500 tiled rectangle (216x208 tile)
40000 trep @ 0.7070 msec ( 1410.0/sec): 500x500 tiled rectangle (216x208 tile)
[root@tanelorn denyaccess]# x11perf -eschertilerect500
x11perf - X11 performance program, version 1.5
The XFree86 Project, Inc server version 40200000 on localhost:11.0
from tanelorn.mythology.gornall.net
Sun Mar 23 11:10:28 2003
Sync time adjustment is 1.1608 msecs.
8000 reps @ 0.6954 msec ( 1440.0/sec): 500x500 tiled rectangle (216x208 tile)
8000 reps @ 0.7964 msec ( 1260.0/sec): 500x500 tiled rectangle (216x208 tile)
8000 reps @ 0.6852 msec ( 1460.0/sec): 500x500 tiled rectangle (216x208 tile)
8000 reps @ 0.6833 msec ( 1460.0/sec): 500x500 tiled rectangle (216x208 tile)
8000 reps @ 0.6844 msec ( 1460.0/sec): 500x500 tiled rectangle (216x208 tile)
40000 trep @ 0.7089 msec ( 1410.0/sec): 500x500 tiled rectangle (216x208 tile)
One other thing. I *like* X. I really couldn't live without the network transparency - editing files in co-located facilities via ssh with X-forwarding is just *so* much nicer than using 'vi'...
Simon.
We host a fair few (60 or so) financial-orientated websites, with an average query-level of some 10-20 queries per second on the database supporting those sites.
:-)
:-( I prefer the idea of hitting a limit in Apache that triggers a script that limits access (dynamic firewalls)
:-)
We have an 8-way cluster of machines to support this (way-overkill for the most part) but recently, we've been (almost) hitting limits... The apache service has logged peaks of 1000 connections/second, with the DB query-level going as high as 70,000/second....
I'm actually fairly happy that the system can more-or-less cope with the load, but nonetheless, I want to make sure (or at least as-sure-as-possible) that we can't be easily DOS'd, so this weekend I'll be writing an Apache module to monitor the number-of-connections-per-second on an IP-by-IP basis, and take a decision to run a script depending on thresholds....
I think stateful firewalls could probably manage it but for historical reasons we're stuck with what we have, and having apache call a bandwidth-limiting script on an IP address that's registered 5000 hits in the last minute (for example) seems reasonable
If there's something that can do this already, I'd like to know - I've found (ntal), but running a script per packet doesn't appeal
Ideas gratefully received
Simon
Herein lies the difference. My browser window always pops up with (I think it's about:blank or something similar) as its' initial URL - anyway the URL area is empty.
:-)
If I want to overwrite existing text, I double-click the offending URL and hit delete first, but in general I don't need to do this.
I use linux as my desktop at home and at work, I'm a programmer by trade. Since we all check in and out via CVS, it's up to me which OS I use. Linux makes my life a lot easier
I have a double-headed machine, normally with (say) 5 or 6 editor windows open - cutting and pasting is a common task, and it's just so much easier to do it all with the mouse, for me, at least. I guess everyone has their own way....
Simon.
Maybe it's just me, but I *like* the X selection system. A windows-bound colleague of mine the other day exclaimed in jealousy when she saw me cutting and pasting between editors... Windows is just so damn clumsy.
To answer your question, I
1) Click on the browser desktop-icon to open a browser window
2) Highlight my URL with the 1st mouse button
3) Position my mouse over the URL bar, and click the middle button
4) There is no (4)
Simple, yes ?
Simon
Well, yes, the data has to be presented, but not necessarily sent out of the FPGA - this is the basis of the 'Platform FPGA' architecture. As long as you can realise your processing within the FPGA, your output may be of significantly less bandwidth.
Pins aren't that limited either - you can get an FPGA with well over 1000 i/o pins, 168 1-clock-cycle 18-bit multipliers in hardware, and as many as will fit in firmware - with 8M "gates", that's a pretty nifty piece of hardware.
Not that you would, but clocking 512 pins at 200MHz gives you 12.8 GByte/sec both in and out...
The way you'd more likely use it though is to embed some of the processing subsequent to the FFT onboard as well, to reduce the output overhead - say you're looking for zero-crossing, or correlation, or anything where the output bandwidth is significantly less than the input (most signal processing tasks are...)
Simon.
I of course meant microseconds (us) not milliseconds (ms).
:-)
MS strikes again
Simon.
I thought about doing this for my own radio telescope. One day I might just get around to it :-)
:-)
:-)
The interesting thing about using an FPGA would be the speed of the FFT. With FFTW (the Fastest Fourier Transform in the West), I get approx 230 us (microseconds) per fft. Using an el-cheapo FFT, best efforts would be on the order of 15us, or approximately 15x faster... Bung several on a few PCBs, and you're talking super-computer speeds
Say you use 16 FPGA implementations, that'd be the equivalent of 240 Athlon 1800XP's... With those sorts of speeds, you could do realtime chirp analysis for doppler effects on an incoming signal. That *would be cool*
Simon.
You're missing the point. You don't try to rewrite a P4 in an FPGA.
You can (for example) do a 4096 point FFT in a few milliseconds on an el-cheapo FPGA (20$ for 1-off pricing) whereas to get almost 1/10th the performance, you could buy a $400 P4/Athlon. Hmmm $20 vs $4000...
Horses for courses - the FPGA is a very low-barrier-to-entry solution, but it's not a general-purpose device - the speed tops out around about 400MHz anyway (and that's not a $20 FPGA!)
Simon.
I think there's a world (no pun intended :-) of difference between an internal market and an external market. The EU is doing more or less what EU-people expect if they "protect" EU-people from external states.
The (horrendously large) french farm subsidies are an internal matter for the EU to sort out, and mainly came from the way in which the EU was set up, with Britain excluded from the EU until "appropriate" safeguards had been made for French farmers.
[History, as far as I recall]
Britain had a far more efficient farming style, wanted to join the EU => France was scared, so as anEU member France lobbied for EU subsidies as a condition for UK entrance. The UK eventually agreed that there were still sufficiently large advantages to be had by membership, and reduced its' original proposed EU payments as well. Britain entered the EU, and France kept their farmers employed.
[/History]
The ideal would be to wean people off subsidies, but I still see the above as the EU "protecting" member states (in this case, France) from external interests (in this case, the UK) . Whether I agree with the subsidies or not isn't really relevant...
As for poverty being the root cause of India & China's resurgence, I don't doubt it's an economic argument that's the cause of the dilemma. I was trying to point out that a practice is being established... Any innate industry feeds from its market, and if the market disappears, so does the industry...
Simon.
Try looking at The EU official website
The EU-15 zone is currently some 378.5 million people. There are another 10 countries currently under consideration for membership which will add a further 74.5 million people. This would make the US approximately 60% of the size.
Even Britain (approx the size of an averaged USA state) has some 59 million people...
Simon.
There are more people under EU law than there are under US law, and the EU is just starting to flex its' muscles a bit more. Negotiation is the key for getting your own way, either for member states or those outside the boundaries. Trying to impose a solution (by anyone, even founder member-states) is becoming more and more difficult.
Has anyone else noticed that plain 'ole numbers are becoming more important over time ? China and India are being cited as the future powerhouses of global commerce; the US and (to a lesser extent) the EU are outsourcing huge chunks of what would have been bread-and-butter work to external countries, etc. Maybe EU expansion isn't such a bad idea after all... Perhaps it'll be Russia next :-)
Simon
Right. You're an astrophysicist who doesn't know the word extraterrestrial. Of course you are.
Simon.
I wouldn't like to be behind the driver when that thing went off :-)
Simon
MM(F)V
Simon.
Yes, they're cool.
They do not (and won't for some time) run anywhere near current CPU speeds. 200MHz in the sort of FPGA he's talking about takes very good knowledge of the internal FPGA architecture, and excellent HDL skills. There's a top end limit (not sure where it is, but it's about 3-400MHz in an FPGA that cost $2k)
There is no way you'll be reconfiguring your CPU any time soon, but having an FPGA resource on-hand would be useful. Reconfiguring for sorenson/MPEG/M-JPEG would be a neat trick.
FPGA's get their speed not from the clock as much as from their inherent parallelism - you can run each of the units slower, but have multiple units where it may not make sense in the general case.
Simon
At which point, you adopt a "spread-spectrum" approach to the data transmission. Chop each item up algorithmically into N blocks (so the split points can be determinable and reproducible across multiple servers), append metadata to the end of each block saying how to get the next from this, and encrypt each block with a key from the previous one. Use changing ports and servers (if it's a true P2P system) for access to each block.
:-)
:-)
The ISP filtering s/w would have to be *damn* good
This doesn't cope with the blocking issue, so the "obvious" thing to do is to coerce the great unwashed into an involuntary P2P network using virus technology to steal bandwidth (disk & net).
There'd be no nasty virus payload (the authors would want the machines to be operating smoothly). The virus might even patch and protect against other virii just to keep it only infected with the P2P s/w!
If the virus can infect (ooh, say, IIS) then it could use HTTP as a transport without affecting normal behaviour.
It's coming, or something like it. It's just a matter of time before the arms race really kicks in.
Or then again, perhaps I've missed something obvious - it's very late over here in the UK
Simon.
In many respects, a simpler culture is far more resilient than a complex one. Increasing the complexity is a bit like walking up a mountain - the safe area to walk on gets smaller the higher you go up. Consider the magnetic flip to be a huge hand reaching down and translocating you a distance horizontally... would you prefer to be higher up the mountain (wheeeeee.....ouch) or farther down ?
In (slightly) more scientific terms, the advances we've made since those cavemen times are built on the premise of incremental change - we talk of "advances", ie: building on the past to get farther. Take away the foundations (communications is the major one, I guess, direction finding, etc.) and see how well everything that depends on them copes. Consider how an economy might react to (for example: the collapse of air traffic), and the subsequent secondary effects. None of this was even slightly worrying to the caveman, but our world is immensely dependent on excellent long-distance communications.
Yes, we have a far and away more complex civilisation than a caveman ever dreamed of. This is a weakness, not a strength. The payoff comes from what we can do with that technology, but if you remove that, you end up with a lot of hungry people in a small space...
I concur with the physics, btw, but you're really overestimating the resilience of our civilisation.
Simon.
I always thought you Americans took that Boston thing far too seriously
Simon
Let's see, they've already transformed 10,000 machines which were previously enslaved to the windows drum over to the free'n'easy linux beat :-)
...) from above in return for bloody coin.
:-)
They've got another 100,000 scheduled for next year. That drum's just going to get louder, and louder, and louder. Can you hear it yet ? What you are hearing, ladies and gentlemen, is the hammering-in of the thin end of the wedge, and I for one can't wait for that wedge to grow.
Windows is the him-use-deep-magick-solve-problem approach, an oligarchy of high priests results with the local priests doling out consolences (note: not solutions
Linux is a meritocracy, where librarians are shown their due worth, knowledge is open to all, and the only currency you need spend is time, the only fear you need have is looking stupid when asking beginner questions. Even then, you are mostly treated well because of the "There but for the grace of [insert deity] go I" mentality.
No, I'm not a librarian, but I much prefer the latter over the former
Simon.
You might be able to do that as well - depends whether you can get Xinerama to work. I'd guess you currently have a :0 and :1 display, thus you can't drag apps. If you run xinerama, you get a single merged :0 display, and can drag to your heart's content...
... which is 2x 1280x1024 displays, next to each other. Makes a great development platform :-)
[simon@atlantis com.saltsw.pto]$ xdpyinfo | grep dimension
dimensions: 2560x1024 pixels (644x241 millimeters)
Simon
Man, if you ever get *any* negative 'K' readings, you're either in for a nobel prize or a Nelson-like 'Haa-Ha!'....
Simon
I could have sworn that my mirror had 5 iso's the other day, but there's only 3 there now... What's missing ?
Simon