The real point is that Linux has a clear division between the kernel and the X server, which lends itself nicely to the idea of having a roaming display. It must be far, far easier to do that in Linux than in XP, with its disgustingly huge number of APIs and ridiculously interdependent architecture.
I have been wondering for a while why no-one had done it. Everything already exists, not necessarily the way it has been done, but with VNC, Bluetooth, X and a few other bits and pieces, any competent hacker with spare time could have cobbled up a prototype. If you have a laptop with a touch screen, a PC, a pair of wireless cards, and VNC, the job is virtually done. Just tidy it up, get rid of the keyboard on the portable end, software in flash so the disk can go...... (I expect the process they followed was nothing like that of course, but it could be done that way if anyone wants to).
I wonder how the development costs of this compared to what Bill spent on his version? 100 times cheaper?
Altogether a very worthwhile achievement, and at the price it deserves to succeed.
Well, I will give it 6 months, and then give it a cautious trial on my main PC. Once bitten........
It will get there in the end, of course. I might even do some testing on a spare PC to see what happens, and report any bugs. No point sitting around and waiting, may as well get involved, as long as there is no risk to my real data.
But, the ultimate goal for everyone has to be to get rid of all M$ code, and having competing groups working in this area might not help.
I must admit that you have a very valid point there. However if it was a Win 2K.dll, it might be good enough for most people. I will never, ever run Win 9x, or any part of it, on anything again, nor will I touch XP, but 2K is mostly OK. Of course you would not want to run anything remotely resembling Windoze on a machine whose function was critical.
A few doubtful things can get into the kernel anyway, some of the more obscure drivers are not very good, and many of us allow a binary Nvidia module, without knowing how good (usually) or bad (sometimes) it might be.
Win 9x is not an emulator, it is not an OS and does not even pretend to be one, except in Bill's imagination. It is a GUI sitting on top of a buggy old 16 bit OS. Of course, if someone wants to be even more pedantic, I will preempt you and suggest that in view of the timing involved, it is actually Bill's feeble attempt at emulation of what he thought a MAC, or maybe a Lisa, was supposed to be.
2K and XP do contain emulators for Messy-DOS and 9x. They also supposedly contain an emulator for Posix, but I have never been able to see the resemblance.
Does it have to be done that way? Is it not possible, given that the Linux kernel can be modified, for a Windoze.dll to be loaded directly in kernel memory space, with a minimal wrapper round it to map linux system calls to the correct dll entry points?
I am not an expert, but all OSs I have ever seen support a fairly standard set of file operations, the mapping should not be overly complex. Or have M$ messed up big time, as usual, and created a messy interface with hundreds of entry points where about 3 would suffice?
I hope you are right. I wonder if there are any other ways of attacking the problem, other than beating up Bill of course, which is best left to Melinda.
It is clearly always possible using simple methods (unless M$ do really stupid things with the disk format, and then it would break on some disks and/or controllers) to read what is on the disk, using Linux, or for that matter an old version of Norton Utilities (when there really were useful utilities). There is no shortage of people who could run simple file writes from Windoze, and read back the changes to see what had really happened. I could do that, given some basic instructions. So. gathering info is not too hard, making sense of huge volumes of it, all the special cases, error recovery and all that sort of thing, possibly working around deliberate obfuscation, may be another matter.
I wonder if it might be better to do a clean-room implementation of the.dll, one team, who absolutely never, ever writes kernel code, disassembles it (they may even do an instruction trace with an emulator if it helps) and writes a spec, while the other re-implements it, without having ever seen the disassembly. That is legal. The key is writing an accurate spec from the results of the disassembly.
I don't know which method the kernel team are using, but IMHO they need more help (people, not competence, which they already have plenty of), only it would have to be meaningful and properly coordinated otherwise it would be a nuisance. Due to the past history of damaged files etc, they really ought to have many thousands of beta, or even alpha, testers on this one, and not pass the code as fit for general use until many gigabytes on thousands of different PCs have been read and written successfully. If someone would publish a to-do list, they might well find that useful assistance would be forthcoming.
Is that safe enough, or complete enough? Having had an NTFS partition badly damaged by Linux soem time ago, I don't really want to try it yet.
I do appreciate the difficulties the kernel team have had with this, it is not their fault that they have to work with an undocumented closed-source file system.
The strange thing about all this is that very many different OSs which have existed over the years have had some capability to read and write "foreign" file systems, either built in or as a third-party driver. Certainly it is standard with Linux, *BSD, even the hated SCO, also MAC in most of its variants, Amiga, Atari, Solaris....... Even many 8-bit computers could read a variety of foreign file systems. The one name missing is M$, absolutely none of their stuff recognises any othe OS at all. (Please correct me if I am wrong!) It is as if Bill arrogantly imagines that there are only Windoze PCs in this universe. The fact is that there are many things that can't be done under Windoze, but are relatively easy under some other OS. Maybe the reverse is true also, but I can't think of an example. It is absolutely normal in this day and age, even without open source, to need to read and write foreign file systems. The one obstacle is the Chief Hacker of Redmond, he will neither interface to other people's file systems (despite having the documentation, and most drivers under BSD licence) nor will he let anyone else do it by denying proper access to his documentation.
One day, when the masses wake up to what they have been denied since Messy-DOS 1, he may realise that his monopolistic actions have in fact shot himself in both feet.
Windows and the old MAC os have no resemblance whatsoever to Unix. Mostly, no preemptive multitasking, always no fork()/exec(), a pathetic memory mismanagement system (Win 9x), a 16-bit DOS core (Win 9x), no true pipes, only simulation by temporary files, mostly no symbolic links, never hard links, absolutely no portability to other hardware platforms, a vile messed-up (like Bill's apology for a brain) set of 70,000 APIs.........
If any of those systems are a form of Unix, then a monkey is in fact a horse.
It generally does not affect the whole system, although no doubt the US could simply switch off GPS any time they wanted. The jamming covers an area which is more or less line of sight from the jamming equipment. From the ground, maybe a 10 or 20 mile radius, or mode depending on how high youcan get it. From the air, several hundred miles. From a satellite, potentially almost half the earth's surface.
GPS is occasionally jammed during military exercises in the UK, it was done some time ago, from the air I think, unfortunately it was one of the rare occasions that I actually tried to use my Garmin GPS, and it said that I was at -50 feet in the Irish Sea, off the coast of Wales. I was in fact in southern Scotland, about 1400 feet up. AFAIK the jamming radius on that occasion was about 100 miles.
The real danger is that civil aircraft will be using these systems as their primary means of navigation, and might not know that anything is wrong. There is the potential for disaster on a huge scale. In fact, given more intelligence than Al Quaeda are capable of, it could be possible to corrupt the course of an airliner flying in bad visibility such that it flew straight into a building, by transmitting suitably corrupt GPS signals. Admittedly, it takes some human error in addition, but we all know that crashes do happen, with monotonous regularity, because the crew do not cross-check their various sources of data.
Not a very good article either! He would have had just as much trouble with any version of Windoze, possibly a lot more.
However he did have a point, although his perspective was quite warped. There really is a need for lots of good developers to be working on ease of installation, and driver compatability, but most of them prefer to do more interesting things.
Lots of silly little issues remain, like why I can't get SuSE 9.0 to automatically install an Nvidia driver. SuSE don't distribute the driver for licensing reasons, Nvidia provide a binary driver and an open source kernel module to go with it, and an installer, but because SuSE has highly automated installation using Yast, which knows nothing of the Nvidia installer, or vice versa, you always get a useless XF86Config file which has to be edited by hand. No big deal for those who know what they are doing, and can boot into text mode to fix it with vi, but a nightmare for a novice, and the first time you forget and use Yast to change graphic modes or screen layout, you have to do it all again. It gets very tedious. Someone needs to bang two heads together, at SuSE and Nvidia. Credit where it is due though, the Nvidia drivers usually work, or get fixed quite quickly, and their example XF86Config file and lots of instructions on its use are a model of clarity, but not for a novice.
Same story with my latest printer, a Minolta colour laser, some nice guy (not at my home PC at the moment so I can't unfortunately give him the credit he deserves, a Google search will find his site) has written a driver, works fine, but no way of adding it to the Yast system, and there are lots of tedious things to do to get it to work in KDE, OpenOffice.org, and I haven't done Gnome yet..... Sad, because the hard bit has been done, but the guy who wrote the driver understandably does not have the resources to test on every distribution. Anyway, printer configuration is a mess, and I think that is only because the wrong terminology is used or the menus are not thought out well. The core of Cups, Foomatic, Omniprint etc is not the problem, it is only clever GUI-based configurators which don't know all the required options that make it difficult. (Same with Windoze of course, there have been a lot of "clever" attempts to do basic things, culminating in the bug-ridden disaster called WinME).
The problem is that there are actually too few open source developers who are content to write boring perl scripts all day, if they are good (and many are very good indeed) they want to be writing complex applications in C or C++. I think they really need to find a way of recruiting assistance in writing documentation for a start. Unfortunately, the guy who has designed something really does need to take time out to instruct someone as to how it really works, so that documentation can be written. I have seen the same problem in a closed-source, supposedly professional environment, where brilliant and highly productive designers would press ahead regardless of the rest of the world, leaving their work undocumented, and creating many future problems as a result.
Maybe it would help to have only one distribution, it would make better use of the limited supply of competent people, but of course none of us wants a monopoly.
Making a genius realise that he needs to also attend to mundane matters has never been easy.
Just looked at your site, it does everything necessary, and is neat, tidy and not needlessly elaborate. It is not asking much to expect people who are deriving income from selling Linux-based products to do at least as much. The small people and large organisations who use Linux mainly do comply with the GPL, why a few should not is without excuse IMHO.
I hope the FSF will teach them a lesson. Of course it may be that when the GPL is validated in court at the expense of a certain Mr. McBride who we all seem to treat with the contempt he deserves, these others will take fright and put themselves in order legally. I certainly hope so.
DeCSS is legal EVERYWHERE except in the USA. If you want it to be legal in the USA, keep annoying your congressman, that is what he is for, to obey the will of the people who elected him. Same for the President, but of course the people did not elect him, not the present incumbent anyway, so I guess that makes him responsible for nothing....
When you purchase a DVD, you purchase the right to play it for your own enjoyment, but certain legislators can't understand the difference between PLAYING it on equipment of YOUR CHOICE and illegally COPYING it. Not the same thing at all. It seems that it would also be illegal in the US only to make your own hardware DVD player. Everywhere else it would just be extremely difficult, very expensive, and a major engineering triumph if someone does build a player at home, rather than simply grafting a processor board etc onto a commercial drive.
I get annoyed because in the UK I have to search around for the DeCSS software because most Linux distributions make one package only, which they also sell in the US.
So, if you are in the US, sadly you are correct, you can't legally use DeCSS. I wonder for how long a stupid law like this will be allowed to stand? Your legal system makes your country look stupid in the eyes of the rest of the world, and if this kind of nonsense is not stopped, it may turn the US into a technological backwater in the end, by stifling development, as your software patents also seem to be doing.
IANAL, but I do wonder why you can play DVDs on Microtrash systems, apparently legally, and I wonder how much influence Bill had on this, to protect his pathetic monopoly? Seems to me that to be reasonable, fair and non-discriminatory, it ought to be legal on any OS, and any hardware, of the consumers choice. After all, if you purchase a book, you can read it anywhere, under natural or artificial light, with or without magnifying glass, etc, surely it should be the same for any copyright material.
...and as it is possible to get free hosting quite easily, there is little excuse IMHO for not hosting it yourself. The GPL is there to be complied with, simple as that. It is a fair, reasonable and non-discriminatory licence. If they don't like it, they should use the BSD kernel, or licence SCO or some proprietary RTOS (at vast expense).
The last drunk TRAIN driver to kill passengers was ar Eltham Well Hall on 11th June 1972. They kill 1000 on the roads every year. A sleeping ROAD VEHICLE driver caused the Great Heck aka Selby crash. There are less than 2 broken rails on the entire network each day, the chances of one affecting a GNER train are not very great. If you actually take a GNER train from London to Edinburgh, it will, apart from station stops, normally drop below 60mph at two places, Newcastle and Morpeth, due to very tight curvature. Newcastle is a station stop anyway. It will cover most of the distance at 100 to 125mph. I know, I have done the journey many times. Have you?
Hat you read the GNER web site carefully, you would see that it is only installed in one coach of one train, and at the speed they would normally pass, you would be unlikely to have time to achieve a connection anyway.
Most GNER journeys are not delayed or cancelled, and that goes for most other operators too. Your point about maintenance is true, that is being reorganised now, but most of what is said about the poor performance of railways in the UK is simply untrue.
The train operator GNER was as much a victim as the unfortunate crew and passengers in both of these accidents in which their trains were involved. They were not at fault in any way. It they as a train operator can provide a facility to attract more people into their comparatively safe, environmentally friendly trains, instead of using filthy, dangerous road transport, they are at liberty to do so. Their probably quite small investment in WiFi comes from an entirely different pot of money to the track maintenance. Don't confuse the issues.
And BTW there were NO OTHERS in that time period which involved passenger injuries, in an average year the railways kill one or two passengers, road 3500. It is 10 or 20 times safer to travel by rail than by road, even in a bad year. About 1 year in 3, no passengers are killed on the railways at all.
Furthermore one of the two accidents involving GNER was clearly caused by an imbecile behind the wheel of a road vehicle! When did you last hear of a train invading the road, causing an accident? Maybe once in 5 years at a level crossing, and even then the road driver inevitably turns out to have been at fault.
One of my other posts said the A1 at Penmaenshiel, which is slightly north of the Scottish border. Thinking a bit more about it, it is likely that in part, when it is not in a deep cutting, the railway that is, the M9 Motorway between Edinburgh and Stirling is visible from the train, likewise the A9 dual carriageway north of Dunblane, for some way towards Perth, and the A9 (single carriageway IIRC, also single-track railway) north of Perth towards Inverness. There is little chance to see the A1 in England.
I have never yet, in any of these places, seen substantial volumes of road traffic overtaking the train.
If you want to race the road traffic, best take the West Coast Main Line, it parallels the M1 for a few miles around Watford Gap, and the M6 through the Lune Gorge (but it only gives one glimpse of the coast).
The very best, IMHO, as a satisfied customer. They may slide to second place when the West Coast Main Line is fixed, and Virgin's shiny new trains can achieve some of their potential, but they are way ahead of the only other major long-distance high-speed operator, First Great Western, who themselves are tolerable. The real rubbish is found in commuter territory, it is amazing what people will actually tolerate, such as the late, unlamented Connex.
So do I, from London, but only as far as Stirling, and despite the critics it is an excellent service. The HST aka IC125 is probably the finest diesel train ever built, they are getting old now, but the coaches are still the favourite of most passengers. Some are almost as old as Concorde IIRC, evidently from an era of real progress in engineering.
Just to further clarify things for the great mass of gas-guzzling, benzene-emitting car drivers, approximately every second electric service to Edinburgh is extended to Glasgow, clearly the best way (and often the cheapest!) to get to Glasgow while work on the West Coast Main Line is in progress. And no, neither diesel nor electric trains emit benzine, a deadly carcinogen, in any more than a trace amount, and a train load of around 500 people produces much less CO2 than 500 people in cars.
What absolute rubbish. The GNER trains regularly run at 125mph, or at the speed limit for each section of line. OK, they do slow to 50 briefly for the Morpeth curve, and also throught Newcastle, but I use them quite a lot and the speed is really close to 125 for most of the route as far as the Scottish border, therafter mostly a maximum of 90 due to curvature, with some stretches of 100 or 110 if they are heading for Glasgow.
IIRC the only place where a major road parallels the East Coast Main Line is near Penmaenshiel, and although it is by no means the fastest section of the route (90, I think) the traffic on the adjacent A1 dual carriageway is always left well behind.
Such ill-informed comment about the true state of railways in the UK tends to come from those who are also Microsoft worshippers, in both cases because they are singularly ill-informed. Need I say more?
Have you actually ever used a GNER train? I do from time to time, either Kings Cross-Edinburgh, which is an electric loco and MK4 coaches, or to Stirling, which is a HST. They invariably run somewhere very near to time, often arrive a little bit early, unless for reasons not of their making. Booked in advance, and avoiding peak times, the fares are very reasonable. The only time I have ever been sat on one that slowed to near walking speed, and then only for about 2 miles, was just after the mining subsidence at Prestonpans, the track in that area has now been diverted so that will not be necessary again.
All in all, GNER provide a very good service, the staff conduct themselves profesionally, the food is good, the seats comfortable, as you might expect since they inherited British Rail's newest and best, and that route was good under BR.
I don't know why people constantly have to knock the railways in the UK, except maybe to appease their own consciences as a result of using their gas-guzzling benzene burners when the train would have been 20 times safer and 30 times better for the environment.
Personally I welcome this initiative, and look forward to it being extended to all their trains, and to second class.
Better to prove that they have illegaly used some GPL code, and force it to be released under the GPL. There would then be no more scope for manipulation by criminal elements like McBride.
I don't think that is correct. Apparently there were several basic problems which were to do with build quality and materials, in particular the steel used for the hull plates, and the rivets, was not up to scratch, and did not behave properly at low temperature. There were no plans at that time to build huge cargo ships.
There were other deficiencies common at the time, for example due to spontaneous combustion, coal in the bunkers was burning, so there may have been strange thermal stresses also involved.
There were undoubtedly economies made in the construction, all too common even today, which for some reason reminds me of Microsoft.
Slashdot Mirror
I have been wondering for a while why no-one had done it. Everything already exists, not necessarily the way it has been done, but with VNC, Bluetooth, X and a few other bits and pieces, any competent hacker with spare time could have cobbled up a prototype. If you have a laptop with a touch screen, a PC, a pair of wireless cards, and VNC, the job is virtually done. Just tidy it up, get rid of the keyboard on the portable end, software in flash so the disk can go...... (I expect the process they followed was nothing like that of course, but it could be done that way if anyone wants to).
I wonder how the development costs of this compared to what Bill spent on his version? 100 times cheaper?
Altogether a very worthwhile achievement, and at the price it deserves to succeed.
It will get there in the end, of course. I might even do some testing on a spare PC to see what happens, and report any bugs. No point sitting around and waiting, may as well get involved, as long as there is no risk to my real data.
But, the ultimate goal for everyone has to be to get rid of all M$ code, and having competing groups working in this area might not help.
A few doubtful things can get into the kernel anyway, some of the more obscure drivers are not very good, and many of us allow a binary Nvidia module, without knowing how good (usually) or bad (sometimes) it might be.
2K and XP do contain emulators for Messy-DOS and 9x. They also supposedly contain an emulator for Posix, but I have never been able to see the resemblance.
I am not an expert, but all OSs I have ever seen support a fairly standard set of file operations, the mapping should not be overly complex. Or have M$ messed up big time, as usual, and created a messy interface with hundreds of entry points where about 3 would suffice?
It is clearly always possible using simple methods (unless M$ do really stupid things with the disk format, and then it would break on some disks and/or controllers) to read what is on the disk, using Linux, or for that matter an old version of Norton Utilities (when there really were useful utilities). There is no shortage of people who could run simple file writes from Windoze, and read back the changes to see what had really happened. I could do that, given some basic instructions. So. gathering info is not too hard, making sense of huge volumes of it, all the special cases, error recovery and all that sort of thing, possibly working around deliberate obfuscation, may be another matter.
I wonder if it might be better to do a clean-room implementation of the .dll, one team, who absolutely never, ever writes kernel code, disassembles it (they may even do an instruction trace with an emulator if it helps) and writes a spec, while the other re-implements it, without having ever seen the disassembly. That is legal. The key is writing an accurate spec from the results of the disassembly.
I don't know which method the kernel team are using, but IMHO they need more help (people, not competence, which they already have plenty of), only it would have to be meaningful and properly coordinated otherwise it would be a nuisance. Due to the past history of damaged files etc, they really ought to have many thousands of beta, or even alpha, testers on this one, and not pass the code as fit for general use until many gigabytes on thousands of different PCs have been read and written successfully. If someone would publish a to-do list, they might well find that useful assistance would be forthcoming.
I do appreciate the difficulties the kernel team have had with this, it is not their fault that they have to work with an undocumented closed-source file system.
The strange thing about all this is that very many different OSs which have existed over the years have had some capability to read and write "foreign" file systems, either built in or as a third-party driver. Certainly it is standard with Linux, *BSD, even the hated SCO, also MAC in most of its variants, Amiga, Atari, Solaris....... Even many 8-bit computers could read a variety of foreign file systems. The one name missing is M$, absolutely none of their stuff recognises any othe OS at all. (Please correct me if I am wrong!) It is as if Bill arrogantly imagines that there are only Windoze PCs in this universe. The fact is that there are many things that can't be done under Windoze, but are relatively easy under some other OS. Maybe the reverse is true also, but I can't think of an example. It is absolutely normal in this day and age, even without open source, to need to read and write foreign file systems. The one obstacle is the Chief Hacker of Redmond, he will neither interface to other people's file systems (despite having the documentation, and most drivers under BSD licence) nor will he let anyone else do it by denying proper access to his documentation.
One day, when the masses wake up to what they have been denied since Messy-DOS 1, he may realise that his monopolistic actions have in fact shot himself in both feet.
If any of those systems are a form of Unix, then a monkey is in fact a horse.
GPS is occasionally jammed during military exercises in the UK, it was done some time ago, from the air I think, unfortunately it was one of the rare occasions that I actually tried to use my Garmin GPS, and it said that I was at -50 feet in the Irish Sea, off the coast of Wales. I was in fact in southern Scotland, about 1400 feet up. AFAIK the jamming radius on that occasion was about 100 miles.
The real danger is that civil aircraft will be using these systems as their primary means of navigation, and might not know that anything is wrong. There is the potential for disaster on a huge scale. In fact, given more intelligence than Al Quaeda are capable of, it could be possible to corrupt the course of an airliner flying in bad visibility such that it flew straight into a building, by transmitting suitably corrupt GPS signals. Admittedly, it takes some human error in addition, but we all know that crashes do happen, with monotonous regularity, because the crew do not cross-check their various sources of data.
However he did have a point, although his perspective was quite warped. There really is a need for lots of good developers to be working on ease of installation, and driver compatability, but most of them prefer to do more interesting things.
Lots of silly little issues remain, like why I can't get SuSE 9.0 to automatically install an Nvidia driver. SuSE don't distribute the driver for licensing reasons, Nvidia provide a binary driver and an open source kernel module to go with it, and an installer, but because SuSE has highly automated installation using Yast, which knows nothing of the Nvidia installer, or vice versa, you always get a useless XF86Config file which has to be edited by hand. No big deal for those who know what they are doing, and can boot into text mode to fix it with vi, but a nightmare for a novice, and the first time you forget and use Yast to change graphic modes or screen layout, you have to do it all again. It gets very tedious. Someone needs to bang two heads together, at SuSE and Nvidia. Credit where it is due though, the Nvidia drivers usually work, or get fixed quite quickly, and their example XF86Config file and lots of instructions on its use are a model of clarity, but not for a novice.
Same story with my latest printer, a Minolta colour laser, some nice guy (not at my home PC at the moment so I can't unfortunately give him the credit he deserves, a Google search will find his site) has written a driver, works fine, but no way of adding it to the Yast system, and there are lots of tedious things to do to get it to work in KDE, OpenOffice.org, and I haven't done Gnome yet..... Sad, because the hard bit has been done, but the guy who wrote the driver understandably does not have the resources to test on every distribution. Anyway, printer configuration is a mess, and I think that is only because the wrong terminology is used or the menus are not thought out well. The core of Cups, Foomatic, Omniprint etc is not the problem, it is only clever GUI-based configurators which don't know all the required options that make it difficult. (Same with Windoze of course, there have been a lot of "clever" attempts to do basic things, culminating in the bug-ridden disaster called WinME).
The problem is that there are actually too few open source developers who are content to write boring perl scripts all day, if they are good (and many are very good indeed) they want to be writing complex applications in C or C++. I think they really need to find a way of recruiting assistance in writing documentation for a start. Unfortunately, the guy who has designed something really does need to take time out to instruct someone as to how it really works, so that documentation can be written. I have seen the same problem in a closed-source, supposedly professional environment, where brilliant and highly productive designers would press ahead regardless of the rest of the world, leaving their work undocumented, and creating many future problems as a result.
Maybe it would help to have only one distribution, it would make better use of the limited supply of competent people, but of course none of us wants a monopoly.
Making a genius realise that he needs to also attend to mundane matters has never been easy.
I hope the FSF will teach them a lesson. Of course it may be that when the GPL is validated in court at the expense of a certain Mr. McBride who we all seem to treat with the contempt he deserves, these others will take fright and put themselves in order legally. I certainly hope so.
When you purchase a DVD, you purchase the right to play it for your own enjoyment, but certain legislators can't understand the difference between PLAYING it on equipment of YOUR CHOICE and illegally COPYING it. Not the same thing at all. It seems that it would also be illegal in the US only to make your own hardware DVD player. Everywhere else it would just be extremely difficult, very expensive, and a major engineering triumph if someone does build a player at home, rather than simply grafting a processor board etc onto a commercial drive.
I get annoyed because in the UK I have to search around for the DeCSS software because most Linux distributions make one package only, which they also sell in the US.
So, if you are in the US, sadly you are correct, you can't legally use DeCSS. I wonder for how long a stupid law like this will be allowed to stand? Your legal system makes your country look stupid in the eyes of the rest of the world, and if this kind of nonsense is not stopped, it may turn the US into a technological backwater in the end, by stifling development, as your software patents also seem to be doing.
IANAL, but I do wonder why you can play DVDs on Microtrash systems, apparently legally, and I wonder how much influence Bill had on this, to protect his pathetic monopoly? Seems to me that to be reasonable, fair and non-discriminatory, it ought to be legal on any OS, and any hardware, of the consumers choice. After all, if you purchase a book, you can read it anywhere, under natural or artificial light, with or without magnifying glass, etc, surely it should be the same for any copyright material.
...and as it is possible to get free hosting quite easily, there is little excuse IMHO for not hosting it yourself. The GPL is there to be complied with, simple as that. It is a fair, reasonable and non-discriminatory licence. If they don't like it, they should use the BSD kernel, or licence SCO or some proprietary RTOS (at vast expense).
The last drunk TRAIN driver to kill passengers was ar Eltham Well Hall on 11th June 1972. They kill 1000 on the roads every year. A sleeping ROAD VEHICLE driver caused the Great Heck aka Selby crash. There are less than 2 broken rails on the entire network each day, the chances of one affecting a GNER train are not very great. If you actually take a GNER train from London to Edinburgh, it will, apart from station stops, normally drop below 60mph at two places, Newcastle and Morpeth, due to very tight curvature. Newcastle is a station stop anyway. It will cover most of the distance at 100 to 125mph. I know, I have done the journey many times. Have you?
A lot less than 30 seconds. The train will cover 1 miles in that time.
Hat you read the GNER web site carefully, you would see that it is only installed in one coach of one train, and at the speed they would normally pass, you would be unlikely to have time to achieve a connection anyway.
Most GNER journeys are not delayed or cancelled, and that goes for most other operators too. Your point about maintenance is true, that is being reorganised now, but most of what is said about the poor performance of railways in the UK is simply untrue.
And BTW there were NO OTHERS in that time period which involved passenger injuries, in an average year the railways kill one or two passengers, road 3500. It is 10 or 20 times safer to travel by rail than by road, even in a bad year. About 1 year in 3, no passengers are killed on the railways at all.
Furthermore one of the two accidents involving GNER was clearly caused by an imbecile behind the wheel of a road vehicle! When did you last hear of a train invading the road, causing an accident? Maybe once in 5 years at a level crossing, and even then the road driver inevitably turns out to have been at fault.
I have never yet, in any of these places, seen substantial volumes of road traffic overtaking the train.
If you want to race the road traffic, best take the West Coast Main Line, it parallels the M1 for a few miles around Watford Gap, and the M6 through the Lune Gorge (but it only gives one glimpse of the coast).
The very best, IMHO, as a satisfied customer. They may slide to second place when the West Coast Main Line is fixed, and Virgin's shiny new trains can achieve some of their potential, but they are way ahead of the only other major long-distance high-speed operator, First Great Western, who themselves are tolerable. The real rubbish is found in commuter territory, it is amazing what people will actually tolerate, such as the late, unlamented Connex.
Just to further clarify things for the great mass of gas-guzzling, benzene-emitting car drivers, approximately every second electric service to Edinburgh is extended to Glasgow, clearly the best way (and often the cheapest!) to get to Glasgow while work on the West Coast Main Line is in progress. And no, neither diesel nor electric trains emit benzine, a deadly carcinogen, in any more than a trace amount, and a train load of around 500 people produces much less CO2 than 500 people in cars.
IIRC the only place where a major road parallels the East Coast Main Line is near Penmaenshiel, and although it is by no means the fastest section of the route (90, I think) the traffic on the adjacent A1 dual carriageway is always left well behind.
Such ill-informed comment about the true state of railways in the UK tends to come from those who are also Microsoft worshippers, in both cases because they are singularly ill-informed. Need I say more?
All in all, GNER provide a very good service, the staff conduct themselves profesionally, the food is good, the seats comfortable, as you might expect since they inherited British Rail's newest and best, and that route was good under BR.
I don't know why people constantly have to knock the railways in the UK, except maybe to appease their own consciences as a result of using their gas-guzzling benzene burners when the train would have been 20 times safer and 30 times better for the environment.
Personally I welcome this initiative, and look forward to it being extended to all their trains, and to second class.
Better to prove that they have illegaly used some GPL code, and force it to be released under the GPL. There would then be no more scope for manipulation by criminal elements like McBride.
There were other deficiencies common at the time, for example due to spontaneous combustion, coal in the bunkers was burning, so there may have been strange thermal stresses also involved.
There were undoubtedly economies made in the construction, all too common even today, which for some reason reminds me of Microsoft.