I did do some reading before posting, so I do have some understanding of what the project has achieved. In this specific case, there is an opportunity to optimise because the virtualisation host is providing most of the services you need including an emulated standard network controller and disk controller.
However it's not correct to draw the conclusion (as the OP did) that this is the case for a general virtualisation workload - if you need access to a wider range of physical devices or you need to co-ordinate service provision across multiple virtual machines, for example, you're going to end up with something that looks a lot like a guest OS, whatever you choose to call it.
I'm not questioning the project's goals or achievements, but the fact that some virtualised servers might have very lightweight requirements for traditional OS services does not justify the proposition "Virtual servers in the future may stop using OSes entirely". And I'm pointing out that, historically, simplified solutions seem gradually and quietly to regrow the complexity they loudly shed.
What part of "there is a specific set of applications of the industrial control type that can run hardwired into interrupt vectors and physical address space" are you trying to contradict?
I have in front of me a Cambridge University Computer Science Tripos examination paper from June 1975. Question 7 reads:
"Computer operating systems are very complicated things whose purpose is to help people to share the use of equipment. They will become largely obsolete as the equipment becomes cheap enough not to need sharing". Discuss the prescience or wrongheadedness of this remark.
This was long before the personal computer which came along later and initially more or less dispensed with the traditional operating system for more or less the reason stated. And as personal computers became more powerful and cheaper and even less intended for sharing, they pretty much all came to run an operating system with multi-user security, process scheduling, protection rings, paging, layered device driver architectures that would be familiar to someone who knew MULTICS or OS/360 with TSO.
Now, there will be a specific set of applications, particularly those whose only I/O operations happen across a network, whose operating system requirements may be sufficiently limited they can be inherited from a parent virtual machine - just as there is a specific set of applications of the industrial control type that can run hardwired into interrupt vectors and physical address space. However, to suggest that virtual servers in general in the future may be able to do away with OSs is contrary to historical precendent. And insane.
... I wonder what other people of my age have been doing so wrong that they still need employment - they've had careers with salaries and conditions that noone is ever going to get in the future and ought to have been banking that while the going was good.
On the rare occasion I stray back into a "real" business to do a bit of consulting, I feel like I'm walking into a kindergarten: it's all competitive attention seeking and fingerpainting (sorry, Powerpoint).
I would feel desperately sorry for a younger generation if they thought they were going to have to be in that environment all their lives - but mainly because it would demonstrate a lack of ambition and foresight. You really ought to have some control over your own destiny by the time you reach your 50s. If you haven't, you've wasted the last 30-odd years.
All the court has decided, it appears, is that the copyright owners don't have a "proprietary" right to the proceeds of infringement. That's a specific form of legal shortcut to seizing assets. The issue of whether there is a valid claim is still proceeding, just not using that specific legal mechanism. No decision has been made on anyone's entitlement to anything, except the entitlement of a copyright owner to make a particular form of legal submission.
IPv6 failed the moment that the IETF decided that Not Invented Here was more important than fixing the problem. Had the original IAB recommendation been adopted, we'd be on to IPv8 by now. The window of opportunity to manage a transition before there really was a "consumer" internet was lost.
It's not just a problem with ISP's penny-pinching - most of the consumer routing kit supplied to date doesn't do IPv6 and to the extent that consumers have any technical knowledge at all, it's a recognition of the pattern "192.168.x.y".
The logisitics of doing this the official way are so horrendous, it's hardly surprising that ISPs are looking at any alternative, however unattractive.
Society is also a genetic construct that has (co-)evolved. If the current shape of society results in a poorer survival rate for people, then either people will die out or people in a different form of society will eventually become more numerous. In any case, we won't know the evolutionary effect of a change in societal behaviour for many generations, so it's probably better to optimise for present well-being rather than contemplate sacrificing (other) people for a hypothetical benefit to future generations that most likely will not occur and which none of us will see.
Just a bit of background to set the context for this.
English* Universities depend very heavily on the income from overseas students as the total funding from English students (fees + government grants) does not, allegedly, meet the costs of the education provided. It's also now the only growth area for student recruitment (applications from English students were down around 10% this year as fees have risen steeply). The last I heard, Newcastle University was building on its campus a college for overseas students of 16+ to improve recruitment rates to the University itself for those same students when they reached 18.
The current government, on the other hand, is committed to substantially reducing immigration levels which it is finding very hard to do - the Eurozone financial situation means that immigration from Europe is increasing (and EU treaties require the free movement of people) and clamping down on overseas students is seen as an easy short-term win. There's been a big argument between Universities and the government about whether students should count in the immigration figures at all (since most of them leave at the end of their courses) which was resolved only in September (http://www.telegraph.co.uk/news/uknews/immigration/9541141/Foreign-students-to-be-marked-out-in-immigration-figures.html) with a compromise which keeps student numbers under very close review.
Universities fear increasingly tight controls on studying in the UK might dissuade students from enrolling and are increasingly starting to open overseas campuses (http://www.timeshighereducation.co.uk/story.asp?storycode=415018) which have the potential additional benefit of tapping the market for those who don't have the resources to relocate for their education. There is talk in some institutions that serving UK students may become an incidental consequence of their academic activities rather than an institutional goal.
It's in the midst of this ongoing policy shift - withdrawing government money from universities then encouraging them to make it up overseas and then tightening up on student visas - that Universities find themselves trapped. They need the money, so they need the visas, so they have to do what the governnment requires to get them. And while government funding for undergraduates may no longer be significant, Univeristies still depend heavily on government funding for their postgraduate programmes, which is where they get their reputations from. So don't expect any crusades from the moral high ground.
*Somewhat different situation in Scotland and Wales
A country the size of Bermuda actually requires very little total tax revenue compared with the US. It only has to support 65,000 people, so if a large number of the world's most profitable organisations put their HQ there, the rate of tax needed across such a large potential tax base is tiny. It simply not possible for a larger country to compete with that.
There is a school of thought that the answer to this is simply to abolish corporation tax altogether. The problem with this is that all returns from investment tend to flow to places that are already capital-rich and little sticks in the places where the business actually takes place - which is precisely the problem with the tax avoidance we see now.
The rich make laws only because the kings that used to make laws - or their proxies - got hacked to pieces if they didn't turn over power. You can make all the laws you want, but when the torches and pitchforks start heading your way, you may find they're not the protection you'd hoped for.
Plutocracy has crossed a line and would be wise to draw back its toes while the plebs are merely hurling insults.
The larger damage is actually caused by the consumers of the oil companies' products not by the oil companies themselves.
Take the entire amount of oil spilled over the course of human history and its contribution to environmental damage is neglible compared with that caused by burning oil and gas.
If US prosecutors really want to hurt some enivronmental criminals, they should confiscate the cars of the entire population.
Some time back in Britain, a bank marketing genius decided that the way to get new customers was to get rid of the old charging model and offer "free" banking. It was such a brilliant wheeze that all the other banks had to follow suit. However, in order to make a profit, banks were then obliged to slap on a whole new range of exceptional and penal charges in the small print and to give their customers the hard sell for a bunch of other financial products that they didn't need (and for which the banks are now paying billions of pounds in compensation). Everyone is agreed that "free" banking is broken, but nobody can be the first to reinstate charges because their customers will all take a hike.
Search engines are the same. Having "free" search engines is a really crazy idea if you think the end user should have some interest in how the results should be selected and presented. But nobody is ever going to pay to use a search engine while the other(s) is/are still free, even if the results are worse.
So we're stuck with a model in which the selection and presentation of results must of commercial necessity be orchestrated for profit and the more people who see those results the more profit is made.
You can argue about the extent to which the orchestration is fair and transparent - and indeed whether fairness and transparency are adequate counterweights - but as long as someone else is paying the conductor you get no say in the performance.
Are you saying that international agreements would be more effectively enforced by making them instead with disparate groups having no effective de facto control over anything in their countries?
And that's without getting started on the dictatorship and corruption that only exist because of the sponsorship of "liberal" western countries...
Bits of the WCML are set to hit capacity. Travel to Manchester at 5pm from London and you'll find the trains largely empty because the 4tph schedule is way in excess of demand.
The original plan for the WMCL upgrade would have allowed for 150mph running - which would have been pretty much sufficient - but was canned on cost grounds (although extremely modest compared with an entire new high-speed line). The (electric) ECML trains have been capable of 150mph for years but the track was never signalled for that line speed, except experimentally. We've already written off higher-speed rail twice when the full costs became apparent.
Now you could argue that the real constraint is the need to mix freight and passenger traffic on the same track. And that certainly constrains capacity as well as operating speed. However, you could relieve that problem by providing additional lower speed routes for freight - and that would be significantly cheaper because you could do without electrification, simplify signalling and you don't have to pass through centres of population with expensive land and the need for expensive engineering solutions - and just do some remediation of bottlenecks on the existing passenger lines.
There's only an economic case for HS2 if the alternative is to do nothing. If you're looking at ways of getting the best return on your rail investment, HS2 is way down the list.
Slashdot was ringing with hurrahs when it was decided that APIs weren't copyrightable in the Google vs Oracle case.
Here we have something that declares itself specifically to be an "API" and suddenly it has to be jealously guarded against the heathens.
As far as I can tell, the GPL being asserted here depends very tenuously on the monolithic nature of the kernel. The enforcement of the GPL against code which merely links, dynamically, to a GPL library seems even more precarious: if you're supplying code using a published API and you're not even distributing the library concerned, it's hard to see exactly what rights are being infringed by the supplier. Or indeed how it's any different from any compiled programme being a "derived work" of the instruction set of the CPU.
When it comes to Linux "IPR", I don't see any real distinction between GPL FUD and Microsoft FUD.
Although they're less effective against things like engine noise, I found shutters helpful for reducing the nighttime intrusion of city centre noise. If you've got some reasonably dense timber board, try covering your window openings with it (inside or outside depending on which is more suitable) and see what difference it makes (you get an additional air gap before the glazing plus the sound absorbency of the board). I don't suggest you board up your windows for good (though, this being Slashdot, I don't discount it as an option), but if it makes a substantial difference you can install something more permanently that hinges and folds to give you access to daylight when required.
Also, if you live on the top floor, noise may be coming in through the roof so insulation above the ceiling may help.
It's a recurrent theme in British politics. Look up Harold Wilson's 1963 "White Heat of Technology" speech and the creation of the Ministry of Technology.
Britain within living memory has been a technology leader in aviation, nuclear, computing, Those were largely developments that came out of the war and declined in the face of a dependence on government money for investment (and in the latter case, an unwillingness to admit even to the existence of the technology).
Private investors aren't interested in long-term investments - the "investment banking" industry has become big largely because it's eschewed actual growth-producing investment for complex financial instruments which are essentially a form of privatised taxation.
There is still a lot of high-tech industry (take Rolls Royce aero engines for example), but it survives and grows pretty much in spite of the business environment. It's no accident that Britain's now successful, productive and growing car industry is owned and financed from Japan, Germany and India.
It may be possible to grow IT-based industries in London, but they won't be owned in London and nor will any IP associated with them. And I'm afraid the government is sufficiently clueless about technology that it might actually feel it needs to encourage businesses like those cited by the article ( Instagram, Skype and Groupon) whereas there is probably a lot to be said for actively discouraging them.
Plus, this seems to be all about exempting businesses from paying their normal dues. I'm all in favour of foreigners spending money in London. I'm not in favour of the government giving it back with interest to encourage them.
I think what people were expecting was what they were told they were going to get, not just in messages directly from the creators but in some extremely selective and breathy press coverage. Any questions about the practicality of delivering what was promised for the price were howled down on any forum on which they were posted by mobs of angry enthusiasts defending the plucky technologists inventing the new British wonder-device, so none of the practical issues were really aired until the device shipped.
By which time the 1080p HD Video was MPEG-4 only and the 35$ board that could be powered from a USB port suddenly needed its own PSU and powered hub for all practical purposes and the USB port wasn't that Universal.
And it's not just a BOM problem - true, the board originally had better power supply arrangements which were dropped to shave a couple of dollars off the board cost and to save people buying extra power adapters (how ironic) and the video codec support was limited for similar reasons - the real hardware issues seem to stem from stuff that could in principle be fixed without impacting the BOM but can't in practice owing to the closed nature of the device firmware.
What the Pi illustrates very nicely is that a working prototype is a long way from a production device and rather than learning that lesson quietly with a low-key limited production run it's being learned in the full glare of publicity and with a growing number of complaints.
Certainly the device is too cheap, but the cheaper the BOM the smaller proportion it is of the overall cost of production - and focussing on the BOM is I think where things have gone wrong. The BOM is just where the costs start.
Well, on the one occasion I flew Concorde JFK-LHR, luxury was not really at the top of the agenda. The *service* was good, but it was also necessary - taking coats from passengers in the private lounge to stow on board gave the impression of service, but was necessary because the aircraft was so small there was nowhere in the passenger cabin where people could have stowed them themselves. The seats were small. The aisle was small. The food selection was small (compared with First Class) because the galleys were small and the food storage areas were small.
There was almost a pioneering air about it - as we climbed out of JFK the captain announced "please don't be alarmed: we're shortly about to turn off the engines [I suspect he meant the afterburners] as part of noise control procedures, but don't worry, they've never failed to reignite yet". The interior walls became noticeably warm as the mach indicator ticked up.
But the big attraction was that Concorde flew during the day (unlike all but a couple of other US/UK flights) and you arrived without the fatigue of 7 hours of confinement in bad air.
And even 35 years back, I got to fly because a conventional aircraft had gone out of service and there was room on Concorde at the last moment to accommodate all the bumped First Class passengers and a good chunk of Business Class - so even then there weren't that many people (the aircraft only had 100 seats) for whom that attraction was worth the price.
Actually, it's not true that Concorde required a long runway in all circumstances.
BA used to use Concorde on some regional services if the regular aircraft was o/s - it used to shuttle up to Newcastle from Heathrow periodically and Newcastle doesn't have a particularly long runway - about 2.3km. Mind you, I doubt it had a full load of fuel on board.
It can't have been an economic proposition but it was seen as good publicity.
When I read computing at Cambridge, they'd just extended the duration of the course from 1 year to 2 - and even that was (as I recall) based on two hours of lectures per day and a couple of harware labs per week. You had to enter as an undergraduate on the assumption you were going to read something else and read CS as an afterthought..
I think this was very useful as it ensured students had a background in something else (like maths or engineering), but also gave them some spare time to attend lectures in other subjects and see how CS might usefully be applied to real world problems.
Slashdot Mirror
...the letters "B-Ark" should be distinctly visible.
I did do some reading before posting, so I do have some understanding of what the project has achieved. In this specific case, there is an opportunity to optimise because the virtualisation host is providing most of the services you need including an emulated standard network controller and disk controller. However it's not correct to draw the conclusion (as the OP did) that this is the case for a general virtualisation workload - if you need access to a wider range of physical devices or you need to co-ordinate service provision across multiple virtual machines, for example, you're going to end up with something that looks a lot like a guest OS, whatever you choose to call it. I'm not questioning the project's goals or achievements, but the fact that some virtualised servers might have very lightweight requirements for traditional OS services does not justify the proposition "Virtual servers in the future may stop using OSes entirely". And I'm pointing out that, historically, simplified solutions seem gradually and quietly to regrow the complexity they loudly shed.
What part of "there is a specific set of applications of the industrial control type that can run hardwired into interrupt vectors and physical address space" are you trying to contradict?
I have in front of me a Cambridge University Computer Science Tripos examination paper from June 1975. Question 7 reads:
"Computer operating systems are very complicated things whose purpose is to help people to share the use of equipment. They will become largely obsolete as the equipment becomes cheap enough not to need sharing". Discuss the prescience or wrongheadedness of this remark.
This was long before the personal computer which came along later and initially more or less dispensed with the traditional operating system for more or less the reason stated. And as personal computers became more powerful and cheaper and even less intended for sharing, they pretty much all came to run an operating system with multi-user security, process scheduling, protection rings, paging, layered device driver architectures that would be familiar to someone who knew MULTICS or OS/360 with TSO.
Now, there will be a specific set of applications, particularly those whose only I/O operations happen across a network, whose operating system requirements may be sufficiently limited they can be inherited from a parent virtual machine - just as there is a specific set of applications of the industrial control type that can run hardwired into interrupt vectors and physical address space. However, to suggest that virtual servers in general in the future may be able to do away with OSs is contrary to historical precendent. And insane.
All the perf of native code with none of the security issues
I have a perpetual motion machine and am seeking investors. I take it you'll be subscribing?
... I wonder what other people of my age have been doing so wrong that they still need employment - they've had careers with salaries and conditions that noone is ever going to get in the future and ought to have been banking that while the going was good.
On the rare occasion I stray back into a "real" business to do a bit of consulting, I feel like I'm walking into a kindergarten: it's all competitive attention seeking and fingerpainting (sorry, Powerpoint).
I would feel desperately sorry for a younger generation if they thought they were going to have to be in that environment all their lives - but mainly because it would demonstrate a lack of ambition and foresight. You really ought to have some control over your own destiny by the time you reach your 50s. If you haven't, you've wasted the last 30-odd years.
All the court has decided, it appears, is that the copyright owners don't have a "proprietary" right to the proceeds of infringement. That's a specific form of legal shortcut to seizing assets. The issue of whether there is a valid claim is still proceeding, just not using that specific legal mechanism. No decision has been made on anyone's entitlement to anything, except the entitlement of a copyright owner to make a particular form of legal submission.
Actually, no.
IPv6 failed the moment that the IETF decided that Not Invented Here was more important than fixing the problem. Had the original IAB recommendation been adopted, we'd be on to IPv8 by now. The window of opportunity to manage a transition before there really was a "consumer" internet was lost.
It's not just a problem with ISP's penny-pinching - most of the consumer routing kit supplied to date doesn't do IPv6 and to the extent that consumers have any technical knowledge at all, it's a recognition of the pattern "192.168.x.y".
The logisitics of doing this the official way are so horrendous, it's hardly surprising that ISPs are looking at any alternative, however unattractive.
Not PC and not relevant either.
Society is also a genetic construct that has (co-)evolved. If the current shape of society results in a poorer survival rate for people, then either people will die out or people in a different form of society will eventually become more numerous. In any case, we won't know the evolutionary effect of a change in societal behaviour for many generations, so it's probably better to optimise for present well-being rather than contemplate sacrificing (other) people for a hypothetical benefit to future generations that most likely will not occur and which none of us will see.
Just a bit of background to set the context for this.
English* Universities depend very heavily on the income from overseas students as the total funding from English students (fees + government grants) does not, allegedly, meet the costs of the education provided. It's also now the only growth area for student recruitment (applications from English students were down around 10% this year as fees have risen steeply). The last I heard, Newcastle University was building on its campus a college for overseas students of 16+ to improve recruitment rates to the University itself for those same students when they reached 18.
The current government, on the other hand, is committed to substantially reducing immigration levels which it is finding very hard to do - the Eurozone financial situation means that immigration from Europe is increasing (and EU treaties require the free movement of people) and clamping down on overseas students is seen as an easy short-term win. There's been a big argument between Universities and the government about whether students should count in the immigration figures at all (since most of them leave at the end of their courses) which was resolved only in September (http://www.telegraph.co.uk/news/uknews/immigration/9541141/Foreign-students-to-be-marked-out-in-immigration-figures.html) with a compromise which keeps student numbers under very close review.
Universities fear increasingly tight controls on studying in the UK might dissuade students from enrolling and are increasingly starting to open overseas campuses (http://www.timeshighereducation.co.uk/story.asp?storycode=415018) which have the potential additional benefit of tapping the market for those who don't have the resources to relocate for their education. There is talk in some institutions that serving UK students may become an incidental consequence of their academic activities rather than an institutional goal.
It's in the midst of this ongoing policy shift - withdrawing government money from universities then encouraging them to make it up overseas and then tightening up on student visas - that Universities find themselves trapped. They need the money, so they need the visas, so they have to do what the governnment requires to get them. And while government funding for undergraduates may no longer be significant, Univeristies still depend heavily on government funding for their postgraduate programmes, which is where they get their reputations from. So don't expect any crusades from the moral high ground.
*Somewhat different situation in Scotland and Wales
Indeed.
A country the size of Bermuda actually requires very little total tax revenue compared with the US. It only has to support 65,000 people, so if a large number of the world's most profitable organisations put their HQ there, the rate of tax needed across such a large potential tax base is tiny. It simply not possible for a larger country to compete with that.
There is a school of thought that the answer to this is simply to abolish corporation tax altogether. The problem with this is that all returns from investment tend to flow to places that are already capital-rich and little sticks in the places where the business actually takes place - which is precisely the problem with the tax avoidance we see now.
The rich make laws only because the kings that used to make laws - or their proxies - got hacked to pieces if they didn't turn over power. You can make all the laws you want, but when the torches and pitchforks start heading your way, you may find they're not the protection you'd hoped for.
Plutocracy has crossed a line and would be wise to draw back its toes while the plebs are merely hurling insults.
The larger damage is actually caused by the consumers of the oil companies' products not by the oil companies themselves.
Take the entire amount of oil spilled over the course of human history and its contribution to environmental damage is neglible compared with that caused by burning oil and gas.
If US prosecutors really want to hurt some enivronmental criminals, they should confiscate the cars of the entire population.
Some time back in Britain, a bank marketing genius decided that the way to get new customers was to get rid of the old charging model and offer "free" banking. It was such a brilliant wheeze that all the other banks had to follow suit. However, in order to make a profit, banks were then obliged to slap on a whole new range of exceptional and penal charges in the small print and to give their customers the hard sell for a bunch of other financial products that they didn't need (and for which the banks are now paying billions of pounds in compensation). Everyone is agreed that "free" banking is broken, but nobody can be the first to reinstate charges because their customers will all take a hike.
Search engines are the same. Having "free" search engines is a really crazy idea if you think the end user should have some interest in how the results should be selected and presented. But nobody is ever going to pay to use a search engine while the other(s) is/are still free, even if the results are worse.
So we're stuck with a model in which the selection and presentation of results must of commercial necessity be orchestrated for profit and the more people who see those results the more profit is made.
You can argue about the extent to which the orchestration is fair and transparent - and indeed whether fairness and transparency are adequate counterweights - but as long as someone else is paying the conductor you get no say in the performance.
Are you saying that international agreements would be more effectively enforced by making them instead with disparate groups having no effective de facto control over anything in their countries?
And that's without getting started on the dictatorship and corruption that only exist because of the sponsorship of "liberal" western countries...
Bits of the WCML are set to hit capacity. Travel to Manchester at 5pm from London and you'll find the trains largely empty because the 4tph schedule is way in excess of demand.
The original plan for the WMCL upgrade would have allowed for 150mph running - which would have been pretty much sufficient - but was canned on cost grounds (although extremely modest compared with an entire new high-speed line). The (electric) ECML trains have been capable of 150mph for years but the track was never signalled for that line speed, except experimentally. We've already written off higher-speed rail twice when the full costs became apparent.
Now you could argue that the real constraint is the need to mix freight and passenger traffic on the same track. And that certainly constrains capacity as well as operating speed. However, you could relieve that problem by providing additional lower speed routes for freight - and that would be significantly cheaper because you could do without electrification, simplify signalling and you don't have to pass through centres of population with expensive land and the need for expensive engineering solutions - and just do some remediation of bottlenecks on the existing passenger lines.
There's only an economic case for HS2 if the alternative is to do nothing. If you're looking at ways of getting the best return on your rail investment, HS2 is way down the list.
Well said.
Slashdot was ringing with hurrahs when it was decided that APIs weren't copyrightable in the Google vs Oracle case.
Here we have something that declares itself specifically to be an "API" and suddenly it has to be jealously guarded against the heathens.
As far as I can tell, the GPL being asserted here depends very tenuously on the monolithic nature of the kernel. The enforcement of the GPL against code which merely links, dynamically, to a GPL library seems even more precarious: if you're supplying code using a published API and you're not even distributing the library concerned, it's hard to see exactly what rights are being infringed by the supplier. Or indeed how it's any different from any compiled programme being a "derived work" of the instruction set of the CPU.
When it comes to Linux "IPR", I don't see any real distinction between GPL FUD and Microsoft FUD.
Although they're less effective against things like engine noise, I found shutters helpful for reducing the nighttime intrusion of city centre noise. If you've got some reasonably dense timber board, try covering your window openings with it (inside or outside depending on which is more suitable) and see what difference it makes (you get an additional air gap before the glazing plus the sound absorbency of the board). I don't suggest you board up your windows for good (though, this being Slashdot, I don't discount it as an option), but if it makes a substantial difference you can install something more permanently that hinges and folds to give you access to daylight when required.
Also, if you live on the top floor, noise may be coming in through the roof so insulation above the ceiling may help.
It's a recurrent theme in British politics. Look up Harold Wilson's 1963 "White Heat of Technology" speech and the creation of the Ministry of Technology.
Britain within living memory has been a technology leader in aviation, nuclear, computing, Those were largely developments that came out of the war and declined in the face of a dependence on government money for investment (and in the latter case, an unwillingness to admit even to the existence of the technology).
Private investors aren't interested in long-term investments - the "investment banking" industry has become big largely because it's eschewed actual growth-producing investment for complex financial instruments which are essentially a form of privatised taxation.
There is still a lot of high-tech industry (take Rolls Royce aero engines for example), but it survives and grows pretty much in spite of the business environment. It's no accident that Britain's now successful, productive and growing car industry is owned and financed from Japan, Germany and India.
It may be possible to grow IT-based industries in London, but they won't be owned in London and nor will any IP associated with them. And I'm afraid the government is sufficiently clueless about technology that it might actually feel it needs to encourage businesses like those cited by the article ( Instagram, Skype and Groupon) whereas there is probably a lot to be said for actively discouraging them.
Plus, this seems to be all about exempting businesses from paying their normal dues. I'm all in favour of foreigners spending money in London. I'm not in favour of the government giving it back with interest to encourage them.
I think what people were expecting was what they were told they were going to get, not just in messages directly from the creators but in some extremely selective and breathy press coverage. Any questions about the practicality of delivering what was promised for the price were howled down on any forum on which they were posted by mobs of angry enthusiasts defending the plucky technologists inventing the new British wonder-device, so none of the practical issues were really aired until the device shipped.
By which time the 1080p HD Video was MPEG-4 only and the 35$ board that could be powered from a USB port suddenly needed its own PSU and powered hub for all practical purposes and the USB port wasn't that Universal.
And it's not just a BOM problem - true, the board originally had better power supply arrangements which were dropped to shave a couple of dollars off the board cost and to save people buying extra power adapters (how ironic) and the video codec support was limited for similar reasons - the real hardware issues seem to stem from stuff that could in principle be fixed without impacting the BOM but can't in practice owing to the closed nature of the device firmware.
What the Pi illustrates very nicely is that a working prototype is a long way from a production device and rather than learning that lesson quietly with a low-key limited production run it's being learned in the full glare of publicity and with a growing number of complaints.
Certainly the device is too cheap, but the cheaper the BOM the smaller proportion it is of the overall cost of production - and focussing on the BOM is I think where things have gone wrong. The BOM is just where the costs start.
Well, on the one occasion I flew Concorde JFK-LHR, luxury was not really at the top of the agenda. The *service* was good, but it was also necessary - taking coats from passengers in the private lounge to stow on board gave the impression of service, but was necessary because the aircraft was so small there was nowhere in the passenger cabin where people could have stowed them themselves. The seats were small. The aisle was small. The food selection was small (compared with First Class) because the galleys were small and the food storage areas were small.
There was almost a pioneering air about it - as we climbed out of JFK the captain announced "please don't be alarmed: we're shortly about to turn off the engines [I suspect he meant the afterburners] as part of noise control procedures, but don't worry, they've never failed to reignite yet". The interior walls became noticeably warm as the mach indicator ticked up.
But the big attraction was that Concorde flew during the day (unlike all but a couple of other US/UK flights) and you arrived without the fatigue of 7 hours of confinement in bad air.
And even 35 years back, I got to fly because a conventional aircraft had gone out of service and there was room on Concorde at the last moment to accommodate all the bumped First Class passengers and a good chunk of Business Class - so even then there weren't that many people (the aircraft only had 100 seats) for whom that attraction was worth the price.
Actually, it's not true that Concorde required a long runway in all circumstances.
BA used to use Concorde on some regional services if the regular aircraft was o/s - it used to shuttle up to Newcastle from Heathrow periodically and Newcastle doesn't have a particularly long runway - about 2.3km. Mind you, I doubt it had a full load of fuel on board.
It can't have been an economic proposition but it was seen as good publicity.
... he was never able to satisfactorily distinguish between "principle" and 'practice".
As in the principle of being opposed to slavery while in practice shagging the property.
When I read computing at Cambridge, they'd just extended the duration of the course from 1 year to 2 - and even that was (as I recall) based on two hours of lectures per day and a couple of harware labs per week. You had to enter as an undergraduate on the assumption you were going to read something else and read CS as an afterthought..
I think this was very useful as it ensured students had a background in something else (like maths or engineering), but also gave them some spare time to attend lectures in other subjects and see how CS might usefully be applied to real world problems.
alternate providers ... are shaking in their pants
I think that proves that enough porn has already been downloaded for their further services to be unnecessary.