Python, at least, has the ability to pull off similar behaviour using its "with" statement. Whether the extra explicit blocks required to do this are good or bad probably depends on point of view, although it's probably rather forced on them by not being a original design feature.
There are some quite nice architectural features in there; whilst they might look like things that are toys for operating systems geeks they are also the kind of thing that makes a nice foundation for building cool stuff that might be more noticeable to the users!
As a multi-server microkernel OS the system is actually split up into a load of different components anyhow. HURD makes it practical (IIRC, assuming nothing's changed in this respect) for users to replace or augment those components - but only for their own programs. So, for instance, a user might develop a network filesystem whilst not requiring any special privileges. Or a user might build a virtual machine / container / chroot to work in, again without their requiring special privileges.
Naturally, this also implies that all the funky hacks involving userspace filesystems are possible - like FUSE on Linux. But that sort of operation is baked into the design and taken further.
The disadvantage of having multi-server microkernel OSes tends to be (reputedly) that they're slower. There are various sides to the argument about whether they *have* to be slower but HURD is, despite its age, not hugely heavily developed - so those problems are not going to have been fixed.
That sort of thing has been done with Linux in various ways - but with substantial disadvantages. Under L4 and Xen there were implementations of running device drivers (for block and network devices) in separate virtual machines from the one running the application. They were restartable and contained only soft state. I worked, in a small way, on the Xen implementation and it was quite enjoyable to sit around restarting the device driver and watching stuff come back. Of course, one advantage of doing things this way is that you can reuse existing drivers.
Another project that was of a similar vintage (around 2003/04) but came slightly earlier (I think) was called "Nooks" and had the advantage of looking more like conventional Linux driver model. You could reuse existing drivers here too but lots of wrappers were needed when interacting with the non-driver portions of the kernel. I was given to understand (admittedly by a Xen developer) that their approach wasn't necessarily very efficient because their kernel-driver switches were still quite expensive and probably fairly frequent.
Anyhow, it's sort of possible to make systems that are more robust to device driver problems with almost-current technology but for various reasons (some subset of needing a hypervisor, cumbersome to set up, memory hungry, performance problems) these mostly don't seem to have taken off yet.
I wanted a cheap MD-Data drive but they weren't available at the time; to me, Minidisk recorders themselves mostly became affordable just around the time that they were ceasing to look the best option. Maybe if I'd been older and had more funds, things would have been different. It's all a bit of a shame though because it was a cool technology that I did want to play with.
Yes, that is important, but it still only works if people are willing to engage enough to hear stuff that the IT guys are saying, which management needs to ensure. If the structures and culture aren't in place then it'll be difficult to get the opportunity to use those communication skills effectively.
Equally well though, I'd expect that part of the job of a well-run IT department is to ensure that poorly thought-out or impractical schemes don't get carried too far, so they ought to push back in those cases. If using the cloud gets around that kind of sanity-checking by the nominated IT experts then that is a problem for the business, whilst still seeming attractive in the short term. Sorting out this trade-off probably requires some reasonably IT-savvy upper management who can make the appropriate tradeoffs and issue guidelines...
Not read the fine article yet but I think these aspects stand regardless...
I'd generally define innovation in technological terms to be "doing things a new way" or just "doing new things". Linux the *kernel*, as opposed to the overall OS, has seen some rather cool developments in various areas (filesystems and storage, virtualisation, other stuff). They don't all have mass-market benefits but they can have massive benefits to the folks who need them.
I have the general *impression* that Linux as a whole OS has been rather less innovative but then I'm not really that acquainted with all of the details. That said, distro package repositories predate Apple's app-store by a long time and basically do the same thing (minus, as is often the case, a really slick UI).
I had the impression that PostgreSQL had some quite unique (at the time, anyhow) features, such as the ability to write stored procedures in a whole host of real programming languages, as well as a domain-specific language.
I was mostly trying to address the technological innovation side of the parent poster's criticisms; there's plenty interesting new developments and research on the storage side taking place in Linux. In real world deployments I'd still generally expect lots of ext3 and ext4 (as Linux is very common in the overall "server market", I'd expect its default filesystem to be all over the place by now). There are also the cluster filesystems (GFS, OCFS2) and, to a certain extent, there's stuff like Lustre (is that entirely open source? I think it might be these days.) that I imagine people are using in random high-end data centres out there.
Virtualisation wise I'm not familiar with the IBM terminology but as far as I know equivalents to all that functionality have been quite established in PC-based virtualisation systems for quite a while now. That said, I'd imagine with IBM's experience that their virtualisation is more dependable than *anyone* else's. IBM were also some of the first (perhaps actually the first) to paravirtualise - modifying the guest OS to run better under virtualisation. I think there are a lot of people out there now using Linux virtualisation in various forms and to a certain extent I would say there's innovative stuff in there already; but there's always more coming and various research projects have done very cool things with it. Folks like VMware also do leading-edge stuff stuff, though, and they're proprietary code.
So I think it depends on whether you count features that are already deployed, or the general innovation in the technology. There's arguably not much impact in innovating without deployment though, so developers do need to get this stuff actually out the door sometimes to really deserve recognition for innovation. I think they're doing that but hope to see more.
The Linux kernel is a relatively fertile centre of innovation in operating systems stuff. Partly that comes in the form of researchers being able to implement their proof-of-concept ideas on an industrial-strength OS rather than a toy (and being able to publish their code in a usable form). Partly that's just in terms of cool features that get implemented in the mainline - my interests are particularly in filesystems and virtualisation and they are where Linux tends to be up there with or ahead of the leaders. Linux-based OSes OTOH; they do have some genuine innovation too but it's maybe less impressive.
I'm not actually sure where Apache is especially innovative because that's not my area.
I'd also add LLVM, gcc, Go, qemu, Android, Xen (admittedly that's a project I've worked on), git, Mercurial (again, personal connections), bzr, darcs... maybe also Python. I'll stop but I think all of these (and more!) have shown some genuine innovation at the time; sometimes there's a commercial backer, sometimes not. I've tried to pick things which were always released to the wider world as open source, if you allow stuff that started closed there's even more.
Xen support got into NetBSD and Solaris more easily, I think, because influential individuals pushed it in there whereas the Linux community had lots of quibbles over the patches and how they should be done correctly. The debate with VMware was a bit confusing and didn't help things get done quickly. RH and IBM and SuSE and others were behind Xen originally but that has gone a bit quieter subsequently.
Part of all of this, though, is due to the Xen team having different priorities to most of those other organisations - especially whilst setting up their own Windows-focused virtualisation product. There's a fair degree of difficult personalities on the Xen and Linux sides too which probably didn't help things go smoothly.
It's partly historical and partly because Xen is structured differently to lots of other virtualisation systems.
"Domain" is to "virtual machine" as "process" is to "program". i.e. it's a running instance of a virtual machine. If you kill a VM and restart it, it's the same VM but a different domain. In practice VM and domain are blurred a bit when people talk, though.
Domain 0 is a bit like the host OS, but for technical reasons it's not exactly.
In *some ways* I like the idea of being able to legally copy, even though I'm against imposing a stupid tax on blank media. To what extent can you "legally copy", though? Can you really just copy any physical media you can get your hands on and it's OK? e.g. rent all the films you want once, then share your copies of those disks with friends?
This case basically sounds like the same people who'd be in favour of the "free market" if the boot was on the other foot, now whinging because nobody is enforcing their business model.
In some ways I'd expect Wikipedia to have more world experts (as well as more 13 year olds) than a traditional encyclopedia. Because the staff who edit a traditional encyclopedia can't necessarily be world experts in all the stuff they must edit - whereas with Wikipedia the real experts can do the edits themselves.
I think his failure in understanding is that used sales help lubricate the market - if people are unable to trade in games that they've already played, they'll have less of their disposable income to spend on new games. And if you're stuck with a game once you've bought it, you're going to be more fussy about which games you get and be less likely to buy them early after release. It's wrong to think of used sales as "sales we should have had", which seems to be an idea gaining traction in the games industry.
A valid complaint in this vein would probably be that the big game stores that are a major channel for selling the new games also (ab)use that position to strategically divert customers into used game deals that are actually pretty poor value for the customer *and* for the publisher / developer. But that's not a problem with used sales per se but with how the big chain stores manipulate them.
From a section of the text article (http://www.bbc.co.uk/news/science-environment-13273348) addressing "Milo", the latest recipient of this type of operation (not the guy shown in the video article): "But Professor Aszmann has faced opposition in some quarters, with senior colleagues even requesting he cancel this latest operation - requests the professor promptly rejected."
The Professor's reason for continuing seemed interesting for me - "Milorad is now 26 years old and he wants to go on with his life. To biologically reconstruct a hand for him would be a never-ending story and in the end he would still have a non-functional hand."
So, they could maybe have tried to fix his real hand but that would have resulted in a lower quality of life due to the process taking longer and having less certain results. The bionic replacement just speeds up the process of getting some functionality back vs trying to fix what's already there.
If someone who's clever enough and can program is still a drag on productivity then it sounds like a problem of technical management in providing appropriate tasks, guidance and training. If you're in need of urgent productive programming (and / or you're a small start-up - *maybe*) then, yes, hire someone with substantial experience so you get returns quickly. Otherwise, it's your job to train them in stuff they might not know. Industry used to be responsible for training and educating workers appropriately beyond their academic career.
A very good step in the right direction, compensating consumers who have been misled. However, I really think that what Sony did requires some kind of punishment. Require them to pay punitive damages to consumers, fine them substantially, do something.
Otherwise, what are we saying? That it's OK to forcibly revoke something somebody's bought so long as you pay for the thing you took away? What Sony did was far more akin to old fashioned theft than piracy *ever* has been. Why? Because they're not getting something for free, they're actively depriving others of valuable things they own. They should be punished for this kind of trick (in a way that they'll notice, rather than just writing off as a minor expense) and / or made to restore the functionality.
This doesn't bother as many people as it should - it's niche functionality, so people don't care, apparently it's OK to swindle as long as it's small numbers of people. Wait a few years and see if you find your music or video playback from non-approved disks and memory cards retroactively disabled or your car satnav is disabled because someone found out how to upload non-approved maps. Then see whether the precedents set on this case look like a good thing.
If you think you've identified an easy way to judge a job applicant, you're probably wrong. A programmer can be good in any language, you need to test his aptitude for logical thinking and learning new things - requires more effort but is more likely to do justice to both you and the applicant.
In the Engines of Light Trilogy by Ken Macleod, one of the major intelligent factions is actually composed of bacterial life living within asteroids all over the galaxy. It's a great series, especially the first book and it has some really interesting takes on a high-tech sci-fi future (e.g. light speed travel but no faster than light communications) as well as creating a universe in which it's possible to playfully "explain" everything from men in black to mass hallucinations, alien abductions, ancient monster myths, etc.
Oh and this research sounds pretty awesome, I hope it leads somewhere as interesting as it sounds like it could!
QT has been Open Source for ages now, although I think possibly the Open Source version for Windows was a bit more recent. The Trolltech team are presumably at Nokia and Nokia have the rights to create closed-source derivatives since they still own the copyrights (thus they can also sell closed licenses). I don't *think* there's a way of making closed-source apps without buying a commercial license (unlike for GTK, interestingly).
But basically QT is perfectly forkable from a community perspective, although it's perhaps worth waiting and seeing what the QT developers themselves choose to do.
The N8 is actually quite impressive, hardware-wise. It's got powerful image processing, can handle HD stuff to the extent it can output to an HD TV, it seems to be quite a slim device and has a fairly nice camera IIRC. Admittedly those aren't all things that'll sell a phone but they do suggest quite decent engineering and investment behind the device's hardware. It's just a shame that it's running Symbian, surely with a better OS it could have been a really impressive product.
I've heard of researchers using Arduino, presumably because hardware's not their speciality but still they need to get a real demonstration device off the ground quickly as a proof of concept. I can imagine this happening in industry too, so even just as a rapid prototyping / concept proving product it could still have quite significant influence outside of hobby development.
Slashdot Mirror
Python, at least, has the ability to pull off similar behaviour using its "with" statement. Whether the extra explicit blocks required to do this are good or bad probably depends on point of view, although it's probably rather forced on them by not being a original design feature.
There are some quite nice architectural features in there; whilst they might look like things that are toys for operating systems geeks they are also the kind of thing that makes a nice foundation for building cool stuff that might be more noticeable to the users!
As a multi-server microkernel OS the system is actually split up into a load of different components anyhow. HURD makes it practical (IIRC, assuming nothing's changed in this respect) for users to replace or augment those components - but only for their own programs. So, for instance, a user might develop a network filesystem whilst not requiring any special privileges. Or a user might build a virtual machine / container / chroot to work in, again without their requiring special privileges.
Naturally, this also implies that all the funky hacks involving userspace filesystems are possible - like FUSE on Linux. But that sort of operation is baked into the design and taken further.
The disadvantage of having multi-server microkernel OSes tends to be (reputedly) that they're slower. There are various sides to the argument about whether they *have* to be slower but HURD is, despite its age, not hugely heavily developed - so those problems are not going to have been fixed.
That sort of thing has been done with Linux in various ways - but with substantial disadvantages. Under L4 and Xen there were implementations of running device drivers (for block and network devices) in separate virtual machines from the one running the application. They were restartable and contained only soft state. I worked, in a small way, on the Xen implementation and it was quite enjoyable to sit around restarting the device driver and watching stuff come back. Of course, one advantage of doing things this way is that you can reuse existing drivers.
Another project that was of a similar vintage (around 2003/04) but came slightly earlier (I think) was called "Nooks" and had the advantage of looking more like conventional Linux driver model. You could reuse existing drivers here too but lots of wrappers were needed when interacting with the non-driver portions of the kernel. I was given to understand (admittedly by a Xen developer) that their approach wasn't necessarily very efficient because their kernel-driver switches were still quite expensive and probably fairly frequent.
Anyhow, it's sort of possible to make systems that are more robust to device driver problems with almost-current technology but for various reasons (some subset of needing a hypervisor, cumbersome to set up, memory hungry, performance problems) these mostly don't seem to have taken off yet.
I wanted a cheap MD-Data drive but they weren't available at the time; to me, Minidisk recorders themselves mostly became affordable just around the time that they were ceasing to look the best option. Maybe if I'd been older and had more funds, things would have been different. It's all a bit of a shame though because it was a cool technology that I did want to play with.
Yes, that is important, but it still only works if people are willing to engage enough to hear stuff that the IT guys are saying, which management needs to ensure. If the structures and culture aren't in place then it'll be difficult to get the opportunity to use those communication skills effectively.
Equally well though, I'd expect that part of the job of a well-run IT department is to ensure that poorly thought-out or impractical schemes don't get carried too far, so they ought to push back in those cases. If using the cloud gets around that kind of sanity-checking by the nominated IT experts then that is a problem for the business, whilst still seeming attractive in the short term. Sorting out this trade-off probably requires some reasonably IT-savvy upper management who can make the appropriate tradeoffs and issue guidelines...
Not read the fine article yet but I think these aspects stand regardless...
I'd generally define innovation in technological terms to be "doing things a new way" or just "doing new things". Linux the *kernel*, as opposed to the overall OS, has seen some rather cool developments in various areas (filesystems and storage, virtualisation, other stuff). They don't all have mass-market benefits but they can have massive benefits to the folks who need them.
I have the general *impression* that Linux as a whole OS has been rather less innovative but then I'm not really that acquainted with all of the details. That said, distro package repositories predate Apple's app-store by a long time and basically do the same thing (minus, as is often the case, a really slick UI).
I had the impression that PostgreSQL had some quite unique (at the time, anyhow) features, such as the ability to write stored procedures in a whole host of real programming languages, as well as a domain-specific language.
I was mostly trying to address the technological innovation side of the parent poster's criticisms; there's plenty interesting new developments and research on the storage side taking place in Linux. In real world deployments I'd still generally expect lots of ext3 and ext4 (as Linux is very common in the overall "server market", I'd expect its default filesystem to be all over the place by now). There are also the cluster filesystems (GFS, OCFS2) and, to a certain extent, there's stuff like Lustre (is that entirely open source? I think it might be these days.) that I imagine people are using in random high-end data centres out there.
Virtualisation wise I'm not familiar with the IBM terminology but as far as I know equivalents to all that functionality have been quite established in PC-based virtualisation systems for quite a while now. That said, I'd imagine with IBM's experience that their virtualisation is more dependable than *anyone* else's. IBM were also some of the first (perhaps actually the first) to paravirtualise - modifying the guest OS to run better under virtualisation. I think there are a lot of people out there now using Linux virtualisation in various forms and to a certain extent I would say there's innovative stuff in there already; but there's always more coming and various research projects have done very cool things with it. Folks like VMware also do leading-edge stuff stuff, though, and they're proprietary code.
So I think it depends on whether you count features that are already deployed, or the general innovation in the technology. There's arguably not much impact in innovating without deployment though, so developers do need to get this stuff actually out the door sometimes to really deserve recognition for innovation. I think they're doing that but hope to see more.
Where did tabs come from? Opera?
The Linux kernel is a relatively fertile centre of innovation in operating systems stuff. Partly that comes in the form of researchers being able to implement their proof-of-concept ideas on an industrial-strength OS rather than a toy (and being able to publish their code in a usable form). Partly that's just in terms of cool features that get implemented in the mainline - my interests are particularly in filesystems and virtualisation and they are where Linux tends to be up there with or ahead of the leaders. Linux-based OSes OTOH; they do have some genuine innovation too but it's maybe less impressive.
I'm not actually sure where Apache is especially innovative because that's not my area.
I'd also add LLVM, gcc, Go, qemu, Android, Xen (admittedly that's a project I've worked on), git, Mercurial (again, personal connections), bzr, darcs ... maybe also Python. I'll stop but I think all of these (and more!) have shown some genuine innovation at the time; sometimes there's a commercial backer, sometimes not. I've tried to pick things which were always released to the wider world as open source, if you allow stuff that started closed there's even more.
Xen support got into NetBSD and Solaris more easily, I think, because influential individuals pushed it in there whereas the Linux community had lots of quibbles over the patches and how they should be done correctly. The debate with VMware was a bit confusing and didn't help things get done quickly. RH and IBM and SuSE and others were behind Xen originally but that has gone a bit quieter subsequently.
Part of all of this, though, is due to the Xen team having different priorities to most of those other organisations - especially whilst setting up their own Windows-focused virtualisation product. There's a fair degree of difficult personalities on the Xen and Linux sides too which probably didn't help things go smoothly.
It's partly historical and partly because Xen is structured differently to lots of other virtualisation systems.
"Domain" is to "virtual machine" as "process" is to "program". i.e. it's a running instance of a virtual machine. If you kill a VM and restart it, it's the same VM but a different domain. In practice VM and domain are blurred a bit when people talk, though.
Domain 0 is a bit like the host OS, but for technical reasons it's not exactly.
...mouldy bit it'll probably still be tasty if you scrape it off a bit.
In *some ways* I like the idea of being able to legally copy, even though I'm against imposing a stupid tax on blank media. To what extent can you "legally copy", though? Can you really just copy any physical media you can get your hands on and it's OK? e.g. rent all the films you want once, then share your copies of those disks with friends?
This case basically sounds like the same people who'd be in favour of the "free market" if the boot was on the other foot, now whinging because nobody is enforcing their business model.
In some ways I'd expect Wikipedia to have more world experts (as well as more 13 year olds) than a traditional encyclopedia. Because the staff who edit a traditional encyclopedia can't necessarily be world experts in all the stuff they must edit - whereas with Wikipedia the real experts can do the edits themselves.
I think his failure in understanding is that used sales help lubricate the market - if people are unable to trade in games that they've already played, they'll have less of their disposable income to spend on new games. And if you're stuck with a game once you've bought it, you're going to be more fussy about which games you get and be less likely to buy them early after release. It's wrong to think of used sales as "sales we should have had", which seems to be an idea gaining traction in the games industry.
A valid complaint in this vein would probably be that the big game stores that are a major channel for selling the new games also (ab)use that position to strategically divert customers into used game deals that are actually pretty poor value for the customer *and* for the publisher / developer. But that's not a problem with used sales per se but with how the big chain stores manipulate them.
From a section of the text article (http://www.bbc.co.uk/news/science-environment-13273348) addressing "Milo", the latest recipient of this type of operation (not the guy shown in the video article):
"But Professor Aszmann has faced opposition in some quarters, with senior colleagues even requesting he cancel this latest operation - requests the professor promptly rejected."
The Professor's reason for continuing seemed interesting for me - "Milorad is now 26 years old and he wants to go on with his life. To biologically reconstruct a hand for him would be a never-ending story and in the end he would still have a non-functional hand."
So, they could maybe have tried to fix his real hand but that would have resulted in a lower quality of life due to the process taking longer and having less certain results. The bionic replacement just speeds up the process of getting some functionality back vs trying to fix what's already there.
If someone who's clever enough and can program is still a drag on productivity then it sounds like a problem of technical management in providing appropriate tasks, guidance and training. If you're in need of urgent productive programming (and / or you're a small start-up - *maybe*) then, yes, hire someone with substantial experience so you get returns quickly. Otherwise, it's your job to train them in stuff they might not know. Industry used to be responsible for training and educating workers appropriately beyond their academic career.
A very good step in the right direction, compensating consumers who have been misled. However, I really think that what Sony did requires some kind of punishment. Require them to pay punitive damages to consumers, fine them substantially, do something.
Otherwise, what are we saying? That it's OK to forcibly revoke something somebody's bought so long as you pay for the thing you took away? What Sony did was far more akin to old fashioned theft than piracy *ever* has been. Why? Because they're not getting something for free, they're actively depriving others of valuable things they own. They should be punished for this kind of trick (in a way that they'll notice, rather than just writing off as a minor expense) and / or made to restore the functionality.
This doesn't bother as many people as it should - it's niche functionality, so people don't care, apparently it's OK to swindle as long as it's small numbers of people. Wait a few years and see if you find your music or video playback from non-approved disks and memory cards retroactively disabled or your car satnav is disabled because someone found out how to upload non-approved maps. Then see whether the precedents set on this case look like a good thing.
If you think you've identified an easy way to judge a job applicant, you're probably wrong. A programmer can be good in any language, you need to test his aptitude for logical thinking and learning new things - requires more effort but is more likely to do justice to both you and the applicant.
*appreciative applause*
In the Engines of Light Trilogy by Ken Macleod, one of the major intelligent factions is actually composed of bacterial life living within asteroids all over the galaxy. It's a great series, especially the first book and it has some really interesting takes on a high-tech sci-fi future (e.g. light speed travel but no faster than light communications) as well as creating a universe in which it's possible to playfully "explain" everything from men in black to mass hallucinations, alien abductions, ancient monster myths, etc.
Oh and this research sounds pretty awesome, I hope it leads somewhere as interesting as it sounds like it could!
QT has been Open Source for ages now, although I think possibly the Open Source version for Windows was a bit more recent. The Trolltech team are presumably at Nokia and Nokia have the rights to create closed-source derivatives since they still own the copyrights (thus they can also sell closed licenses). I don't *think* there's a way of making closed-source apps without buying a commercial license (unlike for GTK, interestingly).
But basically QT is perfectly forkable from a community perspective, although it's perhaps worth waiting and seeing what the QT developers themselves choose to do.
The N8 is actually quite impressive, hardware-wise. It's got powerful image processing, can handle HD stuff to the extent it can output to an HD TV, it seems to be quite a slim device and has a fairly nice camera IIRC. Admittedly those aren't all things that'll sell a phone but they do suggest quite decent engineering and investment behind the device's hardware. It's just a shame that it's running Symbian, surely with a better OS it could have been a really impressive product.
I've heard of researchers using Arduino, presumably because hardware's not their speciality but still they need to get a real demonstration device off the ground quickly as a proof of concept. I can imagine this happening in industry too, so even just as a rapid prototyping / concept proving product it could still have quite significant influence outside of hobby development.