So the absolute worst case scenario here is that installing software in Ubuntu is as easy as installing software on Windows, but chances are it will be much much easier. Basically everything about your post was FUD, and not even intentional FUD, but ignorant FUD. From someone who claims to be a Linux developer (and a Wine developer too?), I can't fathom how you could not know this.
This isn't a case of not knowing, it's a case of having experienced stuff you apparently haven't.
Let me fill you in. What happens if you are not included in a distribution (or just as badly, included but packaged wrong or out of date)? As there's a lot of confusion around what this means, it means you aren't apt-gettable by end users. Not in the repositories.
Well, firstly you have to provide your own packages. That's OK if it's only one distro, but it quickly becomes annoying if it's several. And actually it's one package for each version of each distro. A package for Feisty isn't good enough. You need the last couple of versions as well, because not everybody upgrades at the same time. To do that you need a separate install of each version, and you need to build the package on each install, using multi-boots, or VMware, or chroots, or just relying on volunteers to fill in the gaps for you. So if there are 3 distros you want to support, each with 3 versions in the wild, that's 9 packages you need (therefore 9 independent OS installs). You can try and cheat by reusing older packages on newer distros but sometimes that breaks.
Then you have to tell your users how to install it. Look at the complexity of the page you linked to. This is a light year away from "just download and install it yourself". You have to copy and paste meaningless commands to add "keys" to your "apt list", whatever that is. You have to know which version of Ubuntu you have, although they all look pretty similar (yeah this stuff sounds real basic but people get stuck on simpler things). If you accidentally use the wrong one, you'll get a technical error message that isn't clearly related to your mistake. Or you'll get no error at all and it'll just crash or misbehave in some obscure way.
Inevitably, some people will get this wrong, resulting in additional support load for you.
If you don't have root (family machine?) then you're stuck, of course.
This is the best case scenario. Wine is a big project, and Ubuntu is a big distro, so it works out OK here. Users don't have to wait long after they upgrade before they regain access to their programs. If you use Fedora the situation is worse. If it's a smaller project, again, don't rely on getting a 3rd party repository (actually Wine is pretty rare in doing this).
See, this is what I have problems with. It's the general design of the software distribution scheme that's bogus. It can never work reliably. It's like Microsoft announcing that Vista will only install software you got from Microsoft Download Center... nobody would accept that: it doesn't scale, MS aren't trusted to be impartial, etc. It wouldn't work for Microsoft, so why would it work for anybody else
You're not reading what I am writing, I think. Or I'm not making myself clear. The point is what if it's not in the repositories at all (so you can't apt-get it)
Well, from a practical perspective, every Windows app is guaranteed to have an installer, more or less. But there are very few autopackages in the world, partly because it's hard to make them, and partly because distros constantly change in non-compatible ways and that tends to break them.
If what you say is correct, then fine, my objections are gone. But are you sure about that? It doesn't actually say in the article, but Ubuntu doesn't install lots of software by default. It makes no sense to saying "XYZ won't be shipped with Dell machines" when in distro-parlance to "ship" means to make available in a repository. I don't really care what's installed by default or not because as you say it's very easy to change. But if so, why remark on it, when this is no change of policy over existing Ubuntu?
I wonder where these quotes come from - the linked podcast? I really hope he's being quoted out of context here.
How would you feel if you bought a Vista machine from Dell, and discovered that to install Firefox you had to run regedit, hack up some registry keys, add some magic URLs to a database somewhere pointing to specially crafted Vista-specific Firefox versions on mozilla.org, and then click through a bunch of security warnings before you could install it?
I know exactly what the reaction on Slashdot would be - they'd be raked over the coals for being anti-competitive, forcing the Moz guys to do extra work, requiring loads of knowledge Grandma couldn't be expected to have, etc. But when Canonical does exactly the same thing, it's OK because people who simply "must" have that program can do some technical jiggery-pokery and work around the problem. Amazing.
The focus will be on how Ubuntu sucks at running Windows software, not on how Ubuntu rocks at running Linux software.
Well, that's great as long as there are perfect 1:1 replacements for all the Windows software somebody wants. That isn't true for any serious gamer, for instance, or people who use custom business apps, which basically means every business that uses IT, or anybody with kids who wants to use a particular educational software package.
Hell I'll happily admit I'm biased, because I used to work on Crossover and Wine, but even the MS Office+Wine combination handily beats OpenOffice. Even when not doing anything Wine or software related, I'd use Word/Excel for office tasks on Linux, because it worked a lot better than OpenOffice did, and the small amount of integration OO had into the desktop wasn't a big deal to me compared to things like, not being sluggish, and being able to perfectly import Word docs. Now don't get me wrong, OO has improved a lot since those days and I want to love it, I really do, but I know there are still a lot of people who use MS Office on Linux over OpenOffice just because they prefer it.
This is just a re-run of the ancient debate about whether Win32 emulation is harmful or good. It never interested me, because it assumes an operating system can be a closed world. That's clearly not true and never has been true, if it was, you should argue that MPlayer being able to play non-Ogg codecs is bad and should be pulled, or OpenOffice being able to read.DOCs is bad and should be pulled, or Linux being able to read FAT32 partitions is harmful and should be pulled. It just makes no sense, actually, because if people need that compatibility they'll either use the compatibility layer or they'll just stay with Windows, in which case you haven't even helped them a little bit.
And one of those strengths is that you can still install WINE after you buy the computer despite the decisions made by a large company or single individual.
Only if you are squinting so hard you're blind. Linux is the only desktop operating system in which if your distributor decides to not include software, getting it anyway is extremely difficult. If a package isn't included in Ubuntu, your only option is either to compile it from source (good luck with that if you aren't technical) or using something like an autopackage. Neither Windows nor MacOS X practice this kind of software censorship.
I have to admit, this news really pisses me off. Shuttleworth can't seem to decide what he wants here. For background, I am the creator of autopackage, a framework for writing cross-distro binary installers for Linux. It's kinda like Loki Setup except it's designed for open source software, so it handles dependencies, has GTK, Qt and console frontends, etc. Now I haven't really been involved with this project for some time for various reasons, but back when I was, this whole idea that open source projects might distribute their own binaries was terribly controversial. People wondered what the point was.
Now, I did a presentation at LUGRadio Live last year, in which I laid out the case for autopackage (and klik and zeroinstall), and also talked about a bunch of other issues like malware. One of the issues I raised is that every distribution is a political entity that excludes software for reasons that are, to the non-Linux enthusiast, more or less random. Whether it's to do with the license, or lack of manpower, or because a program isn't UNIXy enough, or simply because the maintainers don't like it, a distribution uses its monopoly on easy software installation to eliminate software from the users world.
At the time I warned that this situation couldn't work long term as Linux scaled up. It makes the distro responsible for all the software that is shipped. More to the point, it harms users, because it forces one groups choice on everybody else, restricting the free market. I warned that while people might find discrimination on the basis of license acceptable, and on the basis of manpower understandable, distros would at some point start discriminating against software for bad reasons. And then what do the authors of the affected software do? They can't tell their users to compile it themselves, because that's too hard and fragile. They can't make their own repositories for every distro out there, that's too much work, and besides users are told not to trust 3rd party repositories because they might mess up the distro, break it or be malware. This was very visible to me, because when an enthusiastic user requested an Ubuntu package of the autopackage runtime (first time installs are awkward without that), it got shot down because an Ubuntu developer didn't think it was useful. A bunch of users did, but he didn't, so tough cookies.
I'm pretty pissed off, because not only was I an autopackage developer but also a Wine developer, and now it's happening again. Once more, both users who want a program and the developers who write it are being screwed over due to the opinions of one guy combined with a bad system. About the best option Wine has now is for the developers to maintain an Ubuntu repository, and for users to be given clear instructions on how to add it, and be told to ignore any warnings about that being a bad idea. If N other distros decide to join in the fun, multiply the effort by N.
Even Microsoft, at the height of their monopolistic practices, never made installing software they didn't like so difficult. This is a big shame for Linux, and as it slowly gets more popular these issues will return again and again.
Well, Apples desktop efforts ARE "doomed" in the sense that they probably won't get big market share anytime soon. They're very visible in the consumer laptop space, so it *feels* like they are doing much better than they really are. But they don't have much presence in business, or in developing countries, or even on the regular desktop market... and these are huge quantities of machines we're talking about here. Apple/MacOS X have been around for some years now and they didn't set the world on fire. To actually start phasing out Windows in serious quantities, you need something that is a fundamental leap or shift, not a slight refinement on an old concept (which is what MacOS is... doesn't make it bad, mind you)
If we insist on dodgy analogies, then it'd be like outlawing physical theft. So now I chop wood all day, and I end up with a nice pile of wood that I intend to take to the market tomorrow and sell. But overnight, somebody walks up to my house and takes all my wood. Now I have worked all day and have nothing to show for it, the wood is gone. So either I can chop wood strictly "on demand" and not try and store it overnight, or I can just build a shed and lock it extra tight, so even though theft is legal, it's really hard to do. Chopping wood "on demand" is really awkward because the place where the wood grows is not the same as the place where people want to buy it, so I'll probably go for the latter. A lot of people seem to think that losing copyright would magically make all software open source. It wouldn't. It'd just lead to stronger locks.
I used to find that argument very convincing but no longer do. The problem with "I must help my neighbour" is it ignores that as society scales sometimes you must do bad for a greater good. What if my neighbour is late for work and asks me for help? Is it OK for me to take a bike propped up against a fence and give it to them, because helping my neighbour in the natural way is theft? No, it isn't, because if everybody did that, our society would collapse (or at least, look totally different).
RMS would say, well, software is different to bikes so the same logic doesn't apply, but I'd disagree. Copyright is a set of tradeoffs. If it means I can't "help my neighbour" by giving them free warez, that's OK, because the alternative is to make retail software (like video games) a non-workable business model, which would result in greater harm to greater numbers of people.
That crowd isn't ethereal. I saw the author arguing with several Slashdotters in an earlier story and they were quite emphatic that all copyright should be abolished. There might not be many of them but they are vocal, and they are here. They're also crazy, but that doesn't mean their "arguments" shouldn't be addressed.
That's like saying if I fix a guy's car and he sells it for a million bucks, I should get a piece of that action.
It's not, because a song is not a physical thing. That's pretty basic.
The simple fact is that without copyright nobody can legally close the source and keep me from it.
Absolutely they can. They can simply not publish it anywhere. What makes you think that having access to a very long number would allow you to force the publication of source code? Remember that if you abolish copyright there is no concept of derived work.
Copyright does the same. Many famous song writers got burned by some obscure copyright holder to a similar tune. The second guy who comes up with the same melody, quite independently from the first is SOL. Screw that. Unacceptable. Please don't try to tell me the same thing doesn't happen in software, or anywhere else affected by these insane laws.
I've never heard of that happening with copyright in software, unless somebody actually copy and pasted code from a different project. You're thinking of patents, which is not the same.
Finally I don't get why you think x86 dominance has anything to do with copyrights or even patents. That's clearly absurd and shows a serious mis-understanding of the technology. x86 chips dominate the market place because they are exceptionally fast, perhaps the fastest chips ever produced, and because they are the most useful of any chip due to the large amount of software produced for them. Maybe you labor under the illusion that if there was no copyright every program would be open source and therefore ported to "better" architectures, but the fact is, nobody gives a shit about SPARC or Alpha on the desktop, and nobody ever has. If there was no copyright all that'd change is that commodity desktop hardware would suddenly sprout much better protection for binary code (more like games consoles) to compensate for the fact that a damages from a code leak can't be reclaimed through the courts.
The idea was "let's create a virtual world where you can do anything". The point was... well actually there was no point. It was just meant to be cool for its own sake, and maybe a way to make money.
Seriously, that more or less sums it up. It's like asking what the point of the web is. It's just a tool that people can use to create things.
Society no longer sees restrictions on copying entertainment to be for its own benefit.
As a member of society, please don't make implications about what I do or do not believe. In fact, I pay for entertainment that I enjoy, because I want to see more of it and I understand that if nobody paid, the creators could either [a] not fund their work or [b] would have to cram it so full of ads that I'd want to throw things against the walls.
In future I'd suggest that before making grandiloquent claims about what society believes, you actually do statistically valid surveys to find out. Oh wait, that'd be hard work...
Erm, how does your proposal fix this? COM has a simple negotiation mechanism where Explorer can say "Do you support FOO?" and the extension can say "Yes, here you go" or "No, I don't". If you get rid of the COM indirection in favor of a simpler plugin system, you'd have to drop this negotiation mechanism or re-implement it, and that is what would be grossly unsafe.
COM is not the problem here. The problem is buggy code. The only policy design that is worth debating here is whether it makes sense to allow 3rd party code to load into a critical process like Explorer. Now, it's easy to say "never!" here but every OS does it. Hell MacOS X and Linux allow the loading of third party code into anything at all. As Raymond has explained many times before, large parts of Windows were designed back in the days when people were basically trusted to do the right thing and not screw up. This is now known to be a bad idea so modern parts of Windows are much more robust against "improvement" by 3rd parties, but the shell dates back to the early 90s and reflects the design mentality of the era.
No, you didn't understand the explanation. This sort of understanding failure is exactly what caused the need for the verclsid program in the first place!
The problem was that this specific shell extension (for an obsolete HP printer) contained a concurrency bug - it tried to synchronize with a thread in its DLL detach function. This is never correct, because in a DLL detach you can't make any assumptions about the liveness of other threads. Because there are buggy shell extensions out there that hang, they had a watchdog thread, but this design can't stop hangs that occur during shutdown of the process itself as this one did (which is presumably a rare kind of bug).
I really find it amazing that despite a clear explanation from Raymond, there are people here trying to blame Microsoft! It just goes to show that programming is hard, I guess....
Its only useful in a closed-source world where you cannot modify programs to suit the new API's.
How much open source work have you actually done? I've done a lot, and this idea is one I see very often in people who haven't done any serious API development work before. The approach of attempting to patch every app when an API changes simply doesn't scale. There's a reason all the important open source APIs (gtk, glibc, alsa, X etc) have "gone stable" in the past 5 years, and it's simply a better approach.
Anyway, ignoring the obvious (!!) problems of scaling such an approach, you are confusing two unrelated things. Microsoft can simply/clean up APIs too - they have done it with DirectX and.NET, but that's irrelevant. The problem here is that there are lots of people in the world writing software who perhaps aren't well qualified, and even the ones that are well qualified make mistakes, even with the implementation of quite simple interfaces like IUnknown. I myself have messed up IUnknown before, in fact.
The root problem that caused the hang was attempting to cleanly handle buggy software. This is a common motif in software, hell, it practically motivated the move from the Windows 9x design to the NT fully protected architecture.
The result is that even Microsoft can't get reasonably trivial things right.
Multi-threading is never trivial.
Not to mention almost all Windows software code being highly complicated compared to equivalent code on other systems.
I worked on Wine for a long time, which implements or maps the Win32 API. The complexity of Linux, Windows and MacOS X are all much the same - they are of the same design era, even OS X which is based on lots of older code at its heart. While the more modern parts of the Linux APIs like GTK+ are better than the Win32 equivalents that's just an age thing: the Win32 API has evolved over a much longer period of time. That means it's uglier (the world has learned a lot about API design since the 80s), but it also means there are far more people out there who know it, better tools support, and critically, more apps that use it!
For what it's worth, I agree with you 100%. The problem is that "computer science" is so vaguely defined it's not even a useful term. There are only a few jobs for computer scientists, and even then, those jobs usually require you to be both a talented software engineer AND a computer scientist. We have a few of those types at Google and I have mad respect for their skillz, but they would never have been hired if they didn't know C++ or Java - it's that simple.
Unfortunately, there is a small but not insignificant part of the software development population who always had a greater love for the mathematical side of computing - in which there aren't many jobs - and were never that keen on the gritty details of how computers actually work. So they end up bitter and take every opportunity to "remind" people that computer science isn't about programming or systems architecture, it's just maths (ie, the part they like). They conveniently ignore that the popular definition of computer science is what's taught on computer science courses, which should be a whole mix of things.
Universities did that originally, but the people who write the P2P software just had to get their free tunes/warez/movies, so they put in workarounds and now it's really hard to throttle this sort of traffic (because it's encrypted or uses unpredictable ports, or whatever). So now P2P is banned completely. Not really surprising if you ask me. Nor an "infringement of free speech" - it's pragmatic network engineering.
Linux is fucked by this kind of thing because (unless a lot changed in the last 12 months) there are no distros that silently install updates without any user intervention being required. Even distros like Ubuntu will pop up a balloon and require you to type in a password to install the updates. But we already know from experience with Windows that nothing short of a fully automatic system will do. If Linux had the popularity of Windows today, this exploit would still be being used 3 years from now.
How did it help NBC? Some other company was taking the ad revenues for a show they paid for. Only if you make some huge assumptions, like people who watch a show on YouTube then go watch it again (with the ads) on an actual TV, does it help NBC. Somebody else posted that SNL ratings didn't change despite the growing popularity of YouTube, so it seems like it didn't really help them.
Now having said that, I suspect a lot of the YouTube traffic is for people outside of the states who can't watch the shows on a regular TV anyway. These days I watch the Daily Show on the comedy central website, because it's not syndicated where I live (or not that I know of, anyway). I don't mind CC doing their own video streaming - it's more or less the same as what you'd see on YouTube anyway, but with ads owned by CC itself.
You shouldn't moderate comments down for being "wrong", this is basic moderator 101 stuff. If a comment is factually wrong, mod up a response that corrects it, or post yourself to do so. If it's "wrong" in the sense that you disagree with it, mod up the opposing viewpoint.
Volume key hacks are not solid. How many times is it necessary to point this out on discussions about AACS? The specification contains a wide variety of traitor-tracing algorithms that let you find a hacked player key given only the released volume keys, or even only the decrypted video itself.
Slashdot Mirror
This isn't a case of not knowing, it's a case of having experienced stuff you apparently haven't.
Let me fill you in. What happens if you are not included in a distribution (or just as badly, included but packaged wrong or out of date)? As there's a lot of confusion around what this means, it means you aren't apt-gettable by end users. Not in the repositories.
Well, firstly you have to provide your own packages. That's OK if it's only one distro, but it quickly becomes annoying if it's several. And actually it's one package for each version of each distro. A package for Feisty isn't good enough. You need the last couple of versions as well, because not everybody upgrades at the same time. To do that you need a separate install of each version, and you need to build the package on each install, using multi-boots, or VMware, or chroots, or just relying on volunteers to fill in the gaps for you. So if there are 3 distros you want to support, each with 3 versions in the wild, that's 9 packages you need (therefore 9 independent OS installs). You can try and cheat by reusing older packages on newer distros but sometimes that breaks.
Then you have to tell your users how to install it. Look at the complexity of the page you linked to. This is a light year away from "just download and install it yourself". You have to copy and paste meaningless commands to add "keys" to your "apt list", whatever that is. You have to know which version of Ubuntu you have, although they all look pretty similar (yeah this stuff sounds real basic but people get stuck on simpler things). If you accidentally use the wrong one, you'll get a technical error message that isn't clearly related to your mistake. Or you'll get no error at all and it'll just crash or misbehave in some obscure way.
Inevitably, some people will get this wrong, resulting in additional support load for you.
If you don't have root (family machine?) then you're stuck, of course.
This is the best case scenario. Wine is a big project, and Ubuntu is a big distro, so it works out OK here. Users don't have to wait long after they upgrade before they regain access to their programs. If you use Fedora the situation is worse. If it's a smaller project, again, don't rely on getting a 3rd party repository (actually Wine is pretty rare in doing this).
See, this is what I have problems with. It's the general design of the software distribution scheme that's bogus. It can never work reliably. It's like Microsoft announcing that Vista will only install software you got from Microsoft Download Center ... nobody would accept that: it doesn't scale, MS aren't trusted to be impartial, etc. It wouldn't work for Microsoft, so why would it work for anybody else
You're not reading what I am writing, I think. Or I'm not making myself clear. The point is what if it's not in the repositories at all (so you can't apt-get it)
Well, from a practical perspective, every Windows app is guaranteed to have an installer, more or less. But there are very few autopackages in the world, partly because it's hard to make them, and partly because distros constantly change in non-compatible ways and that tends to break them.
If what you say is correct, then fine, my objections are gone. But are you sure about that? It doesn't actually say in the article, but Ubuntu doesn't install lots of software by default. It makes no sense to saying "XYZ won't be shipped with Dell machines" when in distro-parlance to "ship" means to make available in a repository. I don't really care what's installed by default or not because as you say it's very easy to change. But if so, why remark on it, when this is no change of policy over existing Ubuntu?
I wonder where these quotes come from - the linked podcast? I really hope he's being quoted out of context here.
Simple question.
How would you feel if you bought a Vista machine from Dell, and discovered that to install Firefox you had to run regedit, hack up some registry keys, add some magic URLs to a database somewhere pointing to specially crafted Vista-specific Firefox versions on mozilla.org, and then click through a bunch of security warnings before you could install it?
I know exactly what the reaction on Slashdot would be - they'd be raked over the coals for being anti-competitive, forcing the Moz guys to do extra work, requiring loads of knowledge Grandma couldn't be expected to have, etc. But when Canonical does exactly the same thing, it's OK because people who simply "must" have that program can do some technical jiggery-pokery and work around the problem. Amazing.
Well, that's great as long as there are perfect 1:1 replacements for all the Windows software somebody wants. That isn't true for any serious gamer, for instance, or people who use custom business apps, which basically means every business that uses IT, or anybody with kids who wants to use a particular educational software package.
Hell I'll happily admit I'm biased, because I used to work on Crossover and Wine, but even the MS Office+Wine combination handily beats OpenOffice. Even when not doing anything Wine or software related, I'd use Word/Excel for office tasks on Linux, because it worked a lot better than OpenOffice did, and the small amount of integration OO had into the desktop wasn't a big deal to me compared to things like, not being sluggish, and being able to perfectly import Word docs. Now don't get me wrong, OO has improved a lot since those days and I want to love it, I really do, but I know there are still a lot of people who use MS Office on Linux over OpenOffice just because they prefer it.
This is just a re-run of the ancient debate about whether Win32 emulation is harmful or good. It never interested me, because it assumes an operating system can be a closed world. That's clearly not true and never has been true, if it was, you should argue that MPlayer being able to play non-Ogg codecs is bad and should be pulled, or OpenOffice being able to read .DOCs is bad and should be pulled, or Linux being able to read FAT32 partitions is harmful and should be pulled. It just makes no sense, actually, because if people need that compatibility they'll either use the compatibility layer or they'll just stay with Windows, in which case you haven't even helped them a little bit.
Only if you are squinting so hard you're blind. Linux is the only desktop operating system in which if your distributor decides to not include software, getting it anyway is extremely difficult. If a package isn't included in Ubuntu, your only option is either to compile it from source (good luck with that if you aren't technical) or using something like an autopackage. Neither Windows nor MacOS X practice this kind of software censorship.
I have to admit, this news really pisses me off. Shuttleworth can't seem to decide what he wants here. For background, I am the creator of autopackage, a framework for writing cross-distro binary installers for Linux. It's kinda like Loki Setup except it's designed for open source software, so it handles dependencies, has GTK, Qt and console frontends, etc. Now I haven't really been involved with this project for some time for various reasons, but back when I was, this whole idea that open source projects might distribute their own binaries was terribly controversial. People wondered what the point was.
Now, I did a presentation at LUGRadio Live last year, in which I laid out the case for autopackage (and klik and zeroinstall), and also talked about a bunch of other issues like malware. One of the issues I raised is that every distribution is a political entity that excludes software for reasons that are, to the non-Linux enthusiast, more or less random. Whether it's to do with the license, or lack of manpower, or because a program isn't UNIXy enough, or simply because the maintainers don't like it, a distribution uses its monopoly on easy software installation to eliminate software from the users world.
At the time I warned that this situation couldn't work long term as Linux scaled up. It makes the distro responsible for all the software that is shipped. More to the point, it harms users, because it forces one groups choice on everybody else, restricting the free market. I warned that while people might find discrimination on the basis of license acceptable, and on the basis of manpower understandable, distros would at some point start discriminating against software for bad reasons. And then what do the authors of the affected software do? They can't tell their users to compile it themselves, because that's too hard and fragile. They can't make their own repositories for every distro out there, that's too much work, and besides users are told not to trust 3rd party repositories because they might mess up the distro, break it or be malware. This was very visible to me, because when an enthusiastic user requested an Ubuntu package of the autopackage runtime (first time installs are awkward without that), it got shot down because an Ubuntu developer didn't think it was useful. A bunch of users did, but he didn't, so tough cookies.
I'm pretty pissed off, because not only was I an autopackage developer but also a Wine developer, and now it's happening again. Once more, both users who want a program and the developers who write it are being screwed over due to the opinions of one guy combined with a bad system. About the best option Wine has now is for the developers to maintain an Ubuntu repository, and for users to be given clear instructions on how to add it, and be told to ignore any warnings about that being a bad idea. If N other distros decide to join in the fun, multiply the effort by N.
Even Microsoft, at the height of their monopolistic practices, never made installing software they didn't like so difficult. This is a big shame for Linux, and as it slowly gets more popular these issues will return again and again.
Well, Apples desktop efforts ARE "doomed" in the sense that they probably won't get big market share anytime soon. They're very visible in the consumer laptop space, so it *feels* like they are doing much better than they really are. But they don't have much presence in business, or in developing countries, or even on the regular desktop market ... and these are huge quantities of machines we're talking about here. Apple/MacOS X have been around for some years now and they didn't set the world on fire. To actually start phasing out Windows in serious quantities, you need something that is a fundamental leap or shift, not a slight refinement on an old concept (which is what MacOS is ... doesn't make it bad, mind you)
If we insist on dodgy analogies, then it'd be like outlawing physical theft. So now I chop wood all day, and I end up with a nice pile of wood that I intend to take to the market tomorrow and sell. But overnight, somebody walks up to my house and takes all my wood. Now I have worked all day and have nothing to show for it, the wood is gone. So either I can chop wood strictly "on demand" and not try and store it overnight, or I can just build a shed and lock it extra tight, so even though theft is legal, it's really hard to do. Chopping wood "on demand" is really awkward because the place where the wood grows is not the same as the place where people want to buy it, so I'll probably go for the latter. A lot of people seem to think that losing copyright would magically make all software open source. It wouldn't. It'd just lead to stronger locks.
I used to find that argument very convincing but no longer do. The problem with "I must help my neighbour" is it ignores that as society scales sometimes you must do bad for a greater good. What if my neighbour is late for work and asks me for help? Is it OK for me to take a bike propped up against a fence and give it to them, because helping my neighbour in the natural way is theft? No, it isn't, because if everybody did that, our society would collapse (or at least, look totally different).
RMS would say, well, software is different to bikes so the same logic doesn't apply, but I'd disagree. Copyright is a set of tradeoffs. If it means I can't "help my neighbour" by giving them free warez, that's OK, because the alternative is to make retail software (like video games) a non-workable business model, which would result in greater harm to greater numbers of people.
That crowd isn't ethereal. I saw the author arguing with several Slashdotters in an earlier story and they were quite emphatic that all copyright should be abolished. There might not be many of them but they are vocal, and they are here. They're also crazy, but that doesn't mean their "arguments" shouldn't be addressed.
It's not, because a song is not a physical thing. That's pretty basic.
Absolutely they can. They can simply not publish it anywhere. What makes you think that having access to a very long number would allow you to force the publication of source code? Remember that if you abolish copyright there is no concept of derived work.
I've never heard of that happening with copyright in software, unless somebody actually copy and pasted code from a different project. You're thinking of patents, which is not the same.
Finally I don't get why you think x86 dominance has anything to do with copyrights or even patents. That's clearly absurd and shows a serious mis-understanding of the technology. x86 chips dominate the market place because they are exceptionally fast, perhaps the fastest chips ever produced, and because they are the most useful of any chip due to the large amount of software produced for them. Maybe you labor under the illusion that if there was no copyright every program would be open source and therefore ported to "better" architectures, but the fact is, nobody gives a shit about SPARC or Alpha on the desktop, and nobody ever has. If there was no copyright all that'd change is that commodity desktop hardware would suddenly sprout much better protection for binary code (more like games consoles) to compensate for the fact that a damages from a code leak can't be reclaimed through the courts.
The idea was "let's create a virtual world where you can do anything". The point was ... well actually there was no point. It was just meant to be cool for its own sake, and maybe a way to make money.
Seriously, that more or less sums it up. It's like asking what the point of the web is. It's just a tool that people can use to create things.
As a member of society, please don't make implications about what I do or do not believe. In fact, I pay for entertainment that I enjoy, because I want to see more of it and I understand that if nobody paid, the creators could either [a] not fund their work or [b] would have to cram it so full of ads that I'd want to throw things against the walls.
In future I'd suggest that before making grandiloquent claims about what society believes, you actually do statistically valid surveys to find out. Oh wait, that'd be hard work ...
Erm, how does your proposal fix this? COM has a simple negotiation mechanism where Explorer can say "Do you support FOO?" and the extension can say "Yes, here you go" or "No, I don't". If you get rid of the COM indirection in favor of a simpler plugin system, you'd have to drop this negotiation mechanism or re-implement it, and that is what would be grossly unsafe.
COM is not the problem here. The problem is buggy code. The only policy design that is worth debating here is whether it makes sense to allow 3rd party code to load into a critical process like Explorer. Now, it's easy to say "never!" here but every OS does it. Hell MacOS X and Linux allow the loading of third party code into anything at all. As Raymond has explained many times before, large parts of Windows were designed back in the days when people were basically trusted to do the right thing and not screw up. This is now known to be a bad idea so modern parts of Windows are much more robust against "improvement" by 3rd parties, but the shell dates back to the early 90s and reflects the design mentality of the era.
No, you didn't understand the explanation. This sort of understanding failure is exactly what caused the need for the verclsid program in the first place!
The problem was that this specific shell extension (for an obsolete HP printer) contained a concurrency bug - it tried to synchronize with a thread in its DLL detach function. This is never correct, because in a DLL detach you can't make any assumptions about the liveness of other threads. Because there are buggy shell extensions out there that hang, they had a watchdog thread, but this design can't stop hangs that occur during shutdown of the process itself as this one did (which is presumably a rare kind of bug).
I really find it amazing that despite a clear explanation from Raymond, there are people here trying to blame Microsoft! It just goes to show that programming is hard, I guess ....
How much open source work have you actually done? I've done a lot, and this idea is one I see very often in people who haven't done any serious API development work before. The approach of attempting to patch every app when an API changes simply doesn't scale. There's a reason all the important open source APIs (gtk, glibc, alsa, X etc) have "gone stable" in the past 5 years, and it's simply a better approach.
Anyway, ignoring the obvious (!!) problems of scaling such an approach, you are confusing two unrelated things. Microsoft can simply/clean up APIs too - they have done it with DirectX and .NET, but that's irrelevant. The problem here is that there are lots of people in the world writing software who perhaps aren't well qualified, and even the ones that are well qualified make mistakes, even with the implementation of quite simple interfaces like IUnknown. I myself have messed up IUnknown before, in fact.
The root problem that caused the hang was attempting to cleanly handle buggy software. This is a common motif in software, hell, it practically motivated the move from the Windows 9x design to the NT fully protected architecture.
Multi-threading is never trivial.
I worked on Wine for a long time, which implements or maps the Win32 API. The complexity of Linux, Windows and MacOS X are all much the same - they are of the same design era, even OS X which is based on lots of older code at its heart. While the more modern parts of the Linux APIs like GTK+ are better than the Win32 equivalents that's just an age thing: the Win32 API has evolved over a much longer period of time. That means it's uglier (the world has learned a lot about API design since the 80s), but it also means there are far more people out there who know it, better tools support, and critically, more apps that use it!
For what it's worth, I agree with you 100%. The problem is that "computer science" is so vaguely defined it's not even a useful term. There are only a few jobs for computer scientists, and even then, those jobs usually require you to be both a talented software engineer AND a computer scientist. We have a few of those types at Google and I have mad respect for their skillz, but they would never have been hired if they didn't know C++ or Java - it's that simple.
Unfortunately, there is a small but not insignificant part of the software development population who always had a greater love for the mathematical side of computing - in which there aren't many jobs - and were never that keen on the gritty details of how computers actually work. So they end up bitter and take every opportunity to "remind" people that computer science isn't about programming or systems architecture, it's just maths (ie, the part they like). They conveniently ignore that the popular definition of computer science is what's taught on computer science courses, which should be a whole mix of things.
Universities did that originally, but the people who write the P2P software just had to get their free tunes/warez/movies, so they put in workarounds and now it's really hard to throttle this sort of traffic (because it's encrypted or uses unpredictable ports, or whatever). So now P2P is banned completely. Not really surprising if you ask me. Nor an "infringement of free speech" - it's pragmatic network engineering.
You just proved my point.
Linux is fucked by this kind of thing because (unless a lot changed in the last 12 months) there are no distros that silently install updates without any user intervention being required. Even distros like Ubuntu will pop up a balloon and require you to type in a password to install the updates. But we already know from experience with Windows that nothing short of a fully automatic system will do. If Linux had the popularity of Windows today, this exploit would still be being used 3 years from now.
3. C/C++ make it really easy to screw up.
How did it help NBC? Some other company was taking the ad revenues for a show they paid for. Only if you make some huge assumptions, like people who watch a show on YouTube then go watch it again (with the ads) on an actual TV, does it help NBC. Somebody else posted that SNL ratings didn't change despite the growing popularity of YouTube, so it seems like it didn't really help them.
Now having said that, I suspect a lot of the YouTube traffic is for people outside of the states who can't watch the shows on a regular TV anyway. These days I watch the Daily Show on the comedy central website, because it's not syndicated where I live (or not that I know of, anyway). I don't mind CC doing their own video streaming - it's more or less the same as what you'd see on YouTube anyway, but with ads owned by CC itself.
You shouldn't moderate comments down for being "wrong", this is basic moderator 101 stuff. If a comment is factually wrong, mod up a response that corrects it, or post yourself to do so. If it's "wrong" in the sense that you disagree with it, mod up the opposing viewpoint.
Volume key hacks are not solid. How many times is it necessary to point this out on discussions about AACS? The specification contains a wide variety of traitor-tracing algorithms that let you find a hacked player key given only the released volume keys, or even only the decrypted video itself.