If MS can leverage their strong market position in the desktop OS area to drive competing commercial producers of media players out of business, that is a freedom issue.
How many people actually believe there is any evidence that Microsoft has "illegally" driven Real or Quicktime out of the media player business? Everyone hates Real software with a passion, Apple's marketing of Quicktime is nonexistent at best, and Microsoft happens to have created a Media Player that is head and shoulders above the competition (for commercial non-free media players) but people still buy into this argument that Microsoft have somehow illegally pushed their competitors out of the business.
The EU's investigation of Media Player bundling is nothing more than an exercise in bureaucracy, a copycat of the Netscape case, but based on far weaker "evidence" much of which comes from serial deceiver, Real Networks. If people want to be gloat that Microsoft has finally gotten some comeuppance, then they should say so, but let's not pretend the Media Player part of this ruling is just and fair, because it's trumped-up bullshit.
The "Microsoft is not innovating" line is something I hear alot on Slashdot, and it's one of a few lines which really illustrates ignorance more than anything else. Anyone who has eye on emerging technologies would realize that Microsoft has a lot of stuff in the works - I don't know if I'd call it 8 billion dollars worth of innovation but then I don't know about all the internal products, research groups and projects that are going on inside the Deathstar. But innovation it is. Microsoft is pushing a vision of standardized programming models, development tools and APIs for all its products.
Just because you don't see tabbed browsing and "save as pdf" in Microsoft's current generation of projects, doesn't mean Microsoft aren't innovating.
there is no unjustified bias. the author clearly stated what the review was about in the introduction, so you don't have a case.
You're wrong. The author made some fairly broad encompassing statements in the introduction, but never introduced a scope that justified only including open source clients as "next generation" clients. The author assumed that the reader would agree with the implicit assertion that closed source, win32 only was bad, and not worthy of attention. There was bias - but no justification, apart from a circular arguement along the lines of "I'm not including closed source, win32 only, because they are closed source, and win32 only." The only mention of price, or hint at justification for the scope of the review is dishonestly hidden away in a FAQ entry that was added after the article was written.
But then, I've addressed this all before - perhaps if you bothered doing a little reading yourself rather than prematurely ejaculating your opinions and misguided venom I wouldn't have to restate this all again.
it is your own fault that you don't even bother to read the intro and then whine and bitch about it. rtfs or stfu.
Would you like some fries to go with that irony, sir? Perhaps you'd like to actually read what I've written and address the points I've raised before you spit out this lame little argument again. If you're not capable of rebutting a proper argument then you're just wasting my time. Bye now.
Thanks for that information. As you may have noticed, I later stated that I use Outlook 2003, so you could probably have assumed that I used the previous version as well. As a result, I'm in no position to judge if there are errors in the other sections so I refrained from saying so, and only commented on the portions with which I was familiar. But then, you aren't actually disagreeing with anything I've written are you? You're just reinforcing my opinion that this article is shoddy, and deserving of criticism. Clearly shoddy on all fronts and with unjustified bias.
Thanks Mr AC. You're clearly an utter jerk, but in this case, you've been a helpful jerk.
Yes, I read everything. I read the review and I read the introduction and I read the title, "Next generation mail clients reviewed", as well.
I read a title which implied the article would review the "next generation" of mail clients, not open or closed source, *nix or win32.
I read an introduction that claimed to be reviewing next generation email clients, again nothing is said about open source, closed source, *nix or Win32.
I read a dismissive explanation as to why some closed source and win32 only clients were not reviewed because they are closed source and win32 only, which struck me both as cyclic and contrary to the stated aims of the review. I also read the rather ad hoc and contradictory explanation that Outlook XP was included "for completeness" as a reference because it is the most used.
I read the individual reviews themselves, and noted that the Outlook review contained more than one factual error.
I read the conclusions, which sought to provide advice to readers without giving the scope of the review, but still attempted to advise windows users despite ignoring many of the clients for the windows platform. I also noted that very little use was made of the "reference" app, Outlook, in the conclusions.
Finally, I read the FAQ, which attempted to justify the non-inclusion of Outlook 2003. Again, it contained factual errors and very clear bias. The reviewer finally admitted that he did not review Outlook 2003 because he didn't have it, which made all his previously justifications smack of dishonesty, IMO. I also thought it very telling that a single FAQ entry gave more insight into the scope and motivation of the article than the introduction or conclusions of the article itself.
And then, after all that, I read your little rant about fanboys, FUD, and hypocrisy, and I laughed. Criticism of this review is more than justified, since it is both biased and factually incorrect on a number of counts. Attempting to dismiss all criticism as kneejerk reactions by fanboys is just wrong. Perhaps you'd realize that if you read the article with more than one eye open?
But then, I use Outlook 2003 and I'm very happy with it. A person who uses and sees the technical merit of Microsoft software can't possibly have a valid opinion can they? They're all "fanboys" who yell "FUD" and "hypocrisy" all the time. It's much easier to think of us that way, isn't it?
There is no FUD here. There is no hypocrisy. There may be bias, but at least he's clear on what he does and why he does it.
That's right. The only thing you can accuse him of is writing an article about one thing and saying it is about something completely different. If he wants to compare an older version of Outlook against the most recent versions of open source products but still claim to be comparing next generation email clients, then that's his prerogative. I mean, when he says "next generation email clients" it's pretty clear that what he means is "next generation open source email clients and one last generation closed source email client that but it's the most widely used so it's ok to include it only as a reference". It comes through in the article heading - right?
I think I can summarize your points concisely: he can exclude anything he wants, and still claim anything he wants, and he shouldn't be open to criticism from Microsoft fanboys, because they are Microsoft fanboys and we can cry "FUD! FUD!" or "Shill! Shill!" and ignore everything they say.
Whether or not you agree with the author on his reasoning is irrelevant - the article is not a review of next generation email clients, because it stupidly includes a review of a last generation email client. The fact is, the article claims to be something it isn't and attempts to explain this inconsistency by making excuses about price and availability in some obscure FAQ entry. Maybe not biased, but poor journalism nonetheless.
There's only so much steam you can blow off before it becomes hot air, no matter how much you try to rationalize it. I've seen way too many posts filled with regurgitated hearsay, assertions and disinformation about Microsoft products to be considered just "a little steam", and more than a couple of those posts have been modded up +5.
I'm sure all the actors are just happy to have taken part in the greatest movie making experience of all time. It's good that LotR has gotten some recognition at the Oscars, but the Oscars has never always been about the best, or most deserving: it's about what and who is fashionable at the time.
This is great news for Open Source, whose goal is to make software cheap and affordable for everybody. Microsoft has been making exorbitant profits from their products for way too long, and I'm glad that China is embracing the new way of Open Source where software is a basic social right of all citizens.
You might see it as great news for Open Source, but I see it is a horribly flawed analogy. Water, gas and electricity are micro-billed services, which we never own. Furthermore, we generally pay arbitrary regulatory bodies for the privilege of using them. Are you advocating a move to a system of leased software licensing - kind of like Microsoft's latest ideas?
"Basic software services" have about as little in common with basic utilities, in terms of complexity, delivery, life-cycle, utility and just about any other measure you can come up with.
Forget the future, what about now? Microsoft ships more than one piece of software that is head and shoulders above its nearest competitor, but people here still discount it all without a second thought, but then go on to expound on the virtues of open source, even though it isn't as good for some things.
You can mod me to hell, but a good part of success in the software industry is a healthy respect for rivals, and a willingness to accept good ideas wherever they come from.
I dunno, maybe because it would be an absolutely enormous undertaking that would require hitherto unforseen coordination between every single government department, exposure to a single enormous risk rather than small, controlled experiments, retraining on a massive scale, adoption of a platform that even leading advocates claim is not quite ready for the desktop, and the total absence of proof that any of the things you've just asserted would actually come true on such a large scale?
I dunno about you, but I'd prefer that the government moved in small steps, and got things right in small steps, rather than taking a big risk, fucking it up, and never trying it again. Or is that just me?
If by "doing their job" you mean "receiving complaints", then yes, they're doing their job. I seriously doubt any action will come out of this particular complaint. The entire basis of the complaint is still under consideration in the US, and I doubt the Australian courts would preempt the US courts, let alone a government body that regulates competition. I'd love to see the ACCC take a stand against litigious predators like SCO, but I doubt this kind of argument will spur the ACCC on. Complain about anti-competitive behaviour, sure, but the ACCC won't act on some unproven legal argument. The ACCC isn't a court of law, it's a Competition and Consumer Commission.
But that's the point: they don't even make any claims about why the status quo is better, or even if it's better, just that the proposed measures are draconian. They don't propose a better solution, just that the EU's solution is bad. They don't even make it clear whether they think the EU should try to control this kind of thing. And for a group that seems to have a fairly large interest in patents and IP, that grabs me as kind of lame. Do they have an agenda or not?
The FUD was in the article, right there next to all the factual errors. Cringley falsely attributes Microsoft's "ever increasing" security problems to.NET, when nothing could be further from the truth. Most of the security problems Microsoft has had recently were not caused by any.NET applications, and were, in fact, caused by things that.NET could actually address. His entire motivation for code obfuscation is just wrong - code obfuscation is not intended for security, but protection of IP. Hell, the guy doesn't even seem to understand virtual machines, but talks about things "being intended for interpretation". And don't tell me the guy is dumbing down - he's just plain dumb.
And those "real-world" issues you cite are actually non-issues that Cringely has made up in his head. It's typical Cringely style - make up some clear and present danger, or otherwise earth shattering motivation to explain some technology that he doesn't quite understand. And then, when something very vaguely related to something he's ranted on about, he revises his claims he predicted it. Very Nostradamus.
Cringely is a charlatan. Don't bother wasting your time on him.
Disputes about patents and trade secrets/confidential information taken out of the scope of the directive altogether. The draconian measures being discussed are completely inappropriate for such complex disputes.
So does the FFII believe that these kind of disputes should be left to the legal systems of individual member nations then or would the FFII prefer to have the EU draft some other directives to handle them uniformly? Or would any directive on this kind of dispute be too draconian according to the FFII?
It's fine to oppose something on principle but the FFII's alert doesn't seem to be proposing any alternative solution, other than "not what you've got", which weakens their stance IMHO. If they want to make a stand on IP, then they should do so instead of just being naysayers.
Why does the moron get space on/. at all? Surely people can see the glaring errors, the ridiculous assertions and the "I'm at the center of the tech universe, so if I happen to have a half-baked idea about something then it must be so!" attitude that Cringely articles reek of.
I feel dirty after reading them. God help the world if Enderle and Cringely ever start working together.
Cringely rarely makes sense about anything. Almost every Cringely article I've read in the last few years contains ridiculous assertions, glaring omissions or blindingly obvious factual errors.
I pity the people who take Cringely articles seriously and the fact that they get semi-regular attention on/. is more than a little worrying.
Are we believing the Microsoft Marketing Machine when they say that their security was not breached? I mean, they've never had security issues before have they?
There's actually been quite a bit of detail released about the potential source of the leak. For example, check out this article. Seems like an awful lot of detail for just the Microsoft Marketing Machine.
I said CAN. That's the whole point. If there exists a one-to-one map, then an pure symbol translation process can perform the compilation.
And as I said, you CAN only have a one to one map in very, very limited cases.
So it is algorithmic.
Of course decompilation is algorithm. If it wasn't algorithmic, we couldn't do it with computers. The point is, with decompilation, you lose all the semantic information conveyed through comments and identifier naming. But this makes total sense, since the binary form is not a design but the result of a design.
So you think there's a specification for how many times the piling machine must work, every movement of the truck axles, there's a specification for when it rains and all the vagaries assoicated working with different weather?
That is such a strawman argument.
1. You don't have to specify an exact number of times a piling machine must work - you specify the when the piling machine should stop. I don't have to specify an exact number of times an application message loop executes before the application is over - it executes until a condition becomes true, or something breaks it.
2. You don't specify every movement of the axles, you specify where you move the truck to, let the operator (hardware) of the truck operate the truck, and let the truck deal with its internals (abstraction). I don't have to program how each pixel appears on the screen every time I write a new X app.
3. You don't have to specify rain - rain is an input so you specify what you do when it rains (most likely, nothing). Just because there is non-determinism in the process doesn't make it non-algorithmic. I don't program the user of an application - I program how the application responds to the user's input.
If the process is algorithmic, why do building schedules run overtime?
Because the inputs are non-deterministic and the hardware is unreliable. Why does my network print job sometimes take half a second to spool and sometimes take half a minute? Because I print different jobs and occaisionally the network traffic is heavy.
You seem to be confusing "algorithmic" with "deterministic". And if your argument is that decompilation is deterministic for a given binary, so the binary equals source, then you are still wrong. A specific decompiler may be deterministic for a given binary, but the total class of decompilers is not. And despite your "CAN be one-to-one" claims, you will virtually never get the original source code back from your binary.
But instead of paying $23.7 million for the Microsoft solution, Munich's city council opted to spend roughly $35.7 million to switch to open source, saying that the higher price would be offset by lower costs and more flexibility in licensing fees and software choices over the long run.
The part about saying that the higher price would be offset... over the long run, says to me that they thought the Microsoft solution was cheaper up front. Pretty straight forward, or am I missing something?
I work next to a guy who has both Linux and OS X on his Powerbook. I'd say he uses Linux 99% of the time and when he does use OS X, it's only for doing presentations, drawing pictures and the like, and he constantly bitches about how unresponsive it is compared to Linux and how this quirk or that quirk is so annoying.
But then, he is a masochist, so my argument is moot.
It is a qualitatively inaccurate comparison. There can be a one-to-one correspondence between C and machine code.
Not unless you compile with full debugging information and no optimizations, and your compiler also saves any in-line comments as well. The relationship between a C program and machine code is actually many-to-many in all but the most trivial cases.
Mapping from C to machine code can be done purely, and reliably by an algorithmic process. Mapping from blueprints to house must be done by human muscle and machinery. It is not algorithmic.
It is algorithmic. I could quite easily write an algorithm for building a house from plan, and once I'd written the algorithm I'd get the hardware (human muscle and machinery) to execute my algorithm.
I think you have fallen for on of the common Open Source argument fallacies - that the source code is the software. Software is the stuff you execute. Source code is the abstraction used to design the software. Sometimes they are the same thing but often not.
Not to mention that, money spent on Linux stays in the local economy instead of feeding a foreign company.
Last time I checked, Suse was owned by Novell, and IBM wasn't a German company. And Microsoft employs more German employees than Suse. Not sure about IBM's German business holdings though so you may be right.
Slashdot Mirror
If MS can leverage their strong market position in the desktop OS area to drive competing commercial producers of media players out of business, that is a freedom issue.
How many people actually believe there is any evidence that Microsoft has "illegally" driven Real or Quicktime out of the media player business? Everyone hates Real software with a passion, Apple's marketing of Quicktime is nonexistent at best, and Microsoft happens to have created a Media Player that is head and shoulders above the competition (for commercial non-free media players) but people still buy into this argument that Microsoft have somehow illegally pushed their competitors out of the business.
The EU's investigation of Media Player bundling is nothing more than an exercise in bureaucracy, a copycat of the Netscape case, but based on far weaker "evidence" much of which comes from serial deceiver, Real Networks. If people want to be gloat that Microsoft has finally gotten some comeuppance, then they should say so, but let's not pretend the Media Player part of this ruling is just and fair, because it's trumped-up bullshit.
The "Microsoft is not innovating" line is something I hear alot on Slashdot, and it's one of a few lines which really illustrates ignorance more than anything else. Anyone who has eye on emerging technologies would realize that Microsoft has a lot of stuff in the works - I don't know if I'd call it 8 billion dollars worth of innovation but then I don't know about all the internal products, research groups and projects that are going on inside the Deathstar. But innovation it is. Microsoft is pushing a vision of standardized programming models, development tools and APIs for all its products.
Just because you don't see tabbed browsing and "save as pdf" in Microsoft's current generation of projects, doesn't mean Microsoft aren't innovating.
there is no unjustified bias. the author clearly stated what the review was about in the introduction, so you don't have a case.
You're wrong. The author made some fairly broad encompassing statements in the introduction, but never introduced a scope that justified only including open source clients as "next generation" clients. The author assumed that the reader would agree with the implicit assertion that closed source, win32 only was bad, and not worthy of attention. There was bias - but no justification, apart from a circular arguement along the lines of "I'm not including closed source, win32 only, because they are closed source, and win32 only." The only mention of price, or hint at justification for the scope of the review is dishonestly hidden away in a FAQ entry that was added after the article was written.
But then, I've addressed this all before - perhaps if you bothered doing a little reading yourself rather than prematurely ejaculating your opinions and misguided venom I wouldn't have to restate this all again.
it is your own fault that you don't even bother to read the intro and then whine and bitch about it. rtfs or stfu.
Would you like some fries to go with that irony, sir? Perhaps you'd like to actually read what I've written and address the points I've raised before you spit out this lame little argument again. If you're not capable of rebutting a proper argument then you're just wasting my time. Bye now.
Thanks for that information. As you may have noticed, I later stated that I use Outlook 2003, so you could probably have assumed that I used the previous version as well. As a result, I'm in no position to judge if there are errors in the other sections so I refrained from saying so, and only commented on the portions with which I was familiar. But then, you aren't actually disagreeing with anything I've written are you? You're just reinforcing my opinion that this article is shoddy, and deserving of criticism. Clearly shoddy on all fronts and with unjustified bias.
Thanks Mr AC. You're clearly an utter jerk, but in this case, you've been a helpful jerk.
Yes, I read everything. I read the review and I read the introduction and I read the title, "Next generation mail clients reviewed", as well.
I read a title which implied the article would review the "next generation" of mail clients, not open or closed source, *nix or win32.
I read an introduction that claimed to be reviewing next generation email clients, again nothing is said about open source, closed source, *nix or Win32.
I read a dismissive explanation as to why some closed source and win32 only clients were not reviewed because they are closed source and win32 only, which struck me both as cyclic and contrary to the stated aims of the review. I also read the rather ad hoc and contradictory explanation that Outlook XP was included "for completeness" as a reference because it is the most used.
I read the individual reviews themselves, and noted that the Outlook review contained more than one factual error.
I read the conclusions, which sought to provide advice to readers without giving the scope of the review, but still attempted to advise windows users despite ignoring many of the clients for the windows platform. I also noted that very little use was made of the "reference" app, Outlook, in the conclusions.
Finally, I read the FAQ, which attempted to justify the non-inclusion of Outlook 2003. Again, it contained factual errors and very clear bias. The reviewer finally admitted that he did not review Outlook 2003 because he didn't have it, which made all his previously justifications smack of dishonesty, IMO. I also thought it very telling that a single FAQ entry gave more insight into the scope and motivation of the article than the introduction or conclusions of the article itself.
And then, after all that, I read your little rant about fanboys, FUD, and hypocrisy, and I laughed. Criticism of this review is more than justified, since it is both biased and factually incorrect on a number of counts. Attempting to dismiss all criticism as kneejerk reactions by fanboys is just wrong. Perhaps you'd realize that if you read the article with more than one eye open?
But then, I use Outlook 2003 and I'm very happy with it. A person who uses and sees the technical merit of Microsoft software can't possibly have a valid opinion can they? They're all "fanboys" who yell "FUD" and "hypocrisy" all the time. It's much easier to think of us that way, isn't it?
There is no FUD here. There is no hypocrisy. There may be bias, but at least he's clear on what he does and why he does it.
That's right. The only thing you can accuse him of is writing an article about one thing and saying it is about something completely different. If he wants to compare an older version of Outlook against the most recent versions of open source products but still claim to be comparing next generation email clients, then that's his prerogative. I mean, when he says "next generation email clients" it's pretty clear that what he means is "next generation open source email clients and one last generation closed source email client that but it's the most widely used so it's ok to include it only as a reference". It comes through in the article heading - right?
I think I can summarize your points concisely: he can exclude anything he wants, and still claim anything he wants, and he shouldn't be open to criticism from Microsoft fanboys, because they are Microsoft fanboys and we can cry "FUD! FUD!" or "Shill! Shill!" and ignore everything they say.
No hypocrisy.
Whether or not you agree with the author on his reasoning is irrelevant - the article is not a review of next generation email clients, because it stupidly includes a review of a last generation email client. The fact is, the article claims to be something it isn't and attempts to explain this inconsistency by making excuses about price and availability in some obscure FAQ entry. Maybe not biased, but poor journalism nonetheless.
There's only so much steam you can blow off before it becomes hot air, no matter how much you try to rationalize it. I've seen way too many posts filled with regurgitated hearsay, assertions and disinformation about Microsoft products to be considered just "a little steam", and more than a couple of those posts have been modded up +5.
I'm sure all the actors are just happy to have taken part in the greatest movie making experience of all time. It's good that LotR has gotten some recognition at the Oscars, but the Oscars has never always been about the best, or most deserving: it's about what and who is fashionable at the time.
This is great news for Open Source, whose goal is to make software cheap and affordable for everybody. Microsoft has been making exorbitant profits from their products for way too long, and I'm glad that China is embracing the new way of Open Source where software is a basic social right of all citizens.
You might see it as great news for Open Source, but I see it is a horribly flawed analogy. Water, gas and electricity are micro-billed services, which we never own. Furthermore, we generally pay arbitrary regulatory bodies for the privilege of using them. Are you advocating a move to a system of leased software licensing - kind of like Microsoft's latest ideas?
"Basic software services" have about as little in common with basic utilities, in terms of complexity, delivery, life-cycle, utility and just about any other measure you can come up with.
Forget the future, what about now? Microsoft ships more than one piece of software that is head and shoulders above its nearest competitor, but people here still discount it all without a second thought, but then go on to expound on the virtues of open source, even though it isn't as good for some things.
You can mod me to hell, but a good part of success in the software industry is a healthy respect for rivals, and a willingness to accept good ideas wherever they come from.
I dunno, maybe because it would be an absolutely enormous undertaking that would require hitherto unforseen coordination between every single government department, exposure to a single enormous risk rather than small, controlled experiments, retraining on a massive scale, adoption of a platform that even leading advocates claim is not quite ready for the desktop, and the total absence of proof that any of the things you've just asserted would actually come true on such a large scale?
I dunno about you, but I'd prefer that the government moved in small steps, and got things right in small steps, rather than taking a big risk, fucking it up, and never trying it again. Or is that just me?
If by "doing their job" you mean "receiving complaints", then yes, they're doing their job. I seriously doubt any action will come out of this particular complaint. The entire basis of the complaint is still under consideration in the US, and I doubt the Australian courts would preempt the US courts, let alone a government body that regulates competition. I'd love to see the ACCC take a stand against litigious predators like SCO, but I doubt this kind of argument will spur the ACCC on. Complain about anti-competitive behaviour, sure, but the ACCC won't act on some unproven legal argument. The ACCC isn't a court of law, it's a Competition and Consumer Commission.
But that's the point: they don't even make any claims about why the status quo is better, or even if it's better, just that the proposed measures are draconian. They don't propose a better solution, just that the EU's solution is bad. They don't even make it clear whether they think the EU should try to control this kind of thing. And for a group that seems to have a fairly large interest in patents and IP, that grabs me as kind of lame. Do they have an agenda or not?
The FUD was in the article, right there next to all the factual errors. Cringley falsely attributes Microsoft's "ever increasing" security problems to .NET, when nothing could be further from the truth. Most of the security problems Microsoft has had recently were not caused by any .NET applications, and were, in fact, caused by things that .NET could actually address. His entire motivation for code obfuscation is just wrong - code obfuscation is not intended for security, but protection of IP. Hell, the guy doesn't even seem to understand virtual machines, but talks about things "being intended for interpretation". And don't tell me the guy is dumbing down - he's
just plain dumb.
And those "real-world" issues you cite are actually non-issues that Cringely has made up in his head. It's typical Cringely style - make up some clear and present danger, or otherwise earth shattering motivation to explain some technology that he doesn't quite understand. And then, when something very vaguely related to something he's ranted on about, he revises his claims he predicted it. Very Nostradamus.
Cringely is a charlatan. Don't bother wasting your time on him.
Above all, FFII would like to see:
Disputes about patents and trade secrets/confidential information taken out of the scope of the directive altogether. The draconian measures being discussed are completely inappropriate for such complex disputes.
So does the FFII believe that these kind of disputes should be left to the legal systems of individual member nations then or would the FFII prefer to have the EU draft some other directives to handle them uniformly? Or would any directive on this kind of dispute be too draconian according to the FFII?
It's fine to oppose something on principle but the FFII's alert doesn't seem to be proposing any alternative solution, other than "not what you've got", which weakens their stance IMHO. If they want to make a stand on IP, then they should do so instead of just being naysayers.
Why does the moron get space on /. at all? Surely people can see the glaring errors, the ridiculous assertions and the "I'm at the center of the tech universe, so if I happen to have a half-baked idea about something then it must be so!" attitude that Cringely articles reek of.
I feel dirty after reading them. God help the world if Enderle and Cringely ever start working together.
Cringely rarely makes sense about anything. Almost every Cringely article I've read in the last few years contains ridiculous assertions, glaring omissions or blindingly obvious factual errors.
/. is more than a little worrying.
I pity the people who take Cringely articles seriously and the fact that they get semi-regular attention on
Are we believing the Microsoft Marketing Machine when they say that their security was not breached? I mean, they've never had security issues before have they?
There's actually been quite a bit of detail released about the potential source of the leak.
For example, check out this article. Seems like an awful lot of detail for just the Microsoft Marketing Machine.
I said CAN. That's the whole point. If there exists a one-to-one map, then an pure symbol translation process can perform the compilation.
And as I said, you CAN only have a one to one map in very, very limited cases.
So it is algorithmic.
Of course decompilation is algorithm. If it wasn't algorithmic, we couldn't do it with computers. The point is, with decompilation, you lose all the semantic information conveyed through comments and identifier naming. But this makes total sense, since the binary form is not a design but the result of a design.
So you think there's a specification for how many times the piling machine must work, every movement of the truck axles, there's a specification for when it rains and all the vagaries assoicated working with different weather?
That is such a strawman argument.
1. You don't have to specify an exact number of times a piling machine must work - you specify the when the piling machine should stop. I don't have to specify an exact number of times an application message loop executes before the application is over - it executes until a condition becomes true, or something breaks it.
2. You don't specify every movement of the axles, you specify where you move the truck to, let the operator (hardware) of the truck operate the truck, and let the truck deal with its internals (abstraction). I don't have to program how each pixel appears on the screen every time I write a new X app.
3. You don't have to specify rain - rain is an input so you specify what you do when it rains (most likely, nothing). Just because there is non-determinism in the process doesn't make it non-algorithmic. I don't program the user of an application - I program how the application responds to the user's input.
If the process is algorithmic, why do building schedules run overtime?
Because the inputs are non-deterministic and the hardware is unreliable. Why does my network print job sometimes take half a second to spool and sometimes take half a minute? Because I print different jobs and occaisionally the network traffic is heavy.
You seem to be confusing "algorithmic" with "deterministic". And if your argument is that decompilation is deterministic for a given binary, so the binary equals source, then you are still wrong. A specific decompiler may be deterministic for a given binary, but the total class of decompilers is not. And despite your "CAN be one-to-one" claims, you will virtually never get the original source code back from your binary.
That's not how I read it. From the article:
... over the long run, says to me that they thought the Microsoft solution was cheaper up front. Pretty straight forward, or am I missing something?
But instead of paying $23.7 million for the Microsoft solution, Munich's city council opted to spend roughly $35.7 million to switch to open source, saying that the higher price would be offset by lower costs and more flexibility in licensing fees and software choices over the long run.
The part about saying that the higher price would be offset
You might not be able to steal copyrighted material by definition but you can steal trade secrets.
I work next to a guy who has both Linux and OS X on his Powerbook. I'd say he uses Linux 99% of the time and when he does use OS X, it's only for doing presentations, drawing pictures and the like, and he constantly bitches about how unresponsive it is compared to Linux and how this quirk or that quirk is so annoying.
But then, he is a masochist, so my argument is moot.
It is a qualitatively inaccurate comparison. There can be a one-to-one correspondence between C and machine code.
Not unless you compile with full debugging information and no optimizations, and your compiler also saves any in-line comments as well. The relationship between a C program and machine code is actually many-to-many in all but the most trivial cases.
Mapping from C to machine code can be done purely, and reliably by an algorithmic process. Mapping from blueprints to house must be done by human muscle and machinery. It is not algorithmic.
It is algorithmic. I could quite easily write an algorithm for building a house from plan, and once I'd written the algorithm I'd get the hardware (human muscle and machinery) to execute my algorithm.
I think you have fallen for on of the common Open Source argument fallacies - that the source code is the software. Software is the stuff you execute. Source code is the abstraction used to design the software. Sometimes they are the same thing but often not.
Not to mention that, money spent on Linux stays in the local economy instead of feeding a foreign company.
Last time I checked, Suse was owned by Novell, and IBM wasn't a German company. And Microsoft employs more German employees than Suse. Not sure about IBM's German business holdings though so you may be right.