Companies get taxed for the goods they buy, the land they own, some of their equipment (taxes on company vehicles), products they sell in most states (sales tax), and wages they pay to employees. The owner of a company can withdraw money directly out of a company without taxation (though I could be wrong), but generally the owner is paid a wage just like everyone else. Since shareholders are owners, them being paid a wage and taxed isn't any different than any other company. The fact is, everything tends to be taxed multiple times which just advantages vertical monopolies. I don't pity shareholders.
I think your definition of measurement needs some work. The problem is, measurement is a side effect of observation but then that side effect can have other side effects (think of the way an electron can perturb an atom which causes it to knock into another atom which...etc). So "measurements" are occuring all the time from the interaction of fields (say a star or comet millions of light years away moving) or particles (spare neutrinos, electrons, photons, etc). The only really surprising thing is what our "measurement" seem to indicate.
It's a good thing to a point. At some point, each manufacturer needs to clarify the difference between their various products other than price or no one will want to buy their products versus someone who does do semi-sane labelling. But, I do agree that up to now it's been like AMD and Intel have each been selling the analogy of different cubical containers where only one dimension is measured and attached as a label (imagine if Pepsi and Coke were selling 2" or 1" bottles (where the 2" or 1" represents the width) -- though at least then you know what Coke and Pepsi are and you can physically try to compare the bottles to get an idea of their relative volumes..and it'd be a hell of a lot easier to use some standard measure). Now, obviously CPUs aren't pop, but there's nothing stopping AMD and Intel from measuring in something meaningful like the time to encode a DVD or the average frame rate of some CPU bound game. Or they could just tell you the Mhz and the CPU core so we could read a few benchmarks and get an idea of the scaling power of Mhz for a core.
>>If you leave the doors to your house open, and a large neon sign over the threshold saying 'WELCOME', you'll be *damned* lucky if your insurer would pay up.
>This is more like just leaving your doors unlocked. There is no protocol for a system to advertise it's vulnerabilities.
No, it's more like your door's lock having a structural defect. And in some cases, (not sure of this one) a web service will happily tell you what version its running which is a nice advertising that the defective lock is present.
>Without regard to whether your doors were locked it is illegal to steal things from your house.
It's not stealing, though. If you walk into a house and drink some water from the faucet or breathe air, you can get in trouble for trespassing and theft of service. And if you pee on the floor, it's vandalism. And if the worm was actually copying things, there'd be copyright infringement (it's actually a good question of if the sasser worm author can sue everyone *else* for copyright infringement). But there's no stealing involved unless the worm actually picked up the computer and carried it out of the place for him to take.
I'd love to see the same penalties applied to breaking into computer systems as as applied to rape, because then I think the people doing these things would have to consider what they're doing and their imipact on their victims. Right now, I think most people like this guy think it's a game, instead of focusing on the incredible amounts of agony that is caused.
I pray you're being a troll. If not, step back for a minute from your weird obsession personification of a computer, and realize that a computer is a product just like a toaster. It isn't an irreplaceable, unique person composed of trillions of cells where a violent physical act is unlikely to be entirely erased from either physical or mental realms. Computers aren't alive and setting the same punishment for abusing either horribly degrades the perceived worth of a human.
Only at the point at which computers are in a position where they can do the same sort of physical harm to a person should there begin to be punishment in the same realm. But that's just harming a person by proxy which is already covered by law (killing a person by a gun is the same as killing a person by strangling them, legally (and the same issue of a question of neglect by using a proxy comes up)). Then comes the question of is the computer maker responsible, and I'd say that they might be if the physical harm was the result of a defect.
While I agree with you on the point that A -> B doesn't mean B -> A, stating A -> B isn't a bad thing in itself. In fact, you bring up the point that if I were to create a program and release it under the GPL, I could not release the source code which would in effect have no real effect (the only thing under which I could be in violating of is copyright and I own the copyright). As for your point about the programmer using GPLed code and the company not knowing, you're right that the company doesn't have to agree to the GPL. Instead, they'd be liable for statutory copyright infringement ($150,000/copy) while the program would be liable for statutory and punitive ($600,000/copy). Unlike stolen goods, you can't claim ignorance of the fact the goods were stolen as an excuse when it comes to copyright--if one could, shell companies would spring up with some contact giving them the materials to pirate until they were shutdown. In any case, the cheapest solution in most cases is to just GPL the program and most GPL authors will call off the suit. The only other economic solution would be to pay for current damages and rewrite the program from some known safe source.
You're right that having duplicate functions in different shared libraries means opposing apps real use of ram is larger than if they both used the same shared library or their shared libraries used a common lower library, but I think you're overlooking another point. Whenever there's a page of a library or a process' code that is going to be swapped out, it can simply be dumped. That means unoften used sections of different shared libraries aren't actually in memory. Also, because the libraries are already on the HD there's no swapout that occurs (though obviously swapin can occur). So the wasted bloat of most libraries matters little. And the often used stuff, which it does take up more ram than if it were unified is still a very miniscule section of ram usage (even at 8MB for qt and 2MB for gtk+ (plus another 1MB for glib and gdk) which cover all gtk+ and qt apps, you're using a amazingly small percentage of any modern system's ram (less than 9% of 128MB); add all the shared libs and you're still not using more than around 32MB which while large provides a massive amount of services for all sorts of apps which would otherwise be much larger). So, unification is good, but it's not nearly the worry that you seem to believe it is.
>Copyright holders hold a number of different rights, several of the most important of which are listed at 17 USC 106. One of them is the right to prevent others from distributing copies of the work.
Yes and no. The problem is, if you give the copyright of a work to someone in the UK, the Berne convention should extend that copyright to the world (and hence back into the US)--just like two people can create the same copyrighted work independently and claim copyright on it. As a result, the transfer of copyright to someone else could effectively make two distributors in a single country. The only way around this is to *not* hand over copyright but instead setup a contract that limits the reach of distribution.
>As it happens, there are exceptions -- limitations -- to that right. But due to the precise language of one of them (17 USC 109) things get a little wonky. What it says is: the owner of a particular copy... lawfully made under this title... is entitled, without the authority of the copyright owner, to sell or otherwise dispose of the possession of that copy or phonorecord.
You're actually taking the line out of context. Here is the whole line.
Notwithstanding the provisions of section 106(3), the owner of a particular copy or phonorecord lawfully made under this title, or any person authorized by such owner, is entitled, without the authority of the copyright owner, to sell or otherwise dispose of the possession of that copy or phonorecord.
So, if allofmp3 is authorized to make copies (it obviously is under some contract), then the user as well as allofmp3 can resell the music. The only way to stop the user from doing it is a contract. And the only way to prevent allofmp3 from doing it outside of the scope of what is wanted is also a contract. Of course, such contracts could convievably be struck down as illegal (by extension of copyright law written under scope of the constitution). So, I'm still not sure where copyright stops it.
> No, you're wrong. Look, a copyright is not a single, indivisible thing. Think of it like a bundle of sticks. One pair of sticks are the right to reproduce and distribute copies in the US. Another pair are the rights to reproduce and distribute copies in the UK. Perry gave Sid the latter pair, but kept the former for himself.
I'm not sure how that prevents importing. Once it's distributed in the UK, why can't it be imported? I don't see anything in US law that prevents it. More importantly, I wonder how Sid ever got the work in the first place. The distribution to Sid from Perry seems to fall under your case too.
> And the way the law is written, first sale only applies to copies that were made legally by whoever had the US rights.
First Sale Doctrine, as I've read it, never makes mention of US rights. It only makes mention of copyright in general. And I don't see mention of dealing with the scenario you mention, either. Or the situation of importing a copy made in the UK into the US when an author has copyright over a work in both countries.
>>Then how can iTunes do it?
>They're given permission to do so by the US copyright holder. And note that they try very hard to only do business where the local rightsholders have given them permission to do so.
I think I said this already. They have a contract with iTunes. It'd seem that at least one of the reasons a contract is in place is because otherwise exporting to other countries would be legal. But maybe I'm wrong. I'd really like to know what part of US law covers importing original or used copies of a copyrighted work for various situations (like the same right holder in one of the two countries or in both).
>>Umm, how do they sell stuff on the internet without charging money?
>They'll make it up in volume.
I know your comment was meant to be funny, but that's actually true to an extent. Just like how ads pay for free tv broadcasts, I wouldn't doubt the volume of people getting free audio online wouldn't provide a nice revenue stream for doing ads. Of course, there's nothing stopping a stream from being 90% ads and having people pay to get 45% ads instead..or was it that cable got you more channels with ads. Anyways, there's this magically thing called the radio which people have been known for years to tape record off of even though it wasn't always the best quality. Streaming audio would probably fit well into that category. I guess it's funny to me how while MS is trying to shift from per item to per time the RIAA is dead-set on doing everything on per item. I guess that's just their little way of saying MS has too little software and the RIAA has way too much music.
However, if you copy it and get a second record, assuming the copying isn't legal (per 17 USC 107 or 1008 or whatever) then you CAN'T resell the second record under the first sale provision (109).
Problem one, you assume the copying isn't legal. The fact is, there's the equivalent of two copying going on. Labels are selling music to allofmp3 and allofmp3 is selling songs to you. The slight difference is more than likely allofmp3 isn't actually getting source copies from the label each time but has some contract. So, allofmp3 has to follow its contract for which it might need for you to sign a contract to remove itself from liability for your illegal actions. And then the legality is placed squarely on the user.
Likewise, if Perry sold his rights in the UK to his close friend Sid Vicious, and Sid was the one making copies in the UK, you couldn't -- as a matter of first sale -- import those copies into the US. There is a good reason for that.
If Perry sells his rights in the UK to Sid, then you can obviously import any copies Sid makes for sale because they're copyrighted. Or do you think the RIAA can't sell US music in the UK because it got hold of the copyright for songs from the authors?
Imagine that there was a small country that bordered the US and could easily ship stuff here. We'll call it Moosylvania. Further, imagine that Moosylvania has no copyright laws at all. This means it's legal for the locals to copy anything they want. If they could freely export it to the US, they'd just do an end run around our copyright laws, and everyone would buy cheap, unauthorized Moosylvanian copies, basically leaving the US copyright holders screwed.
Except Russia has copyright law. And Russia and the US have almost certaintly set up a treaty to deal with copyright law (the whole point of the Berne convention and its cousins, btw, was to unify copyright law enough so that copyright could extend around the globe..and that means importing copyrighted works from Russia should be perfectly legal in itself).
So, for first sale to apply, the copyright holder who made the copy has to be the US copyright holder. If that's not so, even though the copy was made legally over there, it won't qualify, because it would not have been made legally if it had been made over here.
That really doesn't make sense. Of course, it's a good question on whether it's even possible for someone in the US to hand over their copyright in just one country. And if it *was* possible, then it obviously means you can import the song because it's still under copyright (just not a local one). If copyright can extend from the US to the UK, I'm not sure why it can't extend back again.
When you download from this site, there is a master copy in Russia. At the end of the process, there is a master copy in Russia AND a copy on your hard drive. That's two copies, and that already indicates that it's not an import. And the copyright holder has the exclusive right to reproduce his work in the US per 106.
Then how can iTunes do it? Through a contract, of course. And the contract involves making a copy for sale. That copy for sale is paid for before being resold to the consumer. If that weren't true, iTunes wouldn't be paying another company. Or are you saying the contract isn't a valid way of reproducing a work? If that's true, then I hope you don't believe an even less strong legal item called a license (like GPL or the BSD variety) to redistribute works.
That's not how big O notation works. It's actually defined as:
If O(n log n), then
c*f(n)+k < n log n
defines the worst case scenario for time where c and k are constants and f is the function being used. n is always the number of elements being sorted (for sort algorithms). So, the issue is what k and c are in each of the various algorithms. It might be that c and k are huge for heap/merge sort, but with quicksort as O(n^2), it'd take either a hugely massive c and/or k or really small n for quicksort to always be faster than heap/merge sort.
Of course, as was pointed out, quicksort is more trivially to write, is generally n log n performance, and apparently has a lower c value (IIRC, heapsort has a c of 2). So, people use quicksort even though quicksort isn't guaranteed to always have the best time. Heck, people still use bubblesort (which is fine for really small n's as bubblesort's k is really small). Personally, I'd rather sort algorithms (and gc algorithms) be stuffed into system wide libraries and possibly let an outside function chose which to use; gmp which uses several different methods for doing bignum integer math is a good example of just having a library choose the right algorithm for the job; it'd seem smart to have an equivalent sort algorithm which was based on n and either worst or average case as chosen by the programmer.
So, the shoe is on the other foot. And copyright can be twisted to advantage the user (GPL). That doesn't mean every license is a good license.
The fact is, the kernel doesn't arbitarily malfunction when it's tainted. Instead, the taintedness is a great sign to tell the user that they really need to go to the original authors for help since no one else is able to properly debug their proper (and of course, two different modules from two different companies which each taint the kernel creates a problem which no single entity can resolve). Faking the string to not cause taintedness helps no one (in the short term it might help the company, but it might not in the long run; people might pay support money to get bugs fixed in one tainted module). Faking a string in a printer cartridge helps the user to get cheaper ink. It also helps create competition (always a good thing).
Ironically, Lexmark's cases against various clone ink cartridge makers might decide the result of this same type of deception. Faking a string to make some program behave the way you want might be unhelpful and possibly unethical (by misleading users into believing they're using only GPLed code or wasting developers time on problems they can't solve thanks to code they can't see), but it's hard to see how it could be made illegal. Now getting such companies for false advertising...
Most (though not all) companies do have in-house or contracted technical support for user error. Paying MS or any other technical company a glob of money for user error is a huge waste of money. You pay MS or other technical companies for the non-user error technical support like bug reporting, bug fixing, etc. The fact is, though, you really shouldn't have to pay to report or help fix a bug that is the fault of another company. It'd be like finding defects in all the cars you sell and paying the manufacturer to get them fixed.
The only advantage paying a company has, innately, over getting equivalent software for free is when a bug is discovered there's possible motivation for the company to fix the bug so the user might be inclined to buy future versions. However, the only way that works is if the *buyer* actually pays attention to bugs, notices they're not being fixed, and then proceeds to buying from someone else. If there are no buyable alternatives, the next best thing would be to take something that's free and turning the money that would have went to buying software into improvements to the free thing.
So, the best step at the moment would seem for most MS Office buyers, if they're actually seeing problems, to go first to Wordperfect or StarOffice. But technical support for user error is still something that'll be handled in-house.
Windows 9X does have code to allow Windows 9X to use DOS drivers (The 16bit drivers, if you will.) However, This is done by creating a virtual 8086 context, and making calls to int 0x21.
Close, but no cigar. DOS drivers access hardware just as directly as Windows 9x drivers. The functions it exposes are then used by Windows 9x to some level. Drivers can be 32-bit (just has to switch between 32-bit/16-bit mode each call) and run in ring0 (just like DOS programs..v86 mode wouldn't let you run Doom). The reason why DOS drivers were discouraged is because you can't move around DOS drivers in memory through paging (at least not safely), DOS drivers didn't help promote the use of Windows NT (ie, DOS drivers meant people were stuck in 9x), and DOS drivers don't offer as effect an interface or complete an interface to all the features available. And while I'd claim Windows 9x can be a real OS, it's a real OS on a real OS. In that context, X is a real OS on a real OS (X has drivers for various hardware while providing an api).
And I think why people are so upset about Windows 9x, not calling it a real OS, is Microsoft decided to not make it possible to buy a real OS (MS-DOS, DR-DOS, etc) with a real OS (the Windows part of Windows 9x). There are still a lot of people who like DOS in their own way, and cutting out all the other DOSs from competition still seems a bit bad.
Even though I know you're trolling, you do realize that IE, Mozilla, etc transmit a User Agent string which does provide "computer usage data" (look at all those sites which show IE as having X% of the market based on connections to google.com or whereever). Considering that IE is "part of the OS", Microsoft Windows as it currently stands would count. Or is there a section of the Microsoft EULA somewhere that tells you about the User Agent string? Or all the other various places where they get implicit (but not explicit) approval to send microsoft.com or wherever data through one of their programs? And it's not like OSS projects are in the clear, either (I already mentioned Mozilla as an example).
While I don't doubt that, I think it naive to measure all the results off one page which seems focused solely on new releases. It's like only doing surveys in Harlem and talking about how all crimes are committed by blacks. There's *lots* of places online trading all sorts of content. Yes, there are many places trading the most recent stuff in games, movies, etc. And there are places trading things ranging back to the 80s in software (music goes back..well, probably at least the 30s). And while I don't condone the trading of recent stuff, it's harder to be scoldful of people trading in stuff that you can't buy anywhere even if you wanted to (I'm talking about the old stuff, not the prerelease stuff).
So, if it were true that most pirating (which I think is too assumptive that it's most and not many or several) is of recent things, copyright doesn't need to last that long (your suggestion about the timeframe of newness being few months would be somewhat funny to witness as a guide for length of copyright, but not at all helpful...I think it'd have to be minimally a few years). But copyright owners, especially the big conglomerate backed ones, would rather it last forever so they can be assured they'll make the maximal amount of money possible. And I say that for what we get in return, copyright shouldn't last past a much smaller time frame since it is obviously has such a small rate of return for either us, the public, or them, the copyright owner.
This isn't the truth, but an opinionated response with all the bias one can expect from a human. To that end, here is what I wish to say concerning your list of piracy apologists.
(1) I don't personally believe in copying CDs illegally-- but I think we should avoid using unkind words like "piracy" to describe those that do -- instead, we should describe it as an "infringement", much like a parking infringement.
I don't care if they use piracy, content murder, or artist gangbang to describe copyright infringement. I'm sure they don't mind us calling them the son of Satan, either.:) j/k
(2) I don't believe in the record companies emotively abusing the word "theft," but I do believe in emotively abusing words like "information," "sharing," and "Copyright Enforcement Militia."
I'm not sure I understand what you mean by emotively abusing. Are you saying that record companies cry for "stolen works" while hackers cry for "caged information"? Hackers are information hippies. I'm sorry that people here do not always make that clear enough when making their statements against the use of words by the record companies.
(3) I believe that piracy is driven by "overpriced CDs" even though CDs have dropped in price over the years.
Ignoring the issue of whether CDs have dropped in price over the years (which being a non-buyer, leaves me at a disadvantage to answer), I don't think that piracy is driven by overpricing any more than it is driven by any other sole element. I wouldn't be surprised that it is the case, however, for some people.
(4) I believe that piracy is driven by overly long copyright duration, even though most pirated works are recent releases.
Well, technically it is. You can't pirate what isn't copyrighted, so any "overly long" duration causes what would be perfectly legal copying to be piracy. Now, I will admit that at the same time a sizable (but, I wouldn't say "most") percentage of pirated works are recent releases. There's people pirating just about anything under the sun, software wise, since the 80s. The fact is, the actual copyright holder for most such software is unclear, and enforcement is lax as generally the "damage" in lost current or future sales is nonexistant. In such a case, it seems clear that copyright is obviously overly long by at least 75+life years (ie, copyright being over 10/20 seems to not really matter for software).
(5) I believe that illegitimately downloading music is giving the author "free advertising". I don't buy any of the music I download, of course--but lots of other people probably do.
Well, it is free advertising. That doesn't mean the authors any more happy about it than writing graffiti along a lot of buses. I think the point that was trying to be made was that if you d/l and the artist doesn't sue you yet, maybe it's because they don't mind. I think the anonymity of the internet might have something to do with it too.
(6) I believe that ripping off the artists is wrong. The record companies always rip off the artists. Artists support P2P, except the ones that don't (like Metallica), and they don't agree with me, hence they're greedy or their opinion doesn't count or something.
Not to be a complete ass, but a lot of artists are currently in contracts with the record companies and plan future dealings with them. Bad-mouthing them now is unlikely to help them in the future even if it does increase sales a bit (sympathy buyers). At the same time, not all artists are unhappy with their relationship with the record companies, so they're obviously not going to bad-mouth them either. So, I think your comment is obviously leaving out some valid points.
(7) I believe that selling CDs is not a business model, but giving away things for free on the internet is.
I believe that selling CDs is a business model, but having others sell your music for you while you get lots of royalties is
And how long until a user is banned "for pirating movie " because in reality they've made a parody movie insulting the MPAA or are trying to rally against the MPAA? There's a reason why the government (and now days, big business) shouldn't be allowed to just monitor and disable things you pay for at its whim.
ACNS is just a push to work around having to bother providing evidence of wrong doing in a court of law (okay, really, it's just the lawyer fees for intimidating users until they roll over), which costs a lot more time and money. If people are breaking the law, the MPAA should have to find evidence the legal way just like everyone else. I don't see how this is any different than Carnivore. Let the MPAA play by the same rules as every other copyright holder.
> As we learned in Econ 101, it probably comes down to value. Most people do not ascribe value to computer security; they see it as "something the IT guys make us do."
Solution: Make the password embarassing information so they don't dare give it out. There's no one who ever needs to be given the password, anyways, so it works out.
I think the problem is, you don't seem to realize that one of Linux's strengths (and all OSS) is that it can be forked by anyone who so desires to fill some market place. All the people who are talking about Linux taking over the desktop I think assume some distro will come along, build a nice hardcoded setup that most mom + pop users will like, then *get it preinstalled on PCs*. You see, that's the really big step. Most no mom + pop person buys a new machine and buys a new OS separate. In fact, because of how Windows XP's minimal system requirements keep going up every so many years, people are basically forced to buy a new machine when upgrading to a new OS and they upgrade because they think they have to (note: this all is a very broad generalization, but you'll see where I'm going with this in a bit).
So, the desktop market as it stands rests on new machines with new OSs, so the whole question of if hardware X does or doesn't work for a user is really the OEMs (read the distro makers) problem. So, none of what the article talks about really matters. In fact, how the state of events are now with the depth of computer penetration in the developed world (and lesser extend developing world), it's gotten to the point where MS can't make very much more money on new users (corporate users, the big exception, as companies come and go, buy into licensing contracts, etc). So, the push is to get people to buy their new OS.
The problem is, why bother upgrading? Worse yet, do you *really* want people to be upgrading on top of their old version of Windows? Upgrades, really, are the worst thing for programmers. Why? You have to not only sell your product for a lower cost, but the variables go up near infinitely when trying to peaceful update around god knows what that's already installed--it's funny to me, really, that the article kept pointing out that the full install of Linux was the same cost as an upgrade cost of Windows as if an upgrade and a full install were equivalent. So, the push will still be to try to sell new machines where there's less work to further extend Windows to add whatever new features they have plus remain all the backwards compatibility and create an easier installation tool to detect old apps that need compatibility hacks.
But, as computers still become more and more made into commodities, the cost of Windows is a really large burden to the OEM as far as straight costs. I'm sure you've read at least one article pointing that out. And there's still mom + pop who just wants to read email, browse the web, etc. For all of them using Windows who aren't ever going to upgrade by choice, no big deal. Eventually their machine will break down far enough they'll have to buy a new one, and there can be in the market place brand-new Linux based boxes which are locked down tight and don't break very easily at all.
And that's where I see Linux really having the market penetration. Yes, Mac OS X would do the job, but Mac OS X isn't cheap nor is the hardware it runs on. Most of the people using computer these days really only run a very small set of apps that could be dumped on a knoppix cd. Having it all put on the HD instead and having all the modules preset just makes it even easier. I would be surprised if they were hacked to write to disk quickly so users could literally just shut off the computer whenever without fear of dataloss. And java or flash can still be used to show those quirky little toy programs and they can still hypothetically be used to trojan (though there's all sorts of stuff that could be run as root to stop that sort of thing). And the fact is, most people would be pretty okay with that setup.
For the adventurous, you've still got all the other Linux distros and Mac OS X and Windows. And I think Linux really does fill its niche pretty well for the place of applications in the desktop market. It just seems overly pessimistic to think the attitude of non-interested people will matter any more to Linux's adoption than it does now to people using Windows.
But you can't have it both ways; you can't say 'Linux will conquer the desktop world' as many people seem to do, and then simultaneously say '...but we don't give a shit about average mom + pop situations.'
Who is "you"? If you're talking to the parent poster, the parent poster never said that Linux will conquer the desktop world. Yes, there are several on/. who believe that, but how many of them are people actually working in OSS and how many are really just Windows users complaining about MS?
You do realize, btw, that conquering the desktop market is mostly conquering MS, right? They still hold 90%+ of the desktop market. So, for Linux to conquer (or at least, have sizable influence) in the desktop market just means gaining maybe 30% influence.
Most of that 30% will probably be applications (ie, utterly tightened down desktop boxes to do a few tasks) for home or for public locations. The rest, I'd think, would be what the parent poster and I use it for, personal use for the technically minded. Yes, maybe a large part might also be for the preloaded market, where the arguments about "hardware x doesn't work with linux" only matters if x happens to be a firewire or usb device for the non-technical (and with 30% influence, you'd see "For Linux" stickers for the technical for more core components).
So, go Mac OS X and let it grab another 30%. Maybe then we'll see some healthy competition. And maybe some companies will try make the Mac OS X equivalent distro out of Linux. It's just stilly to be bitching about people suggesting Mac OS X. With Windows at 90% penetration, an enemy of an enemy (Mac OS X) is a friend. Until that changes, there's no reason to not suggest Mac OS X from a political position. And from all other positions, it's still mostly a push to use Linux where appropriate cause it's cheaper and Mac OS X because it's easier choosing only Windows when it's necessary (ie, unnecessary software tie-ins). When the market is more balanced, choosing Windows might be more of an option.
Slashdot Mirror
Companies get taxed for the goods they buy, the land they own, some of their equipment (taxes on company vehicles), products they sell in most states (sales tax), and wages they pay to employees. The owner of a company can withdraw money directly out of a company without taxation (though I could be wrong), but generally the owner is paid a wage just like everyone else. Since shareholders are owners, them being paid a wage and taxed isn't any different than any other company. The fact is, everything tends to be taxed multiple times which just advantages vertical monopolies. I don't pity shareholders.
I think your definition of measurement needs some work. The problem is, measurement is a side effect of observation but then that side effect can have other side effects (think of the way an electron can perturb an atom which causes it to knock into another atom which...etc). So "measurements" are occuring all the time from the interaction of fields (say a star or comet millions of light years away moving) or particles (spare neutrinos, electrons, photons, etc). The only really surprising thing is what our "measurement" seem to indicate.
It's a good thing to a point. At some point, each manufacturer needs to clarify the difference between their various products other than price or no one will want to buy their products versus someone who does do semi-sane labelling. But, I do agree that up to now it's been like AMD and Intel have each been selling the analogy of different cubical containers where only one dimension is measured and attached as a label (imagine if Pepsi and Coke were selling 2" or 1" bottles (where the 2" or 1" represents the width) -- though at least then you know what Coke and Pepsi are and you can physically try to compare the bottles to get an idea of their relative volumes..and it'd be a hell of a lot easier to use some standard measure). Now, obviously CPUs aren't pop, but there's nothing stopping AMD and Intel from measuring in something meaningful like the time to encode a DVD or the average frame rate of some CPU bound game. Or they could just tell you the Mhz and the CPU core so we could read a few benchmarks and get an idea of the scaling power of Mhz for a core.
>>If you leave the doors to your house open, and a large neon sign over the threshold saying 'WELCOME', you'll be *damned* lucky if your insurer would pay up.
>This is more like just leaving your doors unlocked. There is no protocol for a system to advertise it's vulnerabilities.
No, it's more like your door's lock having a structural defect. And in some cases, (not sure of this one) a web service will happily tell you what version its running which is a nice advertising that the defective lock is present.
>Without regard to whether your doors were locked it is illegal to steal things from your house.
It's not stealing, though. If you walk into a house and drink some water from the faucet or breathe air, you can get in trouble for trespassing and theft of service. And if you pee on the floor, it's vandalism. And if the worm was actually copying things, there'd be copyright infringement (it's actually a good question of if the sasser worm author can sue everyone *else* for copyright infringement). But there's no stealing involved unless the worm actually picked up the computer and carried it out of the place for him to take.
I'd love to see the same penalties applied to breaking into computer systems as as applied to rape, because then I think the people doing these things would have to consider what they're doing and their imipact on their victims. Right now, I think most people like this guy think it's a game, instead of focusing on the incredible amounts of agony that is caused.
I pray you're being a troll. If not, step back for a minute from your weird obsession personification of a computer, and realize that a computer is a product just like a toaster. It isn't an irreplaceable, unique person composed of trillions of cells where a violent physical act is unlikely to be entirely erased from either physical or mental realms. Computers aren't alive and setting the same punishment for abusing either horribly degrades the perceived worth of a human.
Only at the point at which computers are in a position where they can do the same sort of physical harm to a person should there begin to be punishment in the same realm. But that's just harming a person by proxy which is already covered by law (killing a person by a gun is the same as killing a person by strangling them, legally (and the same issue of a question of neglect by using a proxy comes up)). Then comes the question of is the computer maker responsible, and I'd say that they might be if the physical harm was the result of a defect.
While I agree with you on the point that A -> B doesn't mean B -> A, stating A -> B isn't a bad thing in itself. In fact, you bring up the point that if I were to create a program and release it under the GPL, I could not release the source code which would in effect have no real effect (the only thing under which I could be in violating of is copyright and I own the copyright). As for your point about the programmer using GPLed code and the company not knowing, you're right that the company doesn't have to agree to the GPL. Instead, they'd be liable for statutory copyright infringement ($150,000/copy) while the program would be liable for statutory and punitive ($600,000/copy). Unlike stolen goods, you can't claim ignorance of the fact the goods were stolen as an excuse when it comes to copyright--if one could, shell companies would spring up with some contact giving them the materials to pirate until they were shutdown. In any case, the cheapest solution in most cases is to just GPL the program and most GPL authors will call off the suit. The only other economic solution would be to pay for current damages and rewrite the program from some known safe source.
You're kidding?! I thought January was 0.
You're right that having duplicate functions in different shared libraries means opposing apps real use of ram is larger than if they both used the same shared library or their shared libraries used a common lower library, but I think you're overlooking another point. Whenever there's a page of a library or a process' code that is going to be swapped out, it can simply be dumped. That means unoften used sections of different shared libraries aren't actually in memory. Also, because the libraries are already on the HD there's no swapout that occurs (though obviously swapin can occur). So the wasted bloat of most libraries matters little. And the often used stuff, which it does take up more ram than if it were unified is still a very miniscule section of ram usage (even at 8MB for qt and 2MB for gtk+ (plus another 1MB for glib and gdk) which cover all gtk+ and qt apps, you're using a amazingly small percentage of any modern system's ram (less than 9% of 128MB); add all the shared libs and you're still not using more than around 32MB which while large provides a massive amount of services for all sorts of apps which would otherwise be much larger). So, unification is good, but it's not nearly the worry that you seem to believe it is.
>Copyright holders hold a number of different rights, several of the most important of which are listed at 17 USC 106. One of them is the right to prevent others from distributing copies of the work.
... lawfully made under this title ... is entitled, without the authority of the copyright owner, to sell or otherwise dispose of the possession of that copy or phonorecord.
Yes and no. The problem is, if you give the copyright of a work to someone in the UK, the Berne convention should extend that copyright to the world (and hence back into the US)--just like two people can create the same copyrighted work independently and claim copyright on it. As a result, the transfer of copyright to someone else could effectively make two distributors in a single country. The only way around this is to *not* hand over copyright but instead setup a contract that limits the reach of distribution.
>As it happens, there are exceptions -- limitations -- to that right. But due to the precise language of one of them (17 USC 109) things get a little wonky. What it says is: the owner of a particular copy
You're actually taking the line out of context. Here is the whole line.
Notwithstanding the provisions of section 106(3), the owner of a particular copy or phonorecord lawfully made under this title, or any person authorized by such owner, is entitled, without the authority of the copyright owner, to sell or otherwise dispose of the possession of that copy or phonorecord.
So, if allofmp3 is authorized to make copies (it obviously is under some contract), then the user as well as allofmp3 can resell the music. The only way to stop the user from doing it is a contract. And the only way to prevent allofmp3 from doing it outside of the scope of what is wanted is also a contract. Of course, such contracts could convievably be struck down as illegal (by extension of copyright law written under scope of the constitution). So, I'm still not sure where copyright stops it.
> No, you're wrong. Look, a copyright is not a single, indivisible thing. Think of it like a bundle of sticks. One pair of sticks are the right to reproduce and distribute copies in the US. Another pair are the rights to reproduce and distribute copies in the UK. Perry gave Sid the latter pair, but kept the former for himself.
I'm not sure how that prevents importing. Once it's distributed in the UK, why can't it be imported? I don't see anything in US law that prevents it. More importantly, I wonder how Sid ever got the work in the first place. The distribution to Sid from Perry seems to fall under your case too.
> And the way the law is written, first sale only applies to copies that were made legally by whoever had the US rights.
First Sale Doctrine, as I've read it, never makes mention of US rights. It only makes mention of copyright in general. And I don't see mention of dealing with the scenario you mention, either. Or the situation of importing a copy made in the UK into the US when an author has copyright over a work in both countries.
>>Then how can iTunes do it?
>They're given permission to do so by the US copyright holder. And note that they try very hard to only do business where the local rightsholders have given them permission to do so.
I think I said this already. They have a contract with iTunes. It'd seem that at least one of the reasons a contract is in place is because otherwise exporting to other countries would be legal. But maybe I'm wrong. I'd really like to know what part of US law covers importing original or used copies of a copyrighted work for various situations (like the same right holder in one of the two countries or in both).
>>Umm, how do they sell stuff on the internet without charging money?
>They'll make it up in volume.
I know your comment was meant to be funny, but that's actually true to an extent. Just like how ads pay for free tv broadcasts, I wouldn't doubt the volume of people getting free audio online wouldn't provide a nice revenue stream for doing ads. Of course, there's nothing stopping a stream from being 90% ads and having people pay to get 45% ads instead..or was it that cable got you more channels with ads. Anyways, there's this magically thing called the radio which people have been known for years to tape record off of even though it wasn't always the best quality. Streaming audio would probably fit well into that category. I guess it's funny to me how while MS is trying to shift from per item to per time the RIAA is dead-set on doing everything on per item. I guess that's just their little way of saying MS has too little software and the RIAA has way too much music.
However, if you copy it and get a second record, assuming the copying isn't legal (per 17 USC 107 or 1008 or whatever) then you CAN'T resell the second record under the first sale provision (109).
Problem one, you assume the copying isn't legal. The fact is, there's the equivalent of two copying going on. Labels are selling music to allofmp3 and allofmp3 is selling songs to you. The slight difference is more than likely allofmp3 isn't actually getting source copies from the label each time but has some contract. So, allofmp3 has to follow its contract for which it might need for you to sign a contract to remove itself from liability for your illegal actions. And then the legality is placed squarely on the user.
Likewise, if Perry sold his rights in the UK to his close friend Sid Vicious, and Sid was the one making copies in the UK, you couldn't -- as a matter of first sale -- import those copies into the US. There is a good reason for that.
If Perry sells his rights in the UK to Sid, then you can obviously import any copies Sid makes for sale because they're copyrighted. Or do you think the RIAA can't sell US music in the UK because it got hold of the copyright for songs from the authors?
Imagine that there was a small country that bordered the US and could easily ship stuff here. We'll call it Moosylvania. Further, imagine that Moosylvania has no copyright laws at all. This means it's legal for the locals to copy anything they want. If they could freely export it to the US, they'd just do an end run around our copyright laws, and everyone would buy cheap, unauthorized Moosylvanian copies, basically leaving the US copyright holders screwed.
Except Russia has copyright law. And Russia and the US have almost certaintly set up a treaty to deal with copyright law (the whole point of the Berne convention and its cousins, btw, was to unify copyright law enough so that copyright could extend around the globe..and that means importing copyrighted works from Russia should be perfectly legal in itself).
So, for first sale to apply, the copyright holder who made the copy has to be the US copyright holder. If that's not so, even though the copy was made legally over there, it won't qualify, because it would not have been made legally if it had been made over here.
That really doesn't make sense. Of course, it's a good question on whether it's even possible for someone in the US to hand over their copyright in just one country. And if it *was* possible, then it obviously means you can import the song because it's still under copyright (just not a local one). If copyright can extend from the US to the UK, I'm not sure why it can't extend back again.
When you download from this site, there is a master copy in Russia. At the end of the process, there is a master copy in Russia AND a copy on your hard drive. That's two copies, and that already indicates that it's not an import. And the copyright holder has the exclusive right to reproduce his work in the US per 106.
Then how can iTunes do it? Through a contract, of course. And the contract involves making a copy for sale. That copy for sale is paid for before being resold to the consumer. If that weren't true, iTunes wouldn't be paying another company. Or are you saying the contract isn't a valid way of reproducing a work? If that's true, then I hope you don't believe an even less strong legal item called a license (like GPL or the BSD variety) to redistribute works.
That's not how big O notation works. It's actually defined as:
If O(n log n), then
c*f(n)+k < n log n
defines the worst case scenario for time where c and k are constants and f is the function being used. n is always the number of elements being sorted (for sort algorithms). So, the issue is what k and c are in each of the various algorithms. It might be that c and k are huge for heap/merge sort, but with quicksort as O(n^2), it'd take either a hugely massive c and/or k or really small n for quicksort to always be faster than heap/merge sort.
Of course, as was pointed out, quicksort is more trivially to write, is generally n log n performance, and apparently has a lower c value (IIRC, heapsort has a c of 2). So, people use quicksort even though quicksort isn't guaranteed to always have the best time. Heck, people still use bubblesort (which is fine for really small n's as bubblesort's k is really small). Personally, I'd rather sort algorithms (and gc algorithms) be stuffed into system wide libraries and possibly let an outside function chose which to use; gmp which uses several different methods for doing bignum integer math is a good example of just having a library choose the right algorithm for the job; it'd seem smart to have an equivalent sort algorithm which was based on n and either worst or average case as chosen by the programmer.
Big O Notation
So, the shoe is on the other foot. And copyright can be twisted to advantage the user (GPL). That doesn't mean every license is a good license.
The fact is, the kernel doesn't arbitarily malfunction when it's tainted. Instead, the taintedness is a great sign to tell the user that they really need to go to the original authors for help since no one else is able to properly debug their proper (and of course, two different modules from two different companies which each taint the kernel creates a problem which no single entity can resolve). Faking the string to not cause taintedness helps no one (in the short term it might help the company, but it might not in the long run; people might pay support money to get bugs fixed in one tainted module). Faking a string in a printer cartridge helps the user to get cheaper ink. It also helps create competition (always a good thing).
Ironically, Lexmark's cases against various clone ink cartridge makers might decide the result of this same type of deception. Faking a string to make some program behave the way you want might be unhelpful and possibly unethical (by misleading users into believing they're using only GPLed code or wasting developers time on problems they can't solve thanks to code they can't see), but it's hard to see how it could be made illegal. Now getting such companies for false advertising...
Most (though not all) companies do have in-house or contracted technical support for user error. Paying MS or any other technical company a glob of money for user error is a huge waste of money. You pay MS or other technical companies for the non-user error technical support like bug reporting, bug fixing, etc. The fact is, though, you really shouldn't have to pay to report or help fix a bug that is the fault of another company. It'd be like finding defects in all the cars you sell and paying the manufacturer to get them fixed.
The only advantage paying a company has, innately, over getting equivalent software for free is when a bug is discovered there's possible motivation for the company to fix the bug so the user might be inclined to buy future versions. However, the only way that works is if the *buyer* actually pays attention to bugs, notices they're not being fixed, and then proceeds to buying from someone else. If there are no buyable alternatives, the next best thing would be to take something that's free and turning the money that would have went to buying software into improvements to the free thing.
So, the best step at the moment would seem for most MS Office buyers, if they're actually seeing problems, to go first to Wordperfect or StarOffice. But technical support for user error is still something that'll be handled in-house.
Windows 9X does have code to allow Windows 9X to use DOS drivers (The 16bit drivers, if you will.) However, This is done by creating a virtual 8086 context, and making calls to int 0x21.
Close, but no cigar. DOS drivers access hardware just as directly as Windows 9x drivers. The functions it exposes are then used by Windows 9x to some level. Drivers can be 32-bit (just has to switch between 32-bit/16-bit mode each call) and run in ring0 (just like DOS programs..v86 mode wouldn't let you run Doom). The reason why DOS drivers were discouraged is because you can't move around DOS drivers in memory through paging (at least not safely), DOS drivers didn't help promote the use of Windows NT (ie, DOS drivers meant people were stuck in 9x), and DOS drivers don't offer as effect an interface or complete an interface to all the features available. And while I'd claim Windows 9x can be a real OS, it's a real OS on a real OS. In that context, X is a real OS on a real OS (X has drivers for various hardware while providing an api).
And I think why people are so upset about Windows 9x, not calling it a real OS, is Microsoft decided to not make it possible to buy a real OS (MS-DOS, DR-DOS, etc) with a real OS (the Windows part of Windows 9x). There are still a lot of people who like DOS in their own way, and cutting out all the other DOSs from competition still seems a bit bad.
Even though I know you're trolling, you do realize that IE, Mozilla, etc transmit a User Agent string which does provide "computer usage data" (look at all those sites which show IE as having X% of the market based on connections to google.com or whereever). Considering that IE is "part of the OS", Microsoft Windows as it currently stands would count. Or is there a section of the Microsoft EULA somewhere that tells you about the User Agent string? Or all the other various places where they get implicit (but not explicit) approval to send microsoft.com or wherever data through one of their programs? And it's not like OSS projects are in the clear, either (I already mentioned Mozilla as an example).
While I don't doubt that, I think it naive to measure all the results off one page which seems focused solely on new releases. It's like only doing surveys in Harlem and talking about how all crimes are committed by blacks. There's *lots* of places online trading all sorts of content. Yes, there are many places trading the most recent stuff in games, movies, etc. And there are places trading things ranging back to the 80s in software (music goes back..well, probably at least the 30s). And while I don't condone the trading of recent stuff, it's harder to be scoldful of people trading in stuff that you can't buy anywhere even if you wanted to (I'm talking about the old stuff, not the prerelease stuff).
So, if it were true that most pirating (which I think is too assumptive that it's most and not many or several) is of recent things, copyright doesn't need to last that long (your suggestion about the timeframe of newness being few months would be somewhat funny to witness as a guide for length of copyright, but not at all helpful...I think it'd have to be minimally a few years). But copyright owners, especially the big conglomerate backed ones, would rather it last forever so they can be assured they'll make the maximal amount of money possible. And I say that for what we get in return, copyright shouldn't last past a much smaller time frame since it is obviously has such a small rate of return for either us, the public, or them, the copyright owner.
This isn't the truth, but an opinionated response with all the bias one can expect from a human. To that end, here is what I wish to say concerning your list of piracy apologists.
:) j/k
(1) I don't personally believe in copying CDs illegally-- but I think we should avoid using unkind words like "piracy" to describe those that do -- instead, we should describe it as an "infringement", much like a parking infringement.
I don't care if they use piracy, content murder, or artist gangbang to describe copyright infringement. I'm sure they don't mind us calling them the son of Satan, either.
(2) I don't believe in the record companies emotively abusing the word "theft," but I do believe in emotively abusing words like "information," "sharing," and "Copyright Enforcement Militia."
I'm not sure I understand what you mean by emotively abusing. Are you saying that record companies cry for "stolen works" while hackers cry for "caged information"? Hackers are information hippies. I'm sorry that people here do not always make that clear enough when making their statements against the use of words by the record companies.
(3) I believe that piracy is driven by "overpriced CDs" even though CDs have dropped in price over the years.
Ignoring the issue of whether CDs have dropped in price over the years (which being a non-buyer, leaves me at a disadvantage to answer), I don't think that piracy is driven by overpricing any more than it is driven by any other sole element. I wouldn't be surprised that it is the case, however, for some people.
(4) I believe that piracy is driven by overly long copyright duration, even though most pirated works are recent releases.
Well, technically it is. You can't pirate what isn't copyrighted, so any "overly long" duration causes what would be perfectly legal copying to be piracy. Now, I will admit that at the same time a sizable (but, I wouldn't say "most") percentage of pirated works are recent releases. There's people pirating just about anything under the sun, software wise, since the 80s. The fact is, the actual copyright holder for most such software is unclear, and enforcement is lax as generally the "damage" in lost current or future sales is nonexistant. In such a case, it seems clear that copyright is obviously overly long by at least 75+life years (ie, copyright being over 10/20 seems to not really matter for software).
(5) I believe that illegitimately downloading music is giving the author "free advertising". I don't buy any of the music I download, of course--but lots of other people probably do.
Well, it is free advertising. That doesn't mean the authors any more happy about it than writing graffiti along a lot of buses. I think the point that was trying to be made was that if you d/l and the artist doesn't sue you yet, maybe it's because they don't mind. I think the anonymity of the internet might have something to do with it too.
(6) I believe that ripping off the artists is wrong. The record companies always rip off the artists. Artists support P2P, except the ones that don't (like Metallica), and they don't agree with me, hence they're greedy or their opinion doesn't count or something.
Not to be a complete ass, but a lot of artists are currently in contracts with the record companies and plan future dealings with them. Bad-mouthing them now is unlikely to help them in the future even if it does increase sales a bit (sympathy buyers). At the same time, not all artists are unhappy with their relationship with the record companies, so they're obviously not going to bad-mouth them either. So, I think your comment is obviously leaving out some valid points.
(7) I believe that selling CDs is not a business model, but giving away things for free on the internet is.
I believe that selling CDs is a business model, but having others sell your music for you while you get lots of royalties is
But then Job would be tested by God because of Gill Bates insistence, and all his friends would ask why he has all those boils.
And how long until a user is banned "for pirating movie " because in reality they've made a parody movie insulting the MPAA or are trying to rally against the MPAA? There's a reason why the government (and now days, big business) shouldn't be allowed to just monitor and disable things you pay for at its whim.
ACNS is just a push to work around having to bother providing evidence of wrong doing in a court of law (okay, really, it's just the lawyer fees for intimidating users until they roll over), which costs a lot more time and money. If people are breaking the law, the MPAA should have to find evidence the legal way just like everyone else. I don't see how this is any different than Carnivore. Let the MPAA play by the same rules as every other copyright holder.
> As we learned in Econ 101, it probably comes down to value. Most people do not ascribe value to computer security; they see it as "something the IT guys make us do."
Solution: Make the password embarassing information so they don't dare give it out. There's no one who ever needs to be given the password, anyways, so it works out.
I think the problem is, you don't seem to realize that one of Linux's strengths (and all OSS) is that it can be forked by anyone who so desires to fill some market place. All the people who are talking about Linux taking over the desktop I think assume some distro will come along, build a nice hardcoded setup that most mom + pop users will like, then *get it preinstalled on PCs*. You see, that's the really big step. Most no mom + pop person buys a new machine and buys a new OS separate. In fact, because of how Windows XP's minimal system requirements keep going up every so many years, people are basically forced to buy a new machine when upgrading to a new OS and they upgrade because they think they have to (note: this all is a very broad generalization, but you'll see where I'm going with this in a bit).
So, the desktop market as it stands rests on new machines with new OSs, so the whole question of if hardware X does or doesn't work for a user is really the OEMs (read the distro makers) problem. So, none of what the article talks about really matters. In fact, how the state of events are now with the depth of computer penetration in the developed world (and lesser extend developing world), it's gotten to the point where MS can't make very much more money on new users (corporate users, the big exception, as companies come and go, buy into licensing contracts, etc). So, the push is to get people to buy their new OS.
The problem is, why bother upgrading? Worse yet, do you *really* want people to be upgrading on top of their old version of Windows? Upgrades, really, are the worst thing for programmers. Why? You have to not only sell your product for a lower cost, but the variables go up near infinitely when trying to peaceful update around god knows what that's already installed--it's funny to me, really, that the article kept pointing out that the full install of Linux was the same cost as an upgrade cost of Windows as if an upgrade and a full install were equivalent. So, the push will still be to try to sell new machines where there's less work to further extend Windows to add whatever new features they have plus remain all the backwards compatibility and create an easier installation tool to detect old apps that need compatibility hacks.
But, as computers still become more and more made into commodities, the cost of Windows is a really large burden to the OEM as far as straight costs. I'm sure you've read at least one article pointing that out. And there's still mom + pop who just wants to read email, browse the web, etc. For all of them using Windows who aren't ever going to upgrade by choice, no big deal. Eventually their machine will break down far enough they'll have to buy a new one, and there can be in the market place brand-new Linux based boxes which are locked down tight and don't break very easily at all.
And that's where I see Linux really having the market penetration. Yes, Mac OS X would do the job, but Mac OS X isn't cheap nor is the hardware it runs on. Most of the people using computer these days really only run a very small set of apps that could be dumped on a knoppix cd. Having it all put on the HD instead and having all the modules preset just makes it even easier. I would be surprised if they were hacked to write to disk quickly so users could literally just shut off the computer whenever without fear of dataloss. And java or flash can still be used to show those quirky little toy programs and they can still hypothetically be used to trojan (though there's all sorts of stuff that could be run as root to stop that sort of thing). And the fact is, most people would be pretty okay with that setup.
For the adventurous, you've still got all the other Linux distros and Mac OS X and Windows. And I think Linux really does fill its niche pretty well for the place of applications in the desktop market. It just seems overly pessimistic to think the attitude of non-interested people will matter any more to Linux's adoption than it does now to people using Windows.
yellow journalism
n : sensationalist journalism [syn: {tabloid}]
But you can't have it both ways; you can't say 'Linux will conquer the desktop world' as many people seem to do, and then simultaneously say '...but we don't give a shit about average mom + pop situations.'
/. who believe that, but how many of them are people actually working in OSS and how many are really just Windows users complaining about MS?
Who is "you"? If you're talking to the parent poster, the parent poster never said that Linux will conquer the desktop world. Yes, there are several on
You do realize, btw, that conquering the desktop market is mostly conquering MS, right? They still hold 90%+ of the desktop market. So, for Linux to conquer (or at least, have sizable influence) in the desktop market just means gaining maybe 30% influence.
Most of that 30% will probably be applications (ie, utterly tightened down desktop boxes to do a few tasks) for home or for public locations. The rest, I'd think, would be what the parent poster and I use it for, personal use for the technically minded. Yes, maybe a large part might also be for the preloaded market, where the arguments about "hardware x doesn't work with linux" only matters if x happens to be a firewire or usb device for the non-technical (and with 30% influence, you'd see "For Linux" stickers for the technical for more core components).
So, go Mac OS X and let it grab another 30%. Maybe then we'll see some healthy competition. And maybe some companies will try make the Mac OS X equivalent distro out of Linux. It's just stilly to be bitching about people suggesting Mac OS X. With Windows at 90% penetration, an enemy of an enemy (Mac OS X) is a friend. Until that changes, there's no reason to not suggest Mac OS X from a political position. And from all other positions, it's still mostly a push to use Linux where appropriate cause it's cheaper and Mac OS X because it's easier choosing only Windows when it's necessary (ie, unnecessary software tie-ins). When the market is more balanced, choosing Windows might be more of an option.