Re:So, you programmers ready to give up your jobs?
on
McVoy Strikes Back
·
· Score: 1
I have a feeling that if you were a Ford shareholder and Ford announced that they'd start making most of their money by standing in a small river catching fish with bare hands, you'd want to sell very very quickly before anyone else heard about it. People who buy Ford's shares do so because they are a car company, because they think that a car company is a good way to make money. A smart investor is very much interested in what companies are doing, and that is distinguished by what they actually do. "Making money" is the goal, not the product (unless you're a mint). As the GP said, money is a mechanism by which we keep score and allocate resources and motivate people. By itself, it is worthless. You can't do anything with money if there's nothing to buy.
Re:So, you programmers ready to give up your code?
on
McVoy Strikes Back
·
· Score: 1
Well, release the circuit layout for your custom embedded board, and if it is useful to someone, then maybe they'll use it with your little VB app. Or maybe it will give them an idea for a way to do something else. You never know. Most likely you're right, no one will ever look at it again. So what?
The problem with what you say is that there was apparently no evidence of any encryption of the files in question, or any other relevant files. My tire iron analogy is that just about everyone has a tire iron in their trunk - using the presence of that tire iron to say anything other than that I'd like to be able to change the tire on my car if it goes flat is dumb. Having a ski mask in your car in the middle of winter should also be non-relevant (ooh, ooh, he was wearing ski gloves, he must have wanted to not leave fingerprints!).
It sounds to me as if the standard for relevance are so loose that the following could be admitted: "The defendant had a copy of gcc on his computer. gcc is used to "compile" source code of computer programs so that a computer can understand it. There is a lot of "open source" source code on the Internet, including programs for encrypting and hiding information. Since we didn't find any actual source code on his computer, he must have deleted it. That could indicate that the source code he deleted was a program used to encrypt or hide information. We didn't find any hidden information, so we believe it must be hiding. If we did find it, it might be encrypted (the fact the he hid it would indicate that he'd probably encrypt it as well). Thus, the presence of gcc on a machine without any source code is indicative of guilt."
Now, in this actual case, based on the actual statements, it does sound like it came out that PGP is perfectly legal, has many legitimate uses, and that there were no encrypted files, so the prejudicial nature of the statements was probably minimal. It still sounds irrelevant to me, given the facts that I've read about so far.
Yeah, but this is more akin to saying that the burglar had a tire iron in the trunk of his car... or possibly, that having a full tank of gas meant he was planning to flee the state, thus showing his guilt (innocent people don't plan to flee the state).
He didn't say NSA could break it, he claimed that no one "other than the NSA" could break it. Whether the NSA could break it or not was left unspecified. If I say that NSA can break DES, how am I guilty of divulging classified information? I deduce the information on the basis that NSA isn't incompetent and that DES has been broken by others for quite some time now.
If there was evidence of files being encrypted with PGP as part of the crime, the the existence of PGP on his computer would be relevant. Even if the files were encrypted with something else, you might use the existence of PGP to show he is familiar with encryption technologies. But absent any indication of encryption as part of the crime (or covering it up), I don't see how PGP is any more relevant than a paper shredder or matches and a fireplace. "We didn't find any actual pictures, nor any burnt pictures, but since the defendant has access to a fireplace and has matches, we think that shows his intent to burn the pictures."
I suppose I'm an "Apple fanboy", but my tests show that a dual-processor Opteron system I tested was significantly faster than my dual-processor G5 (both were 1.8GHz with similar amounts of RAM, Opteron was running Linux, G5 running Mac OSX). I suspect that some, but not all of the difference is due to better code generation for the Opteron. The Opteron running 32-bit code was faster than the fastest Intel chips at the time (at least, on my benchmark, which was all I was interested in), the G5 was about as fast as the fastest ones when running 32-bit code as well. My benchmark wasn't very OS dependent.
I don't know enough about the x86-64 architecture to know how happy I am with it, but I don't have any doubts that the PPC architecture is better than IA32, although it isn't quite as clean as it could have been. At least 64-bit processing was designed in from the start (32-bit machines simply didn't implement the 64-bit instructions). Alpha is better than any of them (except for the way PALcode is handled - makes the bootstrap process much more difficult), but that's a lost cause now.
What most developers thought was that it was "too difficult" to program for the Mac. Why, you had to have THREE WHOLE MANUALS, and they were each almost an inch thick (I still have the "phonebook" version, about two inches thick).
The amazing thing about MacWrite and MacPaint was that it fit on a bootable 400K diskette that ran on a machine with 128K of RAM and 64K of ROM, and that included 22K of video frame buffer, all on an 8MHz 68000. And people complain about their poor little old 1.5GHz 256MB systems that only have 20GB of disk space.
I started out with a Lisa - compared to what I could get for the same amount of money in a PC, there was no comparison. True, I gave up color for a few years until the Mac II came out, but writing graphics programs was so much better - on a PC under DOS, each program had to have drivers for each video mode it wanted to support (EGA, VGA, XGA, etc). Each program had to have serial chip drivers and support for upgraded serial chips if you wanted to get any kind of performance (i.e. not dropping characters!). Each program had to include support for as many printers as they could, and if yours wasn't included, you were pretty much out of luck. You had to worry about Extended vs Expanded memory, you had to make sure you didn't step on device or frame buffer memory, you had interrupt conflicts and on and on.
When the Mac II came out, all of a sudden you had color, multiple video card support, on-the-fly color depth and resolution switching, 16MHz 68020 (which was capable of doing virtual memory, if you upgraded the "do-nothing faked" MMU chip). People have more cache memory than that thing had RAM (1MB), and video cards with more memory than that thing had disk space (40MB).
Well, I switched from a Mac II to a 601, and the 68K emulation was definitely faster!
The only programs that had problems with the emulation were ones which used a couple very-rarely-used instructions that Apple left out of the emulation, and programs that didn't play nice with going from MacOS 6 to MacOS 7. In other words, the change to a new OS was more of an issue than the fact that the 68K code was being emulated.
Movie Tickets (for two): 15.00
Fresh Popcorn / OK drink (for two, free refills): $8.50
Getting to watch people in costume whack each other with light sabers, not to mention the Storm Trooper hired for "crowd control": Free
Not standing in line because we had reserved seating: Priceless!
Not that standing in line can't be fun. I've done it for every other Star Wars film (except the first), for Batman, and a few others. But I'll take the reserved seating any time.
I've now seen every Star Wars at the first (local, open to the public) showing. The first one the theater was almost empty, walked in slightly late to see the big cruiser rumbling overhead just before it captures Leia's ship, then of course I had to sit through it a second time since I missed the first few minutes...I'm pretty sure that was the last movie I saw where they let you sit through a show as many times as you wanted - it was common practice to arrive at the theater pretty much any time, watch the last part of the movie, then watch the first part and leave when you got to the part you walked in.
Since they give the directions to the puzzle in advance, most or all of them could also be coded in advance. Also, most of them can be solved fairly easily using brute-force combinatorics. If using a computer was allowed, and a day in advance to read the directions for each puzzle and write some code, I could probably solve every one of those problems very quickly (mostly the time to enter in the specifics of the problem).
And then you watch either a shaky-cam version of it or a pre-release version with timecodes on it. Wow, yeah, that's gonna make me want to spend all that money on a great system, and not go see it in the theater!
They're getting all sorts of ridiculous mechanisms in place to try to ensure that the sacred "digital stream" can't be intercepted, and justify this to Congress and the FCC because "digital is so much better, we need stronger laws to protect it", then also get upset over crappy low quality versions. Which is it, guys? How did this being released on the Internet in ANY WAY "dim the magic" for me? Do they seriously think that someone who would have gone to see this in the theater is going to think "wow, that was such a great quality experience, now I don't have to go to the theater to see it"?
Crack down on the real "piracy", with people selling counterfeit DVDs on the street - those are at least people who probably would have bought the real thing and just want to save a few bucks.
BSA is absolutely right, piracy will continue to grow. More and more companies will take GPL software, base products on it, and refuse to release the source. This MUST be stopped! Thanks, BSA!
Seriously, it does make you wonder how much of the supposedly "pirated" software is actually FLOSS , e.g. that any machine sold without a proprietary operating system is automatically reported as being a pirate copy of Windows.
I'll bet they make a lot more money extorting money from businesses and governments for "piracy" based on poor record keeping than they'd get if they could actually make their software copy-proof, especially when a case could be made that making it copy-proof would actually lower their legitimate sales.
Nice history of gasoline, but I'm not sure how it shows I'm wrong. To get more power out of an engine, you use higher compression, which requires higher "octane rating". If it was cheaper to increase that without adding lead, then why would they take a more expensive route? The consumer doesn't see the individual prices, doesn't care if those 3 cents goes to the gas station, the refinery, or the maker of the additive. All they care is that if they use a lower octane gas, their engine starts misbehaving (and I grant you that many just think it's "better gas", so buy it even when they don't need it - but that doesn't change that they'll still try to find the cheapest way to put out gas that they can label "91" or whatever). It wasn't adding lead that lead to higher priced gasoline, it was the requirement for higher octane gasoline that lead to higher prices - and they would have been even higher if they had done it without lead.
Once they started producing leaded gasoline, it's true that the engines started to be designed around that. Aviation engines still use leaded fuels ("100LL" is 100 octane "low lead" and has more lead in it than leaded auto fuel ever did - some engines have been certified (using a process called "Supplemental Type Certificate", or STC) to use auto fuel, and even then it is usually recommended that you run a tank of 100LL through it every 5 tanks or so). But it if were cheaper to produce fuels without lead, there'd have been a demand to use them, including building engines that didn't need the lead in the first place.
It's odd, isn't it, that many people don't seem to realize that different people have different tastes? Some of it is physiological (different specific receptors for bitterness and smells), some of it is learned. Some of that learning is very deep down (because you were brought up that way), some of it can be picked up later in life. The learning bit also applies to the books you like to read, activities you participate in, music you listen to, movies you watch, religions you believe in.
Adding lead was a cheaper way to bring up the octane rating of the gas. They weren't "saving money" by not adding the lead, then charging more. It actually did cost them more to produce the gasoline without the lead.
The problem with lead, and the reason it is a valid governmental purpose to regulate it, is that it is pervasive - you buy cheap gas and I suffer the consequences - I pay for expensive gas and it doesn't benefit me unless almost everyone else does the same thing.
You don't need to find a file that matches the hash. You just send out that you have a file with that hash, but send a different file. At the end of the transfer, the person finds out that they just downloaded junk.
This can be trivially defeated by having the file hash be a hash of a "hash directory", which gives the hash of individual small parts of the file, allowing you to verify the file as it is being loaded and reject sections that aren't correct. This is how bittorrent works.
Creating new names isn't creating new "English words". Google is a "new English word", because people are actually using it as a verb. Creating a new product called "Farstinium" doesn't create a new word, just a new name. Pentium, AMD, Athlon, Opteron aren't "new words". They're just names, until they end up having a meaning beyond what they name. I suppose you could conceive of a sentence such as "They opteroned those pentiums", meaning to come out with something cheaper, better AND faster over an acknowledged leader, even when the subject matter isn't microprocessors, or specifically those microprocessors, but as far as I know, they're still "just names" at this point.
I don't know what TigerDirect used to come up with, but as many people have pointed out, they still come up higher than Apple's Tiger OS. I'm not sure by what theory anyone could claim that Tiger Direct lost first place to a web site about real tigers because of Apple's branding a version of their OS as Tiger, yet still have Apple come up two places BELOW Tiger Direct. Even if the theory were correct, the facts don't support the case.
As far as what they "should have done": they should have contacted Apple to ask them not to use an infringing mark as soon as they found out about it. That's assuming they think it really was infringing. If that didn't work and they really thought it was infringing, then yes, they should have sued as soon as they found out about it.
If they were merely worried about possibly losing trademark protection, they don't have to sue. They could, for instance, make an agreement with Apple that each one agrees that the other's trademark doesn't infringe on their own. That makes it perfectly clear that they aren't ceding any rights to it.
Re:Its only the bad things we head about?
on
Safari vs. KHTML
·
· Score: 1
Even more to the point, the avowed purpose of the GPL is to allow me, someone with a copy of Safari, the right to modify it by having access to the source code (or, in this case, to modify the LGPL portion of Safari and be able to have the rest of Safari use my changes), and the right to share my changes with anyone I want. The GPL "public commons" aspect of Free Software is a nice side effect, but not the primary reason for it.
From the preamble to the GPL itself:
... the GNU General Public
Licenses are intended to guarantee your freedom to share and change
free software--to make sure the software is free for all its users.
That is the wrong section, since Apple doesn't actually distribute the source with a download of OSX or Safari. In that case, 6c is the applicable term:
Accompany the work with a written offer, valid for at least three years, to give the same user the materials specified in Subsection 6a, above, for a charge no more than the cost of performing this distribution.
(where 6a specifies the source code to the library). Note that this still supports your conclusion: the LGPL specifies "the same user" instead of the GPL "any third party" wording.
I don't see that at all. By that reasoning, Apple should lock down Mac hardware so you can't boot Linux or BSD, because if you choose to do so it isn't as easy to use as OSX.
Technically, it would probably be very difficult to achieve without having iTunes be able to do the actual purchase from the other stores. Apple could license an interface for the stores to be compatible with iTunes directly, with iTunes still managing the iPod and authorization and all that. iTMS would still be the default store enabled by iTunes, a person would have to manually add another store, which should make it clear that it isn't Apple's fault if the other store sucks.
Slashdot Mirror
I have a feeling that if you were a Ford shareholder and Ford announced that they'd start making most of their money by standing in a small river catching fish with bare hands, you'd want to sell very very quickly before anyone else heard about it. People who buy Ford's shares do so because they are a car company, because they think that a car company is a good way to make money. A smart investor is very much interested in what companies are doing, and that is distinguished by what they actually do. "Making money" is the goal, not the product (unless you're a mint). As the GP said, money is a mechanism by which we keep score and allocate resources and motivate people. By itself, it is worthless. You can't do anything with money if there's nothing to buy.
Well, release the circuit layout for your custom embedded board, and if it is useful to someone, then maybe they'll use it with your little VB app. Or maybe it will give them an idea for a way to do something else. You never know. Most likely you're right, no one will ever look at it again. So what?
The problem with what you say is that there was apparently no evidence of any encryption of the files in question, or any other relevant files. My tire iron analogy is that just about everyone has a tire iron in their trunk - using the presence of that tire iron to say anything other than that I'd like to be able to change the tire on my car if it goes flat is dumb. Having a ski mask in your car in the middle of winter should also be non-relevant (ooh, ooh, he was wearing ski gloves, he must have wanted to not leave fingerprints!).
It sounds to me as if the standard for relevance are so loose that the following could be admitted: "The defendant had a copy of gcc on his computer. gcc is used to "compile" source code of computer programs so that a computer can understand it. There is a lot of "open source" source code on the Internet, including programs for encrypting and hiding information. Since we didn't find any actual source code on his computer, he must have deleted it. That could indicate that the source code he deleted was a program used to encrypt or hide information. We didn't find any hidden information, so we believe it must be hiding. If we did find it, it might be encrypted (the fact the he hid it would indicate that he'd probably encrypt it as well). Thus, the presence of gcc on a machine without any source code is indicative of guilt."
Now, in this actual case, based on the actual statements, it does sound like it came out that PGP is perfectly legal, has many legitimate uses, and that there were no encrypted files, so the prejudicial nature of the statements was probably minimal. It still sounds irrelevant to me, given the facts that I've read about so far.
Yeah, but this is more akin to saying that the burglar had a tire iron in the trunk of his car... or possibly, that having a full tank of gas meant he was planning to flee the state, thus showing his guilt (innocent people don't plan to flee the state).
He didn't say NSA could break it, he claimed that no one "other than the NSA" could break it. Whether the NSA could break it or not was left unspecified. If I say that NSA can break DES, how am I guilty of divulging classified information? I deduce the information on the basis that NSA isn't incompetent and that DES has been broken by others for quite some time now.
If there was evidence of files being encrypted with PGP as part of the crime, the the existence of PGP on his computer would be relevant. Even if the files were encrypted with something else, you might use the existence of PGP to show he is familiar with encryption technologies. But absent any indication of encryption as part of the crime (or covering it up), I don't see how PGP is any more relevant than a paper shredder or matches and a fireplace. "We didn't find any actual pictures, nor any burnt pictures, but since the defendant has access to a fireplace and has matches, we think that shows his intent to burn the pictures."
I suppose I'm an "Apple fanboy", but my tests show that a dual-processor Opteron system I tested was significantly faster than my dual-processor G5 (both were 1.8GHz with similar amounts of RAM, Opteron was running Linux, G5 running Mac OSX). I suspect that some, but not all of the difference is due to better code generation for the Opteron. The Opteron running 32-bit code was faster than the fastest Intel chips at the time (at least, on my benchmark, which was all I was interested in), the G5 was about as fast as the fastest ones when running 32-bit code as well. My benchmark wasn't very OS dependent.
I don't know enough about the x86-64 architecture to know how happy I am with it, but I don't have any doubts that the PPC architecture is better than IA32, although it isn't quite as clean as it could have been. At least 64-bit processing was designed in from the start (32-bit machines simply didn't implement the 64-bit instructions). Alpha is better than any of them (except for the way PALcode is handled - makes the bootstrap process much more difficult), but that's a lost cause now.
What most developers thought was that it was "too difficult" to program for the Mac. Why, you had to have THREE WHOLE MANUALS, and they were each almost an inch thick (I still have the "phonebook" version, about two inches thick).
The amazing thing about MacWrite and MacPaint was that it fit on a bootable 400K diskette that ran on a machine with 128K of RAM and 64K of ROM, and that included 22K of video frame buffer, all on an 8MHz 68000. And people complain about their poor little old 1.5GHz 256MB systems that only have 20GB of disk space.
I started out with a Lisa - compared to what I could get for the same amount of money in a PC, there was no comparison. True, I gave up color for a few years until the Mac II came out, but writing graphics programs was so much better - on a PC under DOS, each program had to have drivers for each video mode it wanted to support (EGA, VGA, XGA, etc). Each program had to have serial chip drivers and support for upgraded serial chips if you wanted to get any kind of performance (i.e. not dropping characters!). Each program had to include support for as many printers as they could, and if yours wasn't included, you were pretty much out of luck. You had to worry about Extended vs Expanded memory, you had to make sure you didn't step on device or frame buffer memory, you had interrupt conflicts and on and on.
When the Mac II came out, all of a sudden you had color, multiple video card support, on-the-fly color depth and resolution switching, 16MHz 68020 (which was capable of doing virtual memory, if you upgraded the "do-nothing faked" MMU chip). People have more cache memory than that thing had RAM (1MB), and video cards with more memory than that thing had disk space (40MB).
Well, I switched from a Mac II to a 601, and the 68K emulation was definitely faster!
The only programs that had problems with the emulation were ones which used a couple very-rarely-used instructions that Apple left out of the emulation, and programs that didn't play nice with going from MacOS 6 to MacOS 7. In other words, the change to a new OS was more of an issue than the fact that the 68K code was being emulated.
Movie Tickets (for two): 15.00
Fresh Popcorn / OK drink (for two, free refills): $8.50
Getting to watch people in costume whack each other with light sabers, not to mention the Storm Trooper hired for "crowd control": Free
Not standing in line because we had reserved seating: Priceless!
Not that standing in line can't be fun. I've done it for every other Star Wars film (except the first), for Batman, and a few others. But I'll take the reserved seating any time.
I've now seen every Star Wars at the first (local, open to the public) showing. The first one the theater was almost empty, walked in slightly late to see the big cruiser rumbling overhead just before it captures Leia's ship, then of course I had to sit through it a second time since I missed the first few minutes...I'm pretty sure that was the last movie I saw where they let you sit through a show as many times as you wanted - it was common practice to arrive at the theater pretty much any time, watch the last part of the movie, then watch the first part and leave when you got to the part you walked in.
Since they give the directions to the puzzle in advance, most or all of them could also be coded in advance. Also, most of them can be solved fairly easily using brute-force combinatorics. If using a computer was allowed, and a day in advance to read the directions for each puzzle and write some code, I could probably solve every one of those problems very quickly (mostly the time to enter in the specifics of the problem).
And then you watch either a shaky-cam version of it or a pre-release version with timecodes on it. Wow, yeah, that's gonna make me want to spend all that money on a great system, and not go see it in the theater!
They're getting all sorts of ridiculous mechanisms in place to try to ensure that the sacred "digital stream" can't be intercepted, and justify this to Congress and the FCC because "digital is so much better, we need stronger laws to protect it", then also get upset over crappy low quality versions. Which is it, guys? How did this being released on the Internet in ANY WAY "dim the magic" for me? Do they seriously think that someone who would have gone to see this in the theater is going to think "wow, that was such a great quality experience, now I don't have to go to the theater to see it"?
Crack down on the real "piracy", with people selling counterfeit DVDs on the street - those are at least people who probably would have bought the real thing and just want to save a few bucks.
BSA is absolutely right, piracy will continue to grow. More and more companies will take GPL software, base products on it, and refuse to release the source. This MUST be stopped! Thanks, BSA!
Seriously, it does make you wonder how much of the supposedly "pirated" software is actually FLOSS , e.g. that any machine sold without a proprietary operating system is automatically reported as being a pirate copy of Windows.
I'll bet they make a lot more money extorting money from businesses and governments for "piracy" based on poor record keeping than they'd get if they could actually make their software copy-proof, especially when a case could be made that making it copy-proof would actually lower their legitimate sales.
Nice history of gasoline, but I'm not sure how it shows I'm wrong. To get more power out of an engine, you use higher compression, which requires higher "octane rating". If it was cheaper to increase that without adding lead, then why would they take a more expensive route? The consumer doesn't see the individual prices, doesn't care if those 3 cents goes to the gas station, the refinery, or the maker of the additive. All they care is that if they use a lower octane gas, their engine starts misbehaving (and I grant you that many just think it's "better gas", so buy it even when they don't need it - but that doesn't change that they'll still try to find the cheapest way to put out gas that they can label "91" or whatever). It wasn't adding lead that lead to higher priced gasoline, it was the requirement for higher octane gasoline that lead to higher prices - and they would have been even higher if they had done it without lead.
Once they started producing leaded gasoline, it's true that the engines started to be designed around that. Aviation engines still use leaded fuels ("100LL" is 100 octane "low lead" and has more lead in it than leaded auto fuel ever did - some engines have been certified (using a process called "Supplemental Type Certificate", or STC) to use auto fuel, and even then it is usually recommended that you run a tank of 100LL through it every 5 tanks or so). But it if were cheaper to produce fuels without lead, there'd have been a demand to use them, including building engines that didn't need the lead in the first place.
It's odd, isn't it, that many people don't seem to realize that different people have different tastes? Some of it is physiological (different specific receptors for bitterness and smells), some of it is learned. Some of that learning is very deep down (because you were brought up that way), some of it can be picked up later in life. The learning bit also applies to the books you like to read, activities you participate in, music you listen to, movies you watch, religions you believe in.
Two days later, when UPS shows up:
UPS: "I need an adult signature, are your parents at home?"
Mom: "Billy, what's this?"
Boby: "Oh, crap! It was Billy, it was all his idea!"
Adding lead was a cheaper way to bring up the octane rating of the gas. They weren't "saving money" by not adding the lead, then charging more. It actually did cost them more to produce the gasoline without the lead.
The problem with lead, and the reason it is a valid governmental purpose to regulate it, is that it is pervasive - you buy cheap gas and I suffer the consequences - I pay for expensive gas and it doesn't benefit me unless almost everyone else does the same thing.
Actually, look up "Tragedy of the Commons", it is more applicable.
You don't need to find a file that matches the hash. You just send out that you have a file with that hash, but send a different file. At the end of the transfer, the person finds out that they just downloaded junk.
This can be trivially defeated by having the file hash be a hash of a "hash directory", which gives the hash of individual small parts of the file, allowing you to verify the file as it is being loaded and reject sections that aren't correct. This is how bittorrent works.
Actually, Tiger Direct does sell Tiger branded computers.
Creating new names isn't creating new "English words". Google is a "new English word", because people are actually using it as a verb. Creating a new product called "Farstinium" doesn't create a new word, just a new name. Pentium, AMD, Athlon, Opteron aren't "new words". They're just names, until they end up having a meaning beyond what they name. I suppose you could conceive of a sentence such as "They opteroned those pentiums", meaning to come out with something cheaper, better AND faster over an acknowledged leader, even when the subject matter isn't microprocessors, or specifically those microprocessors, but as far as I know, they're still "just names" at this point.
I don't know what TigerDirect used to come up with, but as many people have pointed out, they still come up higher than Apple's Tiger OS. I'm not sure by what theory anyone could claim that Tiger Direct lost first place to a web site about real tigers because of Apple's branding a version of their OS as Tiger, yet still have Apple come up two places BELOW Tiger Direct. Even if the theory were correct, the facts don't support the case.
As far as what they "should have done": they should have contacted Apple to ask them not to use an infringing mark as soon as they found out about it. That's assuming they think it really was infringing. If that didn't work and they really thought it was infringing, then yes, they should have sued as soon as they found out about it.
If they were merely worried about possibly losing trademark protection, they don't have to sue. They could, for instance, make an agreement with Apple that each one agrees that the other's trademark doesn't infringe on their own. That makes it perfectly clear that they aren't ceding any rights to it.
Even more to the point, the avowed purpose of the GPL is to allow me, someone with a copy of Safari, the right to modify it by having access to the source code (or, in this case, to modify the LGPL portion of Safari and be able to have the rest of Safari use my changes), and the right to share my changes with anyone I want. The GPL "public commons" aspect of Free Software is a nice side effect, but not the primary reason for it.
From the preamble to the GPL itself:
That is the wrong section, since Apple doesn't actually distribute the source with a download of OSX or Safari. In that case, 6c is the applicable term:
(where 6a specifies the source code to the library). Note that this still supports your conclusion: the LGPL specifies "the same user" instead of the GPL "any third party" wording.I don't see that at all. By that reasoning, Apple should lock down Mac hardware so you can't boot Linux or BSD, because if you choose to do so it isn't as easy to use as OSX.
Technically, it would probably be very difficult to achieve without having iTunes be able to do the actual purchase from the other stores. Apple could license an interface for the stores to be compatible with iTunes directly, with iTunes still managing the iPod and authorization and all that. iTMS would still be the default store enabled by iTunes, a person would have to manually add another store, which should make it clear that it isn't Apple's fault if the other store sucks.