"Technical support. He is automatically the ultimate guru on the subject of his software."
Do you expect "technical support" on an embedded BSD or linux system from someone who finds #including <dlfcn.h> and making a call to dlopen() to be "too complicated"? Surely, you jest. Let me rephrase that - Shirley, you jest. You'll be better off asking Apu at the Qwiki-Mart. At least you'll get an authentic-sounding "tech support" accent.
"I read 'too much to bear' as taking too much system resources rather then 'too hard'.
In an embedded system dynamic linking can eat up scarce resources. Static linking is faster to load, faster to run, and takes up less ram."
It can also be used to free up resources. You only load what you need, as you need, then release it. It can also make startup times quicker, again, by only loading what you need at start. Use "RTLD_LAZY" - Runtime-load-lazy, so that you don't try to load everything at startup.
The real reason - the poster thinks its "too hard" to #include <dlfcn.h>, and insert one lousy call to dlopen().
You miss the point about the GPL - anyone is free to use the code, provided that, if they distribute, they have to make the source available. You don't get to pick and choose who gets to use it - you can't say "I don't want [insert evil company] to use my code" - even Microsoft can use GPL'd code (and they do - python is used in WGA) provided they distribute the source either with the binory or upon request.
What if Tivo comes up with a better way to, for example, get rid of the "increstations" - the logos that most TV networks put on their shows... you'd get the source, and you're free to modify it for work with your favourite frame-grabber.
Your own computer contains a lot of closed-source hardware - the cpu, for starters. Video cards are also closed-source hardware. Even if nvidia releases their driver source, its useless to you for programming your SiS video chip.
Also, there are specific situations where there is no way in hell you're going to see GPLv3 software used - medical devices. GPL v2 software would be fine, because you CAN "lock it down a la Tivo". Its turning out that companies that have to meet the Sarbanes-Oxley requirements have to be able to prevent unauthorized mods.
Do you really want to cede the entire medical, defense, and eventually even the networking and telco infrastructure markets? Or would you rather see the "hardening" put in those trickle bach?
Actually, even more recent online manuals of style allude to the old rule of "capitalize after a colon." They state its not followed so much any more, but that's because of a general laxness in a lot of areas nowadays.
Look at how many people can't spell basic words: "grammer", "gramer", "wierd", "thier", "formating", "referer" - yes, that last one is a stinker we're stuck with in the standards, but we really should break down and fix it.
"Tell me, why on earth should I write and distribute my code with a license that allows Tivo to use it without giving me back anything?"
People write code under the BSD license all the time.
Tivo WILL give you back any improvements they make - they have no choice. You can get all the Tivo code from
them.
Here's the Tivo source. You can download gzipped tarballs of all the source directly from the site, or you can even get a CD for $15.
In compliance with the GPL version 2, we are pleased to provide our modifications to the Linux Kernel, as well as a few new commands, and some tools to get you into the code.
Additionally, if you would like a CD-R of our modifications you may send a written request to:
TiVo Inc.
2160 Gold St.
Alviso CA 95002-2160
Attention: Customer Care - GNU/Linux Source Code Request.
You will be charged a $15 fee for reproduction, shipping and handling costs, as allowed by the GPL version 2. Make sure that you include a bank certified check for $15.00. Otherwise, you can download the code for free from below:
So you see, if you decide to sell competing hardware, Tivo uses your GPL v2 code, and makes changes to it, you WILL get a copy of all the changes, to use in your product, just as the GPL provides.
The GPL is a software license, not a hardware license. GPL v3 is trying to blur the lines, and it may backfire.
There are a LOT of developers at Novell - Novell is one of the biggest contributors to kernel code, for example. Imagine being able to "taint" the kernel via this deal... its would be worth at least $25 billion. Remember, linux is the #1 threat to them.
Maybe we just need a "plain-talk" license, like borland used to have.
"
You may use this code any way you want to, provided you're not Microsoft, SCO, or anyone else on our current sh*t list.
"
At least it would be more honest...
Then again, if Microsoft really wanted to change its spots, shake up its corporate culture, and send the right signals for a change, it would start making serious contributions under a true F/LOSS license (not "shared source").
Also, it would mean the removal of all mono-based tools, and a lot of OpenSuse/Suse uses Mono... which was what came to my mind when people started with the "patents fud" stuff.
Fortunately, we have alternatives (except for the GIMP, which really relies heavily on gtk2).
If you were Microsoft, wouldn't it be worth a billion to "contaminate" the top developers at most of your competition, and any projects they've worked on? The FUD alone would be worth billions; the success of any copyright-based attack would be devastating. Can you imagine how hard it would be to have to revert to a code base from 5 years in the past?
Projects are going to have to start vetting their programmers for:
have you ever used "Microsoft Shared Source" or worked on a project that did?
have you ever worked on any project that was developed to help interoperability with Windows under a Microsoft/[insert linux company] program?
have you ever contributed code to any related project?
have you ever worked on an unrelated project with someone who is affected by #1 or #2?
At least when you work on proprietary software, you can "compartmentalize" your knowledge. Write a graphics package for work, work on a completely unrelated f/loss server package on your own time. These "collaborations" are dangerous because of their long-term effects.
Linus has it right - don't look at what the competition is doing, because then they can accuse you of having unclean hands.
I wasn't saying the kernel would fork, but that all the GNU tools would be forked, which is a lot easier to do. Thats why I pointed out that IBM, Intel and Sun have compilers if we ever have to dump gcc.
Forking cat and head (and dumping those stupid.info files - man pages are good enough)... why not ?
The GPL v2 as held to be valid in European courts (against SCO, of all people:-) Linuxtag in Germany got a restraining order against them. Courts look at the intent if the wording is ambiguous; in this case, the wording was sufficient.
"In addition, one particular piece of icing, is that it is actually compatible with more open source licenses, in particular the Apache License."... but its not back-compatible with gpl v2 - and that's a nasty "bug", because GPL v2 is a known. proven quantity - people are comfortable with it. The process around GPL v3 sucked, the constant whining about Microsoft/Novell made the people involved look immature, and it doesn't do what its proponents claim it does - it provides zero protection against patent trolls, for example, who are far more likely to make an attack than Microsoft.
Either it is, or it isn't. Its only in the SCOniverse that the GPL isn't a legally binding document.
Of course, I'm willing to listen to counter-arguments. What is this "icing" you speak of? All I see is some extra verbiage that attempts to deal with the latest round of patent fud (a fud that has already died aborning, claiming Xandros as its latest victim - no big loss there).
There's no "hole." Its a software license, and it works fine in its problem domain - software.
Trying to extend it to hardware (tivo-ization) is what is causing the problems.
Anyone is free to use Tivo's software, strip out the DRM aspects, and market their own brand of hardware, thanks to the GPLv2.
As for the patent "issues" - they've already collapsed. Look at the latest FUD victim - Xandros. Its dropped 10 places on distrowatch in 2 weeks (2 weeks ago it was at #18, now its at #28) . There will never be a Microsoft-Linux patent war. Microsoft has too much to lose by actually naming any supposed patent, and facing patent retaliation from IBM et al.
It wouldn't be that big a deal to fork all the gnu tools.
They're all licensed under v2 or later.
While the copyrights have been assigned to the FSF, the original authors aren't employees of the FSF, and are free to fork their own code, same as you're free to fork it.
If the kernel stays at v2, expect to see a lot of forks; RMS might want to go on and on about GNU/Linux, but the simple fact is that that GNU/HURD isn't ever going to be competition.
Also, there's nothing in v3 that prevents you from using v3 tools to compile and build a v2 kernel. Same with openoffice - the book your writing using oo isn't automatically GPL'd. The gpl isn't "that" viral, and if an attempt is ever made to do so(a GPL v4 ???), it will be the end of any use of the GPL. After all, do you really want your email to be GPL'd, and having to make a copy of the source vailable for "distribution" if you've sent a binary attachment?
IBM, Intel, and Sun all have compilers available, if push comes to shove. Come to think of it, so does Microsoft. Now THAT would be ironic.
The incompatability stems from the incompatability between v2 and v3. v3 is not "backwards compatible"; nor is it compatible with other licenses. That's a problem. Linus has always said that v3 is an over-reaction to a relatively small problem - tivo-ization. Now that we're seeing the MS-Novell and MS-Xandros deals backfiring on all the parties involved, even without the gpl v3, I'd tend to agree.
There's nothing to stop anyone from producing hardware compatible with Tivo's code, minus the "keys".
There's nothing to stop people from switching to another distro when a particular vendor leaves a bad taste in their mouth.
Given that, and that Microsoft would be the loser in any patent war, what's the big rush? GPLv2 isn't broke, but the way some people are reacting, you'd think that Microsoft had managed to coopt all gpl v2 code.
What I'd be more concerned about is that the code written under the "covenants" with Microsoft. And with the status of the people who work on that code. Both the code and the coders will be contaminated, unable to work on related GPL products, the same as if they had partaken of the flavor-aid of Microsoft Shared Source.
Funny how this real threat - contaminated coders - is being overlooked. 5 years from now, Microsoft will be in a position to get injunctions against any distro that uses code touched by them, based on tried and tested copyright law, not patents. That's a real danger, and one that Microsoft will have no fear of retaliation from, unlike a patent war.
Remember, Microsoft has always been very skilled at getting people to look at the wrong hand, just like a magician.
Semicolon = next word is lowercase. Full colon = next word is uppercase.
Practical example - look at how you begin letters:
"Dear Sir or Madame:"
Yadda yadda yadda...
Its the way they taught it in typing class back in the old days of manual typewriters. Just like 2 spaces after each period, 8 spaces == 1 tab, etc. The semi-colon signals "less of an interruption" in the sentence structure.
Hopefully they'll ship with a spelling and grammar checker:
Acording [spelling] to a study published by Forrester Research, 2008 will be the year in wich [spelling] a psychological barrier will be surpassed. By the end of next year there will be 1.000.000.000 computers all [no computer is big enough to be "all around the world"] around the world [CC], a number that will double itself [superfluous "itself"] in just five more years. The reason: the [punctuation - first word after a full colon should be capitalized] emerging markets of countries such as Brazil, China, India or [should be "and"] Russia, which will be responsible of [should be "for"] 775 [missing a few zeroes] new PCs and laptop computers. Part of them, of course, coming from projects like the OLPC's one. [awkward sentence construct, wrong use of "them", superfluous "one"]"
It reads like spam... did anyone actually edit this?
Over time, prices out of line with the true cost of manufacture and distribution are unsupportable. If you price your product too high in relation to its true cost, you invite competitors to undercut you.
When your true costs go down, or a competitor finds a way to change the rules of the game, you have to adjust accordingly. Bittorrent and the mp3 format have both changed the rules AND lowered the base cost, by taking out all the distribution costs. Also, the web has taken out the costs of advertising to help people "find your product."
And yet, the RIAA wants people to believe that they're still entitled to effectively the same cost per song they were getting 15 years ago.
Its unsustainable over the long run...
Point in fact - Apple only keeps 13 cents for each tune. That means that the cost of manufacturing, distributing, and retailer's cut, which used to account for 80% of the $15.00 a music CD was going for (the content and profit were ~ $3.00, or 20 cents a song), are now 13 cents x 15 songs, when downloaded. That's $1.95. At the same profit per song, online music should only cost 33 cents a song, including the 13 cents for Apple's cut.
Of course, the main functions of the music houses were to take care of the manufacturing, distribution, and retailer channels. With the net, they're redundant. Artists selling their wares directly on the net will make as much money at 25 cents a song through Apple, even after Apple's 13 cent cut, than they will with the music houses charging their 87 cents a piece.
Its simple math. The recording houses have failed to pass on the cost savings from the switch from vinyl and cassette to mas-produced CDs at 15 cents a copy, to digital downloads for even less, and they have only themselves to blame for being so uncompetitive now.
Nowadays, everyone has the ability to "press records". No minimum order required. I'm surprised more musicians haven't dumped the record labels.
second-tier acts that come tour your local "tourist trap" bar get 5 figures for a night. The "setup" consists of a few emails and phone calls with the bands' manager, collecting the check, throwing the "small bar" instrument set into a minivan and partying until you can't drink any more.
Sure, they won't become milionaires, but a 6-figure income for 2 nights a week, doing what you like doing, lots of sex, free booze and, for those so inclined, drugs...
Of course, if you hit the A list, you can get a heck of a lot more, but you're more restricted in terms of venues that have the capacity to pay.
A 45-minute album is maybe 60 megabytes in mp3 format. At 10mpbs (1.25 megabytes/sec), I would be able to download that in under a minute at home, provided the server can deliver 10mpbs. I routinely download dvd isos of linux distros at full speed - the latest fedora came in quick (understandable, with over 5000 people in the swarm). Others here pay the extra $15/month and get 20mpbs, so they're looking at half a minute. Of course, that's nothing compared to some european and asian cities where people take 100mpbs as a given...
Of course, with the continuing expansion of video-on-demand, they're also talking about laying fiber to the home. I expect 1gbps within the decade (we've had 100mbps fiber in some areas for 5 years).
"Except, of course, that using an RSS feed would require the people sending the list to pay for the resources, and not the people reading the list..."
Most content management systems let you also send your sites' newest content out as an RSS feed. In postnuke, the default is http://sitename/backend.php.
"Charging some sort of cost - whether it be responding to a whitelist request, paying in CPU cycles to complete a hash, or just flat out paying a quarter of a cent - is the only practical way to fight spam"
I've seen marketers pay up to $3.00 per name for "good" phone numbers (though the usual fee is a lot less). Then they spend even more money having someone phone you. Do you really believe that $0.0025 per email is going to stop them? Heck, the'll gladly pay that to be able to send to every single legit address of the ISP.
Next step - your ISP forwards the spammers messages directly to all members inboxes, whether they want them or not, for a fee. Oops - they already made deals to do exactly that.
"It blows my mind that this shit isn't considered as big of a problem as it needs to be with lawmakers."
Why should you be surprised? Remember, politicians exempted themselves from the "Do-not-call" list. Self-interest before your interests == business as usual.
I remember those $100 per video days. The production houses complained that people were renting the tapes and making copies, and this justified the high prices. Of course, by lowering the price, they more than made up for it by increased revenues. That's what created the "direct-to-video" market.
However, in the case of music, we're talking a 3-minute item, not a 2-hour item. (Unless the song is Iron Butterfly's Ina-Godda-Da-Vida - 17min05sec, or Pink Floyd's "Meddle" 22min).
Downloading a 3-minute song is a matter of seconds. The storage requirements are also minimal. That's the real problem - its so quick and simple that it doesn't seem to have any real value. If it took an hour and filled up a whole blank CD, people would perceive a "legit" downloaded song as having more value. As you say, its a question of milking it for what its worth, which is all about the consumers' perceptions.
Slashdot Mirror
"Technical support. He is automatically the ultimate guru on the subject of his software."
Do you expect "technical support" on an embedded BSD or linux system from someone who finds #including <dlfcn.h> and making a call to dlopen() to be "too complicated"? Surely, you jest. Let me rephrase that - Shirley, you jest. You'll be better off asking Apu at the Qwiki-Mart. At least you'll get an authentic-sounding "tech support" accent.
"I read 'too much to bear' as taking too much system resources rather then 'too hard'.
In an embedded system dynamic linking can eat up scarce resources. Static linking is faster to load, faster to run, and takes up less ram."
It can also be used to free up resources. You only load what you need, as you need, then release it. It can also make startup times quicker, again, by only loading what you need at start. Use "RTLD_LAZY" - Runtime-load-lazy, so that you don't try to load everything at startup.
The real reason - the poster thinks its "too hard" to #include <dlfcn.h>, and insert one lousy call to dlopen().
More info here (with examples).
Lame, lame, lame ... tsk tsk tsk. Where's the "-1 I wish a dingo ate this article" mod?
This is great for spammers. Target their emails better, and htey KNOW that they're going to be seen, and not just silently dumped.
They should be forced by truth-in-advertising laws to call it SPAM-mail.
Better yet, is there any way a user can set it up so that all GoodMail is automatically marked as SPAM? Or better yet, sent back to the sender with a:
That ought to get people to drop GoodMail.
You miss the point about the GPL - anyone is free to use the code, provided that, if they distribute, they have to make the source available. You don't get to pick and choose who gets to use it - you can't say "I don't want [insert evil company] to use my code" - even Microsoft can use GPL'd code (and they do - python is used in WGA) provided they distribute the source either with the binory or upon request.
What if Tivo comes up with a better way to, for example, get rid of the "increstations" - the logos that most TV networks put on their shows ... you'd get the source, and you're free to modify it for work with your favourite frame-grabber.
Your own computer contains a lot of closed-source hardware - the cpu, for starters. Video cards are also closed-source hardware. Even if nvidia releases their driver source, its useless to you for programming your SiS video chip.
Also, there are specific situations where there is no way in hell you're going to see GPLv3 software used - medical devices. GPL v2 software would be fine, because you CAN "lock it down a la Tivo". Its turning out that companies that have to meet the Sarbanes-Oxley requirements have to be able to prevent unauthorized mods.
Do you really want to cede the entire medical, defense, and eventually even the networking and telco infrastructure markets? Or would you rather see the "hardening" put in those trickle bach?
However, GPL v2 is not compatible with GPL v3, because GPL v3 has additional restrictions and requirements that GPL v2 doesn't have.
Actually, even more recent online manuals of style allude to the old rule of "capitalize after a colon." They state its not followed so much any more, but that's because of a general laxness in a lot of areas nowadays.
Look at how many people can't spell basic words: "grammer", "gramer", "wierd", "thier", "formating", "referer" - yes, that last one is a stinker we're stuck with in the standards, but we really should break down and fix it.
So - how many "r"s in referrer?
Not one.
Not 2
4
"Tell me, why on earth should I write and distribute my code with a license that allows Tivo to use it without giving me back anything?"
Here's the Tivo source. You can download gzipped tarballs of all the source directly from the site, or you can even get a CD for $15.
So you see, if you decide to sell competing hardware, Tivo uses your GPL v2 code, and makes changes to it, you WILL get a copy of all the changes, to use in your product, just as the GPL provides.
The GPL is a software license, not a hardware license. GPL v3 is trying to blur the lines, and it may backfire.
There are a LOT of developers at Novell - Novell is one of the biggest contributors to kernel code, for example. Imagine being able to "taint" the kernel via this deal ... its would be worth at least $25 billion. Remember, linux is the #1 threat to them.
I agree.
Maybe we just need a "plain-talk" license, like borland used to have.
" You may use this code any way you want to, provided you're not Microsoft, SCO, or anyone else on our current sh*t list. "
At least it would be more honest ...
Then again, if Microsoft really wanted to change its spots, shake up its corporate culture, and send the right signals for a change, it would start making serious contributions under a true F/LOSS license (not "shared source").
I fear it would tank gnome.
Also, it would mean the removal of all mono-based tools, and a lot of OpenSuse/Suse uses Mono ... which was what came to my mind when people started with the "patents fud" stuff.
Fortunately, we have alternatives (except for the GIMP, which really relies heavily on gtk2).
If you were Microsoft, wouldn't it be worth a billion to "contaminate" the top developers at most of your competition, and any projects they've worked on? The FUD alone would be worth billions; the success of any copyright-based attack would be devastating. Can you imagine how hard it would be to have to revert to a code base from 5 years in the past?
Projects are going to have to start vetting their programmers for:
At least when you work on proprietary software, you can "compartmentalize" your knowledge. Write a graphics package for work, work on a completely unrelated f/loss server package on your own time. These "collaborations" are dangerous because of their long-term effects.
Linus has it right - don't look at what the competition is doing, because then they can accuse you of having unclean hands.
I wasn't saying the kernel would fork, but that all the GNU tools would be forked, which is a lot easier to do. Thats why I pointed out that IBM, Intel and Sun have compilers if we ever have to dump gcc.
Forking cat and head (and dumping those stupid .info files - man pages are good enough) ... why not ?
"In addition, one particular piece of icing, is that it is actually compatible with more open source licenses, in particular the Apache License." ... but its not back-compatible with gpl v2 - and that's a nasty "bug", because GPL v2 is a known. proven quantity - people are comfortable with it. The process around GPL v3 sucked, the constant whining about Microsoft/Novell made the people involved look immature, and it doesn't do what its proponents claim it does - it provides zero protection against patent trolls, for example, who are far more likely to make an attack than Microsoft.
"more legally binding"?
Either it is, or it isn't. Its only in the SCOniverse that the GPL isn't a legally binding document.
Of course, I'm willing to listen to counter-arguments. What is this "icing" you speak of? All I see is some extra verbiage that attempts to deal with the latest round of patent fud (a fud that has already died aborning, claiming Xandros as its latest victim - no big loss there).
There's no "hole." Its a software license, and it works fine in its problem domain - software.
Trying to extend it to hardware (tivo-ization) is what is causing the problems.
Anyone is free to use Tivo's software, strip out the DRM aspects, and market their own brand of hardware, thanks to the GPLv2.
As for the patent "issues" - they've already collapsed. Look at the latest FUD victim - Xandros. Its dropped 10 places on distrowatch in 2 weeks (2 weeks ago it was at #18, now its at #28) . There will never be a Microsoft-Linux patent war. Microsoft has too much to lose by actually naming any supposed patent, and facing patent retaliation from IBM et al.
It wouldn't be that big a deal to fork all the gnu tools.
They're all licensed under v2 or later.
While the copyrights have been assigned to the FSF, the original authors aren't employees of the FSF, and are free to fork their own code, same as you're free to fork it.
If the kernel stays at v2, expect to see a lot of forks; RMS might want to go on and on about GNU/Linux, but the simple fact is that that GNU/HURD isn't ever going to be competition.
Also, there's nothing in v3 that prevents you from using v3 tools to compile and build a v2 kernel. Same with openoffice - the book your writing using oo isn't automatically GPL'd. The gpl isn't "that" viral, and if an attempt is ever made to do so(a GPL v4 ???), it will be the end of any use of the GPL. After all, do you really want your email to be GPL'd, and having to make a copy of the source vailable for "distribution" if you've sent a binary attachment?
IBM, Intel, and Sun all have compilers available, if push comes to shove. Come to think of it, so does Microsoft. Now THAT would be ironic.
The incompatability stems from the incompatability between v2 and v3. v3 is not "backwards compatible"; nor is it compatible with other licenses. That's a problem. Linus has always said that v3 is an over-reaction to a relatively small problem - tivo-ization. Now that we're seeing the MS-Novell and MS-Xandros deals backfiring on all the parties involved, even without the gpl v3, I'd tend to agree.
There's nothing to stop anyone from producing hardware compatible with Tivo's code, minus the "keys".
There's nothing to stop people from switching to another distro when a particular vendor leaves a bad taste in their mouth.
Given that, and that Microsoft would be the loser in any patent war, what's the big rush? GPLv2 isn't broke, but the way some people are reacting, you'd think that Microsoft had managed to coopt all gpl v2 code.
What I'd be more concerned about is that the code written under the "covenants" with Microsoft. And with the status of the people who work on that code. Both the code and the coders will be contaminated, unable to work on related GPL products, the same as if they had partaken of the flavor-aid of Microsoft Shared Source.
Funny how this real threat - contaminated coders - is being overlooked. 5 years from now, Microsoft will be in a position to get injunctions against any distro that uses code touched by them, based on tried and tested copyright law, not patents. That's a real danger, and one that Microsoft will have no fear of retaliation from, unlike a patent war.
Remember, Microsoft has always been very skilled at getting people to look at the wrong hand, just like a magician.
Semicolon = next word is lowercase. Full colon = next word is uppercase.
Practical example - look at how you begin letters:
"Dear Sir or Madame:"
Yadda yadda yadda ...
Its the way they taught it in typing class back in the old days of manual typewriters. Just like 2 spaces after each period, 8 spaces == 1 tab, etc. The semi-colon signals "less of an interruption" in the sentence structure.
Hopefully they'll ship with a spelling and grammar checker:
Acording [spelling] to a study published by Forrester Research, 2008 will be the year in wich [spelling] a psychological barrier will be surpassed. By the end of next year there will be 1.000.000.000 computers all [no computer is big enough to be "all around the world"] around the world [CC], a number that will double itself [superfluous "itself"] in just five more years. The reason: the [punctuation - first word after a full colon should be capitalized] emerging markets of countries such as Brazil, China, India or [should be "and"] Russia, which will be responsible of [should be "for"] 775 [missing a few zeroes] new PCs and laptop computers. Part of them, of course, coming from projects like the OLPC's one. [awkward sentence construct, wrong use of "them", superfluous "one"]"
It reads like spam ... did anyone actually edit this?
Over time, prices out of line with the true cost of manufacture and distribution are unsupportable. If you price your product too high in relation to its true cost, you invite competitors to undercut you.
When your true costs go down, or a competitor finds a way to change the rules of the game, you have to adjust accordingly. Bittorrent and the mp3 format have both changed the rules AND lowered the base cost, by taking out all the distribution costs. Also, the web has taken out the costs of advertising to help people "find your product."
And yet, the RIAA wants people to believe that they're still entitled to effectively the same cost per song they were getting 15 years ago.
Its unsustainable over the long run ...
Point in fact - Apple only keeps 13 cents for each tune. That means that the cost of manufacturing, distributing, and retailer's cut, which used to account for 80% of the $15.00 a music CD was going for (the content and profit were ~ $3.00, or 20 cents a song), are now 13 cents x 15 songs, when downloaded. That's $1.95. At the same profit per song, online music should only cost 33 cents a song, including the 13 cents for Apple's cut.
Of course, the main functions of the music houses were to take care of the manufacturing, distribution, and retailer channels. With the net, they're redundant. Artists selling their wares directly on the net will make as much money at 25 cents a song through Apple, even after Apple's 13 cent cut, than they will with the music houses charging their 87 cents a piece.
Its simple math. The recording houses have failed to pass on the cost savings from the switch from vinyl and cassette to mas-produced CDs at 15 cents a copy, to digital downloads for even less, and they have only themselves to blame for being so uncompetitive now.
Nowadays, everyone has the ability to "press records". No minimum order required. I'm surprised more musicians haven't dumped the record labels.
second-tier acts that come tour your local "tourist trap" bar get 5 figures for a night. The "setup" consists of a few emails and phone calls with the bands' manager, collecting the check, throwing the "small bar" instrument set into a minivan and partying until you can't drink any more.
Sure, they won't become milionaires, but a 6-figure income for 2 nights a week, doing what you like doing, lots of sex, free booze and, for those so inclined, drugs ...
Of course, if you hit the A list, you can get a heck of a lot more, but you're more restricted in terms of venues that have the capacity to pay.
A 45-minute album is maybe 60 megabytes in mp3 format. At 10mpbs (1.25 megabytes/sec), I would be able to download that in under a minute at home, provided the server can deliver 10mpbs. I routinely download dvd isos of linux distros at full speed - the latest fedora came in quick (understandable, with over 5000 people in the swarm). Others here pay the extra $15/month and get 20mpbs, so they're looking at half a minute. Of course, that's nothing compared to some european and asian cities where people take 100mpbs as a given ...
Of course, with the continuing expansion of video-on-demand, they're also talking about laying fiber to the home. I expect 1gbps within the decade (we've had 100mbps fiber in some areas for 5 years).
"Except, of course, that using an RSS feed would require the people sending the list to pay for the resources, and not the people reading the list..."
Most content management systems let you also send your sites' newest content out as an RSS feed. In postnuke, the default is http://sitename/backend.php.
"Charging some sort of cost - whether it be responding to a whitelist request, paying in CPU cycles to complete a hash, or just flat out paying a quarter of a cent - is the only practical way to fight spam"
I've seen marketers pay up to $3.00 per name for "good" phone numbers (though the usual fee is a lot less). Then they spend even more money having someone phone you. Do you really believe that $0.0025 per email is going to stop them? Heck, the'll gladly pay that to be able to send to every single legit address of the ISP.
Next step - your ISP forwards the spammers messages directly to all members inboxes, whether they want them or not, for a fee. Oops - they already made deals to do exactly that.
"It blows my mind that this shit isn't considered as big of a problem as it needs to be with lawmakers."
Why should you be surprised? Remember, politicians exempted themselves from the "Do-not-call" list. Self-interest before your interests == business as usual.
I remember those $100 per video days. The production houses complained that people were renting the tapes and making copies, and this justified the high prices. Of course, by lowering the price, they more than made up for it by increased revenues. That's what created the "direct-to-video" market.
However, in the case of music, we're talking a 3-minute item, not a 2-hour item. (Unless the song is Iron Butterfly's Ina-Godda-Da-Vida - 17min05sec, or Pink Floyd's "Meddle" 22min).
Downloading a 3-minute song is a matter of seconds. The storage requirements are also minimal. That's the real problem - its so quick and simple that it doesn't seem to have any real value. If it took an hour and filled up a whole blank CD, people would perceive a "legit" downloaded song as having more value. As you say, its a question of milking it for what its worth, which is all about the consumers' perceptions.