Thats true, but think about Knuth's work on TeX. That is still in serious use.
One of the circumstances in which software remains current is when it has been widely adopted as a standard for communication. This is also a circumstance in which the proprietry software model fails users especially badly, with a market imperative to sell new versions at odds with the users desire for stability, so he can go about his real business, communicating, and not have to bother keeping his software up to date.
If I thought my GNU software were perpetual I'd but an early expiration to public domain in there. In fact I may
do so anyway for things I open source from now on.
The purpose of copyright is to provide an incentive for creative artists. We all discount the future rather heavily. 20 to 25 years is about the limit. Copyright should be no longer than this.
The purpose of copyleft is freedom. Once copyleft expires, it is back to the bad old days of breaking interoperability for commercial advantage, keeping the source code secret, so that most programming is duplication of work already done, etc. There is no public interest in having copyleft expire.
A rewriting of copyright law, to 20 years from publication, because that provides plenty of incentive, could also provide for two types of public domain.
1)The current public domain, the source of copyright derivative works.
Water mark technology assumes that the watermark doesn't have to survive audible degradation. So you send L+X out the left channel and R-X out the right channel, where X is noise, as loud as the music. Combine them with two resistors (cost $0.02). You have
(L+X)+(R-X) = L+R
You'll need to buy another sound card to get stereo, but RIAA will be happy for you to buy a sound card with automatic watermark detection, because they haven't realised how easy it is to beat. Perhaps they will buy a law limiting persons to one mono sound card per PC:)
I don't believe this post is genuine. Perhaps I've been put on my guard by
PR Firm Fakes Online Posters
but I've also read lots of RMS's own words on the GNU website, and this post doesn't ring true.
One tiny little word, with only seven syllables, says it all. In the past, if the Supreme Court handed down a controversial decision, ordinary folks read about it in the newspapers. Now they go to the Court and read the Judges' own words. Journalists are right to be scared of the internet.
It is our governments, the EU and the USA, that do the most harm, with the Common Agricultural Policy and farm subsidies, and Multi-fibre Agreements.
If our own governments would stop shafting the poor of the world, and let them trade freely, their countries could develop a merchant middle class, and be better able to stand up to the lesser bullies who you cite.
When the peasants in a third world country want a new X-ray machine for the local hospital, they do not refine their own tungsten for the anode, nor wind their own high voltage transformers, nor build their own vaccuum pump. They grow their crops, weave their cloth, and trade.
We in Europe and also the USA make life hard for them by keeping their food and their textiles out of our markets. So they have to turn to other work, such as making trainers for Nike. The real villains are the Protectionists, who wage economic warfare against the poor of the world with trade barriers.
I got rid of my TV. I resented paying money to the BBC for dumbed down programs. The TV news used to really wind me up. They would strike poses like "rising house prices are good" (duh, the cost of housing is a cost folks) or "tariff barriers will protect jobs" (That really worked good in the 1930's didn't it).
Now I've discovered message boards. If some one says something dumb, you can post a refutation. No more passive media for me.
It is not hard to fill the hole in your life when you get rid of your TV. Hiding at the back of the second violins in the Stockbridge and Newtown Cummunity Orchestra does it for me.
Imagine you have three potential readers, willing to pay $15,$10, and $5 respectively.
Without resale you can make one sale at $15. If you want a second sale you can to cut the price to $10.
With resale you can make two sales at $15, because the chap who was only willing to pay $10 can pay $15 and then get $5 back selling it on.
So instead of having to choose between one sale at $15 or two at $10, the publisher gets to make two sales at $15. In micro-economics jargon this increases prices at the high demand, low price end of the demand curve.
You are right, and I am very surprised. BMW had an advertising campaign, aimed at the company car fleet market, saying that although their cars were more expensive to buy new, they held their value better. So when the company sold them after three years, they got more money back. BMW's big claim was that their cars held there value so well that it more than made up for the higher purchase price, thus BMW's were cheap cars for the cost conscious buyer!
In the car market sellers are very well aware that a strong second hand market supports the price at first sale. This makes authors look stupid as well as greedy. Who wants to read a book written by a stupid writer?
CO2 is about 370 parts per million (ppm) while O2 is about 21000 parts per million, so scrubbing out ALL the CO2 still leaves 20630 parts per million O2.
RMS encourages distributors of GPL software to charge what the market will bear
I understand it like this. Imagine that in the future colonists around Alpha Centurii want some GPL software. Hiring a big radio telescope and a powerful transmitter for long enough to send the software might cost $30000. If the radio telescopes available are booked solid, charge $100000 and get rich. Of course, if others notice your wealth and discover how you came by it, they might go into the business themselves. They could build their own transmitting station and charge $70000 for sending the next release. Which still leaves fat enough profits to encourage other to get into the business of distributing GPL software.
Economics is full of unhappy entailments. If you say that GPL software should be distributed for a low fee, it is inherent in this that GPL software will not reach places that are difficult and dangerous to get to. The inhabitants of such places will have to buy non-GPL software. This is not what RMS wants, so he wisely refrains from seeking to limit distribution fees. Is RMS a moralist after the style of Adam Smith, happy enough when greed guides persons, as though by an invisible hand, to serve the common good? It seems so to me.
If it is cheaper to pay a programmer to extend a free package rather than buy a proprietary one...
Mancur Olson wrote a fascinating book, The Logic of Collective Action, all about the provision of public goods. He discusses exactly the problem that the Free Software community faces. Every-one would be better off if they all chipped in money to pay programmers, but since those who don't pay aren't excluded, everyone leaves it to someone else to come up with the money:(
He notes in passing that sometimes the benefit to a single individual is greater than the cost of providing the public good, so it becomes worth his while to pay for it all by himself. You can imagine a shipping magnate paying for a lighthouse for his home port as an example.
This raises the possibility of a critical mass effect in Free Software. If a piece of GPL software is close enough to the needs of a big company it may pay that big company to hire a programmer to close the gap rather than pay for a commercial package. At which point another big company might say, it wouldn't cost too much to add the other feature that we need. Then a third company notices that the GPL code now has the two features it was waiting for, it would be cheaper to hire a programmer to track down that irritating bug than to buy proprietry software, and so on....
Why does a webserver has to be a big program? With a different architecture, couldn't it be a small core that serves the pages, with addition "software components" connected as needed.
A relational database needs an engine, which will be a fairly substantial program, presumably derived from academic research. Does turning that into a software product need lots of coders, paid to code up interfaces and bells and whistles? My intuition is that today, generally speaking, it does, but I see this as much more a product of the proprietry software business than as the origin of that business model.
If you are aiming to make money by selling upgrades, you need to retain control of the upgrade process. If you are only selling an engine that plugs into a generic user interface, the user obtains user interface enhancements across all his applications when he upgrades his user interface. He doesn't come back to you for version 3 to get history lists or whatever. If you sell your engine with well documented hooks for others to connect in their own enhancements, you again undercut your upgrade market. I have an uncomfortable feeling that open source projects tend to tag along behind commercial software, using architectures optimised for extracting money from users, just because "that is the way you write software". I'm sure it could all be different and better, I just cannot work out how:(
I want to learn more about emacs. I'm sure there is a clue in there somewhere. I've only just found out that I can type (cons 'a 'b) into any buffer then press ^U ^X ^E and it evaluates my lisp and adds (a . b) into my buffer. If I've got a database engine from an academic, can I write an emacs major mode to talk to it? Does this let me leverage the functionality already present in emacs, eg emacs diary schedules database updates, lots of cool things become possible with relatively little coding, and the same old user interface I'm already learning.
Notice the phrase "relatively little coding". We've got this big question mark hanging over open source software, of how to pay the coders. How much of this is an artificial concern due to the industry being set up so that it takes acres of code to get anything cool to work?
I don't see proprietry software as the great solution to this problem. You write foobar and sell in at $50. Microsoft brings out foobar-for-windows. While they are charging $50 also, you are still in business. Once they bundle it with Windows XP ultra, you are out of business. No guarantees that you will recoup your development costs before that happens.
Suppose you are in a part of the software market that is still competitive. Five different companies, each selling a similar package for $1000. Lots of programmers are getting paid, but for what? If a new company enters the market it will hire programmers to write its package, that it will try to sell for $1000, but these programmers will not be improving the existing code, they will be employed in repeating work already done, so that the new company can own its own version and split the money available six ways instead of five ways.
What does this do for the users? We saw what it does to the users in the UNIX wars. You get several versions of much the same concept. The programmers break inter-operability between the different versions for commercial advantage by locking in "their" users. I guess that RMS would say "reduce to serfdom" rather than "lock in". Whatever. Users aren't getting value for money when they pay the coders via the proprietry software system.
I'm trying to get back into computer programming after a very lengthy illness. I'm horrified by how insanely complicated it has all become in the past ten years. C++ is twice as big as ANSI Common Lisp and offers a quarter the functionality. XML has 3000 pages of manuals that tell you how to do sod all. The PDF manual is about 1000 pages, and it is just a document format. What does html offer? The core functionallity is the anchor tag so that you can click on links, and reflowing the text so that you can resize your browsers window, but can you write your own browser? No way, there is Java, and Java Script and Flash and endless complication, but can I put maths on my web page? Until very recently, no.
Presumably the point of all this insane complexity is to create barriers to entry, in that programmers have to work in big teams for large companies. It is naive to continue to believe that a programmer can write a program, retain the copyright and join the rentier class living off the royalties. You have to work for a big company who own the copyright to your work, and exploit you just like the big record companies exploit their artists.
There has been no progress in software components since the UNIX pipe was invented. It ought to be possible to write a tiny little program, plug it into your desk top, and hey presto, a fancy program with a full feature user interface. Then users could write their own programs. Somehow we have got trapped into a model of software development that requires big teams of full time coders to get anything done. It looks to me like we have been too successful in paying coders in the past, and have gone down a blind alley
It is the contents page page that comes up orange so stricly speaking it's the links. In the past I've fixed this by saving the source to a file and editing the colours to something sensible, but this time the colours are
I'm glad to find that I was not the only person to find lines running out of his browser window. I also looked at the source, which I found to be slightly more readable; it did not force me to scroll sideways. By the time I had used Lynx to degunk the page, I had lost confidence that it was worth reading.
I discuss window width on my website. Try the buttons that change the number of columns and tell me what you think.
Consumer Broadband and Digital Television Promotion Act
Content Ban Directly Targetting Performances by Amateurs
The bill says
(14) When protected digital content is converted to analog for consumers, it is no longer protected and is subject to
conversion into unprotected digital form that can in turn be copied or redistribute illegally.
(15) As solution to this problem is technologically feasible but will require government action,
It is talking about watermarks. Your CBDTPA-compliant sound card will add "do not play" watermarks to the sound it outputs, and will honour such watermarks in the files sent to it. Ofcourse, false positives will cause endless grief for amateur musicians putting their own performances into the public domain. I presume that the point of
SEC. 8. FEDERAL ADVISORY COMMITTEE ACT EXEMPTION.
The Federal Advisory Committee Act (5 U.S.C. App.) does not apply to any committee, board, commission, council, conference, panel, task force, or other similar group of representatives of digital media devices and representatives of copyright owners convened for the purpose of developing the security system standards and encoding rules described in section 3.
is to shield discussion of false positives from public view.
The Digital rights management of the CBDTPA depends upon watermarking. The sound card on a `copy frustrated' PC checks whether the file it is being asked to output has a watermark indicating that it should not be played. A major danger is that false positives from the watermark check may preventing you from playing your own content. You can repair such a malfunction in either of these two ways.
If you are content with mono output, you can create a mono signal M=L+R. Then, using an arbitary signal X, create new left and right channels L'=M+X, R'=M-X. Finally, combine the left and right channels outputs with two accurately matched resistors. A trim-pot may be needed to prevent residual X being audible in quiet passages.
Alternatively build a simple circuit with two resistors and a diode to give an instantaneous non-linear transfer function f. You will need another one g for the other channel. Digitise a mono test signal T, with f(T) going to the right channel and T going to the left channel. Calculate the inverse of f, say invf. Similarly for g. Now put f and g on the outputs of your sound card and play invf(L) and invg(R). Your speakers recieve f(invf(L))=L and g(invg(R))=R
Why does this work? In the first case, DRM must permit the playing of content with a large added signal, otherwise it would block recording of talking when copyright music was playing in the background. In the second case, the central assumption of music watermarking is that attempts to erase the watermark must preserve the short term magnitude spectrum, otherwise the music will be distorted, but a non-linearity gives rise to harmonics and intermodulation products that change the short term magnitude spectrum. These will shift your own content clear of any accidental matches with copy frustration codes.
The numbers are a crytal ball. If you can see them and read them, they tell you the future.
Back in the dot com boom there was a fashion for internet grocers who delivered by van. I read a newspaper article that told me that it cost twenty pounds an hour to run a van, and vans could do 3 or 4 deliveries an hour in a busy city. Internet grocers were charging five pounds delivery. There was nothing left to pay the stock pickers. Sure enough internet grocers have faded from the scene.
Could musicians cut out the middle man and sell CD's in cooperatively owned record shops at two pounds each? That cures the copyright infringement problem. CD's are pretty, shiny things that are nice to collect, and have much better sound quality than mp3 at the usual bit rates.
Usually the wholesale price is half the retail price. I guess half the price is split equally between shop wages and shop overheads. So of the ten pounds you pay for a CD 2 pound 50 goes to the wages of the shopkeeper. Assuming a wage of five pounds an hour, that is a sale every half hour. That sounds low for Saturday afternoon, high for Monday morning. The trouble is I'm talking out of my arse here. I am dying to hear the real figures from somebody in the business.
Continuing my guess work: if you plan to solve the copyright infringement problem by selling CD's for two pounds each, you need a different kind of record shop, which makes a sale every six minutes. That is possible. News-agents sell magazines at those rates and prices. Imagine lots (say 200) independent labels, each with 50 bands, each bringing out a CD every year. Each week you the latest CD from your favourite record label, just like you buy a weekly magazine. If you are going on a journey, you stop at the station bookstall and pay two pounds for a magazine to read to pass the time or two pounds for a CD to listen to. Perhaps both. You discard them at the end of your journey! If ten million persons buy into that model of music buying the total revenue is a billion a year, with 10000 bands to support, say a 50000 a year per band. Its a living. If only a million persons buy into that model of music buying then most of the musicians can only be semi-professional, much like it is now.
See what a difference some numbers make. Suddenly I'm telling a story that opens up real possibilities. If I was in the business and had some accurate numbers at my finger tips, well...
I was disappointed by the list of business models. There were no back of the envelope calculations to suggest whether a musician could earn his living, or whether he would have to keep his day job.
Slashdot Mirror
Software is rarely current after 5 years
Thats true, but think about Knuth's work on TeX. That is still in serious use.
One of the circumstances in which software remains current is when it has been widely adopted as a standard for communication. This is also a circumstance in which the proprietry software model fails users especially badly, with a market imperative to sell new versions at odds with the users desire for stability, so he can go about his real business, communicating, and not have to bother keeping his software up to date.
If I thought my GNU software were perpetual I'd but an early expiration to public domain in there. In fact I may do so anyway for things I open source from now on.
I've just got to ask: Why?
The purpose of copyright is to provide an incentive for creative artists. We all discount the future rather heavily. 20 to 25 years is about the limit. Copyright should be no longer than this.
The purpose of copyleft is freedom. Once copyleft expires, it is back to the bad old days of breaking interoperability for commercial advantage, keeping the source code secret, so that most programming is duplication of work already done, etc. There is no public interest in having copyleft expire.
A rewriting of copyright law, to 20 years from publication, because that provides plenty of incentive, could also provide for two types of public domain.
1)The current public domain, the source of copyright derivative works.
2)A GNU domain, perpetually copyleft.
Have you read Right to read yet?
I don't believe this post is genuine. Perhaps I've been put on my guard by PR Firm Fakes Online Posters but I've also read lots of RMS's own words on the GNU website, and this post doesn't ring true.
One tiny little word, with only seven syllables, says it all. In the past, if the Supreme Court handed down a controversial decision, ordinary folks read about it in the newspapers. Now they go to the Court and read the Judges' own words. Journalists are right to be scared of the internet.
"Cheat me once, shame on you, cheat me twice, shame on me."
....
If the businessman pays the $500, he still has no rights to the source, and has to pay another $500 next time, and the time after that, and
multi-nationals like Enron
large multi-nationals like Nike
Monsanto
It is our governments, the EU and the USA, that do the most harm, with the Common Agricultural Policy and farm subsidies, and Multi-fibre Agreements.
If our own governments would stop shafting the poor of the world, and let them trade freely, their countries could develop a merchant middle class, and be better able to stand up to the lesser bullies who you cite.
We in Europe and also the USA make life hard for them by keeping their food and their textiles out of our markets. So they have to turn to other work, such as making trainers for Nike. The real villains are the Protectionists, who wage economic warfare against the poor of the world with trade barriers.
Am I a shill for Nike? Read my declaration of interest
I got rid of my TV. I resented paying money to the BBC for dumbed down programs. The TV news used to really wind me up. They would strike poses like "rising house prices are good" (duh, the cost of housing is a cost folks) or "tariff barriers will protect jobs" (That really worked good in the 1930's didn't it).
Now I've discovered message boards. If some one says something dumb, you can post a refutation. No more passive media for me.
It is not hard to fill the hole in your life when you get rid of your TV.
Hiding at the back of the second violins in the
Stockbridge and Newtown Cummunity Orchestra does it for me.
Imagine you have three potential readers, willing to pay $15,$10, and $5 respectively.
Without resale you can make one sale at $15. If you want a second sale you can to cut the price to $10.
With resale you can make two sales at $15, because the chap who was only willing to pay $10 can pay $15 and then get $5 back selling it on.
So instead of having to choose between one sale at $15 or two at $10, the publisher gets to make two sales at $15. In micro-economics jargon this increases prices at the high demand, low price end of the demand curve.
> What very few people are realizing here...
You are right, and I am very surprised. BMW had an advertising campaign, aimed at the company car fleet market, saying that although their cars were more expensive to buy new, they held their value better. So when the company sold them after three years, they got more money back. BMW's big claim was that their cars held there value so well that it more than made up for the higher purchase price, thus BMW's were cheap cars for the cost conscious buyer!
In the car market sellers are very well aware that a strong second hand market supports the price at first sale. This makes authors look stupid as well as greedy. Who wants to read a book written by a stupid writer?
CO2 is about 370 parts per million (ppm)
while O2 is about 21000 parts per million,
so scrubbing out ALL the CO2 still leaves
20630 parts per million O2.
but why wasn't he drawn with a beard?
RMS encourages distributors of GPL software to
charge what the market will bear
I understand it like this. Imagine that in the
future colonists around Alpha Centurii want some GPL
software. Hiring a big radio telescope and a powerful
transmitter for long enough to send the software might cost
$30000. If the radio telescopes available are booked solid,
charge $100000 and get rich. Of course, if others notice
your wealth and discover how you came by it, they might go
into the business themselves. They could build their own
transmitting station and charge $70000 for sending the next
release. Which still leaves fat enough profits to encourage
other to get into the business of distributing GPL software.
Economics is full of unhappy entailments. If you say that
GPL software should be distributed for a low fee, it is
inherent in this that GPL software will not reach places
that are difficult and dangerous to get to. The inhabitants
of such places will have to buy non-GPL software. This is
not what RMS wants, so he wisely refrains from seeking to
limit distribution fees. Is RMS a moralist after the style
of Adam Smith, happy enough when greed guides persons, as
though by an invisible hand, to serve the common good? It
seems so to me.
Mancur Olson wrote a fascinating book, The Logic of Collective Action, all about the provision of public goods. He discusses exactly the problem that the Free Software community faces. Every-one would be better off if they all chipped in money to pay programmers, but since those who don't pay aren't excluded, everyone leaves it to someone else to come up with the money :(
He notes in passing that sometimes the benefit to a single individual is greater than the cost of providing the public good, so it becomes worth his while to pay for it all by himself. You can imagine a shipping magnate paying for a lighthouse for his home port as an example.
This raises the possibility of a critical mass effect in Free Software. If a piece of GPL software is close enough to the needs of a big company it may pay that big company to hire a programmer to close the gap rather than pay for a commercial package. At which point another big company might say, it wouldn't cost too much to add the other feature that we need. Then a third company notices that the GPL code now has the two features it was waiting for, it would be cheaper to hire a programmer to track down that irritating bug than to buy proprietry software, and so on....
A relational database needs an engine, which will be a fairly substantial program, presumably derived from academic research. Does turning that into a software product need lots of coders, paid to code up interfaces and bells and whistles? My intuition is that today, generally speaking, it does, but I see this as much more a product of the proprietry software business than as the origin of that business model.
If you are aiming to make money by selling upgrades, you need to retain control of the upgrade process. If you are only selling an engine that plugs into a generic user interface, the user obtains user interface enhancements across all his applications when he upgrades his user interface. He doesn't come back to you for version 3 to get history lists or whatever. If you sell your engine with well documented hooks for others to connect in their own enhancements, you again undercut your upgrade market. I have an uncomfortable feeling that open source projects tend to tag along behind commercial software, using architectures optimised for extracting money from users, just because "that is the way you write software". I'm sure it could all be different and better, I just cannot work out how :(
I want to learn more about emacs. I'm sure there is a clue in there somewhere. I've only just found out that I can type (cons 'a 'b) into any buffer then press ^U ^X ^E and it evaluates my lisp and adds (a . b) into my buffer. If I've got a database engine from an academic, can I write an emacs major mode to talk to it? Does this let me leverage the functionality already present in emacs, eg emacs diary schedules database updates, lots of cool things become possible with relatively little coding, and the same old user interface I'm already learning.
Notice the phrase "relatively little coding". We've got this big question mark hanging over open source software, of how to pay the coders. How much of this is an artificial concern due to the industry being set up so that it takes acres of code to get anything cool to work?
I don't see proprietry software as the great solution to this problem. You write foobar
and sell in at $50. Microsoft brings out foobar-for-windows. While they are charging $50 also, you are still in business. Once they bundle it with Windows XP ultra, you are out of business. No guarantees that you will recoup your development costs before that happens.
Suppose you are in a part of the software market that is still competitive. Five different companies, each selling a similar package for $1000. Lots of programmers are getting paid, but for what? If a new company enters the market it will hire programmers to write its package, that it will try to sell for $1000, but these programmers will not be improving the existing code, they will be employed in repeating work already done, so that the new company can own its own version and split the money available six ways instead of five ways.
What does this do for the users? We saw what it does to the users in the UNIX wars. You get several versions of much the same concept. The programmers break inter-operability between the different versions for commercial advantage by locking in "their" users. I guess that RMS would say "reduce to serfdom" rather than "lock in". Whatever. Users aren't getting value for money when they pay the coders via the proprietry software system.
I'm trying to get back into computer programming after a very lengthy illness. I'm horrified by how insanely complicated it has all become in the past ten years. C++ is twice as big as ANSI Common Lisp and offers a quarter the functionality. XML has 3000 pages of manuals that tell you how to do sod all. The PDF manual is about 1000 pages, and it is just a document format. What does html offer? The core functionallity is the anchor tag so that you can click on links, and reflowing the text so that you can resize your browsers window, but can you write your own browser? No way, there is Java, and Java Script and Flash and endless complication, but can I put maths on my web page? Until very recently, no.
Presumably the point of all this insane complexity is to create barriers to entry, in that programmers have to work in big teams for large companies. It is naive to continue to believe that a programmer can write a program, retain the copyright and join the rentier class living off the royalties. You have to work for a big company who own the copyright to your work, and exploit you just like the big record companies exploit their artists.
There has been no progress in software components since the UNIX pipe was invented. It ought to be possible to write a tiny little program, plug it into your desk top, and hey presto, a fancy program with a full feature user interface. Then users could write their own programs. Somehow we have got trapped into a model of software development that requires big teams of full time coders to get anything done. It looks to me like we have been too successful in paying coders in the past, and have gone down a blind alley
It is the contents page page that comes up orange so stricly speaking it's the links. In the past I've fixed this by saving the source to a file and editing the colours to something sensible, but this time the colours are
F "
BGCOLOR="#FFFFFF"
TEXT="#000000"
LINK="#0000F
VLINK="#840084"
ALINK="#0000FF"
ie white, black, blue, magenta, blue,
so where has the orange come from, how do I get rid of it?
I'm glad to find that I was not the only person to find lines running out of his browser window. I also looked at the source, which I found to be slightly more readable; it did not force me to scroll sideways. By the time I had used Lynx to degunk the page, I had lost confidence that it was worth reading.
I discuss window width on my website. Try the buttons that change the number of columns and tell me what you think.
Consumer Broadband and Digital Television Promotion Act
Content Ban Directly Targetting Performances by Amateurs
The bill says
(14) When protected digital content is converted to analog for consumers, it is no longer protected and is subject to
conversion into unprotected digital form that can in turn be copied or redistribute illegally.
(15) As solution to this problem is technologically feasible but will require government action,
It is talking about watermarks. Your CBDTPA-compliant sound card will add "do not play" watermarks to the sound it outputs, and will honour such watermarks in the files sent to it. Ofcourse, false positives will cause endless grief for amateur musicians putting their own performances into the public domain. I presume that the point of
SEC. 8. FEDERAL ADVISORY COMMITTEE ACT EXEMPTION.
The Federal Advisory Committee Act (5 U.S.C. App.) does not apply to any committee, board, commission, council, conference, panel, task force, or other similar group of representatives of digital media devices and representatives of copyright owners convened for the purpose of developing the security system standards and encoding rules described in section 3.
is to shield discussion of false positives from public view.
False positives from the watermark detection will
stop you playing content that you have created
yourself, which will be really frustrating.
If you thinks computers are annoying today,
you ain't seen nothing yet!
The Digital rights management of the CBDTPA depends upon
watermarking. The sound card on a `copy frustrated' PC
checks whether the file it is being asked to output has a
watermark indicating that it should not be played. A major
danger is that false positives from the watermark check may
preventing you from playing your own content. You can repair
such a malfunction in either of these two ways.
If you are content with mono output, you can create a mono
signal M=L+R. Then, using an arbitary signal X, create new
left and right channels L'=M+X, R'=M-X. Finally, combine the
left and right channels outputs with two accurately matched
resistors. A trim-pot may be needed to prevent residual X
being audible in quiet passages.
Alternatively build a simple circuit with two resistors and
a diode to give an instantaneous non-linear transfer
function f. You will need another one g for the other
channel. Digitise a mono test signal T, with f(T) going to
the right channel and T going to the left channel. Calculate
the inverse of f, say invf. Similarly for g. Now put f and g
on the outputs of your sound card and play invf(L) and
invg(R). Your speakers recieve f(invf(L))=L and g(invg(R))=R
Why does this work? In the first case, DRM must permit the
playing of content with a large added signal, otherwise it
would block recording of talking when copyright music was
playing in the background. In the second case, the central
assumption of music watermarking is that attempts to erase
the watermark must preserve the short term magnitude
spectrum, otherwise the music will be distorted, but a
non-linearity gives rise to harmonics and intermodulation
products that change the short term magnitude
spectrum. These will shift your own content clear of any
accidental matches with copy frustration codes.
them, they tell you the future.
Back in the dot com boom there was a fashion for
internet grocers who delivered by van. I read a newspaper
article that told me that it cost twenty pounds an hour to
run a van, and vans could do 3 or 4 deliveries an hour in a
busy city. Internet grocers were charging five pounds
delivery. There was nothing left to pay the stock pickers.
Sure enough internet grocers have faded from the scene.
Could musicians cut out the middle man and sell CD's in
cooperatively owned record shops at two pounds each? That
cures the copyright infringement problem. CD's are pretty,
shiny things that are nice to collect, and have much better
sound quality than mp3 at the usual bit rates.
Usually the wholesale price is half the retail price. I
guess half the price is split equally between shop wages and
shop overheads. So of the ten pounds you pay for a CD 2
pound 50 goes to the wages of the shopkeeper. Assuming a
wage of five pounds an hour, that is a sale every half
hour. That sounds low for Saturday afternoon, high for
Monday morning. The trouble is I'm talking out of my arse
here. I am dying to hear the real figures from somebody in
the business.
Continuing my guess work: if you plan to solve the
copyright infringement problem by selling CD's for two
pounds each, you need a different kind of record shop, which
makes a sale every six minutes. That is
possible. News-agents sell magazines at those rates and
prices. Imagine lots (say 200) independent labels, each with
50 bands, each bringing out a CD every year. Each week you
the latest CD from your favourite record label, just like
you buy a weekly magazine. If you are going on a journey,
you stop at the station bookstall and pay two pounds for a
magazine to read to pass the time or two pounds for a CD to
listen to. Perhaps both. You discard them at the end of your
journey! If ten million persons buy into that model of music
buying the total revenue is a billion a year, with 10000
bands to support, say a 50000 a year per band. Its a living.
If only a million persons buy into that model of music
buying then most of the musicians can only be
semi-professional, much like it is now.
See what a difference some numbers make. Suddenly I'm
telling a story that opens up real possibilities. If I was
in the business and had some accurate numbers at my finger
tips, well...
I was disappointed by the list of business models. There were no back of the envelope calculations to suggest whether a musician could earn his living, or whether he would have to keep his day job.