The argument applies only with respect to licensing the DRM Decoder/Decrypter - they surely don't apply with respect to licensing the DRM Encoder. For example, if they licensed the ability for Real Networks' music store to sell FairPlay protected songs, then it is Real Networks that would have the contractual relationship with the music labels, and would need to provide some guarantee over the security of their system.
There are a couple of exceptions, eg. if you're running a Wi-Fi hotspot in a single place such as a coffee shop, or you're running a totally not-for-profit network, then you don't need a carrier licence. See the ACA website for the precise wording of the exceptions..
There's always a choice. Even the choice to postpone or delegate a choice.:)
I can't help but wonder if the South Australian government was to pass a law with a principle stating "software with few bugs should be preferred to software with lots of bugs", then the Initiative for Software Choice would still get irate. They might say the government should not take a stance on the type of software - software with few bugs should be given equal weighting to software with lots of bugs.
To say there is "both types of software" is clever. The nature of the process will always create two types of software: those that meet the requirements (set by the government) and those that don't.
If the ISC took its argument to the logical conclusion, then the government should accept without disadvantage any software that doesn't meet its requirements. Obviously this is going to help Microsoft's submission.:)
1) This is not a law. Its a code of practice, and no-where in the article does it say whether said code is mandatory.
Actually, it became news because it became a law. Or effectively, anyway. The ACA has adopted it, making it part of the conditions of having a telco license.
See the media release from the ACA that says "It requires carriers and carrier service providers to comply..."
It seems to be directed at the bulk SMS operators that pay for a feed into a carrier's messaging system. From a quick flick through the documents, it seems carriers will have to put compliance with this "code of practice" into their contracts with the bulk SMS operators. Carriers that don't will be in breach of their licence.
The cost of SMS is getting lower. According to this page, you can send bulk SMS at 18c / msg (that's only US 12c) in Australia right now.
One thing you neglect to mention is power limits imposed by regulators. At least in Australia, the maximum power level in the 5GHz range is lower than the level in the 2.4GHz range (half?).
Inverse square law.. halving power levels severely reduces the effective range of the transmitters. Also, more power is needed in order to get the same effective range at higher frequencies / bandwidths. Hence the rule of thumb that you'll need about four.11a access points for every one.11b access point.
It's Freedom vs Power when it's Developer vs User
on
Freedom or Power?
·
· Score: 1
Taking this argument from the perspective of a developer of a new project, freedom and power are a bit blurred. My freedom to choose a licence is due to the power provided by copyright laws, etc.
However, from the perspective of a user, this is much clearer cut. If I've just received the source code of a project distributed under a "free" licence, then I must abide by the conditions of that licence when I make further copies or product derivative works.
In this case, my freedom to do what I want with the code is restricted.. I don't have the right to decide on which licence I distribute the work under. The developer has used their power over me to restrict my freedom to choose a licence.
But imagine what would happen if I did have the freedom to choose a licence.. I could take software under a "free" licence and distribute it under a "closed" licence. Obviously, this is not what the original developer wanted when they released their code. The developer has consciously used their power, provided by the copyright laws, to restrict my freedom.
Obviously wireless and wireline are different environments, with totally different technical constraints, business models, and usage patterns.
Just a couple of things... you have to buy spectrum before you launch your wireless service, you have to deploy cells, you have to put together roaming agreements, you have to provide smooth hand-over between cells for moving customers, you have a (small) finite number of simultaneous customers on the one cell (compared with something like fibre), you have conflicting quality of service: voice vs. packet, you have to manage phone numbers (nothing like dynamically allocated IP addresses), weather patterns affect service quality, different countries have non-interoperable standards (making efficiencies of scale harder to reach), etc. etc.
iMode is not the only successful mobile internet service in Japan (contrary to what you might have read in the media). There are two competitors to it doing reasonably well, and one of them is WAP.
Anyway, the "whole new protocol stack" is basically restricted to the mobile network, so from the point of view of content developers it doesn't exist. They use the same web servers as always, just setting up a new MIME type and putting up content in the WAP format. Similar to supporting PNG, Flash, or any other new web content type.
Security is not an absolute - it's a sliding scale defined in terms of effort required to break it. The level of security provided by WAP is not as high as 128-bit secure browsing using Internet Explorer, but that isn't a roadblock. I'm sure you go to sites (such as this one) without an "https:" in the URL, surf with cookies turned on, enter private information at a site or two. The security in WAP is adequate for the majority of applications.
iMode is proprietary, controled by a single corporation, and appears to suffer from scalability problems. Different iMode handsets support different parts of the total set of capabilities, making it just as diverse as the WAP world. Most of the material on programming iMode is written in Japanese, limiting the development community to Japan. It is not "what mobile internet should have been", it was just a kludge to get mobile internet up quickly in Japan.
Firstly, M-Services (not M-Series, as mentioned in the submission) "could include enhanced graphics, music, video, games, ring tones, screen savers and other compelling services" (from the link). Note this is nothing to do with browsing content on the Internet.
Secondly, another quote from the article: "M-Services will leverage other key standardisation efforts like WAP, EMS, MMS and SyncML to bring a consistent user experience for digital content." says Jan Wäreby, CEO, Ericsson Consumer Division
To make an (admittedly poor) analogy, whilst WAP is like Web browsing, M-Services will be like Flash animation.
And anyway, the "failure" of WAP was clearly not due to poor download speeds, inability to use phones for browsing, or problems with using new protocols in mobile phone networks, as had already been claimed in this Slashdot thread. All of these factors are present in Japan, and despite this there is the success of NTT DoCoMo's iMode service, J-Phone's J-sky, or the TU-KA EZweb with in total 37 million users (See the stats at http://www.tca.or.jp/index-e.html)
The WAP caching model is quite different to the web caching model. If multiple 'pages' share the same URL then the 'page' will be loaded from the device's cache rather than the web server. Servlets are one of the few forms of dynamic web content that insist on having a single URL (not counting the ?bla=bla stuff that is tacked on the end).
The solution is to ensure that your 'pages' (I'm quoting this since the correct WAP term is 'decks') are not cached. There are several ways to do this. I suggest that you read the WAP FAQ at wap.colorline.no/wap-faq.
One possibility is that the 7110 crashes in the way that you describe when a WML deck is loaded containing a non-ASCII symbol. An example is the copyright symbol at the bottom of every Slashdot page. As long as you don't keep browsing until you reach the end of a page of Slashdot (it'd take forever using the Google translator anyway) then you should be safe from this problem.
It's stuff like this that results in slashdotters being called crackpots. Here, let's think about the positive aspects of sneaking in and watching movies for free, or shoplifting magazines - if you like the product, it creates awareness of it and makes you likely to buy more of it! Woo hoo!
Obviously you haven't been to a public library recently. It's amazing - you can walking in, walk away with a book, read it, then give it back to someone else can read it. And it's all free. Yet many of the books I've bought were because I borrowed them first from a library and liked them enough to own them.
You seem to claim that using libraries is akin to stealing. (I know this is a major paraphrasing of your view, but I'm making a point) If you used libraries more then you'd appreciate the community benefits of having free access to intellectual works. Libraries also rent out software and music for free these days too.
A simple test for whether piracy (a.k.a copyright infringement) really matters: how did we ever get along without copyright law before?
I suspect the situation for teaching many of the humanities over the net is nearly as bad. The most valuable aspects of these sorts of courses I took were learned in discussion with peers and professors, and in writing papers, and in careful analysis of my papers by others.
I am basically enrolled in a virtual university. It is actually Monash University in Melbourne, Australia. However, I listen to my lectures via a RealAudio stream. I converse with my peers via a UseNet-style newsgroup. Both of these lose something without human contact but gain things from association with a computer too.
While I listen to my lectures, I am connected to the Internet, so I can look up terms on Internet encylopedias and dictionaries and get some background info. Similarly with the newsgroups, but unlike a tutorial discussion, you can look back at weeks of discussion and follow several arguments concurrently. No-one interrupts me mid-thought either!
Admittedly, I do still go to campus every week, but only for a couple of hours, and that's not bad for a 2/3rds loading.
The point is that calendaring isn't like email - you can't operate 100 simulateous calendar accounts. The effort required to simply start up a new calendar account is quite large, since you need to be able to copy ALL of your current calendaring information across into it and keep it up-to-date.
Just about anyone who needs to use a piece of calendar software already does. It probably isn't open source software, and even if some wonder piece of open source software came along that did everything that the best closed piece of calendaring software did, there is the problem with additional calendar accounts that I just mentioned. Somehow you would have to be able to migrate all of your existing appointments, reminders, etc. to the new wonder open-source calendar.
To do this migration, you need to have protocols and formats to facilitate the exchange of calendar information between calendar software. Happily this is exactly what the iCal stuff does. What you are calling "impotent" is actually the very stuff that is required to fulfill your vision of scheduling. Just think where we'd be today with email if nothing like SMTP or MIME was ever standardised. iCal is the critical first step.
Firstly, you probably don't record the news or live sport and then play them later. And if you aren't recording it, then you've got to watch the ads. I don't know a real sports fan who would prefer to watch a delayed telecast than watch the thing live.
Secondly, you probably use the fact that ads are catered for in the scripting of shows, where ads are inserted in an appropriate place (well.. most of the time). This means that you can dash to the kitchen for a drink, do some study, or dash to the toilet. Instead of guessing when the best time to pause would be, I've seen people (and do it myself) play a recorded movie and use the ad breaks rather than skipping them.
Advertisements won't go away. Free to air TV won't die. Certain advertisement slots may get more expensive though.
Although.. infomercials may die - fingers crossed!
Copyright does not need to be abolished, since in practice it already has. When someone can set up a business on the Internet, based in some foreign country without copyright protections, and sell content that is copied from others, then what good is copyright?
The big question is that now that we've lost copyright protection in many ways, how can business continue to work? Although this is a brazen plug, I've written an article on just this topic, and it's quite thorough in its investigation.
Studies of newsgroup activity show that often users can be classified into two groups: those who post stuff, and those who do not. The number of those who do not post are much larger than those who do. This is what the PHOAKS research concluded. For example, a typical distribution on Usenet is around 95% readers only. Slashdot is probably around the same.
You are dead on saying that just using the set of people who post things to rate the articles is not giving a good reflection of the overall readership. But readers don't want to submit anything, since they're readers... The solution is passive rating.
Systems like GroupLens, PHOAKS, and SiteSeer all are set up to extract ratings information without the reader having to be bothered to submit it themselves. For example, by the simple act of reading the post, they have implicitly stated that they thought it would be better than other posts. GroupLens found that the time that is spent reading a post correlates very well with the rating that would've been eventually given anyway.
The problem with any ratings system that allows people to choose the posts that they rate, is that it is in danger of influencing itself - any errors in rating are compounded. In the present slashdot system, if most moderators only read articles that were of a particular score, then they would not be providing useful scores. Topics that are more newsworthy will naturally get better ratings, regardless of the value of the content.
A solution is to normalise the scores so that topics that only have a few ratings score better until more people come along to rate them. This way a good post in a non-newsworthy area will rise to prominance, with a high initial score, until perhaps more people read it and the score goes down.
Passive ratings and normalisation will mean that no explicit moderation is needed, and the scores will be more useful than the current moderation system.
One thing we might ask is why Tim feels the need to start his column by suggesting that he and RMS are perhaps very similar. This comment is not followed up in (or even justified by) the rest of the column. My feel is that he hopes that he'll get more credibility by inferring similarity to a high profile figure in the debate.
It is unfortunate that Tim has fallen victim to the cult of personality that is muddying the waters in the sea of Open Source software. We shouldn't need to engage in the rhetoric of personalities to get our points across.
Granted, this is only a column. It's not as if he's preaching to the heathens rather than the converted, or he was trying to put together a tightly knit theory on software licensing. Anything he says must be taken in this context.
It's just a pity that he didn't think that his comments could stand on their own merits, and needed them prefixed by a RMS reference.
It's interesting that many see a requirement for companies operating with open source licences to 'give back' to the open source community. It is interesting in two ways..
Firstly, the open source community is seen as an entity that has given something to that company. The individual contributions of the expert programmers are not distinguished.
Secondly, it denies ESR's thesis that the reason that programmers do the work is to enhance their reputations. If work on a commercially-driven open source project enhances your reputation, then you've just gotten as much as you can expect to get.
So although the programmers have recieved something in return for their efforts, there's a perception that this is not enough, and that the community itself should benefit somehow. Why? Aren't the programmers who worked on the project happy with their lot?
Perhaps it is a case of the 'freeloaders' in the system (the non-participating programmers) seeing a possible angle to get more for themselves?
Slashdot Mirror
The argument applies only with respect to licensing the DRM Decoder/Decrypter - they surely don't apply with respect to licensing the DRM Encoder. For example, if they licensed the ability for Real Networks' music store to sell FairPlay protected songs, then it is Real Networks that would have the contractual relationship with the music labels, and would need to provide some guarantee over the security of their system.
According to the article itself, it *is* self-balancing. Here are some quotes:
There are a couple of exceptions, eg. if you're running a Wi-Fi hotspot in a single place such as a coffee shop, or you're running a totally not-for-profit network, then you don't need a carrier licence. See the ACA website for the precise wording of the exceptions..
There's always a choice. Even the choice to postpone or delegate a choice. :)
:)
I can't help but wonder if the South Australian government was to pass a law with a principle stating "software with few bugs should be preferred to software with lots of bugs", then the Initiative for Software Choice would still get irate. They might say the government should not take a stance on the type of software - software with few bugs should be given equal weighting to software with lots of bugs.
To say there is "both types of software" is clever. The nature of the process will always create two types of software: those that meet the requirements (set by the government) and those that don't.
If the ISC took its argument to the logical conclusion, then the government should accept without disadvantage any software that doesn't meet its requirements. Obviously this is going to help Microsoft's submission.
Andrew Scott
1) This is not a law. Its a code of practice, and no-where in the article does it say whether said code is mandatory.
Actually, it became news because it became a law. Or effectively, anyway. The ACA has adopted it, making it part of the conditions of having a telco license.
See the media release from the ACA that says "It requires carriers and carrier service providers to comply..."
It seems to be directed at the bulk SMS operators that pay for a feed into a carrier's messaging system. From a quick flick through the documents, it seems carriers will have to put compliance with this "code of practice" into their contracts with the bulk SMS operators. Carriers that don't will be in breach of their licence.
The cost of SMS is getting lower. According to this page, you can send bulk SMS at 18c / msg (that's only US 12c) in Australia right now.
Andrew Scott
One thing you neglect to mention is power limits imposed by regulators. At least in Australia, the maximum power level in the 5GHz range is lower than the level in the 2.4GHz range (half?).
.11a access points for every one .11b access point.
Inverse square law.. halving power levels severely reduces the effective range of the transmitters. Also, more power is needed in order to get the same effective range at higher frequencies / bandwidths. Hence the rule of thumb that you'll need about four
Taking this argument from the perspective of a developer of a new project, freedom and power are a bit blurred. My freedom to choose a licence is due to the power provided by copyright laws, etc.
However, from the perspective of a user, this is much clearer cut. If I've just received the source code of a project distributed under a "free" licence, then I must abide by the conditions of that licence when I make further copies or product derivative works.
In this case, my freedom to do what I want with the code is restricted.. I don't have the right to decide on which licence I distribute the work under. The developer has used their power over me to restrict my freedom to choose a licence.
But imagine what would happen if I did have the freedom to choose a licence.. I could take software under a "free" licence and distribute it under a "closed" licence. Obviously, this is not what the original developer wanted when they released their code. The developer has consciously used their power, provided by the copyright laws, to restrict my freedom.
Andrew Scott
Is this a troll?
Obviously wireless and wireline are different environments, with totally different technical constraints, business models, and usage patterns.
Just a couple of things... you have to buy spectrum before you launch your wireless service, you have to deploy cells, you have to put together roaming agreements, you have to provide smooth hand-over between cells for moving customers, you have a (small) finite number of simultaneous customers on the one cell (compared with something like fibre), you have conflicting quality of service: voice vs. packet, you have to manage phone numbers (nothing like dynamically allocated IP addresses), weather patterns affect service quality, different countries have non-interoperable standards (making efficiencies of scale harder to reach), etc. etc.
Andrew Scott
iMode is not the only successful mobile internet service in Japan (contrary to what you might have read in the media). There are two competitors to it doing reasonably well, and one of them is WAP.
Anyway, the "whole new protocol stack" is basically restricted to the mobile network, so from the point of view of content developers it doesn't exist. They use the same web servers as always, just setting up a new MIME type and putting up content in the WAP format. Similar to supporting PNG, Flash, or any other new web content type.
Security is not an absolute - it's a sliding scale defined in terms of effort required to break it. The level of security provided by WAP is not as high as 128-bit secure browsing using Internet Explorer, but that isn't a roadblock. I'm sure you go to sites (such as this one) without an "https:" in the URL, surf with cookies turned on, enter private information at a site or two. The security in WAP is adequate for the majority of applications.
iMode is proprietary, controled by a single corporation, and appears to suffer from scalability problems. Different iMode handsets support different parts of the total set of capabilities, making it just as diverse as the WAP world. Most of the material on programming iMode is written in Japanese, limiting the development community to Japan. It is not "what mobile internet should have been", it was just a kludge to get mobile internet up quickly in Japan.
Andrew Scott
http://www.gsmworld.com/news/press_2001/press_rele ases_24.html
Firstly, M-Services (not M-Series, as mentioned in the submission) "could include enhanced graphics, music, video, games, ring tones, screen savers and other compelling services" (from the link). Note this is nothing to do with browsing content on the Internet.
Secondly, another quote from the article: "M-Services will leverage other key standardisation efforts like WAP, EMS, MMS and SyncML to bring a consistent user experience for digital content." says Jan Wäreby, CEO, Ericsson Consumer Division
To make an (admittedly poor) analogy, whilst WAP is like Web browsing, M-Services will be like Flash animation.
And anyway, the "failure" of WAP was clearly not due to poor download speeds, inability to use phones for browsing, or problems with using new protocols in mobile phone networks, as had already been claimed in this Slashdot thread. All of these factors are present in Japan, and despite this there is the success of NTT DoCoMo's iMode service, J-Phone's J-sky, or the TU-KA EZweb with in total 37 million users (See the stats at http://www.tca.or.jp/index-e.html)
Andrew Scott
The solution is to ensure that your 'pages' (I'm quoting this since the correct WAP term is 'decks') are not cached. There are several ways to do this. I suggest that you read the WAP FAQ at wap.colorline.no/wap-faq.
One possibility is that the 7110 crashes in the way that you describe when a WML deck is loaded containing a non-ASCII symbol. An example is the copyright symbol at the bottom of every Slashdot page. As long as you don't keep browsing until you reach the end of a page of Slashdot (it'd take forever using the Google translator anyway) then you should be safe from this problem.
It's stuff like this that results in slashdotters being called crackpots. Here, let's think about the positive aspects of sneaking in and watching movies for free, or shoplifting magazines - if you like the product, it creates awareness of it and makes you likely to buy more of it! Woo hoo!
Obviously you haven't been to a public library recently. It's amazing - you can walking in, walk away with a book, read it, then give it back to someone else can read it. And it's all free. Yet many of the books I've bought were because I borrowed them first from a library and liked them enough to own them.
You seem to claim that using libraries is akin to stealing. (I know this is a major paraphrasing of your view, but I'm making a point) If you used libraries more then you'd appreciate the community benefits of having free access to intellectual works. Libraries also rent out software and music for free these days too.
A simple test for whether piracy (a.k.a copyright infringement) really matters: how did we ever get along without copyright law before?
Andrew Scott
I am basically enrolled in a virtual university. It is actually Monash University in Melbourne, Australia. However, I listen to my lectures via a RealAudio stream. I converse with my peers via a UseNet-style newsgroup. Both of these lose something without human contact but gain things from association with a computer too.
While I listen to my lectures, I am connected to the Internet, so I can look up terms on Internet encylopedias and dictionaries and get some background info. Similarly with the newsgroups, but unlike a tutorial discussion, you can look back at weeks of discussion and follow several arguments concurrently. No-one interrupts me mid-thought either!
Admittedly, I do still go to campus every week, but only for a couple of hours, and that's not bad for a 2/3rds loading.
Andrew Scott
The point is that calendaring isn't like email - you can't operate 100 simulateous calendar accounts. The effort required to simply start up a new calendar account is quite large, since you need to be able to copy ALL of your current calendaring information across into it and keep it up-to-date.
Just about anyone who needs to use a piece of calendar software already does. It probably isn't open source software, and even if some wonder piece of open source software came along that did everything that the best closed piece of calendaring software did, there is the problem with additional calendar accounts that I just mentioned. Somehow you would have to be able to migrate all of your existing appointments, reminders, etc. to the new wonder open-source calendar.
To do this migration, you need to have protocols and formats to facilitate the exchange of calendar information between calendar software. Happily this is exactly what the iCal stuff does. What you are calling "impotent" is actually the very stuff that is required to fulfill your vision of scheduling. Just think where we'd be today with email if nothing like SMTP or MIME was ever standardised. iCal is the critical first step.
Firstly, you probably don't record the news or live sport and then play them later. And if you aren't recording it, then you've got to watch the ads. I don't know a real sports fan who would prefer to watch a delayed telecast than watch the thing live.
Secondly, you probably use the fact that ads are catered for in the scripting of shows, where ads are inserted in an appropriate place (well.. most of the time). This means that you can dash to the kitchen for a drink, do some study, or dash to the toilet. Instead of guessing when the best time to pause would be, I've seen people (and do it myself) play a recorded movie and use the ad breaks rather than skipping them.
Advertisements won't go away. Free to air TV won't die. Certain advertisement slots may get more expensive though.
Although.. infomercials may die - fingers crossed!
Copyright does not need to be abolished, since in practice it already has. When someone can set up a business on the Internet, based in some foreign country without copyright protections, and sell content that is copied from others, then what good is copyright?
.com/SiliconValley/Way/1387/copyright.html
The big question is that now that we've lost copyright protection in many ways, how can business continue to work? Although this is a brazen plug, I've written an article on just this topic, and it's quite thorough in its investigation.
Have a look at http://www.geocities
It looks at MP3s, Open Source, Shareware, Project Gutenberg, and what the new rules of the game might be.
Studies of newsgroup activity show that often users can be classified into two groups: those who post stuff, and those who do not. The number of those who do not post are much larger than those who do. This is what the PHOAKS research concluded. For example, a typical distribution on Usenet is around 95% readers only. Slashdot is probably around the same.
You are dead on saying that just using the set of people who post things to rate the articles is not giving a good reflection of the overall readership. But readers don't want to submit anything, since they're readers... The solution is passive rating.
Systems like GroupLens, PHOAKS, and SiteSeer all are set up to extract ratings information without the reader having to be bothered to submit it themselves. For example, by the simple act of reading the post, they have implicitly stated that they thought it would be better than other posts. GroupLens found that the time that is spent reading a post correlates very well with the rating that would've been eventually given anyway.
The problem with any ratings system that allows people to choose the posts that they rate, is that it is in danger of influencing itself - any errors in rating are compounded. In the present slashdot system, if most moderators only read articles that were of a particular score, then they would not be providing useful scores. Topics that are more newsworthy will naturally get better ratings, regardless of the value of the content.
A solution is to normalise the scores so that topics that only have a few ratings score better until more people come along to rate them. This way a good post in a non-newsworthy area will rise to prominance, with a high initial score, until perhaps more people read it and the score goes down.
Passive ratings and normalisation will mean that no explicit moderation is needed, and the scores will be more useful than the current moderation system.
One thing we might ask is why Tim feels the need to start his column by suggesting that he and RMS are perhaps very similar. This comment is not followed up in (or even justified by) the rest of the column. My feel is that he hopes that he'll get more credibility by inferring similarity to a high profile figure in the debate.
It is unfortunate that Tim has fallen victim to the cult of personality that is muddying the waters in the sea of Open Source software. We shouldn't need to engage in the rhetoric of personalities to get our points across.
Granted, this is only a column. It's not as if he's preaching to the heathens rather than the converted, or he was trying to put together a tightly knit theory on software licensing. Anything he says must be taken in this context.
It's just a pity that he didn't think that his comments could stand on their own merits, and needed them prefixed by a RMS reference.
It's interesting that many see a requirement for companies operating with open source licences to 'give back' to the open source community. It is interesting in two ways..
Firstly, the open source community is seen as an entity that has given something to that company. The individual contributions of the expert programmers are not distinguished.
Secondly, it denies ESR's thesis that the reason that programmers do the work is to enhance their reputations. If work on a commercially-driven open source project enhances your reputation, then you've just gotten as much as you can expect to get.
So although the programmers have recieved something in return for their efforts, there's a perception that this is not enough, and that the community itself should benefit somehow. Why? Aren't the programmers who worked on the project happy with their lot?
Perhaps it is a case of the 'freeloaders' in the system (the non-participating programmers) seeing a possible angle to get more for themselves?