I SSH'd into a friend's server and wrote out rm -rf /... just to be funny... I didn't hit enter of course
My cat has the stupid tendency to suddenly jump onto keyboards, often where the enter key is located. You are must be happy not to have a cat like that.
I would not completely rule out the possibility that there may exist some persons on this planet who may be thinking about shutting down the Internet anywhere
The first sound chip I remember was the C64 SID, I still love the PC speaker, I have not forgotten the Adlib, the first Creative card I remember is the Sound Blaster 8-bit, and you call X-Fi old??:)
Note, however, that being anti-luser doesn't mean being anti-newbie. A newbie can be welcome if they are not stupid (ie not lusers), want to learn in the right way, read the docs, and are willing to become powerusers. Lusers, however, are intellectually incapable of even understanding why we call them lusers, and so they never become one of us, therefore no poweruser can accept being encircled by lusers in software community, but newbies that really want and are capable to learn in the right way (ie to edit textfiles etc) are of course welcome.
If you have to edit a text file, your software is not ready for (l)users.
Who wants lusers using the same OS as you? One of the reasons I use GNU/Linux (Debian) is precisely because the user communities are free of lusers, so that I know that whenever I post a message to a mailing list I will get answers from fellow power users.
Lusers tend to infect a software project with their stupidity and naivety. They tend to click on any link they see in their emails, so virus writters target whatever OS the lusers use most. The developers of a piece of software also tend to make their software more suitable for stupid users because they tend to think that accomodating more users is a good thing, thus driving power users away. Unfortunately this currently happens with some GNU/Linux distros. You just have to see that many newer GNU/Linux software projects only work with X and have no command line support, and many websites don't work with text browsers anymore.
Whatever software we use is not only determined by technical merit but also by social factors. We want to use software which is different from anyone else, particularly the lusers and the closed source world. If our OS requires interaction with a command line and editing obscure text files, then we can know for sure that we will never have to deal with a luser in our support mailing lists, etc.
Thus, user-unfriendliness is a filter that we can use intentionally to keep non-powerusers away from our communities. If GNU/Linux ever becomes the preferred OS of lusers I am going to switch to OpenBSD, and if that too gets infected by lusers I will write my own.
And yet, with all the powerful CPUs we have now, CBM still surpasses us in one regard: In the 1980s computers like CBM were sold with a programming language by default, so any new computer owner could (and often had) to learn programming, but today in the 2000s computers are sold with spyware and adware without any mention of programmability, so new computers owners now tend to see the computer as something working automagically like a TV. CBM encourages you to learn programming, today's PCs don't even mention the possibility of programming in the user manuals. Just open a user manual for a new laptop and you see stupid stuff like "the on/off switch is here" and "click start to get going". Most computer owners in this age have absolutely no idea that their computer can do more than downloading p0rn.
Even with GPL there is a big problem with versions and support for later versions. Too many people put code online and say it's GPL, without clarifying whether it's GPLv1, 2, or 3, and whether it is with any later versions or without.:(
It would be great if we could have a licence with an unambiguous name and no versions, so that when we see some code claiming to be under FooPL we can be more assured about it.
"If I write a GFDL article on my wiki and then I post it on Wikipedia under the GFDL as well, everything is fine: I can share content between the two articles taking care to properly update the history and follow the GFDL, so I continue to benefit from modifications to the Wikipedia copy of my article."
Are you quite sure of that? So sure you'd bet a lawsuit on it?
I think that a cautious person should be sure about nothing, of course. Assuming that you mean that GFDL may be untested or too complex/unclear in some respects, and talking hypothetically about the worst that could ever happen, I see absolutely no reason that anyone should feel safe even if using a well-tested and clear-cut licence.
A licence is a piece of paper, and its contents can become life-threatening only when presented in front of an impartial judge. But how could one assume that a case is every going to be presented in front of a judge at all, or that the judge will be impartial? (in some countries impartial judges are a rare luxury).
Chances are that if someone is very angry with someone for any reason, or even no reason, they will probably find a way to do what they want, especially in countries where there is no well-developed legal system and rampant corruption: in some countries judges are too happy to accept bribes or influence by politically well-connected litigants, or perhaps in some countries some people may be used to seeking remedies with violent means outside the courts (I don't speak about any specific country). Even in countries will well-developed justice systems, one can never know what may happen, so just considering that all potential threats will come through the legal system is an incorrect assumption.
So, if you ask me, I personally believe that it is far more important not to give anyone's any reason for becoming angry in the first place, because one can never know how determined or how crazy an opponent can be and what means they have at their disposal. I think one should always consider that the worst is possible.
The best strategy is to act in a way that is not likely to cause any rational potential opponent to require a legal opinion on a licence. This means that one should act within socially acceptable cultural norms that are likely to be acceptable by rational people. Generally, people posting GFDL material probably mostly care about making their work available to the world and not letting people to claim exclusivity on their work, plus about getting proper credit. If these conditions are met, rational people will probably not get angry. With these conditions met, it remains to deal with the irrational potential opponents, who cannot be managed in any way other than increasing one's security anyway (including having a good lawyer ready for various potential threats, plus a good shielded door).
So, in short, my philosophy is: never assume a document is going to save a case, don't make other people feel angry in the first place, and keep any security systems well-working and ready for defence. But the most important than all is not to let other people feel willing to do something bad, as much as possible.
Of course, more important than that is not to get more paranoid than needed:) The probability of a real threat deriving from misunderstandings over a licence designed to give a gift of knowledge to humanity is maybe very low. This doesn't mean that it can't happen to find a crazy person willing to launch a lawsuit or other nasty things over technicalities. It just has a very low probability of happening (of course the probability is culture-dependent, and some in some countries people are more likely to go to courts than in other countries, so for these countries one should place some more attention to licences).
I have always wondered why mobile phones have some empty plastic on the back of the battery pack rather than solar cells. There must surely be some better way to use all that space than leave it as empty plastic.
Wikipedia says that Nupedia started in 2000, while Gnupedia was proposed by RMS in 1999 and started in 2001. It appears RMS had the idea first and had already publicly proposed it when Nupedia started.
about flyers, last time I went to a GNU/Linux convention distro people were giving out flyers, and it contained a notice "this text under GFDL" etc. Just a notice, not the whole GFDL text. People use GFDL because it comes from the trusted FSF people, but they realise GFDL is not good for flyers, so what they do is just to "customise" GFDL to their needs. After all we are all a community, what a law says doesn't mean much if we trust each other and we know that no one from our community is going to sue one of us. of course this is a false sense of security, because you never know what may happen...
GFDL 1.3 says that this relicensing is ok only under certain conditions. One condition is that the site holding the previously GFDL-licensed content must have been a massive collaboration site (MMCS or how they call it), and they specifically say that this includes public wikis. But there is a big problem here: they say that this site is defined as being on a WWW server.
I see no reason why a wiki or any other massive collaboration site should be on a WWW server. And exactly what is WWW? Is it TCP port 80? What if my WWW server is on 8080? Is WWW any server adhering to HTTP? What if I have made private patches to my WWW server so that it uses something which is based on HTTP but different from it and I use it to massively collaborate with other people running the same patches? Couldn't I implement a massive collaboration site using FTP, or some BBS software?
I see no reason why we must be so WWW-centric when talking about the Internet. The Internet is more than the WWW. The Internet is not even specifically TCP/IP: I could wrap another protocol in TCP/IP, or I could make a proxy to another network which is not TCP/IP at all, or I could make patches to my TCP/IP stack so that it becomes very different from TCP/IP and persuade other people use my patches as well.
And when we talk about WWW, do we talk about WWW servers accessible via ICANN's DNS network, via OpenNIC's DNS root, or via IP address? And should a site use HTML or XHTML in order to be considered WWW? Couldn't I serve non-HTML content over HTTP on port 80? Couldn't I implement a public wiki using good old SGML or maybe my own XML variant over HTTP to be used with a custom browser? (in fact now Wikipedia has enabled MediaWiki's write API, I think, so you can play with it using your own tools I think)
WWW-centriness is a bad idea in general. Never assume that if something is on the Internet, then it is on the WWW. Someone could devise a wiki running over email, over BitTorrent, over svn, over anything, maybe even over Gopher or a distributed filesystem (in fact there is wikipediafs in which you can mount Wikipedia or other MediaWiki sites in your filesystem, but I think it's unmaintainable now however).
Of course I know the discussion is more theoretical than practical, because in practice most if not all wikis are on the WWW and it was just Wikipedia and maybe a few other wikis that wanted to switch, and the relicensing provision was made just for them not for any GFDL content. But I still don't like seeing WWW-centriness. In the early days WWW was just a toy and most Internet work was done with other services, not WWW, and there is absolutely no reason to regard WWW as more important than other services.
If I write a GFDL article on my wiki and then I post it on Wikipedia under the GFDL as well, everything is fine: I can share content between the two articles taking care to properly update the history and follow the GFDL, so I continue to benefit from modifications to the Wikipedia copy of my article.
But if suddenly the people who run Wikipedia manage to get FSF do their thing and allow relicensing to CC-BY-SA-3.0 then I am out of luck: Wikipedians will continue making improvements to my article, but they will be now licensed under the CC-BY-SA-3.0 licence, which is incompatible with GFDL. Thus, from the moment Wikipedia switches licences, I will be unable to further benefit from improvements to the Wikipedia copy of my article, and I will not be able to copy the improvements to my copy in my wiki if my wiki is under the GFDL, or if my copy contains substantial contributions by others from my wiki that are under GFDL but were not shared on Wikipedia prior to the relicensing.
A great motivation for using GPL and GFDL is that I am capable of getting back other people's improvements to my code or my articles. If this were not possible, I would use the BSD licence or public domain, or if I were an evil misanthrope I would use standard copyright. But the concept of copyleft as implemented in GFDL and other GNU licences is a great motivation to use these licences so that I can benefit from improvements upon my work by combining these improvements to my copy of the work.
I want to be able to get other people's modifications because I am afraid that if I release my work under the public domain or BSDL then someone who is an evil misanthrope may build upon my work and make improvements to it but not let me have a look to these improvements so that I can combine them with my original work to improve my copy of my work as well (so, imagine writing an article in the public domain, someone translating it to another language but making the translation copyrighted, you won't be able to put the translation on your site, and if you attempt to make a translation yourself you may get sued for copyright infringement for re-translating your own words!).
When I choose an FSF GNU licence for licensing my code or my work, I expect to be able to get other people's modifications under, at the very least, the same licence as I gave my work to the people, because most probably I keep my work under that licence on my wiki/website.
So, now if Wikipedia switches licences, I will lose the ability to transfer content between my GFDL copy and Wikipedia's CC-By-SA-3.0 copy. That would not be an issue if CC-By-SA and GFDL were explicitly compatible. But they are, in fact, excplicitly incompatible (except that GFDL 1.3 no has a short-time window until August 2009 for relicensing to CC-By-SA 3.0 under certain conditions and only for wikis and other MMCSes).
Plus, I see no reason why FSF should accomodate a GFDL user just because they are big. If a person who has written two articles asks FSF to make a future GFDL version permitting relicensing, FSF will probably say no, but now we see a legal person that has actually not written anything but merely hosts content written by others to be able to get FSF do them a favour. This makes me unwilling to allow "or any later versions" when I license under the GPL or GFDL, because I don't know who is going to influence FSF in the future.
However, seeing the matter on a more pragmatic level, I have to agree that the GFDL is a very difficult licence for a wiki, and CC-By-SA 3.0 is much easier to use. Plus, if the relicensing goes forward, I will be partly very happy because I will be able to share Wikipedia articles without including a huge GFDL document plus an extremely huge history everywhere, and there will not be the question of who the most important five authors are, anymore. But if Wikipedia switches licences, I will also be sad because this will make it difficult to keep a local GFDL copy of my work and a Wikipedia CC-By-SA-3.0 copy synchronised. In fact, i
I second that, it's the best way to break monopolies. If no one is offering good service, wake up and start offering the service yourself.
People in communities that do not enjoy fast service can get organised and start a new community ISP. If 100 people get organised in the effort and you find a few richer members of the community who would be willing to sponsor it, then it will be easy to collect the monies needed to get some good business-class connectivity from a backbone ISP with the rights to resell access. After you have the backbone pipe and the resell rights you just start offering the service to the community in your own terms, not the terms of the monopoly players. This is called free market.
$1/GB may seem excessive for fixed broadband, but in mobile broadband prices at 5 EUR per megabyte (MB) are not unheard of (don't know about the US, I'm talking about here in southeastern EU)
Incorrect, business people can have extremely high-volume requirements. Try backing up a server located in the antipodes and you will see how much data you are going to use.
absolute consumer power exists only in a free market, and no place on this planet is a completely free market, so in most places consumer power is limited. If you need service and there is only one provider, you must become their customer (or find a way to not need service).
Many telecommuting employees get their employers to pay for telecom charges, and most telecommuting contractors charge for expenses. Caps can be a terrible problem, though, for the self-employed, because there is nobody else to pay for your traffic if you work alone.
8mbps maximum in Pittsburg, PA? Isn't the Pittsburg a big city? It's the second largest city in PA with more than 2 millions, if I believe Wikipedia. Are you really saying that the maximum DSL speed you can get there is only 8mbps?
In Athens, Greece, EU the maximum you get for DSL is 24mbps. Of course it can be lower depending on line conditions, but the maximum is 24 and a good number of connections do achieve it. This is the maximum of ADSL2+ actually, which is common in many countries. Are you sure there is no such offering in Pittsburg?
Really surprised, I have heard US was behind on broadband, but I thought it was more about ranches in the middle of nowhere having nothing better than dialup or satellite. I didn't know big cities like Pittsburg could be left at 8mbps in 2008 (approaching 2009). Or maybe you are more commonly using cable and DSL is not selling so well there?
3 Mb/s service
I can understand such speeds in the middle of Pacific Ocean, but I find it unacceptable that 3mbps DSL is being sold in big US cities. You must get your government help with the situation and create the right competitive market environment for fast service to thrive.
There is no invention that I have hated more than clamshell packaging. It's stupid and ugly. So much I hate it that I would certainly consider paying more to buy products not placed in clamshell packaging.
Will the users have the option to choose between paying for the extra bandwidth or cutting off or slowing the data link when the cap is exceeded?
Some users prefer to know that in no event they are going to pay more and would prefer a dead datalink to an unexpected bill.
Actually many of these users prefer this because they don't trust the ISPs to do correct billing: there are many stories of companies around the world, both private and state-owned, that send massive bills at random just to collect more revenue whenever their stock price goes down and want to show better results for the next quarter investor's report.
This thing happens regularly around the world with water utilities, power utilities, telephone operators, mobile phone companies, and Internet providers, primarily in countries where political corruption is high and the law doesn't work.
I don't know whether such things happen in the US, but in other countries it is as regular as rain in the winter and many users specifically try to find fixed-price plans in order not to let providers do this to them.
So, if a company wants to attract those users who are cautious, then it should offer an option to either switch off the datalink until the next month or slow it down to 64 or 128kbps when the cap is exceeded, until again next month (or other billing period).
AMD was brave enough to quit using FSBs in PC CPUs and replaced them with HyperTransport. Years later, Intel also says goodbye to FSBs and uses a similar technology. The innovator took all the costs, and now someone with more resources gets the marketshare. After all, the consumers only want a speedy CPU, they don't care who was the innovator, and speedy CPUs are more readily available by whoever has the most resources to build them. It is, therefore, seen that being the innovator is not always a smart movement in the business chessboard, at least not if you cannot build your innovation in sufficient quantity. That said, I congratulate Intel for finally bringing the cores closer to the RAM, which is a much better technical solution than using an FSB. They should, perhaps, have done that much earlier.
Slashdot Mirror
I SSH'd into a friend's server and wrote out rm -rf / ... just to be funny ... I didn't hit enter of course
My cat has the stupid tendency to suddenly jump onto keyboards, often where the enter key is located. You are must be happy not to have a cat like that.
unless we want to shutdown the Internet
I would not completely rule out the possibility that there may exist some persons on this planet who may be thinking about shutting down the Internet anywhere
The first sound chip I remember was the C64 SID, I still love the PC speaker, I have not forgotten the Adlib, the first Creative card I remember is the Sound Blaster 8-bit, and you call X-Fi old?? :)
Note, however, that being anti-luser doesn't mean being anti-newbie. A newbie can be welcome if they are not stupid (ie not lusers), want to learn in the right way, read the docs, and are willing to become powerusers. Lusers, however, are intellectually incapable of even understanding why we call them lusers, and so they never become one of us, therefore no poweruser can accept being encircled by lusers in software community, but newbies that really want and are capable to learn in the right way (ie to edit textfiles etc) are of course welcome.
If you have to edit a text file, your software is not ready for (l)users.
Who wants lusers using the same OS as you? One of the reasons I use GNU/Linux (Debian) is precisely because the user communities are free of lusers, so that I know that whenever I post a message to a mailing list I will get answers from fellow power users.
Lusers tend to infect a software project with their stupidity and naivety. They tend to click on any link they see in their emails, so virus writters target whatever OS the lusers use most. The developers of a piece of software also tend to make their software more suitable for stupid users because they tend to think that accomodating more users is a good thing, thus driving power users away. Unfortunately this currently happens with some GNU/Linux distros. You just have to see that many newer GNU/Linux software projects only work with X and have no command line support, and many websites don't work with text browsers anymore.
Whatever software we use is not only determined by technical merit but also by social factors. We want to use software which is different from anyone else, particularly the lusers and the closed source world. If our OS requires interaction with a command line and editing obscure text files, then we can know for sure that we will never have to deal with a luser in our support mailing lists, etc.
Thus, user-unfriendliness is a filter that we can use intentionally to keep non-powerusers away from our communities. If GNU/Linux ever becomes the preferred OS of lusers I am going to switch to OpenBSD, and if that too gets infected by lusers I will write my own.
Let's assume they contribute patches back and you are the project leader. Would you accept their patches?
And yet, with all the powerful CPUs we have now, CBM still surpasses us in one regard: In the 1980s computers like CBM were sold with a programming language by default, so any new computer owner could (and often had) to learn programming, but today in the 2000s computers are sold with spyware and adware without any mention of programmability, so new computers owners now tend to see the computer as something working automagically like a TV. CBM encourages you to learn programming, today's PCs don't even mention the possibility of programming in the user manuals. Just open a user manual for a new laptop and you see stupid stuff like "the on/off switch is here" and "click start to get going". Most computer owners in this age have absolutely no idea that their computer can do more than downloading p0rn.
Even with GPL there is a big problem with versions and support for later versions. Too many people put code online and say it's GPL, without clarifying whether it's GPLv1, 2, or 3, and whether it is with any later versions or without. :(
It would be great if we could have a licence with an unambiguous name and no versions, so that when we see some code claiming to be under FooPL we can be more assured about it.
"If I write a GFDL article on my wiki and then I post it on Wikipedia under the GFDL as well, everything is fine: I can share content between the two articles taking care to properly update the history and follow the GFDL, so I continue to benefit from modifications to the Wikipedia copy of my article." Are you quite sure of that? So sure you'd bet a lawsuit on it?
I think that a cautious person should be sure about nothing, of course. Assuming that you mean that GFDL may be untested or too complex/unclear in some respects, and talking hypothetically about the worst that could ever happen, I see absolutely no reason that anyone should feel safe even if using a well-tested and clear-cut licence.
A licence is a piece of paper, and its contents can become life-threatening only when presented in front of an impartial judge. But how could one assume that a case is every going to be presented in front of a judge at all, or that the judge will be impartial? (in some countries impartial judges are a rare luxury).
Chances are that if someone is very angry with someone for any reason, or even no reason, they will probably find a way to do what they want, especially in countries where there is no well-developed legal system and rampant corruption: in some countries judges are too happy to accept bribes or influence by politically well-connected litigants, or perhaps in some countries some people may be used to seeking remedies with violent means outside the courts (I don't speak about any specific country). Even in countries will well-developed justice systems, one can never know what may happen, so just considering that all potential threats will come through the legal system is an incorrect assumption.
So, if you ask me, I personally believe that it is far more important not to give anyone's any reason for becoming angry in the first place, because one can never know how determined or how crazy an opponent can be and what means they have at their disposal. I think one should always consider that the worst is possible.
The best strategy is to act in a way that is not likely to cause any rational potential opponent to require a legal opinion on a licence. This means that one should act within socially acceptable cultural norms that are likely to be acceptable by rational people. Generally, people posting GFDL material probably mostly care about making their work available to the world and not letting people to claim exclusivity on their work, plus about getting proper credit. If these conditions are met, rational people will probably not get angry. With these conditions met, it remains to deal with the irrational potential opponents, who cannot be managed in any way other than increasing one's security anyway (including having a good lawyer ready for various potential threats, plus a good shielded door).
So, in short, my philosophy is: never assume a document is going to save a case, don't make other people feel angry in the first place, and keep any security systems well-working and ready for defence. But the most important than all is not to let other people feel willing to do something bad, as much as possible.
Of course, more important than that is not to get more paranoid than needed :) The probability of a real threat deriving from misunderstandings over a licence designed to give a gift of knowledge to humanity is maybe very low. This doesn't mean that it can't happen to find a crazy person willing to launch a lawsuit or other nasty things over technicalities. It just has a very low probability of happening (of course the probability is culture-dependent, and some in some countries people are more likely to go to courts than in other countries, so for these countries one should place some more attention to licences).
I have always wondered why mobile phones have some empty plastic on the back of the battery pack rather than solar cells. There must surely be some better way to use all that space than leave it as empty plastic.
Wikipedia says that Nupedia started in 2000, while Gnupedia was proposed by RMS in 1999 and started in 2001. It appears RMS had the idea first and had already publicly proposed it when Nupedia started.
about flyers, last time I went to a GNU/Linux convention distro people were giving out flyers, and it contained a notice "this text under GFDL" etc. Just a notice, not the whole GFDL text. People use GFDL because it comes from the trusted FSF people, but they realise GFDL is not good for flyers, so what they do is just to "customise" GFDL to their needs. After all we are all a community, what a law says doesn't mean much if we trust each other and we know that no one from our community is going to sue one of us. of course this is a false sense of security, because you never know what may happen...
GFDL 1.3 says that this relicensing is ok only under certain conditions. One condition is that the site holding the previously GFDL-licensed content must have been a massive collaboration site (MMCS or how they call it), and they specifically say that this includes public wikis. But there is a big problem here: they say that this site is defined as being on a WWW server.
I see no reason why a wiki or any other massive collaboration site should be on a WWW server. And exactly what is WWW? Is it TCP port 80? What if my WWW server is on 8080? Is WWW any server adhering to HTTP? What if I have made private patches to my WWW server so that it uses something which is based on HTTP but different from it and I use it to massively collaborate with other people running the same patches? Couldn't I implement a massive collaboration site using FTP, or some BBS software?
I see no reason why we must be so WWW-centric when talking about the Internet. The Internet is more than the WWW. The Internet is not even specifically TCP/IP: I could wrap another protocol in TCP/IP, or I could make a proxy to another network which is not TCP/IP at all, or I could make patches to my TCP/IP stack so that it becomes very different from TCP/IP and persuade other people use my patches as well.
And when we talk about WWW, do we talk about WWW servers accessible via ICANN's DNS network, via OpenNIC's DNS root, or via IP address? And should a site use HTML or XHTML in order to be considered WWW? Couldn't I serve non-HTML content over HTTP on port 80? Couldn't I implement a public wiki using good old SGML or maybe my own XML variant over HTTP to be used with a custom browser? (in fact now Wikipedia has enabled MediaWiki's write API, I think, so you can play with it using your own tools I think)
WWW-centriness is a bad idea in general. Never assume that if something is on the Internet, then it is on the WWW. Someone could devise a wiki running over email, over BitTorrent, over svn, over anything, maybe even over Gopher or a distributed filesystem (in fact there is wikipediafs in which you can mount Wikipedia or other MediaWiki sites in your filesystem, but I think it's unmaintainable now however).
Of course I know the discussion is more theoretical than practical, because in practice most if not all wikis are on the WWW and it was just Wikipedia and maybe a few other wikis that wanted to switch, and the relicensing provision was made just for them not for any GFDL content. But I still don't like seeing WWW-centriness. In the early days WWW was just a toy and most Internet work was done with other services, not WWW, and there is absolutely no reason to regard WWW as more important than other services.
If I write a GFDL article on my wiki and then I post it on Wikipedia under the GFDL as well, everything is fine: I can share content between the two articles taking care to properly update the history and follow the GFDL, so I continue to benefit from modifications to the Wikipedia copy of my article.
But if suddenly the people who run Wikipedia manage to get FSF do their thing and allow relicensing to CC-BY-SA-3.0 then I am out of luck: Wikipedians will continue making improvements to my article, but they will be now licensed under the CC-BY-SA-3.0 licence, which is incompatible with GFDL. Thus, from the moment Wikipedia switches licences, I will be unable to further benefit from improvements to the Wikipedia copy of my article, and I will not be able to copy the improvements to my copy in my wiki if my wiki is under the GFDL, or if my copy contains substantial contributions by others from my wiki that are under GFDL but were not shared on Wikipedia prior to the relicensing.
A great motivation for using GPL and GFDL is that I am capable of getting back other people's improvements to my code or my articles. If this were not possible, I would use the BSD licence or public domain, or if I were an evil misanthrope I would use standard copyright. But the concept of copyleft as implemented in GFDL and other GNU licences is a great motivation to use these licences so that I can benefit from improvements upon my work by combining these improvements to my copy of the work.
I want to be able to get other people's modifications because I am afraid that if I release my work under the public domain or BSDL then someone who is an evil misanthrope may build upon my work and make improvements to it but not let me have a look to these improvements so that I can combine them with my original work to improve my copy of my work as well (so, imagine writing an article in the public domain, someone translating it to another language but making the translation copyrighted, you won't be able to put the translation on your site, and if you attempt to make a translation yourself you may get sued for copyright infringement for re-translating your own words!).
When I choose an FSF GNU licence for licensing my code or my work, I expect to be able to get other people's modifications under, at the very least, the same licence as I gave my work to the people, because most probably I keep my work under that licence on my wiki/website.
So, now if Wikipedia switches licences, I will lose the ability to transfer content between my GFDL copy and Wikipedia's CC-By-SA-3.0 copy. That would not be an issue if CC-By-SA and GFDL were explicitly compatible. But they are, in fact, excplicitly incompatible (except that GFDL 1.3 no has a short-time window until August 2009 for relicensing to CC-By-SA 3.0 under certain conditions and only for wikis and other MMCSes).
Plus, I see no reason why FSF should accomodate a GFDL user just because they are big. If a person who has written two articles asks FSF to make a future GFDL version permitting relicensing, FSF will probably say no, but now we see a legal person that has actually not written anything but merely hosts content written by others to be able to get FSF do them a favour. This makes me unwilling to allow "or any later versions" when I license under the GPL or GFDL, because I don't know who is going to influence FSF in the future.
However, seeing the matter on a more pragmatic level, I have to agree that the GFDL is a very difficult licence for a wiki, and CC-By-SA 3.0 is much easier to use. Plus, if the relicensing goes forward, I will be partly very happy because I will be able to share Wikipedia articles without including a huge GFDL document plus an extremely huge history everywhere, and there will not be the question of who the most important five authors are, anymore. But if Wikipedia switches licences, I will also be sad because this will make it difficult to keep a local GFDL copy of my work and a Wikipedia CC-By-SA-3.0 copy synchronised. In fact, i
what you can do is to build your own networks
I second that, it's the best way to break monopolies. If no one is offering good service, wake up and start offering the service yourself.
People in communities that do not enjoy fast service can get organised and start a new community ISP. If 100 people get organised in the effort and you find a few richer members of the community who would be willing to sponsor it, then it will be easy to collect the monies needed to get some good business-class connectivity from a backbone ISP with the rights to resell access. After you have the backbone pipe and the resell rights you just start offering the service to the community in your own terms, not the terms of the monopoly players. This is called free market.
$1/GB may seem excessive for fixed broadband, but in mobile broadband prices at 5 EUR per megabyte (MB) are not unheard of (don't know about the US, I'm talking about here in southeastern EU)
Many plans also count traffic in both directions toward your quota
I have never heard of a plan that does not count both uploads and downloads. Is there a specific example of a plan that excludes uploads?
Incorrect, business people can have extremely high-volume requirements. Try backing up a server located in the antipodes and you will see how much data you are going to use.
consumer power
absolute consumer power exists only in a free market, and no place on this planet is a completely free market, so in most places consumer power is limited. If you need service and there is only one provider, you must become their customer (or find a way to not need service).
Many telecommuting employees get their employers to pay for telecom charges, and most telecommuting contractors charge for expenses. Caps can be a terrible problem, though, for the self-employed, because there is nobody else to pay for your traffic if you work alone.
"unlimited" data plan (i.e. the one all iPhone uses have), has a 5GB/month cap
In Greece, EU unlimited data plans are either truly unlimited (if business account etc) or have a cap at 30GB.
the maximum possible speeds(up to 8 Mb/s
8mbps maximum in Pittsburg, PA? Isn't the Pittsburg a big city? It's the second largest city in PA with more than 2 millions, if I believe Wikipedia. Are you really saying that the maximum DSL speed you can get there is only 8mbps?
In Athens, Greece, EU the maximum you get for DSL is 24mbps. Of course it can be lower depending on line conditions, but the maximum is 24 and a good number of connections do achieve it. This is the maximum of ADSL2+ actually, which is common in many countries. Are you sure there is no such offering in Pittsburg?
Really surprised, I have heard US was behind on broadband, but I thought it was more about ranches in the middle of nowhere having nothing better than dialup or satellite. I didn't know big cities like Pittsburg could be left at 8mbps in 2008 (approaching 2009). Or maybe you are more commonly using cable and DSL is not selling so well there?
3 Mb/s service
I can understand such speeds in the middle of Pacific Ocean, but I find it unacceptable that 3mbps DSL is being sold in big US cities. You must get your government help with the situation and create the right competitive market environment for fast service to thrive.
There is no invention that I have hated more than clamshell packaging. It's stupid and ugly. So much I hate it that I would certainly consider paying more to buy products not placed in clamshell packaging.
Will the users have the option to choose between paying for the extra bandwidth or cutting off or slowing the data link when the cap is exceeded?
Some users prefer to know that in no event they are going to pay more and would prefer a dead datalink to an unexpected bill.
Actually many of these users prefer this because they don't trust the ISPs to do correct billing: there are many stories of companies around the world, both private and state-owned, that send massive bills at random just to collect more revenue whenever their stock price goes down and want to show better results for the next quarter investor's report.
This thing happens regularly around the world with water utilities, power utilities, telephone operators, mobile phone companies, and Internet providers, primarily in countries where political corruption is high and the law doesn't work.
I don't know whether such things happen in the US, but in other countries it is as regular as rain in the winter and many users specifically try to find fixed-price plans in order not to let providers do this to them.
So, if a company wants to attract those users who are cautious, then it should offer an option to either switch off the datalink until the next month or slow it down to 64 or 128kbps when the cap is exceeded, until again next month (or other billing period).
AMD was brave enough to quit using FSBs in PC CPUs and replaced them with HyperTransport. Years later, Intel also says goodbye to FSBs and uses a similar technology. The innovator took all the costs, and now someone with more resources gets the marketshare. After all, the consumers only want a speedy CPU, they don't care who was the innovator, and speedy CPUs are more readily available by whoever has the most resources to build them. It is, therefore, seen that being the innovator is not always a smart movement in the business chessboard, at least not if you cannot build your innovation in sufficient quantity. That said, I congratulate Intel for finally bringing the cores closer to the RAM, which is a much better technical solution than using an FSB. They should, perhaps, have done that much earlier.