reading of feynman! this is really what he believed, at least later in his life. i should say this: he had very little feeling of social responsibility while at the same time having a great sense of scientific responsibility. there is a great quote even to this effect which i will try to remember now:
what you choose to do in your personal life--whether you cheat on your wife--that is between you and your rabbi, but when it comes to science you'd better be honest!
something like that anyway. if someone else has "Adventures of a curious character" on hand please feel free to correct me, thanx.
But there are still things that you cannot do on the net and will never be able to do on the net, such as grocery shopping, having a sincere discussion with a friend or loved one, and much more. Yet, like this AOL/TW deal, businesses think that it CAN replace all that.
actually, i'm glad that TWX and AOL are merging precicely so that i don't have to have to do this kind of crap that you describe. if my loved ones really cared to have a "sincere discussion" they know which channels to find me on!
However, I was later offended by the author's apparent lack of patience. His comment about "a man...condemned forever to articulate his thoughts at the speed of an imbecile" made me wince. Here he is, one of a privileged few journalists with the opportunity to spend an afternoon with the greatest mind of the last 50 years, and he is focussing on the man's physical disabilities. I nearly stopped reading at that point.
--end--
i also winced reading this, but it was immediately recognizable to me that what the author said was not disparaging. clearly what's implicit is that Hawkings is not an imbecile though he 'talks' very slowly.
anyway, this was really a great article. i've enjoyed reading something of Hawking, but this really put a human face on the man for me. it seems that our author has accomplished his goal.
while i agree wholeheartedly with the Ed's post, i can't imagine that the patent office would ever say "no more software patents". but what about this:
it seems software patents could be made to expire after a much shortened period of time, say 5 years.
patents usually expire 17 or 20 years after they are issued, or much longer (29 years, i think, though i can't find a reference right now) for drugs. however even 17 years is quite a long time to exclude competition in the software industry. i would argue that 5 years or so would be a much more reasonable ammount of time for anyone to have a software patent.
5 years would give any company a reasonable ammount of time to recoup their R&D costs, and would also have the effect of discouraging companies from hoarding software patents in hopes that someone eventually steps on them.
what is it? a rant? you call this a rant? do you call it a rant because you don't know any other nouns to use to describe something you don't agree with? you are so silly.
translation: the author didn't use the sophisticated sorts of arguments that we GPL'ers like.
it's true--this guy is certainly not steeped in the BSD-GPL debate, or else he would have taken care to at least try to make a much more precise argument. _still_, you don't have to make the sort of complicated analysis that/.ers like to make in order to come to an honest and reasonable conclusion as this guy has.
nevermind, the real GPL-BSD debate won't be decided in slashdot threads. it will be decided by people who have good and bad experiences with GPL and with BSD. right now GPL has an advantage because linux is more popular than BSD, and because torvalds has chosen GPL. but the sort of license that we choose is a question that will remain longer than any one program and any one guy.
i am undoing my moderation in this thread in order to respond here, so please listen.
let me preface this by saying that of all of the arguments for GPL that i've read, this is one of the best. it is the quintessential pro-GPL argument. but, the reasoning is selective. i would like to answer two of your points.
first:
once a program is written, to encrypt it by compilation is to destroy some of its value. you would argue that it destroys all of the program's value, but this is not true.
what is not addressed is whether the program will be written in the first place--specifically what will motivate a programmer to write a useful program. money is certainly a more universal motivator than pride of accomplishment. too, money is more readily obtainable when it can be traded for the right to use a program.
second:
"the artistic licence... does not prevent someone from taking the whole body of code and closing it." instead of sarcasm, let me offer this analogy: i am going to take your post, encrypt it, and post it to: http://www.ee.ucr.edu/~jthorpe/document.html
so that nobody else will be able to read your comment or gain any insight from it.
of course, this doesn't work since your post should still be right there above mine. similarly, nobody can subtract from the value of your open-source program by merely compiling it and giving it away.
that being said, neither one of us has yet said enough to answer the implicit question, which is "should we prefer GPL to a completely non-restrictive license" and i leave it as an open question. it is a big question, and shouldn't be decided in half a page.
yea, but what's up with this: the request for moderation was moderated up, but the original post wasn't (!)... was there some moderator hanging around with only one point left?
actually, i think this is a good decision. i mean about easter eggs. it's one thing to, as a programmer, put your name in the credits. it's quite another to attach a doom client to excel--as if it wasn't bloated enough already!
another thought--has anyone ever exploited an easter egg to compromise a system? i'm not saying that they're inherently weak, but if the admin doesn't know what all he's potentially running, how can he know that he's protecting his system?
The GPL enforces sharing, the BSD license expects the other person to want to share.
--end quote--
contraire. BSD expects nothing at all. people who use the BSD are not negotiating for the same kinds of things that people using the GPL are negotiating for. they are simply giving their code to the world. end of story. to say that people expect 'sharing' in return is to miss the point.
give me a break, moderators. since when do people get moderated up for missing the point by so much. what is your point, 'um', that 2048 bytes is a bit long for a symmetric cipher? maybe attacking his company is a little extreme? (and it's the one time i don't have points!)
that it [the transmeta site] uses.gif's, bringing it squarely out of line with the principles of open source.
seriously though, we sorta knew before (from the patents and stuff) that they were aiming at processor emulation. what i want to know is: are they trying to compete with say AMD (intel-compatible motherboards) as i had thought, or are they trying to facilitate a whole new breed of product as their site seems to suggest? if it is the latter, i wonder what good all of this secrecy is. think about how long intel has been promoting their new IA64 and it is still a long way from a sure-thing. is transmeta going to spring a new architecture on the world and change things overnight?
i saw this too. however, based on the reasons cited for excluding the MacOS, i believe that linux would actually have been excluded as well. to recap, [one of] the reason[s] that MacOS was excluded was "a dearth of applications". (the other, of course was that you had to scrap your intel machine, which some people might consider significant)
don't get me wrong, i love linux apps, but emacs simply won't substitute for word, and netscape won't substitute for IE (flamers, here's your cue). and, even more fundamentally, windows users are used to clicking their way through life, while linux users are, even under X, used to using the command prompt (xterms are oo times more popular than command.exe).
still, it hurts to see it in writing, dosen't it:).
okay, then i completely agree with you, about the facts. i would personally choose BSD, and not because i am an unselfish guy, but because i want as many people to use my stuff as possible. i would actually be happy if MS used a piece of code of mine, because (making the dubious assumption that they made a good choice) everyone benefits. also note that i'm not giving a competitive advantage to MS since everyone has access to my code. i just have no good reason to fight against them.
note, by the way, that my own personal code is probably not good enough now for anyone to really care which licence i use, which makes me a kibitzer not a player. but, someday i plan to play the game, and when i do, i won't GPL.
fine. you are right. while we're at it, premitted is not a word.
hehe, but seriously, i realize code has no freedom. however this usage reflects an ambiguity that we incur anyway when we are talking about free code and don't specify precisely what we are talking about.
one sort of freedom that may be applied to code is the public's freedom to use it for any purpose. this sort of freedom is associated with, say, things in the public domain.
another sort of freedom is an author's freedom to see things made by people using stuff he wrote, if they distribute it. this is the sort of freedom that the GPL enforces.
these two are obviously mutually exclusive, so you might want to think about which one of these definitions seems less contrived. if you conclude that the second sense is a more natural and obvious one, by all means, feel free to argue that GPL is actually more free than public domain.
i am glad everyone puts a (tm) by Good Thing because that way i can tell when you're being sarcastic.
but on a more serious note, this RMS's definition of freedom, even if you competely forget about 'gratis' freedom, is a very presumptuous one. it depends on whose freedom you are trying to maximize--RMS decides one of the following things, each less defensible than the last:
1. it is only the author's freedom that is to be maximized. in this case, distribution is unnecessary.
2. it is only the GPL community whose rights are protected. people who do not wish to join this community are not to have the same rights as those who do.
3. distributing proprietary software is not a valid decision in the first place, so restricting this is not really destroying any valid kind of freedom.
i completely agree that the licensing is "too free". let's see what RMS has to say...
--RMS-- The Free Software movement aims to provide certain freedoms for all computer users, including the right to change a program, the right to redistribute copies, and the right to publish modified versions. The most effective way to protect these rights is to deny anyone else the power to take them away from you. --/RMS--
implying, it would seem, that if code were to accidentally be released to (say) the public domain, said rights would be in jeopardy. the nature of this jeopardy: use of code in *gasp* proprietary software by companies we're pretty sure make nothing but crap anyway.
the truth is, the choice of GPL over a non-restrictive licence (if i may make such a contrast) is a tradeoff. you might achieve a larger body of free (whatever that means now) software based on your work, but at the price of turning off other developers. it's everyone's choice.
giving someone several options means nothing about freedom. this is so obvious that i won't bother to construct you an analogy. it is patently obvious that the GPL removes a very important freedom that, say, the BSD does not.
you may try to (as RMS does) argue that the clause is actually protecting the freedom of the code. please examine this argument. from what are you protecting your code? from people who want to use it in proprietary products, right? what kind of threat is this? what exactly have you lost if MS uses your TCP/IP stack in windows and, *gasp* dosen't release windows under GPL? is anyone less free to use your code if they do?
exactly what i thought when i read the article. this is a trend at slashdot lately, by the way. since the article is only about one page long, i can only assume that mr. taco dosen't really even try to read the articles before coming up with a catchy blurb that invariably gets corrected by slashdot readers in the comments section. what's up guys, maybe you can try a little harder...
would it be unfair to give slashdotting McDonell Douglass employees some kind of automatic karma penalty to reflect the massive negative karma that McDonell Douglass earns for this heist?
It means that whatever the local standard, the phone can connect.
it dosen't seem so. according to the article, this chip is to be used by manufacturers to implement their own standards. it dosen't seem that any manufacturers have said that they will actually support multiple protocols in their phones.
while it may be that this will eventually happen, this piece of news is only the first step toward that end.
hmm, thanx for the info. but your answer brings up another technical question: if gif2png is going to decompress the GIF and then re-compress into PNG, it seems that the image quality may still suffer a bit even if the quality of PNG compression is better than LZ for the simple reason that by the time it gets to the user it will have gone through two (lossy) codecs.
Slashdot Mirror
reading of feynman! this is really what he believed, at least later in his life. i should say this: he had very little feeling of social responsibility while at the same time having a great sense of scientific responsibility. there is a great quote even to this effect which i will try to remember now:
what you choose to do in your personal life--whether you cheat on your wife--that is between you and your rabbi, but when it comes to science you'd better be honest!
something like that anyway. if someone else has "Adventures of a curious character" on hand please feel free to correct me, thanx.
sh_
But there are still things that you cannot do on the net and will never be able to do on the net, such as grocery shopping, having a sincere discussion with a friend or loved one, and much more. Yet, like this AOL/TW deal, businesses think that it CAN replace all that.
actually, i'm glad that TWX and AOL are merging precicely so that i don't have to have to do this kind of crap that you describe. if my loved ones really cared to have a "sincere discussion" they know which channels to find me on!
--EricWright wrote--
However, I was later offended by the author's apparent lack of patience. His comment about "a man...condemned forever to articulate his thoughts at the speed of an imbecile" made me wince. Here he is, one of a privileged few journalists with the opportunity to spend an afternoon with the greatest mind of the last 50 years, and he is focussing on the man's physical disabilities. I nearly stopped reading at that point.
--end--
i also winced reading this, but it was immediately recognizable to me that what the author said was not disparaging. clearly what's implicit is that Hawkings is not an imbecile though he 'talks' very slowly.
anyway, this was really a great article. i've enjoyed reading something of Hawking, but this really put a human face on the man for me. it seems that our author has accomplished his goal.
sh_
here's the link i meant to post. and no more signatures :)
while i agree wholeheartedly with the Ed's post, i can't imagine that the patent office would ever say "no more software patents". but what about this:
it seems software patents could be made to expire after a much shortened period of time, say 5 years.
patents usually expire 17 or 20 years after they are issued, or much longer (29 years, i think, though i can't find a reference right now) for drugs. however even 17 years is quite a long time to exclude competition in the software industry. i would argue that 5 years or so would be a much more reasonable ammount of time for anyone to have a software patent.
5 years would give any company a reasonable ammount of time to recoup their R&D costs, and would also have the effect of discouraging companies from hoarding software patents in hopes that someone eventually steps on them.
cheers,
sh_
cheers,
sh_
what is it? a rant? you call this a rant? do you call it a rant because you don't know any other nouns to use to describe something you don't agree with? you are so silly.
translation: the author didn't use the sophisticated sorts of arguments that we GPL'ers like.
/.ers like to make in order to come to an honest and reasonable conclusion as this guy has.
it's true--this guy is certainly not steeped in the BSD-GPL debate, or else he would have taken care to at least try to make a much more precise argument. _still_, you don't have to make the sort of complicated analysis that
nevermind, the real GPL-BSD debate won't be decided in slashdot threads. it will be decided by people who have good and bad experiences with GPL and with BSD. right now GPL has an advantage because linux is more popular than BSD, and because torvalds has chosen GPL. but the sort of license that we choose is a question that will remain longer than any one program and any one guy.
i am undoing my moderation in this thread in order to respond here, so please listen.
let me preface this by saying that of all of the arguments for GPL that i've read, this is one of the best. it is the quintessential pro-GPL argument. but, the reasoning is selective. i would like to answer two of your points.
first:
once a program is written, to encrypt it by compilation is to destroy some of its value. you would argue that it destroys all of the program's value, but this is not true.
what is not addressed is whether the program will be written in the first place--specifically what will motivate a programmer to write a useful program. money is certainly a more universal motivator than pride of accomplishment. too, money is more readily obtainable when it can be traded for the right to use a program.
second:
"the artistic licence... does not prevent someone from taking the whole body of code and closing it." instead of sarcasm, let me offer this analogy: i am going to take your post, encrypt it, and post it to:
http://www.ee.ucr.edu/~jthorpe/document.html
so that nobody else will be able to read your comment or gain any insight from it.
of course, this doesn't work since your post should still be right there above mine. similarly, nobody can subtract from the value of your open-source program by merely compiling it and giving it away.
that being said, neither one of us has yet said enough to answer the implicit question, which is "should we prefer GPL to a completely non-restrictive license" and i leave it as an open question. it is a big question, and shouldn't be decided in half a page.
cheers,
jeremy
yea, but what's up with this: the request for moderation was moderated up, but the original post wasn't (!)... was there some moderator hanging around with only one point left?
actually, i think this is a good decision. i mean about easter eggs. it's one thing to, as a programmer, put your name in the credits. it's quite another to attach a doom client to excel--as if it wasn't bloated enough already!
another thought--has anyone ever exploited an easter egg to compromise a system? i'm not saying that they're inherently weak, but if the admin doesn't know what all he's potentially running, how can he know that he's protecting his system?
DaemonDownTheHall wrote:
The GPL enforces sharing, the BSD license expects the other person to want to share.
--end quote--
contraire. BSD expects nothing at all. people who use the BSD are not negotiating for the same kinds of things that people using the GPL are negotiating for. they are simply giving their code to the world. end of story. to say that people expect 'sharing' in return is to miss the point.
sh_
no, the last prime timstamp is a good 10 days away. it is 23:59:59 29/11/1999, although nobody had previously mentioned the last prime timestamp.
give me a break, moderators. since when do people get moderated up for missing the point by so much. what is your point, 'um', that 2048 bytes is a bit long for a symmetric cipher? maybe attacking his company is a little extreme? (and it's the one time i don't have points!)
sh_
sometimes, a cigar is just a cigar!
that it [the transmeta site] uses .gif's, bringing it squarely out of line with the principles of open source.
seriously though, we sorta knew before (from the patents and stuff) that they were aiming at processor emulation. what i want to know is: are they trying to compete with say AMD (intel-compatible motherboards) as i had thought, or are they trying to facilitate a whole new breed of product as their site seems to suggest? if it is the latter, i wonder what good all of this secrecy is. think about how long intel has been promoting their new IA64 and it is still a long way from a sure-thing. is transmeta going to spring a new architecture on the world and change things overnight?
i saw this too. however, based on the reasons cited for excluding the MacOS, i believe that linux would actually have been excluded as well. to recap, [one of] the reason[s] that MacOS was excluded was "a dearth of applications". (the other, of course was that you had to scrap your intel machine, which some people might consider significant)
don't get me wrong, i love linux apps, but emacs simply won't substitute for word, and netscape won't substitute for IE (flamers, here's your cue). and, even more fundamentally, windows users are used to clicking their way through life, while linux users are, even under X, used to using the command prompt (xterms are oo times more popular than command.exe).
still, it hurts to see it in writing, dosen't it
cheers
sh_
okay, then i completely agree with you, about the facts. i would personally choose BSD, and not because i am an unselfish guy, but because i want as many people to use my stuff as possible. i would actually be happy if MS used a piece of code of mine, because (making the dubious assumption that they made a good choice) everyone benefits. also note that i'm not giving a competitive advantage to MS since everyone has access to my code. i just have no good reason to fight against them.
note, by the way, that my own personal code is probably not good enough now for anyone to really care which licence i use, which makes me a kibitzer not a player. but, someday i plan to play the game, and when i do, i won't GPL.
sh_
fine. you are right. while we're at it, premitted is not a word.
hehe, but seriously, i realize code has no freedom. however this usage reflects an ambiguity that we incur anyway when we are talking about free code and don't specify precisely what we are talking about.
one sort of freedom that may be applied to code is the public's freedom to use it for any purpose. this sort of freedom is associated with, say, things in the public domain.
another sort of freedom is an author's freedom to see things made by people using stuff he wrote, if they distribute it. this is the sort of freedom that the GPL enforces.
these two are obviously mutually exclusive, so you might want to think about which one of these definitions seems less contrived. if you conclude that the second sense is a more natural and obvious one, by all means, feel free to argue that GPL is actually more free than public domain.
cheers,
sh_
i am glad everyone puts a (tm) by Good Thing because that way i can tell when you're being sarcastic.
but on a more serious note, this RMS's definition of freedom, even if you competely forget about 'gratis' freedom, is a very presumptuous one. it depends on whose freedom you are trying to maximize--RMS decides one of the following things, each less defensible than the last:
1. it is only the author's freedom that is to be maximized. in this case, distribution is unnecessary.
2. it is only the GPL community whose rights are protected. people who do not wish to join this community are not to have the same rights as those who do.
3. distributing proprietary software is not a valid decision in the first place, so restricting this is not really destroying any valid kind of freedom.
cheers,
sh_
i completely agree that the licensing is "too free". let's see what RMS has to say...
--RMS--
The Free Software movement aims to provide certain freedoms for all computer users, including the right to change a program, the right to redistribute copies, and the right to publish modified versions. The most effective way to protect these rights is to deny anyone else the power to take them away from you.
--/RMS--
implying, it would seem, that if code were to accidentally be released to (say) the public domain, said rights would be in jeopardy. the nature of this jeopardy: use of code in *gasp* proprietary software by companies we're pretty sure make nothing but crap anyway.
the truth is, the choice of GPL over a non-restrictive licence (if i may make such a contrast) is a tradeoff. you might achieve a larger body of free (whatever that means now) software based on your work, but at the price of turning off other developers. it's everyone's choice.
cheers,
sh_
giving someone several options means nothing about freedom. this is so obvious that i won't bother to construct you an analogy. it is patently obvious that the GPL removes a very important freedom that, say, the BSD does not.
you may try to (as RMS does) argue that the clause is actually protecting the freedom of the code. please examine this argument. from what are you protecting your code? from people who want to use it in proprietary products, right? what kind of threat is this? what exactly have you lost if MS uses your TCP/IP stack in windows and, *gasp* dosen't release windows under GPL? is anyone less free to use your code if they do?
cheers,
sh_
exactly what i thought when i read the article. this is a trend at slashdot lately, by the way. since the article is only about one page long, i can only assume that mr. taco dosen't really even try to read the articles before coming up with a catchy blurb that invariably gets corrected by slashdot readers in the comments section. what's up guys, maybe you can try a little harder...
sh_
would it be unfair to give slashdotting McDonell Douglass employees some kind of automatic karma penalty to reflect the massive negative karma that McDonell Douglass earns for this heist?
It means that whatever the local standard, the phone can connect.
it dosen't seem so. according to the article, this chip is to be used by manufacturers to implement their own standards. it dosen't seem that any manufacturers have said that they will actually support multiple protocols in their phones.
while it may be that this will eventually happen, this piece of news is only the first step toward that end.
cheers,
sh_
hmm, thanx for the info. but your answer brings up another technical question: if gif2png is going to decompress the GIF and then re-compress into PNG, it seems that the image quality may still suffer a bit even if the quality of PNG compression is better than LZ for the simple reason that by the time it gets to the user it will have gone through two (lossy) codecs.