look, it seems that if PNG is to preserve the benefits that GIF provides, it must use some kind of fast and strong compression algorithm like Lempel-Ziv. If the compression is fundamentally different from GIF, then what kind of compression rates and speed are we expecting from the new algorithm? if it is fundamentally the same compression algorithm, what legal benefits does anyone gain from using this software? even according to the page, there is no consensus about whether this new program is even legal.
well, i don't think i would be nearly as afraid of the courts as you would like me to be. in fact, it seems that i have 30 days to fix the situation (i suppose this means stop distributing the code) before you can do anything about it, is that right?
but i digress. here's the point i want to make, and i will have to go back to the original story:
recall, Derek said that he his GPL'd code was included in the source release of the DVD decryption algorithm. if it happened that the guys who reverse-engineered the thing in the first place wanted to release their code under another license (BSD, commercial, whatever), nobody, not even Derek can stop them. you simply cannot play these kinds of games to get control of people's code.
well, the premise of the post that i responded to was that the GPL'd part of the software, that is the validation algorithm, was bundled with the release of the code, but the code was not actually part of the program.
okay, consider this: i write a program A, and say, release it closed source. sometime later, i (perhaps on accident) include a some GPL code B, and release again, closed source. someone points this out. fine, i withdraw the new version. do you claim that i have lost my rights to distribute A closed source? i'm not actually sure what the GPL says (please somebody tell me). but actually it dosen't matter. program A would still be mine. the GPL could say that all of my work for the rest of my life, and the work of my decendants must henceforth be released under GPL. there is a point after which it simply ceases to matter what the GPL says. sorry.
so, what if source is released once not under GPL, and then the author decides to GPL the code later? i would think that i can just take the uninfected code from the first release and freely use that. it seems that the question of whether or not it *will* be GPL'd is moot.
interesting that raymond is so quick to make that particular analogy (open-source--media) and so quick to dismiss the analogy (indeed, the whole article) between open-source and the scientific community written by nikolai bezroukov (anyone remembers that?)
well, maybe he had a bad day that day. at any rate, this interview sounds more like the professional eric. again, not like the eric that threw a tantrum when publicly challenged by bruce perens on that whole apple licence thing.
gee, it just struck me--maybe if people would simply stop doubting eric's cleverness, we could actually benefit from his discerning (There's nothing funny about the popularity of "Dilbert." Companies should take that more seriously than they seem to. ) social commentary.
i am not sure, but now i have the idea that this works with anything written with microsoft foundation classes or something like that. (i coule look it up, but at this point i'm too lazy...) it's not a bad deal for developers since you don't have to build your apps to be i18n aware...
anyway, this is a huge thing for linux right now... when i went to japan in '97, windows was already deployed there in people's homes. that's the entire system, every program written with support for japanese. and now, W2k is supposed to be written "languageless" which certainly sounds like a good thing. anyway, i'm talking too much about windows here, but that's prolly because windows is definitely out in front on this front (to make a pun and mix a metaphor). oh well, maybe the unix guys, and especially the free unix guys, will get into the game soon. here's hoping.
i know. i wasn't responding to your post in particular there. nevermind.
about needing japanese version of windows--no, you don't. i downloaded some program (i can't find it now, and i'm not at my home computer, but it also lets you view japanese characters on webpages), installed it, and could immediately write to a lot of apps including notepad.
about your point about inputting text into any program in any language (which i think is the main point for both of us), i definitely agree. i don't know if it should be x's job, or be built into the kernel (i don't like this way), or if it should simply be a command line program that uses pipes to talk to any application (like by redirecting stdin).
it seems that a lot of people have the idea that internationalization means language translation: certainly it does not.
the critical step right now is to get standard input methods working for a broad range of languages that works for all programs. i have used a japanese language input methods in windows, and it works uniformly in word, on netscape forms, and even in notepad, IIRC.
by contrast, emacs 20.x has its own kind of input method, which cannot be used by other programs. a better (?!?) solution: redirect standard input to a program that converts keyboard input to the language of your choice. this is how UNIX is engineered in the first place, and it's a far less klugy way than windows! but, for now, windows works much better!!
as a commercial software developer, what reason do i have to use GPL stuff? the world simply dosen't work on free software now, and ultimately there may be no path from where we are now to a world with plenty of perfect GPL software.
GPL, as opposed to BSD, LGPL et cetera, provides the worst sort of incompatibility: it does not, and will never work with the commercial software distribution model. programmers who simply want to have their stuff *used* should licence under a less restrictive licence than GPL.
that, in the first place, it forbids companies to build proprietary stuff based on GPL stuff.
what is the worst that could happen?
fragmentation? yes, it's possible, but if your goal is 'world domination' would you rather have chinese people using a proprietary ver. of linux or windoze? it's easy to change linux distros coampared to learning linux after using MS...
developers have a harder time writing chinese programs? if that happens, who is ultimately hurt--seems it would be Blue Point.
choosing a good licence is a tough decision, and ultimately everyone who wrote linux must have their wishes honored. i do respect this. but, i would love to see programmers choose a licence that truly allows free redistribution (under any terms) of the code. or, here's an idea, release your code to the public domain!
"the GPL goes a step further than other FRLs by stating that the GPL automatically applies to the totality of any program that incorporates GPL'ed code."
i'm thrilled that VA is finally having thier IPO. hey, i'm all about the money, and i plan to invest in VA, and let me tell you, it's *only* because they're gonna make money!
god, if everyone talks like this, it's no wonder that the mainstream media keeps mistaking open source guys for communists!
please, nobody take this too seriously--THIS IS NOT A FLAME, OKAY:) i just think that you techies should try and make some money off your work, that's all. hehe, okay, 'nuff said.
you know what--i never saw sun even saying that they were 'open sourcing' their code, or that they were going to make it 'open source TM' compliant. only that they were publishing their code under their own licence. if 'community source' sounds too close to 'open source', too bad.
hey, i'm all for linux, and open source. but i do find it sad that such a popular licence, the GPL, isn't actually freely redistributable. i know that some other (let's leave them unnamed) licences are. it just means that there will be less high-quality commercial software out there for me to use on my linux box.
by comparison, the GPL restricts companies from developing software based on GPL stuff and selling it--even if consumers would readily pay for it. please learn that it is not always that what is good for companies is bad for people!
--end quote--
okay, this is a mis-statement--fine. what i mean is that a company cannot build on a GPL thing and then claim that they own the final product. why not? it doesn't change the fact that what that company built on is *still* out there, and available for free!
guys, the slashdot community is not so huge that they can dictate what it means for something to be 'open source'. some of you may think that it is not enough for a company to merely publish their code. for you, they must give up all of their rights to the intellectual property which they paid for the development of. this is truly disingenuous.
by comparison, the GPL restricts companies from developing software based on GPL stuff and selling it--even if consumers would readily pay for it. please learn that it is not always that what is good for companies is bad for people!
well, although i agree that the guy to whom you are replying doesn't have a clue... as far as i understand, there is no way in general to "try every key at the same time". i feel that it is more likely there is some more specific algorithm that is able to factor numbers or something like that.
for example, it is not the case that every NP complete problem can be solved (in polynomial time) by a quantum computer (yet!).
...If Microsoft wants to clone something first done in the Linux community, they're free to do so. If they want to embrace and extend, copy source code and make it proprietary, or spread FUD about Linux, they're not.
(emphasis added)
come on, think about this: it's not like anyone can make the code proprietary, is it? i mean, after all the code is still available to anyone (if they can find it), same as it would if microsoft hadn't used it in the first place.
and what if microsoft does embarace and extend, like their java thing, which by the way there would be nothing illegal about it if java was open source ?? it's *microsoft* who puts itself out of the loop by these actions. what, like there's no competition from IBM and sun? please!
Second, this is not creation like God did. In order for God to have created life, He had to be outside the universe, because if He was in the universe He would be living and someone would have had to create Him, and if THAT someone was in the universe, he would have had to be created by someone... ad infinitum.
(end quote)
you seem to have a semi-clear idea about this, but what does it mean to exist, but not be in this universe? i don't think this is as intuitive as, it seems, you do.
you said: Second, this is not creation like God did. In order for God to have created life, He had to be outside the universe, because if He was in the universe He would be living and someone would have had to create Him, and if THAT someone was in the universe, he would have had to be created by someone... ad infinitum. (end quote) you seem to have a semi-clear idea about this, but what does it mean to exist, but not be in this universe? i don't think this is as intuitive as, it seems, you do.
technically, channel capacity is proportional to the bandwidth times the log of the snr, which means that channel capacity is still not that bad, even if snr is ridiculously low.
this is, of course, if you can find error-correcting codes that nearly achieve channel capacity with low output error probability, and can be decoded fast enough. there are such codes though, like so-called turbo codes.
this particular argument bugs the hell out of me. first of all, how does anyone know what is the shortest sequence of amino acids that will self-replicate (in a given environment)? furthermore, how does one even know that the MO by which life came about would have to involve such a thing? but, suppose even that this is indeed the MO, and cheritably suppose that all of the assumptions that this mathematician made were correct. you still have the equivalent of the following argument: consider the following two explanations for your own existence: 1. you were concieved by your mom and dad. 2. you were put here by aliens yesterday. now, the probability that your mom and dad would even have met in the first place is certainly something like 1,000,000 to 1 against, right? so you have someth,ing like a 1/1000,000 chance of being here under that hypothesis. therefore it is 1,000,000 times more likely that you were created yesterday by aliens.
hey man, every single fricking law on the books in the US has a "reasonableness" clause somewhere in it. i read the story and i don't understand that the guy actually lost $.7M, so what's the big deal, or would you rather see the courts upholding more "unreasonable" stuff so that you can gossip about it?
do you really think that all of the gold in the world is as valuable as all of the cash in the world? if not, what good is it to gurantee your ability to trade cash in for gold when you can't possibly do it?
besides, i personally have little use for gold itself--the best i can hope for is that someone else will be willing to trade me for something that i do want--but that's just exactly what i hope for when i earn my little green pieces of paper.
Slashdot Mirror
look, it seems that if PNG is to preserve the benefits that GIF provides, it must use some kind of fast and strong compression algorithm like Lempel-Ziv. If the compression is fundamentally different from GIF, then what kind of compression rates and speed are we expecting from the new algorithm? if it is fundamentally the same compression algorithm, what legal benefits does anyone gain from using this software? even according to the page, there is no consensus about whether this new program is even legal.
sh_
write something that MS can use.
sh_
(moderators: please use 'troll' or 'flaimbait' to downgrade)
well, i don't think i would be nearly as afraid of the courts as you would like me to be. in fact, it seems that i have 30 days to fix the situation (i suppose this means stop distributing the code) before you can do anything about it, is that right?
but i digress. here's the point i want to make, and i will have to go back to the original story:
recall, Derek said that he his GPL'd code was included in the source release of the DVD decryption algorithm. if it happened that the guys who reverse-engineered the thing in the first place wanted to release their code under another license (BSD, commercial, whatever), nobody, not even Derek can stop them. you simply cannot play these kinds of games to get control of people's code.
got class. later.
sh_
well, the premise of the post that i responded to was that the GPL'd part of the software, that is the validation algorithm, was bundled with the release of the code, but the code was not actually part of the program.
okay, consider this: i write a program A, and say, release it closed source. sometime later, i (perhaps on accident) include a some GPL code B, and release again, closed source. someone points this out. fine, i withdraw the new version. do you claim that i have lost my rights to distribute A closed source? i'm not actually sure what the GPL says (please somebody tell me). but actually it dosen't matter. program A would still be mine. the GPL could say that all of my work for the rest of my life, and the work of my decendants must henceforth be released under GPL. there is a point after which it simply ceases to matter what the GPL says. sorry.
sh_
so, what if source is released once not under GPL, and then the author decides to GPL the code later? i would think that i can just take the uninfected code from the first release and freely use that. it seems that the question of whether or not it *will* be GPL'd is moot.
sh_
interesting that raymond is so quick to make that particular analogy (open-source--media) and so quick to dismiss the analogy (indeed, the whole article) between open-source and the scientific community written by nikolai bezroukov (anyone remembers that?)
well, maybe he had a bad day that day. at any rate, this interview sounds more like the professional eric. again, not like the eric that threw a tantrum when publicly challenged by bruce perens on that whole apple licence thing.
gee, it just struck me--maybe if people would simply stop doubting eric's cleverness, we could actually benefit from his discerning (There's nothing funny about the popularity of "Dilbert." Companies should take that more seriously than they seem to. ) social commentary.
&
sh_
yes, thank you. it was AsianSuite.
i am not sure, but now i have the idea that this works with anything written with microsoft foundation classes or something like that. (i coule look it up, but at this point i'm too lazy...) it's not a bad deal for developers since you don't have to build your apps to be i18n aware...
anyway, this is a huge thing for linux right now... when i went to japan in '97, windows was already deployed there in people's homes. that's the entire system, every program written with support for japanese. and now, W2k is supposed to be written "languageless" which certainly sounds like a good thing. anyway, i'm talking too much about windows here, but that's prolly because windows is definitely out in front on this front (to make a pun and mix a metaphor). oh well, maybe the unix guys, and especially the free unix guys, will get into the game soon. here's hoping.
sh_
I didn't make any reference(s) to translation.
i know. i wasn't responding to your post in particular there. nevermind.
about needing japanese version of windows--no, you don't. i downloaded some program (i can't find it now, and i'm not at my home computer, but it also lets you view japanese characters on webpages), installed it, and could immediately write to a lot of apps including notepad.
about your point about inputting text into any program in any language (which i think is the main point for both of us), i definitely agree. i don't know if it should be x's job, or be built into the kernel (i don't like this way), or if it should simply be a command line program that uses pipes to talk to any application (like by redirecting stdin).
sh_
guys,
it seems that a lot of people have the idea that internationalization means language translation: certainly it does not.
the critical step right now is to get standard input methods working for a broad range of languages that works for all programs. i have used a japanese language input methods in windows, and it works uniformly in word, on netscape forms, and even in notepad, IIRC.
by contrast, emacs 20.x has its own kind of input method, which cannot be used by other programs. a better (?!?) solution: redirect standard input to a program that converts keyboard input to the language of your choice. this is how UNIX is engineered in the first place, and it's a far less klugy way than windows! but, for now, windows works much better!!
what do you guys think?
sh_
as a commercial software developer, what reason do i have to use GPL stuff? the world simply dosen't work on free software now, and ultimately there may be no path from where we are now to a world with plenty of perfect GPL software.
GPL, as opposed to BSD, LGPL et cetera, provides the worst sort of incompatibility: it does not, and will never work with the commercial software distribution model. programmers who simply want to have their stuff *used* should licence under a less restrictive licence than GPL.
-sh_
that, in the first place, it forbids companies to build proprietary stuff based on GPL stuff.
what is the worst that could happen?
fragmentation? yes, it's possible, but if your goal is 'world domination' would you rather have chinese people using a proprietary ver. of linux or windoze? it's easy to change linux distros coampared to learning linux after using MS...
developers have a harder time writing chinese programs? if that happens, who is ultimately hurt--seems it would be Blue Point.
choosing a good licence is a tough decision, and ultimately everyone who wrote linux must have their wishes honored. i do respect this. but, i would love to see programmers choose a licence that truly allows free redistribution (under any terms) of the code. or, here's an idea, release your code to the public domain!
"the GPL goes a step further than other FRLs by stating that the GPL automatically applies to the totality of any program that incorporates GPL'ed code."
i'm thrilled that VA is finally having thier IPO. hey, i'm all about the money, and i plan to invest in VA, and let me tell you, it's *only* because they're gonna make money!
:) i just think that you techies should try and make some money off your work, that's all. hehe, okay, 'nuff said.
god, if everyone talks like this, it's no wonder that the mainstream media keeps mistaking open source guys for communists!
please, nobody take this too seriously--THIS IS NOT A FLAME, OKAY
-sh
you know what--i never saw sun even saying that they were 'open sourcing' their code, or that they were going to make it 'open source TM' compliant. only that they were publishing their code under their own licence. if 'community source' sounds too close to 'open source', too bad.
hey, i'm all for linux, and open source. but i do find it sad that such a popular licence, the GPL, isn't actually freely redistributable. i know that some other (let's leave them unnamed) licences are. it just means that there will be less high-quality commercial software out there for me to use on my linux box.
cheers
-sh_mmer
i wrote:
by comparison, the GPL restricts companies from developing software based on GPL stuff and selling it--even if consumers would readily pay for it. please learn that it is not always that what is good for companies is bad for people!
--end quote--
okay, this is a mis-statement--fine. what i mean is that a company cannot build on a GPL thing and then claim that they own the final product. why not? it doesn't change the fact that what that company built on is *still* out there, and available for free!
guys, the slashdot community is not so huge that they can dictate what it means for something to be 'open source'. some of you may think that it is not enough for a company to merely publish their code. for you, they must give up all of their rights to the intellectual property which they paid for the development of. this is truly disingenuous.
by comparison, the GPL restricts companies from developing software based on GPL stuff and selling it--even if consumers would readily pay for it. please learn that it is not always that what is good for companies is bad for people!
well, although i agree that the guy to whom you are replying doesn't have a clue... as far as i understand, there is no way in general to "try every key at the same time". i feel that it is more likely there is some more specific algorithm that is able to factor numbers or something like that.
for example, it is not the case that every NP complete problem can be solved (in polynomial time) by a quantum computer (yet!).
-sh_mmer
...If Microsoft wants to clone something first done in the Linux community, they're free to do so. If they want to embrace and extend, copy source code and make it proprietary, or spread FUD about Linux, they're not.
(emphasis added)
come on, think about this: it's not like anyone can make the code proprietary, is it? i mean, after all the code is still available to anyone (if they can find it), same as it would if microsoft hadn't used it in the first place.
and what if microsoft does embarace and extend, like their java thing, which by the way there would be nothing illegal about it if java was open source ?? it's *microsoft* who puts itself out of the loop by these actions. what, like there's no competition from IBM and sun? please!
...where *is* this URL?
you said:
Second, this is not creation like God did. In order for God to have created life, He had to be outside the universe, because if He was in the universe He would be living and someone would have had to create Him, and if THAT someone was in the universe, he would have had to be created by someone... ad infinitum.
(end quote)
you seem to have a semi-clear idea about this, but what does it mean to exist, but not be in this universe? i don't think this is as intuitive as, it seems, you do.
you said: Second, this is not creation like God did. In order for God to have created life, He had to be outside the universe, because if He was in the universe He would be living and someone would have had to create Him, and if THAT someone was in the universe, he would have had to be created by someone... ad infinitum. (end quote) you seem to have a semi-clear idea about this, but what does it mean to exist, but not be in this universe? i don't think this is as intuitive as, it seems, you do.
technically, channel capacity is proportional to the bandwidth times the log of the snr, which means that channel capacity is still not that bad, even if snr is ridiculously low.
this is, of course, if you can find error-correcting codes that nearly achieve channel capacity with low output error probability, and can be decoded fast enough. there are such codes though, like so-called turbo codes.
this particular argument bugs the hell out of me. first of all, how does anyone know what is the shortest sequence of amino acids that will self-replicate (in a given environment)? furthermore, how does one even know that the MO by which life came about would have to involve such a thing? but, suppose even that this is indeed the MO, and cheritably suppose that all of the assumptions that this mathematician made were correct. you still have the equivalent of the following argument: consider the following two explanations for your own existence: 1. you were concieved by your mom and dad. 2. you were put here by aliens yesterday. now, the probability that your mom and dad would even have met in the first place is certainly something like 1,000,000 to 1 against, right? so you have someth,ing like a 1/1000,000 chance of being here under that hypothesis. therefore it is 1,000,000 times more likely that you were created yesterday by aliens.
hey man, every single fricking law on the books in the US has a "reasonableness" clause somewhere in it. i read the story and i don't understand that the guy actually lost $.7M, so what's the big deal, or would you rather see the courts upholding more "unreasonable" stuff so that you can gossip about it?
guys,
do you really think that all of the gold in the world is as valuable as all of the cash in the world? if not, what good is it to gurantee your ability to trade cash in for gold when you can't possibly do it?
besides, i personally have little use for gold itself--the best i can hope for is that someone else will be willing to trade me for something that i do want--but that's just exactly what i hope for when i earn my little green pieces of paper.