I can just see the customers now. "Curse you for releasing the source! See if I ever pay for an upgrade of this software again!" Actually, if I were such a customer, it would irritate me a lot less than the possibly more common announcement "New major version! Registration fee is $39.95, or FREE for all customers who purchased the old version within the past year. Too bad about you early supporters who forked over the dough for that same version 14 months ago."
2 or more open source products?
on
GPL FAQ
·
· Score: 1
Yep, but they were all fairly small programs, so I don't know if they count as "product".
Re:Things RMS didn't forsee in 1984
on
GPL FAQ
·
· Score: 1
the GPL removes some rights that are held intrinsically by the author of copyright.
Um, no. The author of copyright is granting specific limited rights to other people, e.g. the right to create derivative works provided that some funky restrictions are obeyed. The author of copyright still has his original rights, e.g. the right to create derivative works without obeying the restrictions he imposes on what he's granted to others; and, indeed, the right to grant other people the ability to do things which are otherwise his own exclusive rights.
This is irrelevant to your main point but worth noting because most questions people ask have obvious answers (e.g., can I take code that I wrote single-handedly, and reuse it in a closed-source program?) but, if a person starts reasoning from the "my rights have been removed" perspective, he will tend to arrive, with unshakeable conviction, at the wrong obvious answers. heh heh.
how can you expect everyone to come up with a unique expression describing how an
airplane wing works?
Easy. The same way I have come to expect hundreds of people on/. to come up with high-scoring posts that are uniquely worded, yet say nothing that hasn't already been said in an earlier post or the article itself. (Present company excluded:)
The key insight here (no, it's not hexapodia) is that you can't just read the GPL. It relies on copyright law. There is a fairly readable faq with two sections of particular relevance to this problem. Read it and you will know as much as me.
"What is copyright" describes the exclusive rights of you the copyright holder, such as "to prepare derivative works based upon the work", which you can do yourself and can authorize others to do (or not authorize, as you see fit).
By authorizing you're not giving up anything other than, of course, the ability to whine when someone does something that you actually authorized them to do.:-)
"Transfer of copyright" describes what you have to do in order to transfer some or all of your exclusive rights to another entity. Seems that it has to be in writing and signed by you in order for you to "lose" your rights and for the other entity to gain the "exclusiveness" of them, otherwise you still have those rights too (and they get non-exclusive rights, I guess, which is maybe the same as being authorized).
If you write some source code and don't put it in the public domain and don't sign a piece of paper that says that your code belongs to someone else now, well, it seems to me that you still have the original full set of rights, and not whatever limited/restricted set you may have authorized for other people. You may freely decide to only use the restricted set, but that is your decision, and I cannot see anything that binds you to it.
But they have to release the super set [proprietary] version to the users as well
No, they don't have to (as others have already stated). The company can release and maintain an inferior open-source version of their proprietary product, without infecting the proprietary version in any way. Infection of the proprietary version can occur if the company is foolish enough to incorporate code that someone else holds the copyright to. To avoid that, all the company has to do is to declare "we won't accept any patches into either of our two versions unless the author of the patch signs this-here legal form", either a form to place the patch in public domain or to transfer the copyright to the company.
[suppose the open source version] has tons of build problems and explosive bugs... Then what?
Bah. Then programmers with a clue (PWAC) ask for the source code for the open-source version; fix the bugs, send in their patches with a "these patches are Public Domain*" notice; and if the patches are not incorporated into the next release of the open-source version, and all other polite courses of action have been exhausted, the PWAC can always fork the code and maintain their own version under that same open-source license. (Yes, it's an option, this is discussed in someone else's comment thread.)
*see, then you can still use them in the forked version if need be, which you couldn't do if you signed them over to the company and the company decided not to stick them in the open-source version.
This comment is GPL-independent, which is fortunate, because I can't figure out what your last three sentences mean.
It seems to me that the way to keep attracting new people to clubs, hobbies, amateur radio, open source development, or whatever, is for the existing groups of people to be Doing Stuff Enthusiastically. Even if the stuff a club is working on has been done before, done better, done with fewer wires, or could be done by your dog with one paw tied behind its back, it is doing something that the members are interested in. Sure, to be doing something beneficial or novel is worth something too [and is what I would expect to see in/. articles if I were less cynical], but not as much as attitude.
The way to drive away new people is to sit around and say "Yeah, I heard the guys up the street are hacking on this 'new' thing, but it's not very novel or clever or useful to me and it's also not what we're Really About" in such a way as to imply "By the way, feel free to express any blue-sky ideas that you have, so that we can... help you, yes, that's it, for your own good." If this sort of thing went on regularly in open source project groups... well, actually, my planet-sized ego could withstand it trivially, but the point is that not everyone has "What do you care what other people think" tattooed on their eyelids, and we would have this whole uncool DDT eggshells, squashed little downy fledgling geeks, thing going, and eventually a bunch of projects would dry up and blow away, resulting in poisoned streams downwind and the eventual demise of the ecosphere. Oh, the embarrassment!
Since you seem to be fond of the BSD license, I'll make you a deal. I'll write some BSD-licensed code, if you'll use a spelling checker. We can bid on grammar another time.
James Aspnes, Richard Beigel, Merrick Furst, and Steven Rudich published a paper on this kind of voting puzzle in 1993 (i.e. it's been floating around for at least five years longer than the NYTimes article suggests). On page 2 of their paper, we see:
Let
n be an odd number of women. Let each have a uniformly chosen random bit on her forehead.
That is to say, a red hat or a blue hat!
They wish to vote on the parity of the bits.
Since each can see everyone else's foreheads but her own, in plain English that means that each of them wants to guess what parity her own bit (that is, what color her own hat) is.
So if this puzzle appeals to you and you like a bit of math, you can go read the paper over lunch:-)
Sure it's informative: by negating the misinformation in its parent (and, as a bonus, reminding people that the way to get a clue is RTFM or in this case RTFL). Learning by negative examples eh.
Some other possiblities that we shouldn't neglect:
(3) Hacks happen and you don't hear about them.
(4) Hacks happen and you hear about them, but don't consider them clever because they're too
similar to historical hacks.
(5) No toad sexing
(6) Hacks happen and you hear about them, but
don't consider them hacks because they're not
similar enough to historical hacks.
(7) Do not lick
(8) By gum, the world really has gone downhill since MIT started admitting those round women
Well, it's not news to me (though I agree that it's cool) because I get Sky & Telescope. But it was interesting to me to see it show up here in a non-disjoint, but seemingly pretty different, n% of the population.
This leads me to wonder: how big is the overlap, anyway? Honk if you can find a constellation other than Orion, Ursa Major, or.. hm.. Southern Cross.
Moties and pollution, two great tastes that..
on
Solar Sails
·
· Score: 1
I'm reminded of The Mote in God's Eye and the poor little Moties trying to escape their planet on a solar sail powered craft.
Poor little Moties, heh. Now there would be a culture that the word "fatalism" properly describes... and a planet that better exemplifies "despoiled".
Not only did they use a solar sail, but they also powered it for a number of years with a massive laser (so powerful that humans living in the destination solar system saw the Moties' star change color while the laser was on). A character later commented that this was like being able to leave your engines at home. A neat trick if it worked, eh.
I guess the next step up in travel is leaving yourself at home, e.g. Genesis Quest: Giant lasers not enough for you? How about modulating a star..
Yeah, some of us counter-picketed because, hey, what else is there to do after finals week; but the scientologists turned out to be fundamentally incapable of grasping the scintillating absurdity of our slogans. Bummer, eh.
Based on the recent past, I expected this thread to contain a lot of whinging about how "Google took down Deja and my life lies in a shattered ruin about my feet because now I can't post anymore".
I can't tell you how pleased I am to instead see whinging about how "Google plans to let people post". I could understand this being a tragedy and commiserate if, say, the September of 1993 had actually ended.
I'm also tickled to see the phrase "quality web publishing" used with a serious face.;-)
(Actually, it does make a nice change from
"The People demand that Google open source [sic] the archives!" vs. "I'm going to sue for copyright violation!")
I could say something on topic, but I think it's all (and more than all) been said. Besides, I'm giving up "Arguing about the GPL" for Lent, and figure I should make an early start.
...it would be nice if licenses like this were distributed with their own hello-world example "applications" that exercise individual features.
Then us geek types could be more sure of understanding it.
% hello_world
Alice develops a GPL library.
Bill Gates wants to link his code to it.
segmentation violation - core dumped
1) Read Section 3 of the GPLv2. If the library is "normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs" and you are not distributing that library, you are off the hook, otherwise trouble.
2) Read Section 2 paragraph b
"You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License." If you did not distribute or publish the derived package you would be off the hook, but since you say "commercial" presumably it is to be distributed in return for money. Trouble.
3a,3b,4a) If you have read this far in the post you should have read section 3. "For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable." All modules means ALL modules. Trouble.
4b) Same as the answer to Question 1, if bison is something normally distributed with the major components of the OS you are ok, otherwise trouble.
5) Dynamic linking is hard.. let's go shopping! My best understanding of this case is that you are not in trouble. (You could be in trouble if the original non-GPL object had not existed, but since it does exist I think the fact that a plug-compatible GPL equivalent object exists is irrelevant. I figure I have a 50% chance of being wrong, so someone better correct me.)
I know it's hard to understand licenses and official fine print like that, but it is still important to read them.:-)
Well, you know what they say - Throw the first one away!
(I didn't get my Visor for its USB connectivity anyway. Got a serial cradle immediately for use with Linux. But wireless is tempting. Hmm, can people port/clone classic multiplayer games like M.U.L.E. now?)
I think it's safe to say that, for any randomly selected/. poster, at least 2/3 of his* acquaintances are computer geeks.
Thus it could be the case that lute-playing is independent of profession, and one would still find that 2/3 of one's lute-playing acquaintances were computer geeks.
* extrapolating from my weekday environment, I conclude that only 12% of the world's population is female, therefore the need for gender-neutral pronouns is much less than previously supposed.
Another vital point is keeping the documentation up to date!
Suppose that I write some pretty decent comments every time I write a new function. Then some night around 2am when I'm adding some little feature or stalking a bug, I don't remember to change the comment along with the code. Now the comment may be worse than nothing - it could be actively misleading. Even if no one else ever reads this code, I am going to be pretty ticked at myself when I try to figure out what is going on in this function six months later.
This is why undisciplined people like me would welcome newbies to look over the code and say "Yo, I think you have some documentation rot here, let me fix that." We have plenty of comments in our code. It's just that half of them are wrong.
Also, never drink and code, but that's another story entirely.
Slashdot Mirror
I can just see the customers now. "Curse you for releasing the source! See if I ever pay for an upgrade of this software again!" Actually, if I were such a customer, it would irritate me a lot less than the possibly more common announcement "New major version! Registration fee is $39.95, or FREE for all customers who purchased the old version within the past year. Too bad about you early supporters who forked over the dough for that same version 14 months ago."
Yep, but they were all fairly small programs, so I don't know if they count as "product".
Um, no. The author of copyright is granting specific limited rights to other people, e.g. the right to create derivative works provided that some funky restrictions are obeyed. The author of copyright still has his original rights, e.g. the right to create derivative works without obeying the restrictions he imposes on what he's granted to others; and, indeed, the right to grant other people the ability to do things which are otherwise his own exclusive rights.
This is irrelevant to your main point but worth noting because most questions people ask have obvious answers (e.g., can I take code that I wrote single-handedly, and reuse it in a closed-source program?) but, if a person starts reasoning from the "my rights have been removed" perspective, he will tend to arrive, with unshakeable conviction, at the wrong obvious answers. heh heh.
Excellent. [holds up "Hi Mom!" sign]
Easy. The same way I have come to expect hundreds of people on /. to come up with high-scoring posts that are uniquely worded, yet say nothing that hasn't already been said in an earlier post or the article itself. (Present company excluded :)
Go go gadget Google Groups... possibly not the right HTTP-jiving article, though...
"What is copyright" describes the exclusive rights of you the copyright holder, such as "to prepare derivative works based upon the work", which you can do yourself and can authorize others to do (or not authorize, as you see fit). By authorizing you're not giving up anything other than, of course, the ability to whine when someone does something that you actually authorized them to do. :-)
"Transfer of copyright" describes what you have to do in order to transfer some or all of your exclusive rights to another entity. Seems that it has to be in writing and signed by you in order for you to "lose" your rights and for the other entity to gain the "exclusiveness" of them, otherwise you still have those rights too (and they get non-exclusive rights, I guess, which is maybe the same as being authorized).
If you write some source code and don't put it in the public domain and don't sign a piece of paper that says that your code belongs to someone else now, well, it seems to me that you still have the original full set of rights, and not whatever limited/restricted set you may have authorized for other people. You may freely decide to only use the restricted set, but that is your decision, and I cannot see anything that binds you to it.
No, they don't have to (as others have already stated). The company can release and maintain an inferior open-source version of their proprietary product, without infecting the proprietary version in any way. Infection of the proprietary version can occur if the company is foolish enough to incorporate code that someone else holds the copyright to. To avoid that, all the company has to do is to declare "we won't accept any patches into either of our two versions unless the author of the patch signs this-here legal form", either a form to place the patch in public domain or to transfer the copyright to the company.
[suppose the open source version] has tons of build problems and explosive bugs ... Then what?
Bah. Then programmers with a clue (PWAC) ask for the source code for the open-source version; fix the bugs, send in their patches with a "these patches are Public Domain*" notice; and if the patches are not incorporated into the next release of the open-source version, and all other polite courses of action have been exhausted, the PWAC can always fork the code and maintain their own version under that same open-source license. (Yes, it's an option, this is discussed in someone else's comment thread.)
*see, then you can still use them in the forked version if need be, which you couldn't do if you signed them over to the company and the company decided not to stick them in the open-source version.
This comment is GPL-independent, which is fortunate, because I can't figure out what your last three sentences mean.
The way to drive away new people is to sit around and say "Yeah, I heard the guys up the street are hacking on this 'new' thing, but it's not very novel or clever or useful to me and it's also not what we're Really About" in such a way as to imply "By the way, feel free to express any blue-sky ideas that you have, so that we can... help you, yes, that's it, for your own good." If this sort of thing went on regularly in open source project groups ... well, actually, my planet-sized ego could withstand it trivially, but the point is that not everyone has "What do you care what other people think" tattooed on their eyelids, and we would have this whole uncool DDT eggshells, squashed little downy fledgling geeks, thing going, and eventually a bunch of projects would dry up and blow away, resulting in poisoned streams downwind and the eventual demise of the ecosphere. Oh, the embarrassment!
Since you seem to be fond of the BSD license, I'll make you a deal. I'll write some BSD-licensed code, if you'll use a spelling checker. We can bid on grammar another time.
So if this puzzle appeals to you and you like a bit of math, you can go read the paper over lunch :-)
Chuckle. My first thought was: Many people already believe this, and that's why we have so much light pollution.
Sure it's informative: by negating the misinformation in its parent (and, as a bonus, reminding people that the way to get a clue is RTFM or in this case RTFL). Learning by negative examples eh.
'Cause even Morpheus knows that the best way to get past a locked door.. is to find another way in that isn't locked.
(3) Hacks happen and you don't hear about them.
(4) Hacks happen and you hear about them, but don't consider them clever because they're too similar to historical hacks.
(5) No toad sexing
(6) Hacks happen and you hear about them, but don't consider them hacks because they're not similar enough to historical hacks.
(7) Do not lick
(8) By gum, the world really has gone downhill since MIT started admitting those round women
This leads me to wonder: how big is the overlap, anyway? Honk if you can find a constellation other than Orion, Ursa Major, or.. hm.. Southern Cross.
Poor little Moties, heh. Now there would be a culture that the word "fatalism" properly describes... and a planet that better exemplifies "despoiled".
Not only did they use a solar sail, but they also powered it for a number of years with a massive laser (so powerful that humans living in the destination solar system saw the Moties' star change color while the laser was on). A character later commented that this was like being able to leave your engines at home. A neat trick if it worked, eh.
I guess the next step up in travel is leaving yourself at home, e.g. Genesis Quest: Giant lasers not enough for you? How about modulating a star..
Yeah, some of us counter-picketed because, hey, what else is there to do after finals week; but the scientologists turned out to be fundamentally incapable of grasping the scintillating absurdity of our slogans. Bummer, eh.
I can't tell you how pleased I am to instead see whinging about how "Google plans to let people post". I could understand this being a tragedy and commiserate if, say, the September of 1993 had actually ended.
I'm also tickled to see the phrase "quality web publishing" used with a serious face. ;-)
(Actually, it does make a nice change from "The People demand that Google open source [sic] the archives!" vs. "I'm going to sue for copyright violation!")
http://www.webgurus.com/matic/graphics/49255.gif
I could say something on topic, but I think it's all (and more than all) been said. Besides, I'm giving up "Arguing about the GPL" for Lent, and figure I should make an early start.
% hello_world
Alice develops a GPL library.
Bill Gates wants to link his code to it.
segmentation violation - core dumped
2) Read Section 2 paragraph b "You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License." If you did not distribute or publish the derived package you would be off the hook, but since you say "commercial" presumably it is to be distributed in return for money. Trouble.
3a,3b,4a) If you have read this far in the post you should have read section 3. "For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable." All modules means ALL modules. Trouble.
4b) Same as the answer to Question 1, if bison is something normally distributed with the major components of the OS you are ok, otherwise trouble.
5) Dynamic linking is hard.. let's go shopping! My best understanding of this case is that you are not in trouble. (You could be in trouble if the original non-GPL object had not existed, but since it does exist I think the fact that a plug-compatible GPL equivalent object exists is irrelevant. I figure I have a 50% chance of being wrong, so someone better correct me.)
I know it's hard to understand licenses and official fine print like that, but it is still important to read them. :-)
(I didn't get my Visor for its USB connectivity anyway. Got a serial cradle immediately for use with Linux. But wireless is tempting. Hmm, can people port/clone classic multiplayer games like M.U.L.E. now?)
Thus it could be the case that lute-playing is independent of profession, and one would still find that 2/3 of one's lute-playing acquaintances were computer geeks.
* extrapolating from my weekday environment, I conclude that only 12% of the world's population is female, therefore the need for gender-neutral pronouns is much less than previously supposed.
Suppose that I write some pretty decent comments every time I write a new function. Then some night around 2am when I'm adding some little feature or stalking a bug, I don't remember to change the comment along with the code. Now the comment may be worse than nothing - it could be actively misleading. Even if no one else ever reads this code, I am going to be pretty ticked at myself when I try to figure out what is going on in this function six months later.
This is why undisciplined people like me would welcome newbies to look over the code and say "Yo, I think you have some documentation rot here, let me fix that." We have plenty of comments in our code. It's just that half of them are wrong.
Also, never drink and code, but that's another story entirely.