Oh, I forgot. Borland does, and always did. Others don't.
I'm too lazy to type Sun's license agreement, but apparently it seeks to regulate distribution of compiled binaries only if the binaries happen to use "Tools.h++". This makes perfect sense to me. Tools.h++ is a library distributed as source code (by Rogue Wave I think). Software that uses Tools.h++ is a derivative work of Tools.h++. Compilers have nothing to do with it.
Oh well. Apparently all those companies that release closed-source stuff for Linux don't have legal departments, or use some super-sikr1t non-GPLed compiler. And the whole *BSD family is screwed, too.
Nice FUD (or else your lawyers are incompetent, to put it mildly). Compiler output is not a derivative work of the compiler.
Imagine that it is. Then no one can use any compiler that does not give an explicit permission to copy compiled binaries in its license. Now check out any compiler license from Microsoft, Borland, or Sun. Is the permission there? Check out your Notepad and Word licenses too. Do they give you an explicit permission to copy text files that were processed with these tools? No? You're SOL. --
Beer (?), n. [OE. beor, ber, AS. beór; akin to Fries. biar, Icel. bjrr, OHG. bior, D. & G. bier, and possibly E. brew. 93, See Brew.]
1. A fermented liquor made from any malted grain, but commonly from barley malt, with hops or some other substance to impart a bitter flavor. Beer has different names, as small beer, ale, porter, brown stout, lager beer, according to its strength, or other qualities. See Ale.
Ariane 5, a $500,000,000 rocket, crashed because of a software-related problem. And the software was not from M$. Far from it.
Please don't flame me on this. I know why Ariane 5 crashed. I've read the report. You might just as well say it was a hardware-related problem, or a harware-to-software-interface-related problem. Whatever. --
This is pretty interesting. Consider this situation.
Suppose I write a piece of free software and upload it to some.org FTP server. Suppose, for the sake of argument, I forget to disclaim liability in my license.
Somebody else fixes 2 killer bugs and 14 minor ones, introducing 6 more in the process.
Yet somebody else adds two new features (and some new bugs to go with them).
Somebody from Microsoft needs this software to do his job. He downloads it, finds a minor bug, fixes it, introduces another one, doesn't put his name anywhere, and uploads the new version (and logs at the.org remember it was uploaded from a Microsoft computer).
Yet somebody else rewrites everything from C to C++, leaving all existing bugs intact and of course introducing some more.
Finally, someone else submits a one-line fix that allows the whole thing to be compiled with an ancient version of g++ on some piece of obscure hardware.
Again, for the sake of argument, suppose that up-to-date binaries for x86 Linux and Windows 9x are generated automatically by said.org for each source release.
Now the only logical conclusion that I'm able to make is that everyone involved is equally liable. Which, frankly, doesn't make any sense whatsoever to me. Probably laws just don't obey the rules of logic. --
And pray tell me, who's the bearer of the primary liability in the case of, say, oh, Linux kernel?
If you buy commercial software it's a clear cut. The bearer is whoever gets your money.
Now, if you buy Red Hat Linux x.y, the entity that gets your money is Red Hat, Inc. So one may have luck suing them. Indeed, if you decide to sue someone because your Microsoft software is faulty, that "someone" is likely to be Microsoft, not individual developers(s) that produced faulty code.
If you download that same piece of software from freefsckingdownloadsofthefsckingday.org, it's not clear at all why Torvalds/Cox/whoever would suddenly become bearers of primary liability. It looks like the operator of the aforementioned.org will bear liability in this case, as it's not really different from Red Hat.
And said operator is, by the way, a $50/year nonprofit.
This "solution" does not work for people that get real money in exchange for their wares (either directly or through a subsidiary or maybe through the Satan himself), because the buck must stop somewhere. If there's no buck, there's no place to stop either.
I propose a solution for independent developers who want to release free software without risking lawsuits, not for the people who're in the business of free software. I didn't state it in the very beginning, so I'm a dumbfsck, Ok? --
Create a non-profit organization or two (or two hundred, for that matter) which will assume all legal responsibility for (some part of) free-{speech,beer} software. Such an organization will have a yearly budget of about, oh, $50.00 (I feel generous today). So some Joe Schmoe decides to sue such an organization, and it wents belly up. Who fscking cares? Another one will be created tomorrow.
If I were an American, I would do it today. But I'm not, so I leave this task to you.
One small personal request: if you do create such an organization, please name it FUNT. I can't make up a phrase that FUNT would stand for, so please do it yourself if you feel like it. May seem strange to you, but don't worry, it makes a perfect sense. --
Not even that I'm afraid. Loading the software into RAM is copying and not fair use, and needs license. Or so the courts in the US want us to believe. You are right wrt the book though. --
If you have a piece of code with no license attached to it, you can do whatever you want with it.
Nope. Suppose a drunk MS developer have lost a CD-R with the next version of Windows on it (with sources), and you've found it. No license on that piece of software. Can't use it in any lawful way though (except as a coaster) because it's still copyrighted by Microsoft. Even simpler: you buy a book. There's no license. There's still copyright. --
By releasing your Goof 0.3 you don't give up your copyright, so Microsoft cannot accuse you of using "their" software. It's still yours. They might have a license to use it, and if GPL is struck down, they might be able to get away with releasing binary-only versions of it. But it will not affect your Goof 0.3.5, unless you incorporate their code into it.
GPL might one day become invalid, but it doesn't mean all software released under it will suddenly become public domain. It will become available under a BSD-like license (GPL with your obligations struck down), or not available under any reasonable license at all. Or so I understand it. --
See, GPL allows you to do things that otherwise would be considered copyrigt infringement. Suppose GPL is struck down. What happens with you, the customer? You've left with a piece of code with no license attached to it. You can use it as is, or throw it away, but you can't make a copy or modify it. So if you want to make copies, you download the new version under a fixed license.
Unless of course the court decides that provisions of GPL that allow you to make copies remain intact, while nullifying your obligations. I however don't believe it can be done. --
IANAL Now there's a whole load of courts in the US alone, and what one of them says is not necessarily picked up by others. Plus, for any given loophole discovered by the court, a patch could be developed. If the court declares the whole concept of GPL invalid, it's another matter. But how on earth can it do that? --
They were absolutely sure that people in Iran can't type. Now they think that same people can't run od(1). And they call themselves Intelligence agency. How cool. --
I have to repeat this over and over: Don't confuse Microsoft with Microsoft Research. This is like confusing, say, US Government with NASA.
Simon Peyton Jones and other nice people from MS Research release a Haskell compiler called GHC. It is open-source, multiplatform (Wintel, Linux and many more), and uses GCC as a backend. Did the previous sentence surprise you? They designed C-- because C does not serve well as a backend for a functional language like Haskell.
In most dictionaries words are already divided into syllables and annotated with their grammatical roles. To write a good haiku generator, one needs to add a list of associations to each word. Then have a (simplified) haiku grammar that goes like this:
and start generating. Make sure that syllable count is right, and words are more or less associated with each other. This is of course easier said than done. --
Slashdot Mirror
I'm too lazy to type Sun's license agreement, but apparently it seeks to regulate distribution of compiled binaries only if the binaries happen to use "Tools.h++". This makes perfect sense to me. Tools.h++ is a library distributed as source code (by Rogue Wave I think). Software that uses Tools.h++ is a derivative work of Tools.h++. Compilers have nothing to do with it.
Oh well. Apparently all those companies that release closed-source stuff for Linux don't have legal departments, or use some super-sikr1t non-GPLed compiler. And the whole *BSD family is screwed, too.
--
One word: DeCSS.
--
Imagine that it is. Then no one can use any compiler that does not give an explicit permission to copy compiled binaries in its license. Now check out any compiler license from Microsoft, Borland, or Sun. Is the permission there? Check out your Notepad and Word licenses too. Do they give you an explicit permission to copy text files that were processed with these tools? No? You're SOL.
--
I protest against being moderated offtopic! This is about fireworks, and I just blew up his argument! Justice! I demand justice! Oh, whatever.
--
--
Please don't flame me on this. I know why Ariane 5 crashed. I've read the report. You might just as well say it was a hardware-related problem, or a harware-to-software-interface-related problem. Whatever.
--
This is a DoS attack. Sue him.
--
OTOH, what will you do if you see your credit card number posted on a site hosted in, uh, say, Kazakhstan?
--
- Suppose I write a piece of free software and upload it to some
.org FTP server. Suppose, for the sake of argument, I forget to disclaim liability in my license. - Somebody else fixes 2 killer bugs and 14 minor ones, introducing 6 more in the process.
- Yet somebody else adds two new features (and some new bugs to go with them).
- Somebody from Microsoft needs this software to do his job. He downloads it, finds a minor bug, fixes it, introduces another one, doesn't put his name anywhere, and uploads the new version (and logs at the
.org remember it was uploaded from a Microsoft computer). - Yet somebody else rewrites everything from C to C++, leaving all existing bugs intact and of course introducing some more.
- Finally, someone else submits a one-line fix that allows the whole thing to be compiled with an ancient version of g++ on some piece of obscure hardware.
Again, for the sake of argument, suppose that up-to-date binaries for x86 Linux and Windows 9x are generated automatically by saidNow the only logical conclusion that I'm able to make is that everyone involved is equally liable. Which, frankly, doesn't make any sense whatsoever to me. Probably laws just don't obey the rules of logic.
--
If you buy commercial software it's a clear cut. The bearer is whoever gets your money.
Now, if you buy Red Hat Linux x.y, the entity that gets your money is Red Hat, Inc. So one may have luck suing them. Indeed, if you decide to sue someone because your Microsoft software is faulty, that "someone" is likely to be Microsoft, not individual developers(s) that produced faulty code.
If you download that same piece of software from freefsckingdownloadsofthefsckingday.org, it's not clear at all why Torvalds/Cox/whoever would suddenly become bearers of primary liability. It looks like the operator of the aforementioned .org will bear liability in this case, as it's not really different from Red Hat.
And said operator is, by the way, a $50/year nonprofit.
This "solution" does not work for people that get real money in exchange for their wares (either directly or through a subsidiary or maybe through the Satan himself), because the buck must stop somewhere. If there's no buck, there's no place to stop either.
I propose a solution for independent developers who want to release free software without risking lawsuits, not for the people who're in the business of free software. I didn't state it in the very beginning, so I'm a dumbfsck, Ok?
--
Create a non-profit organization or two (or two hundred, for that matter) which will assume all legal responsibility for (some part of) free-{speech,beer} software. Such an organization will have a yearly budget of about, oh, $50.00 (I feel generous today). So some Joe Schmoe decides to sue such an organization, and it wents belly up. Who fscking cares? Another one will be created tomorrow.
If I were an American, I would do it today. But I'm not, so I leave this task to you.
One small personal request: if you do create such an organization, please name it FUNT. I can't make up a phrase that FUNT would stand for, so please do it yourself if you feel like it. May seem strange to you, but don't worry, it makes a perfect sense.
--
Not even that I'm afraid. Loading the software into RAM is copying and not fair use, and needs license. Or so the courts in the US want us to believe. You are right wrt the book though.
--
--
GPL might one day become invalid, but it doesn't mean all software released under it will suddenly become public domain. It will become available under a BSD-like license (GPL with your obligations struck down), or not available under any reasonable license at all. Or so I understand it.
--
Unless of course the court decides that provisions of GPL that allow you to make copies remain intact, while nullifying your obligations. I however don't believe it can be done.
--
IANAL
Now there's a whole load of courts in the US alone, and what one of them says is not necessarily picked up by others. Plus, for any given loophole discovered by the court, a patch could be developed. If the court declares the whole concept of GPL invalid, it's another matter. But how on earth can it do that?
--
Wait, it should be split in two very soon.
--
They were absolutely sure that people in Iran can't type. Now they think that same people can't run od(1). And they call themselves Intelligence agency. How cool.
--
The nice point is that M$(R) allows them to continue their work on something that is open source and runs on Linux.
--
Simon Peyton Jones and other nice people from MS Research release a Haskell compiler called GHC. It is open-source, multiplatform (Wintel, Linux and many more), and uses GCC as a backend. Did the previous sentence surprise you? They designed C-- because C does not serve well as a backend for a functional language like Haskell.
Haskell is a cool language BTW.
--
Oh my. You are right. Delta is derivative of theta. How dumb of me.
--
You mean delta function, don't you?</nitpicking>
--
Now censor this.
--
haiku ::= sentence sentence sentence ::= noun-group verb-group ::= noun | adjective noun-group | ... ::= verb | adverb verb-group | ...
sentence
noun-group
verb-group
and start generating. Make sure that syllable count is right, and words are more or less associated with each other. This is of course easier said than done.
--
Can a program write
a good, poetic haiku?
When hell freeze over.
--