1) If it's a standard C instruction, then any C compiler can handle it. 2) If it's a routine in that system's libc, then as libc would be shipped with the OS, there's no problem (e.g. GPL apps linking with Solaris's libc). 3) If it's a routine which is not in a standard library shipping with the OS, then the source must be provided.
You missed this part:...unless that component itself accompanies the executable.
Since the compiled executable would contain the binary block, the distributor would need to supply the source *even if* that binary block was part of the 'standard' platform.
A compiler that inserted a binary block into an executable *would* be in violation of the GPL, since it specifically states that:
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. However, as a special exception, the source code distributed need not include anything that 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, unless that component itself accompanies the executable.
Since the binary block would accompany the executable, source code would have to be distributed in order to satisfy this requirement.
They're under no requirement to release their toolchain, so yes, in theory, they could port GPL'd software to another language for which only they have a suitable compiler, and then release only the source code to the software.
Sort of ignores the fact that to do a port to a language that's different enough that no-one else can compile it using extant compilers would be such an effort that they might just as well write the whole thing themselves from scratch.
*Sigh*... no, they don't. First, go and read the damn license, OK?
The problem this time round is that Linksys have released some kernel source code, but the code they have released is not sufficient to build a working kernel (minus the binary drivers for their wireless stuff, which don't have to be released under the GPL because of Linus's binary module exception).
Effectively, they've just given back the community exactly what the community gave them, without contributing the changes they made that the GPL requires be released under the GPL.
- 1 first post - 5 complaints about the submitter's lack of English skills and/or the editors' failure to correct same - 2 comments on how fluoride is not good for you - 1 comment making a double entendre about "getting stuff for free" - 1 Gentoo fanboy comment - Exactly zero comments about the article itself
I predict that these proportions will be true for this article no matter how many comments it collects.
Slashdot Mirror
Yes, strange as it may seem, I do know that after 13 years in Japan...
You're looking at it from the wrong angle...
1) If it's a standard C instruction, then any C compiler can handle it.
2) If it's a routine in that system's libc, then as libc would be shipped with the OS, there's no problem (e.g. GPL apps linking with Solaris's libc).
3) If it's a routine which is not in a standard library shipping with the OS, then the source must be provided.
It's a misspelling. It should be 'Asahi Shimbun'.
In fact, he's #defined as FP_INFINITE (by Christians anyway).
Agnostics say he's FP_NAN.
Atheists argue that he's FP_ZERO.
*sigh*
No, what I meant is, "If he's old enough, he might still use that term to describe the main component of the system".
Is that clear enough?
Depends on how old he is - the main unit was sometimes called the CPU when your memory, permanent stoarge, etc. were in separate boxes...
Pity both the shop location and his name are up there - I give him about a week before the boys in blue come calling.
You missed this part: ...unless that component itself accompanies the executable.
Since the compiled executable would contain the binary block, the distributor would need to supply the source *even if* that binary block was part of the 'standard' platform.
A compiler that inserted a binary block into an executable *would* be in violation of the GPL, since it specifically states that:
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. However, as a special exception, the source code distributed need not include anything that 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, unless that component itself accompanies the executable.
Since the binary block would accompany the executable, source code would have to be distributed in order to satisfy this requirement.
That's not a compiler; that's a translator, and it'd be getting awfully close to obfuscation (which the GPL *does* prohibit).
They're under no requirement to release their toolchain, so yes, in theory, they could port GPL'd software to another language for which only they have a suitable compiler, and then release only the source code to the software.
Sort of ignores the fact that to do a port to a language that's different enough that no-one else can compile it using extant compilers would be such an effort that they might just as well write the whole thing themselves from scratch.
*Sigh*... no, they don't. First, go and read the damn license, OK?
The problem this time round is that Linksys have released some kernel source code, but the code they have released is not sufficient to build a working kernel (minus the binary drivers for their wireless stuff, which don't have to be released under the GPL because of Linus's binary module exception).
Effectively, they've just given back the community exactly what the community gave them, without contributing the changes they made that the GPL requires be released under the GPL.
Grab the m3u file, cut and paste the URL therein to your favorite downloader, and you have an mp3 file.
No, no, it should be "I, for one, welcome our non-evil overlords."
You see? Much better.
Depends on where the mole is, I suppose... ;)
...that the first third-party UHDV production will be pr0n?
In the corner crying like a baby, that's where.
Did you even realize that italics means that it's the *submitter*, and not the *poster*, who wrote that part?
Not all management people are evil you know.
Yeah, some of them are dead.</joke>
I think you're misunderstanding the article. It was anti-spam services that got DDoSed, not spammers.
The first ten comments consisted of:
- 1 first post
- 5 complaints about the submitter's lack of English skills and/or the editors' failure to correct same
- 2 comments on how fluoride is not good for you
- 1 comment making a double entendre about "getting stuff for free"
- 1 Gentoo fanboy comment
- Exactly zero comments about the article itself
I predict that these proportions will be true for this article no matter how many comments it collects.
Would that make her J-Floe?
Well, no... Bluetooth is a protocol, whereas RF headsets are effectively just a transmitter/receiver pair; no protocol involved.
It mentions on the site that the video cards are intact.
What a waste...
Wireless headsets here have tended to go down the RF route rather than Bluetooth.