"If you do not provide source with the binaries, you must provide with the binaries an offer to supply the source to anyone who requests it."
It depends on the version of the GPL. For GPL3, you're incorrect:
You may convey a covered work in object code form... provided that you also convey the machine-readable Corresponding Source...[if distributed in a physical product, then] accompanied by a written offer,... to give anyone who possesses the object code... a copy of the Corresponding Source... [If you convey] the object code by offering access from a designated place (gratis or for a charge), and offer equivalent access to the Corresponding Source in the same way through the same place at no further charge.
So, there is no requirement to offer source to anyone, only those who legitimately possess the object code.
Contrast that with GPL2:
Accompany it with a written offer... to give any third party...
You're simply wrong. It becomes economically unrecoverable at present. If silver can't be cheaply mined from the ground in 20 years, and there's still a demand for it, prices will rise until a source is found. One ready source is all those bags of investment coins stashed somewhere. Silver is not "running out."
Today, we celebrate the first glorious anniversary of the Information Purification Directives. We have created, for the first time in all history, a garden of pure ideology — where each worker may bloom, secure from the pests purveying contradictory truths. Our Unification of Thoughts is more powerful a weapon than any fleet or army on earth. We are one people, with one will, one resolve, one cause. Our enemies shall talk themselves to death, and we will bury them with their own confusion. We shall prevail!
Most of those underprovisioned netbooks shouldn't run this, then. Hell, my phone has 768 MiB of RAM. Can't afford the ~$30 for a 2 GiB DIMM?
Of course, if this "benchmark" is supposed to be so you can figure out what's best for your netbook, then it shouldn't have been run on a machine with 3 GiB of RAM. How much of the footprint is in active use, and how much can be put into swap? What's the performance difference when running on a low end device? Perhaps one pulls more code into memory when more memory is available for performance reasons.
It all goes back to a simple "this 'benchmark' sucks."
Don't you think it's bad when your Win XP system needs 16 times more RAM than the Mac OS 7 minimum requirements? Or that Mac OS 7 needs more than 256 times more RAM than MS-DOS, or that MS-DOS needs 16 times more RAM than a KIM-1 running Microchess?
but since when has memory footprint been a "benchmark?" Really, we're talking roughly half a gig here, and who's running these on a system without at least 2? The system used for this has 3 GiB, so the largest user consumes just 20% of RAM - so what? If a couple hundred meg is that important, use the CLI. Perhaps there's a design tradeoff - use more RAM for faster performance. You sure can't tell from this "benchmark."
This is just an advertisement for a "look at me" blog.
Sending an email ensures that the unique info necessary to re-register gets to the correct person (unless their email account has _already_ been hacked, which they should already know about and have taken care of). And of course, anyone who was on the PSN and hasn't already changed their other passwords (assuming they reused their PSN one) is a fool.
"this strategy document will govern how the U.S. behaves on the international stage — including around big issues like internet governance and internet freedom."
Let me summarize: "We'll use a bunch of flowery language like "freedom" and "democratic," but in the end, we'll do WTF we want, while you'll bend over and like it."
What do you expect from a company who's mascot's a fucking rat?
Mickey's not a rat, but is a rodent. More interestingly, Disney supports slavery. Did you ever wonder why the dog Goofy is Mickey's peer, but the dog Pluto is his slave?
They don't own all copyrights for all code within Android. Examples include the http-apache and sqlite stuff, which Google uses under the Apache license. There's also the alsa stuff, under the LGPL. Presumably, they don't have any separate licenses for this stuff, since they include the Apache/LGPL ones in the source tree. There are many more examples. If any of those were GPL, then Google would have to release source code. The whole premise of the summarized post was that Google derived Android from code with GPL-like licenses, and therefore should have to release Android source beyond the kernel.
So your presumption that Google has copyright to all the Android code is incorrect, which invalidates your claim that "Even if the code was released under GPLv3 (and it isn't) they would be under no obligation to release the code because they own it and can do whatever they like," because they don't in fact own it.
It is of course all moot, since only the kernel is GPL, and the foreign code in Android isn't "must provide source."
If it were a derivative of a GPL'd work (it isn't), they would have to release the source code including their additions/changes, at the time the binary was publicly distributed, regardless of their holding copyright to the additions/changes.
Since he apparently can't find or read Apache's FAQ, which plainly states, with regard to their license:
It does not require you to: include the source of the Apache software itself, or of any modifications you may have made to it, in any redistribution you may assemble that includes it...
RF based WLAN is shared media, there's no way around that. Unlicensed spectrum is always subject to interference, so can't be considered reliable. So, there's limited, unreliable bandwidth to share with multiple devices. It makes zero sense to use all wireless clients, just as it makes zero sense to deploy Ethernet hubs. That is, unless you don't really care about performance or reliability.
Slashdot Mirror
"you can't place requirements on redistribution of the object code"
Not in theory, but in fact you can from a practical sense. See Red Hat.
"What would happen if everyone did it?"
Then why would you do it any differently?
It depends on the version of the GPL. For GPL3, you're incorrect:
So, there is no requirement to offer source to anyone, only those who legitimately possess the object code.
Contrast that with GPL2:
You don't know the difference between an OS platform, and an application which runs on it.
You're right. If you remove the movie studio oligarchies, then there won't be any major motion pictures; to pay for, or to pirate. Problem solved!
You're simply wrong. It becomes economically unrecoverable at present. If silver can't be cheaply mined from the ground in 20 years, and there's still a demand for it, prices will rise until a source is found. One ready source is all those bags of investment coins stashed somewhere. Silver is not "running out."
"How about silver, which projections have shown that we will be OUT OF in twenty years? "
Really? Alchemy works? What is all the silver getting transmuted to?
Lots of that around. Most 50 year old cars are made of the stuff.
And you'll see why 2011 will be like "1984."
Most of those underprovisioned netbooks shouldn't run this, then. Hell, my phone has 768 MiB of RAM. Can't afford the ~$30 for a 2 GiB DIMM?
Of course, if this "benchmark" is supposed to be so you can figure out what's best for your netbook, then it shouldn't have been run on a machine with 3 GiB of RAM. How much of the footprint is in active use, and how much can be put into swap? What's the performance difference when running on a low end device? Perhaps one pulls more code into memory when more memory is available for performance reasons.
It all goes back to a simple "this 'benchmark' sucks."
Don't you think it's bad when your Win XP system needs 16 times more RAM than the Mac OS 7 minimum requirements? Or that Mac OS 7 needs more than 256 times more RAM than MS-DOS, or that MS-DOS needs 16 times more RAM than a KIM-1 running Microchess?
Now, GET OFF MY LAWN!
but since when has memory footprint been a "benchmark?" Really, we're talking roughly half a gig here, and who's running these on a system without at least 2? The system used for this has 3 GiB, so the largest user consumes just 20% of RAM - so what? If a couple hundred meg is that important, use the CLI. Perhaps there's a design tradeoff - use more RAM for faster performance. You sure can't tell from this "benchmark."
This is just an advertisement for a "look at me" blog.
Whoosh.
Sending an email ensures that the unique info necessary to re-register gets to the correct person (unless their email account has _already_ been hacked, which they should already know about and have taken care of). And of course, anyone who was on the PSN and hasn't already changed their other passwords (assuming they reused their PSN one) is a fool.
"That's some pretty specious logic."
No more specious than that which says growing crops on your own land, for your own use is interstate commerce. (see Wickard v. Filburn)
Let me summarize: "We'll use a bunch of flowery language like "freedom" and "democratic," but in the end, we'll do WTF we want, while you'll bend over and like it."
What do you expect from a company who's mascot's a fucking rat?
Mickey's not a rat, but is a rodent. More interestingly, Disney supports slavery. Did you ever wonder why the dog Goofy is Mickey's peer, but the dog Pluto is his slave?
This is Facebook, so the two factors are username and password.
They don't own all copyrights for all code within Android. Examples include the http-apache and sqlite stuff, which Google uses under the Apache license. There's also the alsa stuff, under the LGPL. Presumably, they don't have any separate licenses for this stuff, since they include the Apache/LGPL ones in the source tree. There are many more examples. If any of those were GPL, then Google would have to release source code. The whole premise of the summarized post was that Google derived Android from code with GPL-like licenses, and therefore should have to release Android source beyond the kernel.
So your presumption that Google has copyright to all the Android code is incorrect, which invalidates your claim that "Even if the code was released under GPLv3 (and it isn't) they would be under no obligation to release the code because they own it and can do whatever they like," because they don't in fact own it.
It is of course all moot, since only the kernel is GPL, and the foreign code in Android isn't "must provide source."
If it were a derivative of a GPL'd work (it isn't), they would have to release the source code including their additions/changes, at the time the binary was publicly distributed, regardless of their holding copyright to the additions/changes.
"global warming will disrupt the economy"
I'm investing in temperature futures, you insensitive clod.
"I've seen chips use tools at the zoo."
Yeah, but they had help from the fish.
"Actually, yes, it is illegal to leave keys in your car ignition."
What state are you referring to? Please provide a legal citation to support your claim.
SO, Not Yet.
RF based WLAN is shared media, there's no way around that. Unlicensed spectrum is always subject to interference, so can't be considered reliable. So, there's limited, unreliable bandwidth to share with multiple devices. It makes zero sense to use all wireless clients, just as it makes zero sense to deploy Ethernet hubs. That is, unless you don't really care about performance or reliability.