That would imply that I should have the right to use emacs in the original environment on which Stallman wrote it. Which I certainly do not
Actually, assuming you have the rights to such a platform, then yes, you should have the right to use emacs in that environment.
Or are you saying that if you had a PDP-6 with ITS, you would expect not to have the right to run emacs, or to be restricted to a particular, vendor-approved version of emacs? If so, you don't know your history at all -- ITS source code is available on the web, such as it is. (If I remember right, the reason ITS was ditched was that it was not portable, and large chunks were assembly.)
Or wait -- you were assuming that a modern distribution of emacs must be able to run on the PDP-6? Or that it should come with a PDP-6 in the box? In other words, it's not enough that you get the software for free, you want hardware, too?
It would also imply that GPL software is linked to a particular hardware/OS combination, which certainly is not the case.
Not at all. All it means is that if you distribute a particular hardware/OS along with a chunk of GPL software, you may not use the hardware/OS to artificially restrict which version of that GPL software will run.
It does not mean that you cannot port software at all. How the fuck did you arrive at that conclusion, troll?
It just means that if you port it, you can only port it to a platform that allows you to run modified versions of that GPL'd software. And this can only possibly apply to someone who manufactures hardware/OSes, because anyone who is not the manufacturer, yet managed to do the port, implies that the platform is open enough to allow any GPL3 software that anyone can port.
It's actually insanely straightforward, if you understand the GPL and LGPL v2. For instance, the LGPLv2 requires that you either dynamically link with the LGPL'd library, or provide the object files (at least) to your own software -- thus ensuring that your software will work with any version of the LPGL libraries it uses, assuming a compatible API.
I predict GPL 4 within 5 years, and it'll be even more coercive and restrictive.
I'll bite. What about any version of the GPL is the least bit coercive? And what about the GPLv3 is at all restrictive, unless your name is Tivo?
RMS may be a zealot, but he's not the one trying to force you to use The One OS with The One Version of PowerDVD in order to play hi-def content.
Never going to happen, even if I wanted to. I don't own the source of everything I link to.
Right, so use LGPL.
In any case, that comment was not meant for you, personally.
As far as I can tell, the hardcore RMS philosophy is that producers of proprietary software, like me, are evil bastards; and that losing our contributions is a small price to pay for causing us inconvenience.
You're half right. They do believe that proprietary software is evil. That doesn't necessarily make you an evil bastard, it just means you do something evil.
The other half is not true. The philosophy has nothing to do with you, and everything to do with what they consider to be theirs. And yes, it is also about tempting you to go entirely open, but the core point of it is, if proprietary software is evil, they don't want to be a part of it.
Personally, I probably wouldn't mind my stuff being used in proprietary software, especially if it was under LGPL, but maybe even under BSD. There are other things that bother me, though -- like DRM systems and voting machines. In any case, I consider it possible to develop evil stuff that is proprietary -- spyware, for instance -- and relatively impossible to do that with open source. Specifically, if I GPL something, I can be sure that it will never be used directly in a piece of spyware that even attempts to be legal.
Sorry, but MY hardcore philosophy says: No karma if you require a return; Only altruists need apply.:)
So I can get your software for free? After all, paying you cash would be a "return", wouldn't it?
I suggest you might get more money if you did it that way. You get nothing (per reader) if people borrow the book from a friend or the library. Give it to them as a free download, and more people might read your stuff and become the sort of fans who will want a nice hardcover for their bookshelf.
I think I'm much better off charging money for it up front, in every form. I'll sell you a cheap, un-DRM'd copy, though -- and distribute it with the book -- and look the other way for people to download it.
The reason is, I find that the younger generation has less of a problem with reading something on a computer than, say, the Baby Boomers. But people do occasionally tend to "legitimize" the things they like, especially if they know their copy is illegal.
In any case, I ask again: Are you doing this with your software?
Never going to happen, even if I wanted to.
You are starting to get the picture, though, right?
It's not intended to be an assault against proprietary software. It's a deal, a contract, and a way to promote free software.
When they protest something, they do it openly and directly -- DefectiveByDesign, for example.
No GPL charting library will ever possibly be so amazing that we're going to be motivated to open-source all the market analysis stuff just to use it.
Maybe an LGPL charting library, then, although I realize that the hardcore RMS/GNU philosophy is that LGPL is only for situations where competing libraries exist.
I can see that, though in my world-view giving unknown strangers a free ride is a bonus; it's good for my karma or something.
I just rest easier if karma is inherent in the system. Kind of like BitTorrent, actually.
You can say "Feel free to use this, and if you make improvements, I'd love to hear about them!", or you can say "Feel free to use this and if you make improvements, you must give me the source to them and any other software you link this with or I'll sue your ass!"
Let me put this another way: I occasionally toy with writing fiction. I haven't gotten anything published yet, but if I did, I know I would prefer not to publish it under "Feel free to read this, and if you like it, I'd love for you to send me some money!"
Don't think of it as a threat, think of it as a deal. Give me the source [and anything you link it with, if GPL instead of LGPL], and you get all this stuff for free!
While I could keep these to myself, cackling evilly, there's no reason to since this code is obviously not the core of my business in any case.
Except that many organizations don't, not because they're evil, but because they have a corporate policy centered around fears of others "stealing" their stuff (there's that word again), and maybe because their lawyers tell them it's easier not to.
The nice thing about the GPL here is, if there's something sufficiently attractive under the GPL, I've seen corporations decide it simply makes sense to release all their stuff GPL'd. This may or may not apply to you -- it's happened to me more with Drupal, for instance, where none of the software I wrote was the core of the company's business.
And as a developer, I find myself much more likely to license something as GPLv3. I can always relicence it later, if I keep ownership of the code, but for now, it means no one gets a free ride. BSD may make it more likely in some practical or ethical sense that I get compensated somehow, but GPL ensures that I get your code or you don't get mine, and I'm OK with that.
We've never paid a dime to enhance any thing GPL, simply because we can't use it.
I'm confused. Not even indirectly?
Specifically: Not even the way Oracle pays for Linux development, so they can ship their product for Linux? There are other ways to use something without directly incorporating the code into your own product. Even LGPL, for instance.
People think of the GPL as ensuring they'll get compensated if others use their code. But that's not really its goal
I have to say, I agree with that assessment of Zelda. We don't judge whether something is a masterpiece based on whatever era it was made it. We judge whether it's a masterpiece by whether it still amazes us, and challenges us to make anything even close to that good.
For example: Gutenberg's printing press was a marvel of technology, for its time. But no one even uses the same class of technology anymore, and no one travels from around the world to see it. Macchu Picchu, on the other hand, is impressive enough that even if we have the technology to do something similar now -- and I'm not convinced of that -- the artwork and the mathematics is still impressive, and really unmatched anywhere else.
Ocarina of Time was such a game. Mario Bros, at least to me, was not.
Wolfenstein 3-D also lacks gameplay, which is why people who like old-school games still play Doom.
Also, technology does, in fact, allow new styles of gameplay. Quake 3 would not have been possible on the Doom engine. Headshots still aren't done on Quake 3.
Seriously, football strategy can be sophisticated? Soccer? You're trying to get this round thing into that big square thing. Football? This oddly-shaped thing goes to that end of the field.
But you're right, actually. Football strategy is sophisticated. And so is paintball strategy, and Counter-Strike strategy, and even Halo strategy.
What makes their capture points more "imaginary" than football, as far as strategy goes?
I'm looking for a better word there, and I can't think of one.
I suppose I should have said "Someone can ship your code in their product without compensating you in any way, even with their source code."
And I should point out that there is actually a history of fragmentation because of this. Someone releases a chunk of code under a BSD-like license, and suddenly there are 15 incompatible proprietary flavors of it, which can never be reconciled at the code level. It's already enough of a problem for Compiz/Beryl -- Compiz is BSD-ish, Beryl is a Compiz fork that went GPL.
They didnt use ODF because ODF and OOXML are not isomorphic. You cannot do a lossless translation of one to another.
Which is really their fault for not speaking up when ODF was being drafted. I believe they were on that committee.
But you didn't answer my question. I asked "Why did they extend ODF instead of inventing OOXML?" And you said "Because OOXML supports more stuff." Well, duh. I'm asking, why did they go to the trouble to roll their own, new spec, instead of improving the existing one.
In addition, what if MS wants to add a new feature to Office (ie, word, excel, etc) that is not supported in ODF.
In that case, they extend ODF. And keep in mind, ODF is intended to become a bit like ASCII or UTF8 -- to cover the problem domain so completely and generically that no other standard is needed. I believe this to be possible.
If they only use ODF, then all of a sudden they are limited to what new innovations they can create around the office apps.
Kindly name one "innovation" you can think of that requires a change of office format.
Makes it hard to differentiate your product.
Well, the point is, we're asking them to differentiate on real features, not merely the "feature" of "We control the document format; if you switch to our competition, your documents might not convert properly."
They did fully spec out what they've got.
Oh? Then tell me, exactly what does "Auto Space Like Word 95" do? Where is that listed in the spec?
In the backwards compatibility cases, you're explicitly instructed in the spec to do what they do, and upconvert it to something else.
Except the problem is, they don't upconvert. Not really. They just slap an "AutoSpaceLikeWord95" tag on it and call it "standardized".
What would be nice is if the spec provided a generic way to define how an autospace should work (for example). Then, AutoSpaceLikeWord95 would be a document-specific style, and entirely transparent to any implementation -- and the implementations would be simpler, too. The only place where the word processor would require specific knowledge of AutoSpaceLikeWord95 would be when converting to/from an actual Word95 document, which seems a lot more sensible to me than carrying the baggage around in the word processor and the spec for the next hundred years.
Oh, well put, Anonymous Coward. If I was not convinced of your argument by now, that bit certainly changed my mind. </sarcasm>
My point is simple: if Massachusetts is able to define a simple word that everyone knows the meaning of, that's had the same meaning since it was originally coined
You know, "Freedom" used to mean "All white males get to do whatever they want, at the expense of everyone else." I'm sure as hell glad that the government redefines words sometimes. Sometimes they need a new definition.
For that matter, this isn't even new. The 1913 Webster's Dictionary provides this definition for marriage: "Any intimate or close union."
that can probably trace its origins back to the origin of speech itself
You don't have to go that far to find a shockingly different definition. Polygamy was quite popular in the Bible. Is that a form of marriage you endorse?
It's simple. I should be allowed to pick any format I want to use. The government shouldn't dictate the format I want to use.
Which is not what this is about. This is about what format the government dictates for itself.
The government shouldn't dictate the formats I can communicate with them using.
Really, if they can't dictate that, what can they dictate?
You're telling me that the government should support a particular monopoly, and buy a particular product, just in case you want to send documents to them in that form?
Gee, I guess we better buy them WordPerfect licenses... hell, why not Acrobat? You might send them a PDF to edit... Or hell, maybe some Flash source code. Better have them buy Flash.
You know, why don't we all send our notes scribbled on the walls of Quake 4 maps!
I'll concede that they should accept any document in a widely-implemented industry standard. But sorry, I don't accept OpenXML as a standard, and with good reason, I think. And the binary Microsoft Office is right out.
In any case, it's moot. This legislation only concerns what format they use internally, for archival purposes. They may still accept all of the above formats, meaning this doesn't affect you at all, dipshit. It's entirely an internal affair, and it is as much their right to regulate as what dress code government employees should wear. (And if you're going to complain about that, then Christ, who's whining now?)
WebKit, which I believe is LGPL, and either way the code for it is released by Apple.
What does Apple releasing it have to do with it?
The reason it's LGPL is it's derived from KHTML. Apple does not have the original copyright, therefore, they are not allowed to re-license it. Of course, the LGPLv2 won't cause problems for them...
The hardware is not GPL'd and there is no inherent claim or promise that you can do what you want with it.
Oh please. There's certainly an expectation, or there should be.
My desktop is built from all kinds of hardware that is not GPL'd or anything like it. I don't have schematics, or even specs to some things -- my video card is run by a proprietary driver, even on Linux.
I also have a Windows partition. Yet I run Firefox on it. Certainly Windows is not GPL'd -- as a matter of fact, neither is Firefox, but it's a relevant example. I can and do run GPL'd software on Windows, and I do expect to be able to recompile it and run a modified version without Windows stepping in and stopping me.
For that matter, I can run a modified kernel and expect my nVidia driver to still work, even though the hardware is not GPL'd. It may not work anyway -- after all, the nVidia driver is buggy enough with the vanilla kernel -- but nVidia won't actually stop me from trying.
You have the source code and you are entitled to rewrite the Apple firmware to do whatever you want, with whatever means of hacking. Of course this means that it will any longer work with Apple's system.
Why shouldn't it?
An it makes no logical sense to tie the software copyright to a particular piece of hardware.
The GPLv3 manages to prevent Tivoisation without tying the software to a particular piece of hardware. It simply requires that if you distribute that software, you must provide a way for the user to run modified versions -- including, for instance, if you distribute it with some hardware, your hardware must support running modified versions, with no restrictions not already placed on the original version.
It's all kind of like how the LGPL prevents a sort of software tivoisation -- you may link against this library, but only if you either do it dynamically, or provide at least all of the object files to your program, so that users may relink them statically -- and it explicitly says that your program must still function with another version of that library, providing the API is compatible.
After all you have the source code and can make it run on ANY piece of hardware, why should the license apply to an unforseen piece of hardware.
It doesn't. It only applies to the software.
What this says is, Apple can distribute whatever hardware they want. We can do nothing about this. However, they cannot distribute our code with their hardware, unless said hardware provides the freedoms we ask for.
The issue here seems to be that there is some implicit requirement that hardware vendors make it EASY to hack their hardware
I'm sorry, no. I have no desire to solder an iPhone. This is NOT about hardware hacking.
It's about software hacking. I'd like very much to be able to modify the software on an iPhone.
And I see a lot of BSD-license zealots making similar arguments. Let me make it clear for anyone who doesn't like the terms of the GPL, even the GPLv3: It's take it or leave it. We're not forcing you to change your hardware, or release your source code, or anything of the sort. You're free to lock down your product however you want. You're just not free to do it using our code.
It's simple: If you want people to write you software for free, there's a price. You have to give back.
Linus has decided that the GPLv2 is enough, because the only freedom he wants is freedom to read the source.
But the intent of the GPL is not only to give you freedom to read the source, but freedom to modify it and use it in its intended environment. That means not only giving us Webkit source code, but letting us load our custom versions of Webkit onto the iPhone.
For example, I am not free to kill someone. However, I consider it a good thing that no one is free to kill me -- and that increases my own freedom.
I consider the GPL to be similar to that. It always has to be a tradeoff, but it's a lot closer to freedom than most other licenses. BSD could be seen as pretty much absolutely free, but with a price -- someone can steal your code -- someone can kill you.
People are making a mountain out of a molehill based on the corner case of importing a document from wordperfect of many years ago
Except it's not a corner case. The spec is absolutely FILLED with this kind of shit. And this is exactly the kind of thing that would cause problems when trying to switch to an alternate implementation: "Look, we tried OpenOffice, but it screwed up the formatting on some of our documents."
For all intents and purposes its open, people are just nitpicking over the fact that importing files from long ago and having the description for how a few obscure formatting issues should be handled is a little vague.
"A little vague" might be OK. Except it's beyond vague. It's non-existant.
If they're really about open standards, answer me this: Why didn't they use ODF? Or, failing that, why didn't they create a better standard, or fully spec out the one they've got?
If they can manage to define "marriage" as "a union between a man and a man or a woman and a woman - oh yeah, or that other one, man and woman"
Isn't that about choice? No one's forcing you to marry either a man or a woman.
But even if you are going to be that closed-minded, WTF does marriage have to do with office formats? (I can just see you sitting there sipping a beer... "Yyyep... *burp*... aah... y'know, that them there Open shit's fer faggots.")
why would you be surprised that they're forcing a format on anyone doing business with them
Except they're not. From what I understand, this only reflects what format they would use internally for archiving, which means that it's as irrelevant to anyone doing business with them as whether they use FAT32 or NTFS on their workstations.
while claiming it improves choice?
Well, it does. More on that in a bit...
And forcing OpenOffice at that.
Except it's not. It forces you to choose between OpenOffice, KOffice, AbiWord, GNUmeric, StarOffice, NeoOffice, ajaxWrite, Google Docs, TextEdit, TextMaker, Zoho Writer, eZ publish, Scribus, Visioo Writer, EditGrid... hell, I even wrote a ruby script to read OpenDocument. Didn't take that long.
And if you're that attached to Microsoft Office, there is a third-party project to add ODF support to it, also.
Oh noes, how restricting to be forced to use only those programs! Please. With OpenXML, you get MS Office, Gnumeric (very limited), and maybe OpenOffice. And if you've read the spec, it seems doubtful anything but MS Office can implement it fully.
So tell me again how this is forcing anything on anyone? Really, it's funny. Just as much of a laugh as the idea that gay marriage hurts anyone.
It can be statically linked, provided you distribute your program as object files (not necessarily source code). I do not know if it places any requirements on the format of said files, though (or the compiler used).
Here's a question: Does this protect from a certain bullshit kind of Tivo-ization, where the library is loaded dynamically by the program, but only after the program runs a checksum on the library?
Assuming you already have Vista. They no longer offer free XP 64-bit to 32-bit users, and when they did, the deal sucked. They nuked your 32-bit license upon installing 64-bit, so if you went to 64-bit and didn't have a driver for something, you couldn't downgrade.
If they're doing that again, thanks, but I'll stick with XP.
Apple's universal binaries are cool but there is the tradeoff of having binary data for absolutely everything in formats for 2-3 architectures.
Well, what's cool is, generally developers actually at least have to compile stuff for all platforms. And the tradeoff is close to nil -- the binaries aren't what takes the most space -- although a part of me is glad I'm using Linux and wasting almost no space on that.
So, you're going to make the cheat developers jump ahead. Instead of injecting their software into the wall code, they're going to have to inject their software into the anticheat code, to pass keystrokes into it as it shoots the guy down the alley..
Except without access to the wall and the guy, it won't know where to shoot.
Why waste money with this? Blizzard's Warden is pretty robust from an anticheat stand.
That's Blizzard's. I somehow doubt that it would work for other games, and if it would, you'd have to license it from Blizzard. And it causes lag. And really, for what -- so they can stop people from auto-crafting and auto-hunting, which isn't possible in a good game (auto-hunting isn't, in some better MMOs).
You can clap all the DRM and encryption on this issue you want,but it won't do any good.
It kind of has, though. At least, it seems to me that cheating doesn't work the way piracy does, it works more the way antivirus does. (Which is not the best kind of security, but it sort of works.) Basically, either you'll have like two people cheating for a long time, and it's a lot of work for two people to crack the anti-cheat stuff -- their time is better spend practicing to where they don't have to cheat -- or you have 20,000 people cheating, and if that many people can find the cheat, so can your developers, and then you just screen for that specific cheat.
At the end of the day, the PC still has to keep stuff in ram, unencrypted, for the CPU to be able to run it.
Missing the point of trusted computing. Unless you have custom-manufactured RAM, it's useless to have unencrypted stuff in RAM that you can get to. And I'm not even convinced that would help.
Until Intel makes some kind of RSA type on the fly on-cpu encryption (which I'm going to patent, btw)
Except then you'll have it unencrypted in the CPU registers! Ohnoes!
Or, more realistically, the cache. But even at that level, consider that you'd need a damned good crypto chip to do RSA faster than the CPU can deal with the data. It'd probably have to be better than your CPU anyway!
Slashdot Mirror
Actually, assuming you have the rights to such a platform, then yes, you should have the right to use emacs in that environment.
Or are you saying that if you had a PDP-6 with ITS, you would expect not to have the right to run emacs, or to be restricted to a particular, vendor-approved version of emacs? If so, you don't know your history at all -- ITS source code is available on the web, such as it is. (If I remember right, the reason ITS was ditched was that it was not portable, and large chunks were assembly.)
Or wait -- you were assuming that a modern distribution of emacs must be able to run on the PDP-6? Or that it should come with a PDP-6 in the box? In other words, it's not enough that you get the software for free, you want hardware, too?
Not at all. All it means is that if you distribute a particular hardware/OS along with a chunk of GPL software, you may not use the hardware/OS to artificially restrict which version of that GPL software will run.
It does not mean that you cannot port software at all. How the fuck did you arrive at that conclusion, troll?
It just means that if you port it, you can only port it to a platform that allows you to run modified versions of that GPL'd software. And this can only possibly apply to someone who manufactures hardware/OSes, because anyone who is not the manufacturer, yet managed to do the port, implies that the platform is open enough to allow any GPL3 software that anyone can port.
It's actually insanely straightforward, if you understand the GPL and LGPL v2. For instance, the LGPLv2 requires that you either dynamically link with the LGPL'd library, or provide the object files (at least) to your own software -- thus ensuring that your software will work with any version of the LPGL libraries it uses, assuming a compatible API.
I'll bite. What about any version of the GPL is the least bit coercive? And what about the GPLv3 is at all restrictive, unless your name is Tivo?
RMS may be a zealot, but he's not the one trying to force you to use The One OS with The One Version of PowerDVD in order to play hi-def content.
Maybe an LGPL charting library, then, although I realize that the hardcore RMS/GNU philosophy is that LGPL is only for situations where competing libraries exist.
I just rest easier if karma is inherent in the system. Kind of like BitTorrent, actually.
Let me put this another way: I occasionally toy with writing fiction. I haven't gotten anything published yet, but if I did, I know I would prefer not to publish it under "Feel free to read this, and if you like it, I'd love for you to send me some money!"
Don't think of it as a threat, think of it as a deal. Give me the source [and anything you link it with, if GPL instead of LGPL], and you get all this stuff for free!
Except that many organizations don't, not because they're evil, but because they have a corporate policy centered around fears of others "stealing" their stuff (there's that word again), and maybe because their lawyers tell them it's easier not to.
The nice thing about the GPL here is, if there's something sufficiently attractive under the GPL, I've seen corporations decide it simply makes sense to release all their stuff GPL'd. This may or may not apply to you -- it's happened to me more with Drupal, for instance, where none of the software I wrote was the core of the company's business.
And as a developer, I find myself much more likely to license something as GPLv3. I can always relicence it later, if I keep ownership of the code, but for now, it means no one gets a free ride. BSD may make it more likely in some practical or ethical sense that I get compensated somehow, but GPL ensures that I get your code or you don't get mine, and I'm OK with that.
I'm confused. Not even indirectly?
Specifically: Not even the way Oracle pays for Linux development, so they can ship their product for Linux? There are other ways to use something without directly incorporating the code into your own product. Even LGPL, for instance.
What is its goal, then?
And what do you do when it does break?
(Not if. When.)
I have to say, I agree with that assessment of Zelda. We don't judge whether something is a masterpiece based on whatever era it was made it. We judge whether it's a masterpiece by whether it still amazes us, and challenges us to make anything even close to that good.
For example: Gutenberg's printing press was a marvel of technology, for its time. But no one even uses the same class of technology anymore, and no one travels from around the world to see it. Macchu Picchu, on the other hand, is impressive enough that even if we have the technology to do something similar now -- and I'm not convinced of that -- the artwork and the mathematics is still impressive, and really unmatched anywhere else.
Ocarina of Time was such a game. Mario Bros, at least to me, was not.
Wolfenstein 3-D also lacks gameplay, which is why people who like old-school games still play Doom.
Also, technology does, in fact, allow new styles of gameplay. Quake 3 would not have been possible on the Doom engine. Headshots still aren't done on Quake 3.
You've obviously never played paintball.
Seriously, football strategy can be sophisticated? Soccer? You're trying to get this round thing into that big square thing. Football? This oddly-shaped thing goes to that end of the field.
But you're right, actually. Football strategy is sophisticated. And so is paintball strategy, and Counter-Strike strategy, and even Halo strategy.
What makes their capture points more "imaginary" than football, as far as strategy goes?
I'm looking for a better word there, and I can't think of one.
I suppose I should have said "Someone can ship your code in their product without compensating you in any way, even with their source code."
And I should point out that there is actually a history of fragmentation because of this. Someone releases a chunk of code under a BSD-like license, and suddenly there are 15 incompatible proprietary flavors of it, which can never be reconciled at the code level. It's already enough of a problem for Compiz/Beryl -- Compiz is BSD-ish, Beryl is a Compiz fork that went GPL.
Which is really their fault for not speaking up when ODF was being drafted. I believe they were on that committee.
But you didn't answer my question. I asked "Why did they extend ODF instead of inventing OOXML?" And you said "Because OOXML supports more stuff." Well, duh. I'm asking, why did they go to the trouble to roll their own, new spec, instead of improving the existing one.
In that case, they extend ODF. And keep in mind, ODF is intended to become a bit like ASCII or UTF8 -- to cover the problem domain so completely and generically that no other standard is needed. I believe this to be possible.
Kindly name one "innovation" you can think of that requires a change of office format.
Well, the point is, we're asking them to differentiate on real features, not merely the "feature" of "We control the document format; if you switch to our competition, your documents might not convert properly."
Oh? Then tell me, exactly what does "Auto Space Like Word 95" do? Where is that listed in the spec?
Except the problem is, they don't upconvert. Not really. They just slap an "AutoSpaceLikeWord95" tag on it and call it "standardized".
What would be nice is if the spec provided a generic way to define how an autospace should work (for example). Then, AutoSpaceLikeWord95 would be a document-specific style, and entirely transparent to any implementation -- and the implementations would be simpler, too. The only place where the word processor would require specific knowledge of AutoSpaceLikeWord95 would be when converting to/from an actual Word95 document, which seems a lot more sensible to me than carrying the baggage around in the word processor and the spec for the next hundred years.
Fair enough. But I think there's a decent argument for your software being called "non-free", then.
Oh, well put, Anonymous Coward. If I was not convinced of your argument by now, that bit certainly changed my mind. </sarcasm>
You know, "Freedom" used to mean "All white males get to do whatever they want, at the expense of everyone else." I'm sure as hell glad that the government redefines words sometimes. Sometimes they need a new definition.
For that matter, this isn't even new. The 1913 Webster's Dictionary provides this definition for marriage: "Any intimate or close union."
You don't have to go that far to find a shockingly different definition. Polygamy was quite popular in the Bible. Is that a form of marriage you endorse?
Which is not what this is about. This is about what format the government dictates for itself.
Really, if they can't dictate that, what can they dictate?
You're telling me that the government should support a particular monopoly, and buy a particular product, just in case you want to send documents to them in that form?
Gee, I guess we better buy them WordPerfect licenses... hell, why not Acrobat? You might send them a PDF to edit... Or hell, maybe some Flash source code. Better have them buy Flash.
You know, why don't we all send our notes scribbled on the walls of Quake 4 maps!
I'll concede that they should accept any document in a widely-implemented industry standard. But sorry, I don't accept OpenXML as a standard, and with good reason, I think. And the binary Microsoft Office is right out.
In any case, it's moot. This legislation only concerns what format they use internally, for archival purposes. They may still accept all of the above formats, meaning this doesn't affect you at all, dipshit. It's entirely an internal affair, and it is as much their right to regulate as what dress code government employees should wear. (And if you're going to complain about that, then Christ, who's whining now?)
MSFT: I would like to draw the court's attention to what I am calling Exhibit $$$. I have here a large sack of money.
Judge: Well, in that case...
What does Apple releasing it have to do with it?
The reason it's LGPL is it's derived from KHTML. Apple does not have the original copyright, therefore, they are not allowed to re-license it. Of course, the LGPLv2 won't cause problems for them...
Oh please. There's certainly an expectation, or there should be.
My desktop is built from all kinds of hardware that is not GPL'd or anything like it. I don't have schematics, or even specs to some things -- my video card is run by a proprietary driver, even on Linux.
I also have a Windows partition. Yet I run Firefox on it. Certainly Windows is not GPL'd -- as a matter of fact, neither is Firefox, but it's a relevant example. I can and do run GPL'd software on Windows, and I do expect to be able to recompile it and run a modified version without Windows stepping in and stopping me.
For that matter, I can run a modified kernel and expect my nVidia driver to still work, even though the hardware is not GPL'd. It may not work anyway -- after all, the nVidia driver is buggy enough with the vanilla kernel -- but nVidia won't actually stop me from trying.
Why shouldn't it?
The GPLv3 manages to prevent Tivoisation without tying the software to a particular piece of hardware. It simply requires that if you distribute that software, you must provide a way for the user to run modified versions -- including, for instance, if you distribute it with some hardware, your hardware must support running modified versions, with no restrictions not already placed on the original version.
It's all kind of like how the LGPL prevents a sort of software tivoisation -- you may link against this library, but only if you either do it dynamically, or provide at least all of the object files to your program, so that users may relink them statically -- and it explicitly says that your program must still function with another version of that library, providing the API is compatible.
It doesn't. It only applies to the software.
What this says is, Apple can distribute whatever hardware they want. We can do nothing about this. However, they cannot distribute our code with their hardware, unless said hardware provides the freedoms we ask for.
I'm sorry, no. I have no desire to solder an iPhone. This is NOT about hardware hacking.
It's about software hacking. I'd like very much to be able to modify the software on an iPhone.
And I see a lot of BSD-license zealots making similar arguments. Let me make it clear for anyone who doesn't like the terms of the GPL, even the GPLv3: It's take it or leave it. We're not forcing you to change your hardware, or release your source code, or anything of the sort. You're free to lock down your product however you want. You're just not free to do it using our code.
It's simple: If you want people to write you software for free, there's a price. You have to give back.
Linus has decided that the GPLv2 is enough, because the only freedom he wants is freedom to read the source.
But the intent of the GPL is not only to give you freedom to read the source, but freedom to modify it and use it in its intended environment. That means not only giving us Webkit source code, but letting us load our custom versions of Webkit onto the iPhone.
For example, I am not free to kill someone. However, I consider it a good thing that no one is free to kill me -- and that increases my own freedom.
I consider the GPL to be similar to that. It always has to be a tradeoff, but it's a lot closer to freedom than most other licenses. BSD could be seen as pretty much absolutely free, but with a price -- someone can steal your code -- someone can kill you.
Except it's not a corner case. The spec is absolutely FILLED with this kind of shit. And this is exactly the kind of thing that would cause problems when trying to switch to an alternate implementation: "Look, we tried OpenOffice, but it screwed up the formatting on some of our documents."
"A little vague" might be OK. Except it's beyond vague. It's non-existant.
If they're really about open standards, answer me this: Why didn't they use ODF? Or, failing that, why didn't they create a better standard, or fully spec out the one they've got?
Isn't that about choice? No one's forcing you to marry either a man or a woman.
But even if you are going to be that closed-minded, WTF does marriage have to do with office formats? (I can just see you sitting there sipping a beer... "Yyyep... *burp* ... aah ... y'know, that them there Open shit's fer faggots.")
Except they're not. From what I understand, this only reflects what format they would use internally for archiving, which means that it's as irrelevant to anyone doing business with them as whether they use FAT32 or NTFS on their workstations.
Well, it does. More on that in a bit...
Except it's not. It forces you to choose between OpenOffice, KOffice, AbiWord, GNUmeric, StarOffice, NeoOffice, ajaxWrite, Google Docs, TextEdit, TextMaker, Zoho Writer, eZ publish, Scribus, Visioo Writer, EditGrid... hell, I even wrote a ruby script to read OpenDocument. Didn't take that long.
And if you're that attached to Microsoft Office, there is a third-party project to add ODF support to it, also.
Oh noes, how restricting to be forced to use only those programs! Please. With OpenXML, you get MS Office, Gnumeric (very limited), and maybe OpenOffice. And if you've read the spec, it seems doubtful anything but MS Office can implement it fully.
So tell me again how this is forcing anything on anyone? Really, it's funny. Just as much of a laugh as the idea that gay marriage hurts anyone.
I don't think there's a geek in the world with the sheer physical strength to work with that interface for very long.
Oh, how I wish some of my day job bosses had ever written a single line of code in their lives...
It can be statically linked, provided you distribute your program as object files (not necessarily source code). I do not know if it places any requirements on the format of said files, though (or the compiler used).
Here's a question: Does this protect from a certain bullshit kind of Tivo-ization, where the library is loaded dynamically by the program, but only after the program runs a checksum on the library?
Isn't it already? They seem to think so.
Assuming you already have Vista. They no longer offer free XP 64-bit to 32-bit users, and when they did, the deal sucked. They nuked your 32-bit license upon installing 64-bit, so if you went to 64-bit and didn't have a driver for something, you couldn't downgrade.
If they're doing that again, thanks, but I'll stick with XP.
Well, what's cool is, generally developers actually at least have to compile stuff for all platforms. And the tradeoff is close to nil -- the binaries aren't what takes the most space -- although a part of me is glad I'm using Linux and wasting almost no space on that.
Except without access to the wall and the guy, it won't know where to shoot.
That's Blizzard's. I somehow doubt that it would work for other games, and if it would, you'd have to license it from Blizzard. And it causes lag. And really, for what -- so they can stop people from auto-crafting and auto-hunting, which isn't possible in a good game (auto-hunting isn't, in some better MMOs).
It kind of has, though. At least, it seems to me that cheating doesn't work the way piracy does, it works more the way antivirus does. (Which is not the best kind of security, but it sort of works.) Basically, either you'll have like two people cheating for a long time, and it's a lot of work for two people to crack the anti-cheat stuff -- their time is better spend practicing to where they don't have to cheat -- or you have 20,000 people cheating, and if that many people can find the cheat, so can your developers, and then you just screen for that specific cheat.
Missing the point of trusted computing. Unless you have custom-manufactured RAM, it's useless to have unencrypted stuff in RAM that you can get to. And I'm not even convinced that would help.
Except then you'll have it unencrypted in the CPU registers! Ohnoes!
Or, more realistically, the cache. But even at that level, consider that you'd need a damned good crypto chip to do RSA faster than the CPU can deal with the data. It'd probably have to be better than your CPU anyway!