Latest news on the Debian packaging effort can be found here. There is also a mailing list, debian-openoffice, if you're interested in helping with this project.
There are other forms of mutation than genetic. A mutation is simply a change (with an implication of outside forces and permanence). Granted, the usage was a bit ambiguous -- years of bad scifi, with "mutant monsters", and years of annoying arguments with annoying religious freaks has caused other uses of the word "mutate" to fade from the public eye.
But to claim that this change is not a mutation simply reveals the speaker's own ignorance of the English language. The word "mutation" is used in other technical fields than genetics, in contexts that have nothing to do with inheritance. On the other hand, the widespread confusion over the term here at/. certainly points out a problem. Journalists should strive to be clear, and apparently a lot of people are unclear on the actual meaning of "mutation".
So, the article was wrong because it used the term in a way many people don't expect, but it wasn't technically wrong.
The word mutation has other uses than in genetics. "To mutate" simply means to change. In fact, my dictionary lists "CHANGE" as a synonym for "mutation". On the other hand, it's true that common usage has become strongly tied to genetics. However this merely makes the headlines misleading, not incorrect. Which makes the parent post overrated (as well as wrong -- the change is a mutation, just not a genetic mutation.)
The deformation of crystal lattices under extremes of pressure and temperature is often referred to as mutation. Obviously, crystals don't have any form of genetics.
A loop scope "commentParserDatabaseCursor" is an idiocy.
That would tend to depend on how obvious it is what the code is doing. If you start your loop with "i = commentParserDatabase.startingCursor", then surely "i" is sufficient. If it's not obvious from context what "i" is, though, then maybe you should go ahead and use "commentParserDatabaseCursor" instead. Or maybe you just need a comment like: "iterate over the comment database".
Bottom line, programming is neither an art nor a science, but a craft that requires both good technical skills and good communications skills. If you have only the former, you may write fast, solid but unmaintainable code. If you only have the latter, you can write easily maintainable but fundamentally broken code which may be easier to toss than to fix, even though it's clear.
We have what we have to settle for because there is no other choice. ... Sorry, but neither of those options appeals to me.
That is a choice! I have used nothing but Linux for nearly three years now. Before that, I dual-booted, but between the improvements in the available software for Linux, the improvements in Wine, and my aquisition of a playstation (later upgraded to PS2), I stopped. My new machine has never ever had Windows of any flavor anywhere near it.
I'm not saying you're wrong for your choice (choice tends to be a personal matter), but the fact that you made one choice does not mean that the other choices aren't choices. They're just choices that you have rejected.
While your ballpark figures are probably in the ballpark, keep in mind that most statistics about "what browsers visit our site" are generated from the browsers agent string. I haven't used Windows in nearly three years (and never used it much), but one of my browsers claims to be IE, because that claim tends to provide a more hassle-free browsing experience. Another claims to be "telnet/plan9", just to shake up anyone who actually looks at the minority hits, and possibly remind them that what a browser claims to be is not necessarily what it is.
If there's no "innovation" or originality in free software, then why does MS complain so loudly about the GPL? After all, if the code's so boring and derivative, why do they care what license it uses?
"Your code sucks, and I'm really upset that you won't let me use it." Sorta reminds me of the old joke: "the food at this restaurant is lousy, and the portions are too small."
Ok, MS complains about the GPL, because it doesn't allow them to use the code in their proprietary, commercial products. Then they release their own code under this new license that doesn't allow people to use the code in their commercial products, period.
So, tell ya what, MS. How 'bout if I dual license my software under the GPL and your new "shared-source" license. Your choice. If you find the GPL too evil, you can choose to use my code under the terms of your wonderful shared-source license. That means no commercial use, but surely you don't care about that, or you wouldn't be promoting this license in the first place.
Unless you're being hypocritical, surely you'll applaud this move on my part. And nice guys like you would never be hypocritical, would you?
Crazies, people with tinfoil hats, astroturfers, arrogant rich bastards, clueless rich twits, and, of course, people with stolen credit cards.
Overall, I suspect that the total volume of posting would drop dramatically, but I'm not sure that the percentage of igorant, offtopic and clueless posts would drop all that much.
Well, again, IANAL, but as I understand it, the reason that fair use is not technically considered a "right" is that it doesn't have hard and fast boundaries. You are allowed to copy some material for, e.g. review purposes, but there are no precise rules on how much you can copy for these purposes (AFAIK). If the copyright holder takes you to court, then the court decides if what you did was fair use or not.
Fredric Brown once wrote the world's shortest horror story. It was two sentences long. How much of that should a reviewer be allowed to quote in his review? Fair use is fuzzy, and rights are never fuzzy.
In any case, I'm on your side. I realize this is quibbling. But I'm trying to be consistent. I loudly decry claims that copyright violation is theft. If I'm going to accuse the other side of being incorrect or inconsistent (when they refer to copyright violators as "thieves" and "pirates"), then I have to be correct and consistent myself.
Technically, I believe that's correct: fair use is a defense, not a right. In other words, it's up to you to show that your copying falls under fair use. (I think -- IANAL.)
However, as long as we're being technical, copyright violation is not theft, either. Funny how Disney is so hung up on the former technical quibble, but is perfectly happy to ignore the second (more-than-a-quibble, IMO).:-)
Besides, if you do migrate away from Microsoft, emacs doesn't have a built-in spell checker, AFAIK.
Not only does emacs have built-in spell checking, but it can use the spell checker to do word completion.
And what does LaTeX have to do with a secretary's ability to use staroffice/openoffice/koffice/abiword/etc.? The only semi-reasonable argument I've seen for using MSOffice, rather than the actual alternatives like this is that MS file formats are a sort of de-facto standard. And that argument is undermined by the incompatibilities MS introduces with each new release (as you mention).
This will have a huge effect in the long run, since crypto isn't just used for encryption.
I think you're missing Mr. Coward's point. Crypto was already available to Debian users. To most, this change will be all-but-transparent, due to the magic of apt-get.
Yes, but it's the US gummit doing the restricting. Nor is this issue specific to Debian: any distro which includes crypto-enabled software (mozilla, galeon, even mutt) is going to have the same issues. If you want to sell a modern, non-crippled Linux distro of any type from the US, you're either going to have to:
a) sell only to US citizens, or b) do the paperwork.
Which makes life hard for CD distributors
Apparently, the US gummint doesn't care. If I were a US-based CD vendor, I'd definitely complain to my gummint, but I'm not.
and is in contradiction with the GPL.
No, the GPL has nothing to do with it. The GPL addresses copyright issues. Other legal issues, like patents and other gummint regulations, are outside the scope of the GPL.
Very nearly true. The main end-user effect will be on the bandwidth-challenged, who will find ssh and SSL-enabled versions of mozilla, galeon, mutt, evolution, and ghod knows what else on their CDs in the future. These people will end up saving a lot of download time (and possibly money if they pay by the minute for being online).
The flip-side of this is that CD vendors in the US might be slightly more reluctant to jump through the hoops necessary to distribute Debian on CD. However, the same hoops are going to be required for any other distros that include non-crippled SSL-enabled apps and the like, so I don't imagine this is going to be a major problem.
Easy to arrange. Release some code under the GPL. Get a friend to build a possibly-derivative work, using the sort of linking you want to test. Have friend release this new work under a proprietary license. Sue friend. Once a judgement has been handed down, settle out of court for some trivial sum, then repeat to test the next type of linking.
Of course, you (and friend) will probably be out-of-pocket for all court costs, and may be facing accusations of "wasting the court's time" and "filing frivolous lawsuits", but heck, you said you wanted to know, surely the information is worth spending a little money on?
Me, frankly, I don't really care, I'm happy to err on the side of caution, and save my money.:-)
No need to run BIND either. You can just make up a hosts file with your name mappings, and drop that in the appropriate places (/etc/hosts on *nix, c:\windows\hosts on a windows box).
Advantages: no BIND process running, with its overheads and potential security issues. Fewer points of failure (each machine does its own resolving,
Disadvantages: more work when something changes. You have to manually replicate the hosts file to all system whenever machines are added or removed. For anything but the smallest networks, this is a huge problem, and the reason that (almost) no-one uses this approach any more. Still for a small-enough SOHO network, the simplicity and reliability of this approach may well make it worthwile.
What's clear is that NuSphere violated the GPL and lost their rights as per article 4. What's not clear is whether (and why) article 6 no longer applies. Article 6 would seem to allow them to regain their rights by receiving a new copy of the code. The copyright holder's explicit permission would seem to be automatically granted by article 6!
I would further note that article 4 speaks of an action ("any attempt to" blah, blah) which results in automatic termination of the license. I think it's hard to argue that in the absence of an action (an actual or attempted violation), any termination of a new license obtained via article 6 would occur.
But what the heck, I'm not a lawyer or a judge, and your opinion, my opinion, and even Mr. Moglen's opinion won't matter a hill of beans if the judge disagrees.
Obtaining the code from a different source wouldn't help, as your right to redistribute has been withdrawn by the copyright holder, and applies over all copies.
Yes, but according to section 6 of the GPL, you automatically recieve a license from the original licensor each time you recieve "the Program".
I don't think "figured it out" is the right term. It wasn't actually in the original movie, and it definitely wasn't in the novel. But it was, indeed, an obvious (and IMO extremely cheesy) way they could have gone. We already had the idea that replicants might not know they're replicants (from Rachel). No need to beat it into the ground by making Deckard a replicant too.
Watching the movie for the first time, I was afraid they were going to make Deckard a replicant, and I was very glad they didn't take the easy, cheap way out.
Look at it this way -- even if you drop the original cheesy ending, if Deckard is a replicant, then he's been lied to / brainwashed / programmed, and plus, he has a kinship with Rachel, and so there's no reason for him not to save her. So, even with the original ending dropped, you can still assume he does The Right Thing(TM). If, however, she's a replicant, and he's human, then the new ending becomes far more ambiguous and powerful.
Or to put it another way, if Deckard is human, we have a powerful movie about how humans react to the unhuman. If he's a replicant, we have a cheesy Hollywood thriller about robots killing each other.
Honestly, making Deckard a replicant is a mistake that no one over the (mental) age of eight should have made. Phil Dick didn't make that mistake. Unfortunately Ridley Scott ain't so smart.
The OSI folks reject such licenses absolutely. The FSF calls them "semi-free", and admits that they're better than proprietary licenses. Yet, somehow, it's the FSF that gets called "fanatic".:-)
"The only real conditions are that you don't then try to sell it"
Then it's not open source. It's in the category which the FSF calls "semi-free", and admits is better than not-at-all-free, which Debian calls "non-free" but still puts on their servers for download, and which the fanatics at OSI simply reject as not-open-source, period.
The problems with the licensing were pointed out in what quickly became a huge flamewar within hours of the initial KDE Call For Developers on Usenet (yes, this was long enough ago that serious developers still used Usenet).
So, the core KDE team knew about the problems before they had one line of code written!
(Frankly, I don't give a crap about KDE or GNOME -- they both suck, IMO -- but revisionist history annoys me.)
Although RH (and other flavors of Linux) are commonly referred to as "distributions", calling RH a distributor is a bit misleading. Normally, a distributor is one who takes manufactured/published goods from a plant or press and physically delivers them to various retail outlets. Distibutors typically do not produce. RH produces CDs and books. Companies that produce CDs and books are normally referred to as "publishers", and this certainly seems like the most appropriate term to use for RH.
Slashdot Mirror
Latest news on the Debian packaging effort can be found here. There is also a mailing list, debian-openoffice, if you're interested in helping with this project.
There are other forms of mutation than genetic. A mutation is simply a change (with an implication of outside forces and permanence). Granted, the usage was a bit ambiguous -- years of bad scifi, with "mutant monsters", and years of annoying arguments with annoying religious freaks has caused other uses of the word "mutate" to fade from the public eye.
/. certainly points out a problem. Journalists should strive to be clear, and apparently a lot of people are unclear on the actual meaning of "mutation".
But to claim that this change is not a mutation simply reveals the speaker's own ignorance of the English language. The word "mutation" is used in other technical fields than genetics, in contexts that have nothing to do with inheritance. On the other hand, the widespread confusion over the term here at
So, the article was wrong because it used the term in a way many people don't expect, but it wasn't technically wrong.
The word mutation has other uses than in genetics. "To mutate" simply means to change. In fact, my dictionary lists "CHANGE" as a synonym for "mutation". On the other hand, it's true that common usage has become strongly tied to genetics. However this merely makes the headlines misleading, not incorrect. Which makes the parent post overrated (as well as wrong -- the change is a mutation, just not a genetic mutation.)
The deformation of crystal lattices under extremes of pressure and temperature is often referred to as mutation. Obviously, crystals don't have any form of genetics.
A file scope "i" is an abomination.
Absolutely.
A loop scope "commentParserDatabaseCursor" is an idiocy.
That would tend to depend on how obvious it is what the code is doing. If you start your loop with "i = commentParserDatabase.startingCursor", then surely "i" is sufficient. If it's not obvious from context what "i" is, though, then maybe you should go ahead and use "commentParserDatabaseCursor" instead. Or maybe you just need a comment like: "iterate over the comment database".
Bottom line, programming is neither an art nor a science, but a craft that requires both good technical skills and good communications skills. If you have only the former, you may write fast, solid but unmaintainable code. If you only have the latter, you can write easily maintainable but fundamentally broken code which may be easier to toss than to fix, even though it's clear.
We have what we have to settle for because there is no other choice.
...
Sorry, but neither of those options appeals to me.
That is a choice! I have used nothing but Linux for nearly three years now. Before that, I dual-booted, but between the improvements in the available software for Linux, the improvements in Wine, and my aquisition of a playstation (later upgraded to PS2), I stopped. My new machine has never ever had Windows of any flavor anywhere near it.
I'm not saying you're wrong for your choice (choice tends to be a personal matter), but the fact that you made one choice does not mean that the other choices aren't choices. They're just choices that you have rejected.
While your ballpark figures are probably in the ballpark, keep in mind that most statistics about "what browsers visit our site" are generated from the browsers agent string. I haven't used Windows in nearly three years (and never used it much), but one of my browsers claims to be IE, because that claim tends to provide a more hassle-free browsing experience. Another claims to be "telnet/plan9", just to shake up anyone who actually looks at the minority hits, and possibly remind them that what a browser claims to be is not necessarily what it is.
If there's no "innovation" or originality in free software, then why does MS complain so loudly about the GPL? After all, if the code's so boring and derivative, why do they care what license it uses?
"Your code sucks, and I'm really upset that you won't let me use it." Sorta reminds me of the old joke: "the food at this restaurant is lousy, and the portions are too small."
Ok, MS complains about the GPL, because it doesn't allow them to use the code in their proprietary, commercial products. Then they release their own code under this new license that doesn't allow people to use the code in their commercial products, period.
So, tell ya what, MS. How 'bout if I dual license my software under the GPL and your new "shared-source" license. Your choice. If you find the GPL too evil, you can choose to use my code under the terms of your wonderful shared-source license. That means no commercial use, but surely you don't care about that, or you wouldn't be promoting this license in the first place.
Unless you're being hypocritical, surely you'll applaud this move on my part. And nice guys like you would never be hypocritical, would you?
I mean, who wants to pay to crapflood?
Crazies, people with tinfoil hats, astroturfers, arrogant rich bastards, clueless rich twits, and, of course, people with stolen credit cards.
Overall, I suspect that the total volume of posting would drop dramatically, but I'm not sure that the percentage of igorant, offtopic and clueless posts would drop all that much.
Space is 3 K, not 0 K. Anything below that, and the universe itself starts providing a "decent heat source". :)
Well, again, IANAL, but as I understand it, the reason that fair use is not technically considered a "right" is that it doesn't have hard and fast boundaries. You are allowed to copy some material for, e.g. review purposes, but there are no precise rules on how much you can copy for these purposes (AFAIK). If the copyright holder takes you to court, then the court decides if what you did was fair use or not.
Fredric Brown once wrote the world's shortest horror story. It was two sentences long. How much of that should a reviewer be allowed to quote in his review? Fair use is fuzzy, and rights are never fuzzy.
In any case, I'm on your side. I realize this is quibbling. But I'm trying to be consistent. I loudly decry claims that copyright violation is theft. If I'm going to accuse the other side of being incorrect or inconsistent (when they refer to copyright violators as "thieves" and "pirates"), then I have to be correct and consistent myself.
Technically, I believe that's correct: fair use is a defense, not a right. In other words, it's up to you to show that your copying falls under fair use. (I think -- IANAL.)
:-)
However, as long as we're being technical, copyright violation is not theft, either. Funny how Disney is so hung up on the former technical quibble, but is perfectly happy to ignore the second (more-than-a-quibble, IMO).
Besides, if you do migrate away from Microsoft, emacs doesn't have a built-in spell checker, AFAIK.
Not only does emacs have built-in spell checking, but it can use the spell checker to do word completion.
And what does LaTeX have to do with a secretary's ability to use staroffice/openoffice/koffice/abiword/etc.? The only semi-reasonable argument I've seen for using MSOffice, rather than the actual alternatives like this is that MS file formats are a sort of de-facto standard. And that argument is undermined by the incompatibilities MS introduces with each new release (as you mention).
This will have a huge effect in the long run, since crypto isn't just used for encryption.
I think you're missing Mr. Coward's point. Crypto was already available to Debian users. To most, this change will be all-but-transparent, due to the magic of apt-get.
This restricts people from selling debian.
Yes, but it's the US gummit doing the restricting. Nor is this issue specific to Debian: any distro which includes crypto-enabled software (mozilla, galeon, even mutt) is going to have the same issues. If you want to sell a modern, non-crippled Linux distro of any type from the US, you're either going to have to:
a) sell only to US citizens, or
b) do the paperwork.
Which makes life hard for CD distributors
Apparently, the US gummint doesn't care. If I were a US-based CD vendor, I'd definitely complain to my gummint, but I'm not.
and is in contradiction with the GPL.
No, the GPL has nothing to do with it. The GPL addresses copyright issues. Other legal issues, like patents and other gummint regulations, are outside the scope of the GPL.
Very nearly true. The main end-user effect will be on the bandwidth-challenged, who will find ssh and SSL-enabled versions of mozilla, galeon, mutt, evolution, and ghod knows what else on their CDs in the future. These people will end up saving a lot of download time (and possibly money if they pay by the minute for being online).
The flip-side of this is that CD vendors in the US might be slightly more reluctant to jump through the hoops necessary to distribute Debian on CD. However, the same hoops are going to be required for any other distros that include non-crippled SSL-enabled apps and the like, so I don't imagine this is going to be a major problem.
Easy to arrange. Release some code under the GPL. Get a friend to build a possibly-derivative work, using the sort of linking you want to test. Have friend release this new work under a proprietary license. Sue friend. Once a judgement has been handed down, settle out of court for some trivial sum, then repeat to test the next type of linking.
:-)
Of course, you (and friend) will probably be out-of-pocket for all court costs, and may be facing accusations of "wasting the court's time" and "filing frivolous lawsuits", but heck, you said you wanted to know, surely the information is worth spending a little money on?
Me, frankly, I don't really care, I'm happy to err on the side of caution, and save my money.
cheers
No need to run BIND either. You can just make up a hosts file with your name mappings, and drop that in the appropriate places (/etc/hosts on *nix, c:\windows\hosts on a windows box).
:)
Advantages: no BIND process running, with its overheads and potential security issues. Fewer points of failure (each machine does its own resolving,
Disadvantages: more work when something changes. You have to manually replicate the hosts file to all system whenever machines are added or removed. For anything but the smallest networks, this is a huge problem, and the reason that (almost) no-one uses this approach any more. Still for a small-enough SOHO network, the simplicity and reliability of this approach may well make it worthwile.
Anyway, it Works For Me(TM).
What's clear is that NuSphere violated the GPL and lost their rights as per article 4. What's not clear is whether (and why) article 6 no longer applies. Article 6 would seem to allow them to regain their rights by receiving a new copy of the code. The copyright holder's explicit permission would seem to be automatically granted by article 6!
I would further note that article 4 speaks of an action ("any attempt to" blah, blah) which results in automatic termination of the license. I think it's hard to argue that in the absence of an action (an actual or attempted violation), any termination of a new license obtained via article 6 would occur.
But what the heck, I'm not a lawyer or a judge, and your opinion, my opinion, and even Mr. Moglen's opinion won't matter a hill of beans if the judge disagrees.
Obtaining the code from a different source wouldn't help, as your right to redistribute has been withdrawn by the copyright holder, and applies over all copies.
Yes, but according to section 6 of the GPL, you automatically recieve a license from the original licensor each time you recieve "the Program".
I don't think "figured it out" is the right term. It wasn't actually in the original movie, and it definitely wasn't in the novel. But it was, indeed, an obvious (and IMO extremely cheesy) way they could have gone. We already had the idea that replicants might not know they're replicants (from Rachel). No need to beat it into the ground by making Deckard a replicant too.
Watching the movie for the first time, I was afraid they were going to make Deckard a replicant, and I was very glad they didn't take the easy, cheap way out.
Look at it this way -- even if you drop the original cheesy ending, if Deckard is a replicant, then he's been lied to / brainwashed / programmed, and plus, he has a kinship with Rachel, and so there's no reason for him not to save her. So, even with the original ending dropped, you can still assume he does The Right Thing(TM). If, however, she's a replicant, and he's human, then the new ending becomes far more ambiguous and powerful.
Or to put it another way, if Deckard is human, we have a powerful movie about how humans react to the unhuman. If he's a replicant, we have a cheesy Hollywood thriller about robots killing each other.
Honestly, making Deckard a replicant is a mistake that no one over the (mental) age of eight should have made. Phil Dick didn't make that mistake. Unfortunately Ridley Scott ain't so smart.
The OSI folks reject such licenses absolutely. The FSF calls them "semi-free", and admits that they're better than proprietary licenses. Yet, somehow, it's the FSF that gets called "fanatic". :-)
"The only real conditions are that you don't then try to sell it"
Then it's not open source. It's in the category which the FSF calls "semi-free", and admits is better than not-at-all-free, which Debian calls "non-free" but still puts on their servers for download, and which the fanatics at OSI simply reject as not-open-source, period.
The problems with the licensing were pointed out in what quickly became a huge flamewar within hours of the initial KDE Call For Developers on Usenet (yes, this was long enough ago that serious developers still used Usenet).
So, the core KDE team knew about the problems before they had one line of code written!
(Frankly, I don't give a crap about KDE or GNOME -- they both suck, IMO -- but revisionist history annoys me.)
Although RH (and other flavors of Linux) are commonly referred to as "distributions", calling RH a distributor is a bit misleading. Normally, a distributor is one who takes manufactured/published goods from a plant or press and physically delivers them to various retail outlets. Distibutors typically do not produce. RH produces CDs and books. Companies that produce CDs and books are normally referred to as "publishers", and this certainly seems like the most appropriate term to use for RH.