abolishing copyright (at least for software) and no GPL,
he'd choose the latter. He created the GPL as the best thing he could think of, since abolishing copyright is not within his powers. In a sense, the GPL is a simulation of abolished copyright. [BSDish licenses are not, because they don't have the level playing field that abolished copyright would provide. (i.e., "I can use yours and you can use mine")]
Here's an idea, derived from other bright people (imagine that!):
Anyone that wants a copyright has to declare its value, and pay an annual property tax on it of (say) one percent. They can declare any value they like, but only once, and only within a year of creation.
The kicker: total income for the work, from licensing, sales, infringement penalties, etc., are capped at ten times the amount declared. Upon payment (which cannot be refused) from any source, it goes into the public domain.
Not quite as good as getting rid of copyright altogether, but still would solve 99% of the nonsense we're seeing.
I won't try to convince you that I know Perl, but I did use it regularly for a number of years. Here's a list of Perl puzzles that I wrote a few years ago--knock yourself out.
This is exactly why I don't find myself using Python much. It's not so much the language, but the attitude of its fans, or to be more precise, fanboys. Most utterances from Python fanboys seem to be about pissing on other languages with Perl being the frequent target.
I wouldn't have said it if it weren't true. Like it or not, I have to read and fix Perl code all of the time, and I wish nothing more than that that code were readable and reasonably well-written. It's not. Maybe I'm just really unlucky.
The perceived issues with Perl, justified or not, are irrelevant.
If the issues are justified, they're hardly irrelevant, no?
along came windows and blew that out-of-the water - but not in ways that you might anticipate. what really blew everything away was the use of MSRPC aka DCE/RPC with some small but highly strategic enhancements.
Wow!!! OMG I can't believe I missed this!!! I've been living in the past and didn't even know it. I'm going to say it three times just so I don't forget: MSRPC, MSRPC, MSRPC is the future!
this was so successfully deployed in Windows NT that nobody really even knows that it exists. even the _internal_ teams inside microsoft often assume that an API is "direct" instead of networked, resulting in 2nd level APIs that repeatedly send 10mb of unintialised _crap_ over-the-wire.
This seems like a relatively minor problem, though. In these days of gigabit ethernet, who cares whether every function call comes with a 10MB network overhead?
all of these have varying degrees of sophistication _well_ beyond "pipes and text file formats" that have developers throwing up their arms either in horror or to scratch their heads.
I know what you mean--that pipe character is spooky. It almost looks like a tiny axe-murderer, only without arms (which is even scarier, if you think about it).
and it's exactly the ignorance and the lack of appreciation for these powerful technologies that leaves you, cthonicdaemon, in the situation that you are presently in.
I'm intrigued by what I've read of.NET. It does sound like a desirable thing, at least when viewed from afar.
Unfortunately, though, it's from Microsoft and they have patents and god knows what else to garrote anyone that uses it, any time they like. And they have a considerable track record of doing just that sort of thing. You'd really have to be a fool to walk into that one at this point...
Like any sufficiently-powerful language, inexperienced programmers -- and especially very *clever* inexperienced programmers -- can make a huge mess.
True enough, but I've found that because Python is not yet that popular, it's still drawing mainly above-average programmers. Maybe 2% of the Python code I see is awful, versus over 90% for Perl. I think part of this is due to Perl's inherent flaws, but surely another part is due to the fact that every Joe learning to program is pointed at Perl (which can't end well, and doesn't).
This sounds like a half-baked (and incorrect) jab at Python's syntax. Sounds like someone's skill set isn't quite as current as he thinks...:-)
That notwithstanding, I also find after decades that programming involves a great deal of janitorial work. I'm trying to convince myself that this is a good thing, as it means that I will always be able to find a job cleaning up these messes. But, it does kind of suck cleaning up vomit written in Perl, C++, and (god help me) RPG...
Git is a better source code control system than Visual Source Safe. cp is a better source code control system than Visual Source Safe. Heck, rm is a better source code control system than Visual Source Safe.
In the U.S., less than five percent of cases go to trial. That means that less than five percent of people ever test the presumption of innocence. Why? Maybe because they're guilty . ..
A little research will uncover the answer. Say the police break down your door one early morning, shoot your dog, and cuff you and your family. They have an informant that says you're involved in the meth trade. They take it to the DA, who can see it's bullshit, but DA's are measured in pleas and convictions, so he offers you a plea. You can cop to some minor class-D felony and three months in county. Or you can take it to court and put yourself at the mercy of 12 random citizens and/or a judge. Win or lose, you're out your job, your house, quite possibly your spouse, and your life savings. (Not to mention other details not suitable for this family publication.)
If you've got any sense at all, you take the plea. Why? Maybe "because you're guilty...".
That's how the presumption of innocence really works.
A: Several confusing postponements, followed by a Flag Day that's actually smeared over several days, followed by several months of "Hey, whenever, dude...".
Yes, Intel must have written proper code-generation for the Itanium and worked it into their existing compilers, or perhaps they wrote some new ones. Either way, they could have chosen to contribute that to gcc as well. Those costs would have been quite modest, compared to the cost of developing the chip (not to mention developing the chip and having it fail).
I looked once at the Intel compilers, thinking it might give a nice 2x or 3x speed boost to a CPU-critical program I was working on. I rapidly realized that figuring out their (non-gcc-compatible) compiler options, how to generate shared libraries, how to generate output that would link correctly with gcc-generated output, etc., was going to be a serious undertaking. Not to mention that source for the compiler isn't available, so bug fixing becomes a nightmare. All of this just isn't worth it for a minor performance bump, especially in a world where the tide is rising exponentially anyway.
Well, for the many people who have been tortured, it would be nice to have a medical palliative, yes. Currently their lives may well be more or less destroyed.
(insert Guantanimo/extraordinary-rendition slam here, etc.)
It doesn't matter whether it's "great" or not. What matters is that (a) everyone has access to it, and (b) Intel could have added decent Itanium support at almost no cost.
Unless their plan was for Itanium to fail, they were quite foolish not to avail themselves of this option.
I'm impressed how Perens feels he can just brush away all disagreement with the GPL v3 because "Linus had a personal issue with it, some I'm ignoring that".
When considered against the backdrop of the entire space of Open Source (-ish) licenses, the differences between versions 2 and 3 of the GPL barely warrant a footnote. They are completely immaterial for almost all projects, and the only real question is whether or not you feel that Stallman's tweaks (which were based on lengthy consultation with the community) are worth following. Virtually everyone who's using the GPL in the first place would be sympathetic to the goals of the FSF, and therefore ought to go with the latest GPL.
As always, the real question is whether you
choose a license that gives for-profit corporations carte blanche to use your work, without compensation (e.g., MIT/BSD), or
choose a license that requires some sort of quid pro quo (GPL).
I prefer to think of the first option as the "communist" option and the second as the "capitalist" option.:-)
...your chip sucks. Given that there was nothing other than greed that prevented Intel from making gcc the best Itanium compiler on the planet, I'd say they just plain screwed the pooch on this one.
Slashdot Mirror
If RMS were given a choice between
he'd choose the latter. He created the GPL as the best thing he could think of, since abolishing copyright is not within his powers. In a sense, the GPL is a simulation of abolished copyright. [BSDish licenses are not, because they don't have the level playing field that abolished copyright would provide. (i.e., "I can use yours and you can use mine")]
Check out his writings...
Here's an idea, derived from other bright people (imagine that!):
Anyone that wants a copyright has to declare its value, and pay an annual property tax on it of (say) one percent. They can declare any value they like, but only once, and only within a year of creation.
The kicker: total income for the work, from licensing, sales, infringement penalties, etc., are capped at ten times the amount declared. Upon payment (which cannot be refused) from any source, it goes into the public domain.
Not quite as good as getting rid of copyright altogether, but still would solve 99% of the nonsense we're seeing.
I won't try to convince you that I know Perl, but I did use it regularly for a number of years. Here's a list of Perl puzzles that I wrote a few years ago--knock yourself out.
This is exactly why I don't find myself using Python much. It's not so much the language, but the attitude of its fans, or to be more precise, fanboys. Most utterances from Python fanboys seem to be about pissing on other languages with Perl being the frequent target.
I wouldn't have said it if it weren't true. Like it or not, I have to read and fix Perl code all of the time, and I wish nothing more than that that code were readable and reasonably well-written. It's not. Maybe I'm just really unlucky.
The perceived issues with Perl, justified or not, are irrelevant.
If the issues are justified, they're hardly irrelevant, no?
along came windows and blew that out-of-the water - but not in ways that you might anticipate. what really blew everything away was the use of MSRPC aka DCE/RPC with some small but highly strategic enhancements.
Wow!!! OMG I can't believe I missed this!!! I've been living in the past and didn't even know it. I'm going to say it three times just so I don't forget: MSRPC, MSRPC, MSRPC is the future!
this was so successfully deployed in Windows NT that nobody really even knows that it exists. even the _internal_ teams inside microsoft often assume that an API is "direct" instead of networked, resulting in 2nd level APIs that repeatedly send 10mb of unintialised _crap_ over-the-wire.
This seems like a relatively minor problem, though. In these days of gigabit ethernet, who cares whether every function call comes with a 10MB network overhead?
all of these have varying degrees of sophistication _well_ beyond "pipes and text file formats" that have developers throwing up their arms either in horror or to scratch their heads.
I know what you mean--that pipe character is spooky. It almost looks like a tiny axe-murderer, only without arms (which is even scarier, if you think about it).
and it's exactly the ignorance and the lack of appreciation for these powerful technologies that leaves you, cthonicdaemon, in the situation that you are presently in.
With a dusty debugging skills? ;-)
I'm intrigued by what I've read of .NET. It does sound like a desirable thing, at least when viewed from afar.
Unfortunately, though, it's from Microsoft and they have patents and god knows what else to garrote anyone that uses it, any time they like. And they have a considerable track record of doing just that sort of thing. You'd really have to be a fool to walk into that one at this point...
It's not FUD, it's annoyance
I admit that you did succeed in annoying me... ;-)
every problem looks like a nail that takes 18 months to pound in...
(and 18 months to pull back out when you figure out it's crooked)
Like any sufficiently-powerful language, inexperienced programmers -- and especially very *clever* inexperienced programmers -- can make a huge mess.
True enough, but I've found that because Python is not yet that popular, it's still drawing mainly above-average programmers. Maybe 2% of the Python code I see is awful, versus over 90% for Perl. I think part of this is due to Perl's inherent flaws, but surely another part is due to the fact that every Joe learning to program is pointed at Perl (which can't end well, and doesn't).
written in a language that uses tabs for scoping
This sounds like a half-baked (and incorrect) jab at Python's syntax. Sounds like someone's skill set isn't quite as current as he thinks... :-)
That notwithstanding, I also find after decades that programming involves a great deal of janitorial work. I'm trying to convince myself that this is a good thing, as it means that I will always be able to find a job cleaning up these messes. But, it does kind of suck cleaning up vomit written in Perl, C++, and (god help me) RPG...
If I wanted the answer to be correct, I wouldn't. And don't.
e.g., $ perl -e 'print "this should give an error"' > /dev/full but it doesn't
Your post is not nonsense. The responder is just grinding his axe...
Two months later she went out drinking margaritas at lunchtime with the CFO. And ...never came back. Neither of 'em.
So, was it like murder-suicide, or what?
I think I could mount several thousand wall receptacles on a reasonable-sized wall. (Better yet, POE...)
Git is a better source code control system than Visual Source Safe. cp is a better source code control system than Visual Source Safe. Heck, rm is a better source code control system than Visual Source Safe.
In the U.S., less than five percent of cases go to trial. That means that less than five percent of people ever test the presumption of innocence. Why? Maybe because they're guilty . . .
A little research will uncover the answer. Say the police break down your door one early morning, shoot your dog, and cuff you and your family. They have an informant that says you're involved in the meth trade. They take it to the DA, who can see it's bullshit, but DA's are measured in pleas and convictions, so he offers you a plea. You can cop to some minor class-D felony and three months in county. Or you can take it to court and put yourself at the mercy of 12 random citizens and/or a judge. Win or lose, you're out your job, your house, quite possibly your spouse, and your life savings. (Not to mention other details not suitable for this family publication.)
If you've got any sense at all, you take the plea. Why? Maybe "because you're guilty...".
That's how the presumption of innocence really works.
A: Several confusing postponements, followed by a Flag Day that's actually smeared over several days, followed by several months of "Hey, whenever, dude...".
If Alberto Gonzales can get a job, I'm sure you'll have no trouble...
That way it won't take us two posts to figure out that you're talking out of your ass.
New to the Internet are you? ;-)
Yes, Intel must have written proper code-generation for the Itanium and worked it into their existing compilers, or perhaps they wrote some new ones. Either way, they could have chosen to contribute that to gcc as well. Those costs would have been quite modest, compared to the cost of developing the chip (not to mention developing the chip and having it fail).
I looked once at the Intel compilers, thinking it might give a nice 2x or 3x speed boost to a CPU-critical program I was working on. I rapidly realized that figuring out their (non-gcc-compatible) compiler options, how to generate shared libraries, how to generate output that would link correctly with gcc-generated output, etc., was going to be a serious undertaking. Not to mention that source for the compiler isn't available, so bug fixing becomes a nightmare. All of this just isn't worth it for a minor performance bump, especially in a world where the tide is rising exponentially anyway.
Well, for the many people who have been tortured, it would be nice to have a medical palliative, yes. Currently their lives may well be more or less destroyed.
(insert Guantanimo/extraordinary-rendition slam here, etc.)
It doesn't matter whether it's "great" or not. What matters is that (a) everyone has access to it, and (b) Intel could have added decent Itanium support at almost no cost.
Unless their plan was for Itanium to fail, they were quite foolish not to avail themselves of this option.
I'm impressed how Perens feels he can just brush away all disagreement with the GPL v3 because "Linus had a personal issue with it, some I'm ignoring that".
When considered against the backdrop of the entire space of Open Source (-ish) licenses, the differences between versions 2 and 3 of the GPL barely warrant a footnote. They are completely immaterial for almost all projects, and the only real question is whether or not you feel that Stallman's tweaks (which were based on lengthy consultation with the community) are worth following. Virtually everyone who's using the GPL in the first place would be sympathetic to the goals of the FSF, and therefore ought to go with the latest GPL.
As always, the real question is whether you
I prefer to think of the first option as the "communist" option and the second as the "capitalist" option. :-)
...your chip sucks. Given that there was nothing other than greed that prevented Intel from making gcc the best Itanium compiler on the planet, I'd say they just plain screwed the pooch on this one.
US$398 to fix security problems with their software...