But the question I draw from this is: why not relax the GPL restrictions a bit for embedded applications?
To which, I'm afraid, the only reply is: "Why not go write your own closed kernel - or actually pay money for one someone else has already written ?"
The whole point of the GPL is that, in return for the millions of lines of code you receive, you are expected to return the few hundred/thousand you produce. If you don't want to share, no-one is making you.
Re:You don't understand the spirit then.
on
Abusing the GPL?
·
· Score: 1
My source code is source code A transformed into B. A is not a script, nor an IDL, it is a transformation. Hence I am not required to distribute the originals. I am only required to distribute the transformation results.
If the transformation is an automatic process (e.g. an obfuscator) then the "source" is the input to the obfuscator, even if there are other tools (compiler etc) further down the chain.
"Source" equals "the bits that the humans generate"
There is no true need to rely on the DMCA for this argument, it comes down to a license: you bought the product, you agreed to the license. Don't buy it and don't play it if you disagree with said license.
If the text of the license was presented before any money changed hands, then I would agree with you. Seeing as the text of most EULAs is presented after money changes hands, I would say that (morally at least, the legalities are probably all fucked up and IANAL) the "license agreement" is an attempt to modify the contract *after* the actual agreement has happened.
Personally, I feel no moral obligation to comply with such "agreements".
I've done a bunch of those too amongst others, and my personal fave for "nicest ISA" is the ARC. Pretty much the nicest bits of MIPS mixed with the nicest bits of ARM.
" Both "America" and "Alabama" would come before "Argentina" in an alphabetical list."
America is a continent (or 2)
Alabama is a state
Argentina is a country
Re:Well, that's Itanium, and see what it got them.
on
Inside Intel
·
· Score: 1
" They haven't successfully introduced a really new architecture since the i860/i960, and that was YEARS ago. "
The i860 and i960 were very different architectures, the i860 was a fp number cruncher and the i960 was an early superscalar-ish architecture (co-developed, iirc, with Siemens, or someone else beginning with S, anyway).
The i960 was a beauty to write code for. i860 wasn't.
Or, if that's too much of a problem or you're running LookOut! or the suchlike, just dump
all email containing the string "!!!" into a "suspected shite" folder. Catches about 99% of all
my incoming spam with practically no effort on my part.
Please realise that at least one of these has to be a variable. 90% of my
grievances with management usually come down to them presenting us
with a spec where all 3 are constants.
In real-life applications, n can get to some quite non-trivial values. However, in real-life applications, n can also stay in the range of values where O(kn) > O(n*n).
It's just a question of knowing what the code you're writing will actually be doing.
Well, the ABC *was* digital but it was very much a special purpose machine that performed a single task, which qualifies it as a calculator rather than a computer. In terms of functionality, it offered little more than Babbage's machines.
Why are you confusing Java developers with dorks? Big O notation applies in all programming languages, an O(n) implementation in Java will beat an O(n*n) in C any day.
Any day with an application that generates sufficiently high values of n, that is.
I wish I had time to respond to this more fully, as you make a well reasoned case. However, it's late, and I've got deadlines... so this is a quickie
Well written OO code is truely a thing of beauty.
I would say: "well designed code is truly a thing of beauty." - but the quality of the design is independent of the implementation language. Maybe our different viewpoints stem from the fact that I regard language as an implementation issue and you (appear) to regard language as a design issue.
Performance problems with Java are somewhat mis-stated
It's not really Java's poor performance I was talking about, but more that it makes people less aware of performance issues (I dislike the STL for similar reasons... it encourages people to write code that they have no understanding of the behaviour of).
Yes, we all were taught that low-level details are unimportant, appropriate algorithm choice is everything. However the O(whatever(n)) viewpoint only applies for large values of n, which people often forget.
What we should be trying to teach is the style of thinking required to write good code.
Exactly. Which is why Java is not a good teaching language.
Code can be good from a structural/maintainability aspect, from an algorithmic/elegance aspect, or from a performance aspect. Being able to write code quickly (which is all Java does for you) does not make the code good, it just makes it cheap.
If you're teaching programming, you need to teach algorithms, structure and performance issues - all of which are far more important than the syntactic quirks of the particular language.
Structure is not something that should be taught at the machine anyway, it's a pencil-and-paper issue. Algorithms can be taught using a conventional imperative language, but it's not the best choice, and Java hides too much of the machine mechanics from the user for them ever to get a clue about performance.
Maybe if people learnt to write good code, rather than get something running fast, software performance would start to catch up with hardware performance.
IIRC, US citizens *are* bound by US law even when outside the country. This has (again, IIRC) been used to arrest people who have been using child prostitutes in Asia whilst on holiday.
Yup, I'd add my support to this option. Tai Chi is the business when it comes to fixing lower back problems. I've been doing it on and off for a few years now, and have grown about two inches over that time due to improved posture.
Still got to get rid of that coder's gut though...
How come saying Worldforge is good merits +3 informative, and saying Worldforge is bad merits "flamebait". Two sides of the same coin eh.
Well, for a start, the +3 post I think you're referring to didn't say "Worldforge is good", it said "Worldforge is here." No value judgement was made, other than calling the people on the Worldforge team "nice" - which is my personal experience from interacting with them on IRC and in person. As for the "flamebait" post, I can't find the post you're referring to in this article - care to post a link ?
I still say pointing to Worldforge as a comparable alternative to Star Wars Galaxies, or UO for that matter, is laughable at best.
At present, you're probably right - it's an ongoing project, not "finished" yet, but moving forwards quickly. People said the same about Linux, gcc, and all those other good things once, you know...
Slashdot Mirror
But the question I draw from this is: why not relax the GPL restrictions a bit for embedded applications?
To which, I'm afraid, the only reply is: "Why not go write your own closed kernel - or actually pay money for one someone else has already written ?"
The whole point of the GPL is that, in return for the millions of lines of code you receive, you are expected to return the few hundred/thousand you produce. If you don't want to share, no-one is making you.
My source code is source code A transformed into B. A is not a script, nor an IDL, it is a transformation. Hence I am not required to distribute the originals. I am only required to distribute the transformation results.
If the transformation is an automatic process (e.g. an obfuscator) then the "source" is the input to the obfuscator, even if there are other tools (compiler etc) further down the chain.
"Source" equals "the bits that the humans generate"
No such thing as the "British Commonwealth" anymore, just the "Commonwealth". The "British" part got dropped many many moons ago.
Jack, fuck off..
I'll let you specify standards for my PC when I get to specify minimum nipple counts for your movies
Beowulf cluster? That's called a government isn't it?
No, but close. The phrase you're thinking of for "government" does indeed include "cluster", but at the beginning rather than the end.
There is no true need to rely on the DMCA for this argument, it comes down to a license: you bought the product, you agreed to the license. Don't buy it and don't play it if you disagree with said license.
If the text of the license was presented before any money changed hands, then I would agree with you. Seeing as the text of most EULAs is presented after money changes hands, I would say that (morally at least, the legalities are probably all fucked up and IANAL) the "license agreement" is an attempt to modify the contract *after* the actual agreement has happened.
Personally, I feel no moral obligation to comply with such "agreements".
I've done a bunch of those too amongst others, and my personal fave for "nicest ISA" is the ARC. Pretty much the nicest bits of MIPS mixed with the nicest bits of ARM.
" Both "America" and "Alabama" would come before "Argentina" in an alphabetical list."
America is a continent (or 2)
Alabama is a state
Argentina is a country
" They haven't successfully introduced a really new architecture since the i860/i960, and that was YEARS ago. "
The i860 and i960 were very different architectures, the i860 was a fp number cruncher and the i960 was an early superscalar-ish architecture (co-developed, iirc, with Siemens, or someone else beginning with S, anyway).
The i960 was a beauty to write code for. i860 wasn't.
"Ask about sharing music and you get a 50:50 split of people who think its great, and people who think its a first step twards lawless anarchy!"
Sounds like a tautology to me...
Or, if that's too much of a problem or you're running LookOut! or the suchlike, just dump
all email containing the string "!!!" into a "suspected shite" folder. Catches about 99% of all
my incoming spam with practically no effort on my part.
Certainly looks like one to me.
I think I'll submit an ask-slashdot: "I'm looking for a new editor. Which should I choose, vi or emacs ?"
Duh.
A slightly different three things:
If x=time, y=features/quality and z=resources
Please realise that at least one of these has to be a variable. 90% of my
grievances with management usually come down to them presenting us
with a spec where all 3 are constants.
In real-life applications, n can get to some quite non-trivial values. However, in real-life applications, n can also stay in the range of values where O(kn) > O(n*n).
It's just a question of knowing what the code you're writing will actually be doing.
Well, the ABC *was* digital but it was very much a special purpose machine that performed a single task, which qualifies it as a calculator rather than a computer. In terms of functionality, it offered little more than Babbage's machines.
You're right, it does deserve a mention, though.
More accurately, "we're US-centric" does not justify propogating a known lie.
A slightly more detailed chronology.
Well, to be accurate:
"The Manchester Machine (aka Manchester Mk 1)" (1949) was the *second* stored program computer, and was a general purpose machine.
"The Baby" (1947) (also from Manchester) was the *first* stored program computer, and was a general purpose machine.
"Colossus" (1943) was built at Bletchley Park, and was neither a stored program computer nor a general purpose machine.
"ENIAC" (1945) was built at the University of Pennsylvania and was (almost) a general purpose machine, but not a stored program computer.
"Ferranti Mark 1" (February 1951) was the world's *first* commercial computer.
"UNIVAC" (March 1951) was the world's *second* commercial computer.
(I'm not familiar enough with Zuse's contributions to place them accurately, but will acknowledge that they exist)
Why are you confusing Java developers with dorks? Big O notation applies in all programming languages, an O(n) implementation in Java will beat an O(n*n) in C any day.
Any day with an application that generates sufficiently high values of n, that is.I wish I had time to respond to this more fully, as you make a well reasoned case. However, it's late, and I've got deadlines... so this is a quickie
Well written OO code is truely a thing of beauty.
I would say: "well designed code is truly a thing of beauty." - but the quality of the design is independent of the implementation language. Maybe our different viewpoints stem from the fact that I regard language as an implementation issue and you (appear) to regard language as a design issue.
Performance problems with Java are somewhat mis-stated
It's not really Java's poor performance I was talking about, but more that it makes people less aware of performance issues (I dislike the STL for similar reasons ... it encourages people to write code that they have no understanding of the behaviour of).
Yes, we all were taught that low-level details are unimportant, appropriate algorithm choice is everything. However the O(whatever(n)) viewpoint only applies for large values of n, which people often forget.
What we should be trying to teach is the style of thinking required to write good code.
Exactly. Which is why Java is not a good teaching language.
Code can be good from a structural/maintainability aspect, from an algorithmic/elegance aspect, or from a performance aspect. Being able to write code quickly (which is all Java does for you) does not make the code good, it just makes it cheap.
If you're teaching programming, you need to teach algorithms, structure and performance issues - all of which are far more important than the syntactic quirks of the particular language.
Structure is not something that should be taught at the machine anyway, it's a pencil-and-paper issue. Algorithms can be taught using a conventional imperative language, but it's not the best choice, and Java hides too much of the machine mechanics from the user for them ever to get a clue about performance.
Maybe if people learnt to write good code, rather than get something running fast, software performance would start to catch up with hardware performance.
Considering that they won your War of Independence for you, it seems you're about even now.
IIRC, US citizens *are* bound by US law even when outside the country. This has (again, IIRC) been used to arrest people who have been using child prostitutes in Asia whilst on holiday.
Yup, I'd add my support to this option. Tai Chi is the business when it comes to fixing lower back problems. I've been doing it on and off for a few years now, and have grown about two inches over that time due to improved posture.
Still got to get rid of that coder's gut though...
How come saying Worldforge is good merits +3 informative, and saying Worldforge is bad merits "flamebait". Two sides of the same coin eh.
Well, for a start, the +3 post I think you're referring to didn't say "Worldforge is good", it said "Worldforge is here." No value judgement was made, other than calling the people on the Worldforge team "nice" - which is my personal experience from interacting with them on IRC and in person. As for the "flamebait" post, I can't find the post you're referring to in this article - care to post a link ?
I still say pointing to Worldforge as a comparable alternative to Star Wars Galaxies, or UO for that matter, is laughable at best.
At present, you're probably right - it's an ongoing project, not "finished" yet, but moving forwards quickly. People said the same about Linux, gcc, and all those other good things once, you know...