Since egcs stands for "Experimental GNU Compiler System", you are correct. However, a compiler is not an essential part of an OS. The majority of systems in the world don't have a C/C++ compiler. It's essential for *building* Linux, yes, but then the C compiler RMS used to start building gcc before it was bootstrap was likewise essential, but isn't credited.
Personally I'm happy writing GNU/Linux where people want me to and it won't cause confusion among the non-cogniscenti, but I'd really rather say Linux in conversation.
Note that since it's small, you can put something over it. Macsense makes a Bondi blue clip-on thing that transforms it into a reasonably ergonomic mouse. Also, now that USB is becoming mainstream on PCs, other USB mice are available.
If you want an int return type, use an int. The problem with this multi-purpose return type thing is it's just begging for misinterpretation of the return value.
You would get far fewer errors, I think, with an AreStringsEqual or even a strequ function, rather than using strcmp when all you care about is equivalence.
>Java lacks templates and overloaded operators, which makes it a fairly useless language for generic programming.
Note that there is a Generic Java spec as well as the Pizza effort that add templates to Java. I don't know if this stuff can be hooked into the egcs system easily, but if it is, one of your (and my) two complaints would be gone.
This is almost exactly the same as giving a ~6% discount to companies that use Intel chips exclusively. (This may also be monopolistic behaviour.) What Intel does with their revenues is pretty irrelevant.
>The first problem: "Run simultanously on many processors" has been answered by SMP.
According to the SMP FAQ, Linux's SMP support has been tested on 4 processor systems and theoretically should support up to 16 processors. While this is a reasonable number, it probably falls short of being "many."
Most especially, I missed it under the list of reasons why it hasn't gotten many outside contributions. I personally have a couple of ideas about managing bookmarks that I'd like to add, but the code hasn't gotten to that point yet. I think this isn't a failure of Mozilla/Open Source, just a standard requirement for it to work, and there's nothing that could be done to fix it.
(The ideas? Have a "delete last bookmark" menu option, for when a website goes away, and a "replace last bookmark" menu option for the various redirects/restructurings. I'd also like to see an IE-like system for filing bookmarks, with the ability to rename the URL and add new directories as the URL is being filed. Also, I'd like to have a way to go "Back" and "Forward" without reloading, and to resize without reloading. Furthermore, I'd like a mode that doesn't load images until requested, either all images at once or per image on request.)
I've found that since I started watching the series, while reading the strip I hear the actors' voices for the characters. The biggest incongruity is the PHB, since the PHB in the series is mainly clueless, while the one in the strip is more caustic.
His analysis misses the key thing you need for attracting developers: something that (at least to a point) works. It's much harder to get outsiders to get involved in the start-up development.
Any truth to the rumour that a bald guy and a geeky-looking guy who referred to each other as "Bill" and "Steve" were seen fleeing the area, laughing about how much they loved to compete?:-)
>There is a standardized help system. It's called man.
Fully half the man pages I look at (including, for example, ls) say they are obsolete and no longer being updated, and that I need to look at some other documentation. How exactly to do that (or even what to look at to learn how to do that) is not explained.
I would love to see all these howtos, man pages, texinfo, etc. downloadable in one big HTML bundle.
"rich latencies"?!?!, "Communitarian Software"
on
The Power of Openness
·
· Score: 1
>Anybody want to call it "Communitarian Software"?
If "free" is a problem, why not "freed software"? (As in "freed from restrictions.")
>Primarily, he brings up what I think will be the most disappointing issue for companies releasing open source software. You probably wont *get* that many contributors.
If you take "The Cathedral and the Bazaar" as a good explanation of the process under which open source works, then the problem isn't related to being corporate at all: it's related to not having a reasonably functional program to start with. If I have to wait at least a year before my mods are of any use to me, that's a long time to wait for my itch to be scratched.
>I'm starting to think that Richard actually hates Linus so much that any word that sounds like Linus is repugnant to him. Either that, or this is another hoax.
I doubt it's so much hate of Linus as it is vehement disagreement with Linus's publicly stated position on intellectual property. I could see that it would be rather galling if you spent years on a project that in large part was meant to spread your philosophy, and then the other fellow comes along, provides a key component to the project, is the person most closely associated with the project -- including the name tie-in -- but he doesn't buy in to your philosophy.
I must admit that Richard is consistent with his philosophy: no job, no kids, few possessions -- none of those things that restrict the freedoms of most of us at some point in our lives.
>GCC and G++ may be the main C/C++ compilers available for Linux right now, but who's to say that Imprise/Borland won't port over C++ Builder one day?
Metrowerks CodeWarrior for Linux is already in beta. If it's as good as their Mac compiler, it has better template handling than egcs.
Also, a compiler is a tool for building an OS, not an OS itself. You use hammers to build houses, but a hammer is not part of the house. Likewise gzip and tar are tools for installing an OS, again not part of the OS itself.
>In the spirit of April 1st questioning everything... Are you sure Linux isn't what he had in mind all along?
Even if was, it's irrelevant, since Linus was perfectly within his rights to name the kernel whatever he wanted. Naming the whole system after the kernel was a shortcut taken thereafter.
Subsuming Linux into GNU, however, seems like *intentionally* trying to transfer the credit the other way.
>I personally like not having to worry about restrictions being placed on me by other people.
But you mention working. Doesn't your employer place restrictions on you, basically requiring your butt be on a chair at their place for ~40 hours a week? Doesn't he/she specify what you must working on during that time? In exchange, you get something of value (salary), just as you get the use of the software in exchange for agreeing not to distribute it and paying a share of the development costs.
If you don't like the deal, don't take it.
And ever see "No trespassing" signs? Physical property (real estate) imposes a heck of a lot of restrictions on freedom for all of us.
P.S. I wonder how many IP-haters would change their mind once it became apparent that movies like Star Wars would no longer be made without IP laws...
>Take Cell Phones for example. Initially, they sold the hardware. Now, they give away the hardware, and make all their money in service and support.
Try to get the phone *without* the service and support. It isn't free in any meaningful way. You buy the package. The phone may not have a big pricetag on it, but you're paying for the phone with your payment for the service contract.
>Reciprocity. I believe that if they can benefit from using my code, I should benefit from theirs. Seems only fair.
But the person using your code to run (say) a commercial website is also benefiting from your code, and you're not benefiting at all. Why is it only programmers who must share?
(from another message) >The new proprietary version would effectively replace the old free version.
If you knew about a free program and a trivially better proprietary version, which would you use? Not to mention the free version would probably be improved to match shortly thereafter if the proprietary one has any ideas worth taking.
Since the invention of the poking stick, everyone has built on someone else's work.
>The old harmony project list archive included a few threads of the form "You know, I don't think we should hurt troll by continuing harmony, since they're being so cool in letting people use their code."
Maybe so, but Schaller claimed that it was that attitude which killed Harmony, which is a much stronger statment.
As a proprietary software developer, I don't encourage that attitude. If free software folks can do what I do, then I'm not earning my keep for my customers, and I should be doing something else. There's a lot of software that hackers just don't enjoy working on, and that's where us proprietary guys should make our money.
>However, I do not want my GPL work to be a free lunch for whoever wants to proprietarize it.
I ask this again and again, and no one ever answers it: why should it bother you so much that someone might make money off of modifications to code you wrote, but it doesn't bother you that someone makes money using that code?
Despite the frequently mis-stated claim, your code is *not* being taken away -- it's still just as free as it ever was. It's the work done by someone else based on it that's not free.
>mm... so I guess you like RMS so much that everything he says is an order in your eyes, although it's meant as a suggestion ?
Don't be ridiculous. You can make demands without having any way to force compliance. If I say "I demand that you apologize for your response to my slashdot posting", is that a suggestion or a demand? Webster's defines "demand" in part as "something claimed as due" -- which fits here perfectly.
>IMHO, a good way to give credit to something is calling it by its name...
I do, I call it the GCC -- for Gnu C Compiler.
>The name "GNU" covers the entire system
No it doesn't. If in 1982 I had decided we should have a free Unix-like OS (I first used Unix in 1980...maybe I did come up with that idea!), would that make everything that comes after the Eccles Operating System? I think not. GNU is a part, not the whole.
>The number of users of the GNU system has increased a lot because of the Linux kernel.
That's a deceptively phrased fact. No one used a GNU operating system before Linux because there wasn't one.
>...I just don't understand why you want to call it "Linux" without the "GNU/".
A) Because it's shorter, B) because people know what I'm talking about, whereas if you just call it GNU like Richard has, no one knows what you're talking about, C) it sounds like I'm talking about a particular distribution, D) I hate the "gn" sound, E) without a kernel, you don't have an operating system at all, so it isn't the GNU operating system anyway, F) I dislike demanding horn-tooting, G) I disagree with the "no proprietary software" stance of the FSF, especially RMS's remarks about piracy (particularly since that is "free beer, not free speech."), H) I dislike the belittling of the difficulty of writing a kernel, given that the FSF hasn't managed to build a stable one over the past 16 years, I) I dislike this belief that GNU is somehow entitled to be the overarching name of everything.
Oh, and I call Microsoft's operating systems "Windows", unless I need to clarify what version.
>Do you remember all the excitement when the MS-Office for Linux rumour spread? I got to see some comments/mails like "YES! We are winning!". Or the excitement with Oracle supporting Linux? Why should we care?
Even if you don't like those applications or any proprietary applications, the above actions are indicative of the growing importance of Linux. At least some of us like to work on Linux and open source because we want our programming to help as many people as possible; this is a sign that this is succeeding.
Oh, and plenty of us are perfectly happy with proprietary applications, some of us make them. I can like the work of "Habitat for Humanity" without condemning homebuilding companies as immoral; likewise I can appreciate freed* software without labelling proprietary vendors as immoral.
>As for our "demand"; we don't make demands, we make suggestions.
When RMS says it, it sounds rather like a demand to me.
>I assume that you want to give the GNU project credit, otherwise this discussion is just silly.
Of course, gcc was essential to the development of Linux, and the rest of the tools are extremely useful. The question is how is this credit to be given.
>So when you then call a system "Linux", you assume that all users will automatically pick up on and understand that there are quite a few GNU utilities they are using.
Don't be absurd.
When I say Linux, people who have never heard of it before don't instantly think of GNU. Nor do they think of Linus, Alan Cox, the X project, BSD, etc. It's a name, not a list of credits. Yes, the name is derived from Linus's name, but that seems more an accident of history, not an attempt by Mr. Torvalds to usurp all the credit.
The GNU/Hurd project has been assisted greatly by the existence of Linux (at least if the FSF goal of using free software rather than proprietary is being followed). Yet there is no insistence that it be called Linux/GNU/Hurd, nor will there be.
>All philosophies that I've seen have at least one fatal flaw, and Objectivism is no exception.
You're right about that, although wrong about the exception. The flaw is that essentially, we live in a world that is consistent with Objectivist rules already.
You don't believe me? Look at any deed. Who grants that deed? At least in the U.S., it originates in the government. What are the rules of the government? The laws of the country. They are restrictions on the use of the land controlled by the U.S. Government, a super-HOA contract if you like. Don't like the restrictions on the use of your property? Go somewhere that doesn't have those restrictions. Can't find such a place? Hey, that's not our problem.
Objectivism assumes absolute ownership of property, which simply isn't the case.
Slashdot Mirror
Since egcs stands for "Experimental GNU Compiler System", you are correct. However, a compiler is not an essential part of an OS. The majority of systems in the world don't have a C/C++ compiler. It's essential for *building* Linux, yes, but then the C compiler RMS used to start building gcc before it was bootstrap was likewise essential, but isn't credited.
Personally I'm happy writing GNU/Linux where people want me to and it won't cause confusion among the non-cogniscenti, but I'd really rather say Linux in conversation.
>the mouse on the iMac is terrible.
Note that since it's small, you can put something over it. Macsense makes a Bondi blue clip-on thing that transforms it into a reasonably ergonomic mouse. Also, now that USB is becoming mainstream on PCs, other USB mice are available.
If you want an int return type, use an int. The problem with this multi-purpose return type thing is it's just begging for misinterpretation of the return value.
You would get far fewer errors, I think, with an AreStringsEqual or even a strequ function, rather than using strcmp when all you care about is equivalence.
>Java lacks templates and overloaded operators, which makes it a fairly useless language for generic programming.
Note that there is a Generic Java spec as well as the Pizza effort that add templates to Java. I don't know if this stuff can be hooked into the egcs system easily, but if it is, one of your (and my) two complaints would be gone.
This is almost exactly the same as giving a ~6% discount to companies that use Intel chips exclusively. (This may also be monopolistic behaviour.) What Intel does with their revenues is pretty irrelevant.
>The first problem: "Run simultanously on many processors" has been answered by SMP.
According to the SMP FAQ, Linux's SMP support has been tested on 4 processor systems and theoretically should support up to 16 processors. While this is a reasonable number, it probably falls short of being "many."
Most especially, I missed it under the list of reasons why it hasn't gotten many outside contributions. I personally have a couple of ideas about managing bookmarks that I'd like to add, but the code hasn't gotten to that point yet. I think this isn't a failure of Mozilla/Open Source, just a standard requirement for it to work, and there's nothing that could be done to fix it.
(The ideas? Have a "delete last bookmark" menu option, for when a website goes away, and a "replace last bookmark" menu option for the various redirects/restructurings. I'd also like to see an IE-like system for filing bookmarks, with the ability to rename the URL and add new directories as the URL is being filed. Also, I'd like to have a way to go "Back" and "Forward" without reloading, and to resize without reloading. Furthermore, I'd like a mode that doesn't load images until requested, either all images at once or per image on request.)
I've found that since I started watching the series, while reading the strip I hear the actors' voices for the characters. The biggest incongruity is the PHB, since the PHB in the series is mainly clueless, while the one in the strip is more caustic.
His analysis misses the key thing you need for attracting developers: something that (at least to a point) works. It's much harder to get outsiders to get involved in the start-up development.
Any truth to the rumour that a bald guy and a geeky-looking guy who referred to each other as "Bill" and "Steve" were seen fleeing the area, laughing about how much they loved to compete? :-)
>There is a standardized help system. It's called man.
Fully half the man pages I look at (including, for example, ls) say they are obsolete and no longer being updated, and that I need to look at some other documentation. How exactly to do that (or even what to look at to learn how to do that) is not explained.
I would love to see all these howtos, man pages, texinfo, etc. downloadable in one big HTML bundle.
>Anybody want to call it "Communitarian Software"?
If "free" is a problem, why not "freed software"? (As in "freed from restrictions.")
>Primarily, he brings up what I think will be the most disappointing issue for companies releasing open source software. You probably wont *get* that many contributors.
If you take "The Cathedral and the Bazaar" as a good explanation of the process under which open source works, then the problem isn't related to being corporate at all: it's related to not having a reasonably functional program to start with. If I have to wait at least a year before my mods are of any use to me, that's a long time to wait for my itch to be scratched.
>I'm starting to think that Richard actually hates Linus so much that any word that sounds like Linus is repugnant to him. Either that, or this is another hoax.
I doubt it's so much hate of Linus as it is vehement disagreement with Linus's publicly stated position on intellectual property. I could see that it would be rather galling if you spent years on a project that in large part was meant to spread your philosophy, and then the other fellow comes along, provides a key component to the project, is the person most closely associated with the project -- including the name tie-in -- but he doesn't buy in to your philosophy.
I must admit that Richard is consistent with his philosophy: no job, no kids, few possessions -- none of those things that restrict the freedoms of most of us at some point in our lives.
>GCC and G++ may be the main C/C++ compilers available for Linux right now, but who's to say that Imprise/Borland won't port over C++ Builder one day?
Metrowerks CodeWarrior for Linux is already in beta. If it's as good as their Mac compiler, it has better template handling than egcs.
Also, a compiler is a tool for building an OS, not an OS itself. You use hammers to build houses, but a hammer is not part of the house. Likewise gzip and tar are tools for installing an OS, again not part of the OS itself.
>In the spirit of April 1st questioning everything ... Are you sure Linux isn't what he had in mind all along?
Even if was, it's irrelevant, since Linus was perfectly within his rights to name the kernel whatever he wanted. Naming the whole system after the kernel was a shortcut taken thereafter.
Subsuming Linux into GNU, however, seems like *intentionally* trying to transfer the credit the other way.
>I personally like not having to worry about restrictions being placed on me by other people.
But you mention working. Doesn't your employer place restrictions on you, basically requiring your butt be on a chair at their place for ~40 hours a week? Doesn't he/she specify what you must working on during that time? In exchange, you get something of value (salary), just as you get the use of the software in exchange for agreeing not to distribute it and paying a share of the development costs.
If you don't like the deal, don't take it.
And ever see "No trespassing" signs? Physical property (real estate) imposes a heck of a lot of restrictions on freedom for all of us.
P.S. I wonder how many IP-haters would change their mind once it became apparent that movies like Star Wars would no longer be made without IP laws...
>Take Cell Phones for example. Initially, they sold the hardware. Now, they give away the hardware, and make all their money in service and support.
Try to get the phone *without* the service and support. It isn't free in any meaningful way. You buy the package. The phone may not have a big pricetag on it, but you're paying for the phone with your payment for the service contract.
>Reciprocity. I believe that if they can benefit from using my code, I should benefit from theirs. Seems only fair.
But the person using your code to run (say) a commercial website is also benefiting from your code, and you're not benefiting at all. Why is it only programmers who must share?
(from another message)
>The new proprietary version would effectively
replace the old free version.
If you knew about a free program and a trivially better proprietary version, which would you use? Not to mention the free version would probably be improved to match shortly thereafter if the proprietary one has any ideas worth taking.
Since the invention of the poking stick, everyone has built on someone else's work.
>The old harmony project list archive included a few threads of the form "You know, I don't think we should hurt troll by continuing harmony, since they're being so cool in letting people use their code."
Maybe so, but Schaller claimed that it was that attitude which killed Harmony, which is a much stronger statment.
As a proprietary software developer, I don't encourage that attitude. If free software folks can do what I do, then I'm not earning my keep for my customers, and I should be doing something else. There's a lot of software that hackers just don't enjoy working on, and that's where us proprietary guys should make our money.
>However, I do not want my GPL work to be a free lunch for whoever wants to proprietarize it.
I ask this again and again, and no one ever answers it: why should it bother you so much that someone might make money off of modifications to code you wrote, but it doesn't bother you that someone makes money using that code?
Despite the frequently mis-stated claim, your code is *not* being taken away -- it's still just as free as it ever was. It's the work done by someone else based on it that's not free.
>mm... so I guess you like RMS so much that everything he says is an order in your eyes, although it's meant as a suggestion ?
Don't be ridiculous. You can make demands without having any way to force compliance. If I say "I demand that you apologize for your response to my slashdot posting", is that a suggestion or a demand? Webster's defines "demand" in part as
"something claimed as due" -- which fits here perfectly.
>IMHO, a good way to give credit to something is calling it by its name...
I do, I call it the GCC -- for Gnu C Compiler.
>The name "GNU" covers the entire system
No it doesn't. If in 1982 I had decided we should have a free Unix-like OS (I first used Unix in 1980...maybe I did come up with that idea!), would that make everything that comes after the Eccles Operating System? I think not. GNU is a part, not the whole.
>The number of users of the GNU system has increased a lot because of the Linux kernel.
That's a deceptively phrased fact. No one used a GNU operating system before Linux because there wasn't one.
>...I just don't understand why you want to call it "Linux" without the "GNU/".
A) Because it's shorter, B) because people know what I'm talking about, whereas if you just call it GNU like Richard has, no one knows what you're talking about, C) it sounds like I'm talking about a particular distribution, D) I hate the "gn" sound, E) without a kernel, you don't have an operating system at all, so it isn't the GNU operating system anyway, F) I dislike demanding horn-tooting, G) I disagree with the "no proprietary software" stance of the FSF, especially RMS's remarks about piracy (particularly since that is "free beer, not free speech."), H) I dislike the belittling of the difficulty of writing a kernel, given that the FSF hasn't managed to build a stable one over the past 16 years, I) I dislike this belief that GNU is somehow entitled to be the overarching name of everything.
Oh, and I call Microsoft's operating systems "Windows", unless I need to clarify what version.
>Do you remember all the excitement when the MS-Office for Linux rumour spread? I got to see some comments/mails like "YES! We are winning!". Or the excitement with Oracle supporting Linux? Why should we care?
Even if you don't like those applications or any proprietary applications, the above actions are indicative of the growing importance of Linux. At least some of us like to work on Linux and open source because we want our programming to help as many people as possible; this is a sign that this is succeeding.
Oh, and plenty of us are perfectly happy with proprietary applications, some of us make them. I can like the work of "Habitat for Humanity" without condemning homebuilding companies as immoral; likewise I can appreciate freed* software without labelling proprietary vendors as immoral.
* "freed" as in freed from restrictions.
>As for our "demand"; we don't make demands, we make suggestions.
When RMS says it, it sounds rather like a demand to me.
>I assume that you want to give the GNU project credit, otherwise this discussion is just silly.
Of course, gcc was essential to the development of Linux, and the rest of the tools are extremely useful. The question is how is this credit to be given.
>So when you then call a system "Linux", you assume that all users will automatically pick up on and understand that there are quite a few GNU utilities they are using.
Don't be absurd.
When I say Linux, people who have never heard of it before don't instantly think of GNU. Nor do they think of Linus, Alan Cox, the X project, BSD, etc. It's a name, not a list of credits. Yes, the name is derived from Linus's name, but that seems more an accident of history, not an attempt by Mr. Torvalds to usurp all the credit.
The GNU/Hurd project has been assisted greatly by the existence of Linux (at least if the FSF goal of using free software rather than proprietary is being followed). Yet there is no insistence that it be called Linux/GNU/Hurd, nor will there be.
>All philosophies that I've seen have at least one fatal flaw, and Objectivism is no exception.
You're right about that, although wrong about the exception. The flaw is that essentially, we live in a world that is consistent with Objectivist rules already.
You don't believe me? Look at any deed. Who grants that deed? At least in the U.S., it originates in the government. What are the rules of the government? The laws of the country. They are restrictions on the use of the land controlled by the U.S. Government, a super-HOA contract if you like. Don't like the restrictions on the use of your property? Go somewhere that doesn't have those restrictions. Can't find such a place? Hey, that's not our problem.
Objectivism assumes absolute ownership of property, which simply isn't the case.