In this sense the rise of PHP has mystified me. Why the need for a novel language to do web scripting
mod_php is installed at most Apache based web hosting services, mod_perl practically never.
You are mixing up cause and effect. What you describe is the effect. And the cause is what was asked for.
It is not that PHP has become widespread because mod_php is installed more often than mod_perl (which is not necessarily true by itself), but mod_php is installed so often, because PHP is so popular. Of course, meanwhile the number of installations help PHP's popularity.
As for the original question, I do not have an answer. Maybe PHP has a lower entry threshold? Maybe it's because it's more similar to C/Java? Probably for some other reason.
Besides, regarding web pages, you can have both, PHP and Perl either as stand-alone (=CGI) or as Apache module.
At the very least an API as you define it should be able to form a firewall between GPL and non-GPL code (note - not necessarily commercial - I may want to distribute my code under BSD license, but it still needs to be untainted by GPL for that to be an option).
Without this provision companies will continue to avoid GPL code for fear of losing core business logic.
Sorry, but who cares?;)
More seriously, the LGPL is for such things. If an author did not put the code under LGPL, but GPL. Tough. Live with it. Either comply or look for another source for your needs.
PS: Couldn't the kernel linking question also be resolved this way (i.e. using LGPL instead of GPL). Sorry about my ignorance, but I am not into this kernel issue.
In the meantime, simply use the right mouse button to choose the menu point in question.
I can reproduce the bug, but did not notice it until now, because I only work with the right button to use the menu. Meanwhile I found, that I am fastest, if I do not even press the button a second time, but simply, press and hold the right button, choose the menu and then let go.
It might be interesting to note that traceroutes do end up in China, so it looks like the packets are making it there unmolested, but the web server on the other end is what's making the redirect:
Or it simply means, that they are not changing all traffic, but only TCP to port 80 and therefore a traceroute goes through.
After reading this discussion, it seems I am nearly the only one who religiously uses Ctrl-Insert and Shift-Insert for copy and paste. Last I checked Dvorak doesn't break that.
Maybe that's because it requires two hands (or one hand in an uncommon position). That would force me to take my right hand from the mouse, when I am using it (and often I just used it to mark the phrase I want to copy).
Locally, well, partially. For me, it is more dependend of whom I talk to, I would say. If I tell the URL my mom, I would spell it out, even avoiding shorthands like "amp" and "bang".
or are you mostly talking to smarter people than I generally am?
Don't think so. Would you have been able to figure out the URL? If so, why should I say more?;-)
More seriously, I was talking about my co-workers, most of them are very computer-literate. That was, because the original was about code. If I dictate someone a C++ program, I presume he knows C++. Therefore, for my simple example, I presumed that the audience roughly knows how URLs are structured.
Go and tell someone (not a oh-it's-a-computer-I-lose-common-sense-person) 100 URLs. When you are on the last, there is a good chance that you don't need to tell them the punctuation any more.
In the same way, say "function main" (in a C content) to someone who knows C and he will know to write
Not code though. Saying the names for various brackets is way too time consuming -- even if you go with square, round and squiggly rather than proper names.
Usually my co-workers know quite good what I mean, when I tell them what to type at command line or in an editor, from the context, even if I leave out the punktuation. That would be what I expect from the ideal machine, too. A simplistic example, if I tell you (in an URL context):
"virtual com port 8080 query c g i <pause> i d equals 10 level equals 3" (might be a bit off, because I am not a native English speaker, but you get the idea)
you would probably know that the correct URL should be
http://virtual.com:8080/query.cgi?id=10&level=3
Of course, the context is important. If I say the same while you are sitting in front or a C++ program, your interpretion would be a bit different.;-)
Finally a little question for the P2P junks out there: many people claim they get to learn new kinds of music by P2P sharing. I won't say it isn't true, but how? You still need a handle to search new stuff? You just type in random keywords, or what? Just curious, because I'd like to broaden my musical horizonts a bit.
First, if you hear one song on radio or so and manage to remember the interpret(?) or tile, you can relatively easy search for other songs and test listen (meanwhile that works with quite some online shops, too, e.g. amazon).
The second, and probably what people claim, is the possibility (at least in edonkey) to view the file list of others. I.e. download some songs, then, if you notice someone having several of the songs, choose their file list and browse for other stuff (in edonkey you have to add them as friend first).
Oh, come on! What else is new? They explicitly state that they are related:
NEW YORK (AP) -- A Web browser project run primarily by volunteers and backed by America Online is making one last stab at challenging the dominance of Microsoft Corp.
The group released its Mozilla 1.0 package this month -- some four years after AOL's Netscape unit launched the project. (AOL Time Warner is the parent company of CNN.com.)
And I wonder whether those people modding this up to 5 (Informative) did bother to read the article?
They released it with the GPL, which is a document telling you that you can recieve the source changes they've made if you have the binary.[...]only the the written offer of availability is necessary per 3(b); which does not specify when or how the source is to be delivered, leaving a gaping hole, imho, [...]
It seems you are right, that 3b) not enforcing a timeline could be a hole. But NuSphere has not complied to the prerequisites for 3b) and therefore cannot claim to comply with the GPL by that point.
3. You may copy and distribute the Program (or a work based on it [...] provided that you also do one of the following: [...] b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; [...]
Sorry, but I cannot see how packaging the source with the GPL (which was already there) could be interpreted as complying with 3b), which you seem to imply.
Just for curiosity: In which version Linus changed that?
The COPYING (2.4.17) file says, that only the GPLv2 counts, unless stated otherwise. I just checked 2.4.17 and it has the (at your option) any later version clause in many, many source files. Unless you mean Linus changed these notices, your point is lapsed. Regardless, if Linus makes (or has made) such a change, I think there is an argument that he is on the safe side with a change to GPLv2 only. The copyright notice in the kernel files reads:
...you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2, or (at your option) any later version.
Therefore I can choose which version of the GPL I want to use to redistribute the kernel. Furthermore, the GPLv2 requires me to put in (keep) an appropriate copyright notice, a disclaimer and notices about the license.
So if I redistribute it under GPLv4, what forbids me to change the notice to something like either version 4, or (at your option) any later version?
If there is some clause indeed (I did not found any), you are right. If not, there is nothing to prevent me from putting it under GPLv2 only following the same reasoning, is there?
No, your implication, that writing code means learning how to write it well. Surely, you will improve in your coding skill, but that's not the same as activlely searching for and learning how to do something right.
Especially that he wrote it alone (if he really did?!?), as you say, would rather explain why code could be bad. If you have to work together with others, you will faster get new ideas how to solve problems more elegantly.
Btw, I don't mean to know whether the original poster knows C++ well, I just wanted to put your comment into a different light.
...but at high load levels you will be creating and destroying instantiations of Apache quite regularly.
This is only true with an inappropriate configuration. If you want, you can even tell Apache to never fork again any child after the initial start-up (except for CGIs, of course, which were not subject of the discussed).
More practically you just set the limit higher, how much requests a child is allowed to answer before it dies.
If you have the source, you can make changes to it. [...]. For all those who bitch about those shortcomings, well, now you have the source to a fast database engine and you can do something about it.
The source of MySQL was available before - I don't know for how long, but at least a year. It was/is not GPL'ed, however.
Yes, I can spell "code fork" -- but if enough people start building on the old MySQL source, maybe they will eventually catch up to current versions of MySQL in terms of features,
No need to 'catch up'. The source of the newest version is available, too. Just do to it what you want to and send the patches in.
performance, and reliability and this might give TCX more reason to go all the way with the GPL.
Well, if the GPL is the only concern you have, then you are right and you probably have to double that work if you can.
I, personally, see the main advantage of open source in the ability to improve/fix the source and you already have this with MySQL and the way TcX handles it.
>... but it has a long way to go before it even reaches beta-level.
Sigh. It seems it must be repeated over and over again:
The current Milestones of Mozilla ARE DECLARED TO BE PRE-BETA.
If you don't know that it means exactly what you described, just don't touch it.
PS: Did not intent to be offensive, but please people, read the label before you open a package and wonder about the contents.
PPS: Why shouldn't it be made of thousand little files? It's a pre-beta! There are still many things that will change (though, I would not mind the number files it has now and know some reasons to keep it as it is...)
Slashdot Mirror
In this sense the rise of PHP has mystified me. Why the need for a novel language to do web scripting
mod_php is installed at most Apache based web hosting services, mod_perl practically never.
You are mixing up cause and effect. What you describe is the effect. And the cause is what was asked for.
It is not that PHP has become widespread because mod_php is installed more often than mod_perl (which is not necessarily true by itself), but mod_php is installed so often, because PHP is so popular. Of course, meanwhile the number of installations help PHP's popularity.
As for the original question, I do not have an answer. Maybe PHP has a lower entry threshold? Maybe it's because it's more similar to C/Java? Probably for some other reason.
Besides, regarding web pages, you can have both, PHP and Perl either as stand-alone (=CGI) or as Apache module.
Agreed. Bruce clarification would help a lot.
At the very least an API as you define it should be able to form a firewall between GPL and non-GPL code (note - not necessarily commercial - I may want to distribute my code under BSD license, but it still needs to be untainted by GPL for that to be an option).
;)
Without this provision companies will continue to avoid GPL code for fear of losing core business logic.
Sorry, but who cares?
More seriously, the LGPL is for such things. If an author did not put the code under LGPL, but GPL. Tough. Live with it. Either comply or look for another source for your needs.
PS: Couldn't the kernel linking question also be resolved this way (i.e. using LGPL instead of GPL). Sorry about my ignorance, but I am not into this kernel issue.
In the meantime, simply use the right mouse button to choose the menu point in question.
I can reproduce the bug, but did not notice it until now, because I only work with the right button to use the menu. Meanwhile I found, that I am fastest, if I do not even press the button a second time, but simply, press and hold the right button, choose the menu and then let go.
It might be interesting to note that traceroutes do end up in China, so it looks like the packets are making it there unmolested, but the web server on the other end is what's making the redirect:
Or it simply means, that they are not changing all traffic, but only TCP to port 80 and therefore a traceroute goes through.
After reading this discussion, it seems I am nearly the only one who religiously uses Ctrl-Insert and Shift-Insert for copy and paste. Last I checked Dvorak doesn't break that.
Maybe that's because it requires two hands (or one hand in an uncommon position). That would force me to take my right hand from the mouse, when I am using it (and often I just used it to mark the phrase I want to copy).
You figure this is a locality-based thing
;-)
Locally, well, partially. For me, it is more dependend of whom I talk to, I would say. If I tell the URL my mom, I would spell it out, even avoiding shorthands like "amp" and "bang".
or are you mostly talking to smarter people than I generally am?
Don't think so. Would you have been able to figure out the URL? If so, why should I say more?
More seriously, I was talking about my co-workers, most of them are very computer-literate. That was, because the original was about code. If I dictate someone a C++ program, I presume he knows C++. Therefore, for my simple example, I presumed that the audience roughly knows how URLs are structured.
Go and tell someone (not a oh-it's-a-computer-I-lose-common-sense-person) 100 URLs. When you are on the last, there is a good chance that you don't need to tell them the punctuation any more.
In the same way, say "function main" (in a C content) to someone who knows C and he will know to write
int main( int argc, char *argv[] )
{
}
Not code though. Saying the names for various brackets is way too time consuming -- even if you go with square, round and squiggly rather than proper names.
;-)
Usually my co-workers know quite good what I mean, when I tell them what to type at command line or in an editor, from the context, even if I leave out the punktuation. That would be what I expect from the ideal machine, too. A simplistic example, if I tell you (in an URL context):
"virtual com port 8080 query c g i <pause> i d equals 10 level equals 3"
(might be a bit off, because I am not a native English speaker, but you get the idea)
you would probably know that the correct URL should be
http://virtual.com:8080/query.cgi?id=10&level=3
Of course, the context is important. If I say the same while you are sitting in front or a C++ program, your interpretion would be a bit different.
Finally a little question for the P2P junks out there: many people claim they get to learn new kinds of music by P2P sharing. I won't say it isn't true, but how? You still need a handle to search new stuff? You just type in random keywords, or what? Just curious, because I'd like to broaden my musical horizonts a bit.
First, if you hear one song on radio or so and manage to remember the interpret(?) or tile, you can relatively easy search for other songs and test listen (meanwhile that works with quite some online shops, too, e.g. amazon).
The second, and probably what people claim, is the possibility (at least in edonkey) to view the file list of others. I.e. download some songs, then, if you notice someone having several of the songs, choose their file list and browse for other stuff (in edonkey you have to add them as friend first).
Oh, come on! What else is new? They explicitly state that they are related:
NEW YORK (AP) -- A Web browser project run primarily by volunteers and backed by America Online is making one last stab at challenging the dominance of Microsoft Corp.
The group released its Mozilla 1.0 package this month -- some four years after AOL's Netscape unit launched the project. (AOL Time Warner is the parent company of CNN.com.)
And I wonder whether those people modding this up to 5 (Informative) did bother to read the article?
They released it with the GPL, which is a document telling you that you can recieve the source changes they've made if you have the binary.[...]only the the written offer of availability is necessary per 3(b); which does not specify when or how the source is to be delivered, leaving a gaping hole, imho, [...]
It seems you are right, that 3b) not enforcing a timeline could be a hole. But NuSphere has not complied to the prerequisites for 3b) and therefore cannot claim to comply with the GPL by that point.
3. You may copy and distribute the Program (or a work based on it [...] provided that you also do one of the following:
[...]
b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange;
[...]
Sorry, but I cannot see how packaging the source with the GPL (which was already there) could be interpreted as complying with 3b), which you seem to imply.
Just for curiosity: In which version Linus changed that?
...you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2, or (at your option) any later version.
The COPYING (2.4.17) file says, that only the GPLv2 counts, unless stated otherwise. I just checked 2.4.17 and it has the (at your option) any later version clause in many, many source files. Unless you mean Linus changed these notices, your point is lapsed. Regardless, if Linus makes (or has made) such a change, I think there is an argument that he is on the safe side with a change to GPLv2 only. The copyright notice in the kernel files reads:
Therefore I can choose which version of the GPL I want to use to redistribute the kernel. Furthermore, the GPLv2 requires me to put in (keep) an appropriate copyright notice, a disclaimer and notices about the license.
So if I redistribute it under GPLv4, what forbids me to change the notice to something like either version 4, or (at your option) any later version?
If there is some clause indeed (I did not found any), you are right. If not, there is nothing to prevent me from putting it under GPLv2 only following the same reasoning, is there?
No, your implication, that writing code means learning how to write it well. Surely, you will improve in your coding skill, but that's not the same as activlely searching for and learning how to do something right.
Especially that he wrote it alone (if he really did?!?), as you say, would rather explain why code could be bad. If you have to work together with others, you will faster get new ideas how to solve problems more elegantly.
Btw, I don't mean to know whether the original poster knows C++ well, I just wanted to put your comment into a different light.
Well, of course, I mean
"it does not apply to 'multimedia' DVDs".
It *may* very well apply to DVDs where computer programs are stored on.
Aehem, the paragraph you quoted is talking about "computer programs", not music data. It just does not apply to DVDs.
Please prove me wrong if I am.
This is only true with an inappropriate configuration. If you want, you can even tell Apache to never fork again any child after the initial start-up (except for CGIs, of course, which were not subject of the discussed).
More practically you just set the limit higher, how much requests a child is allowed to answer before it dies.
Why compile MySQL yourself? You did not suggest anything what TcX does not do already.
The binaries (Intel) you can get from www.mysql.com are already compiled with pgcc -O6 and statically linked.
The source of MySQL was available before - I don't know for how long, but at least a year. It was/is not GPL'ed, however.
Yes, I can spell "code fork" -- but if enough people start building on the old MySQL source, maybe they will eventually catch up to current versions of MySQL in terms of features,
No need to 'catch up'. The source of the newest version is available, too. Just do to it what you want to and send the patches in.
performance, and reliability and this might give TCX more reason to go all the way with the GPL.
Well, if the GPL is the only concern you have, then you are right and you probably have to double that work if you can.
I, personally, see the main advantage of open source in the ability to improve/fix the source and you already have this with MySQL and the way TcX handles it.
Sigh. It seems it must be repeated over and over again:
The current Milestones of Mozilla
ARE DECLARED TO BE PRE-BETA.
If you don't know that it means exactly what you described, just don't touch it.
PS: Did not intent to be offensive, but please people, read the label before you open a package and wonder about the contents.
PPS: Why shouldn't it be made of thousand little files? It's a pre-beta! There are still many things that will change (though, I would not mind the number files it has now and know some reasons to keep it as it is...)