One Peeved Off Theme Maker

[Brad]

I might have had a little too much of "Jack Daniel's" when I wrote this parody of a "Charlie Daniels" song.


The Dev Went Down to Georgia
http://bradpotter.wordpress.com/2009...wn-to-georgia/

[chipbennett]

What CSS editor would that be?

They purposely didn't include one in WPMU, since everyone is using the same set of theme files, as we've discussed in this thread, and wordpress.com has a paid feature to allow adding CSS to the theme you're using.

Sorry, yes; I'm referring to the paid feature.

Hmm, wait now; this clarification - that the WPMU CSS editor adds CSS to a theme being used, but that added CSS itself doesn't become part of the theme - could have some interesting repercussions.

Regardless, I will concede that WPMU doesn't particularly convey themes. (I would still argue that it is a form of distribution; but as you have shown, that term is irrelevant.)

Note: a new question is below, and isn't intended to detract from the above concession.

I use "particularly" above, only because of a philosophical question, regarding rendering an HTML/CSS/PHP resource in general, and the CSS part of that resource specifically:

Given that anyone using a web browser can find and make a copy of a CSS file being used by any web site, isn't the mere publishing of the file on a web server inherently enabling someone to make a copy of that file?

If so, wouldn't every web site that uses a CSS file thus be conveying that CSS file, since the user is enabled to make a copy?

Thoughts?

[Ryan]

The same policy does not seem to apply to themes:

http://wordpress.org/extend/plugins/about/

There are only a few restrictions:

Your plugin must be GPL Compatible.
It can’t do anything illegal, or be morally offensive (that’s subjective, we know).
You have to actually use the subversion repository we give you in order for your plugin to show up on this site. The WordPress Plugins Directory is a hosting site, not a listing site.

I looked this up after seeing the following plugin in the repository:
http://wordpress.org/extend/plugins/thesis-openhook/

[dancole]

I use "particularly" above, only because of a philosophical question, regarding rendering an HTML/CSS/PHP resource in general, and the CSS part of that resource specifically:

Given that anyone using a web browser can find and make a copy of a CSS file being used by any web site, isn't the mere publishing of the file on a web server inherently enabling someone to make a copy of that file?

If so, wouldn't every web site that uses a CSS file thus be conveying that CSS file, since the user is enabled to make a copy?

Thoughts?

The point of the GPL is to make the source code available to anyone who receives a complied program. With programs such as GIMP or Firefox, this is easy to understand, because there is a clear difference between the source code, the program, and the output. However, with WordPress and WordPress Themes, it becomes harder for people not familiar with the GPL concept to understand.

The HTML and CSS viewed through the browser is the output of a compiled program. In this case, the PHP, HTML, and CSS source code are compiled for each visitor upon request for a webpage. In the case of WordPress Mu and its themes, the compiled program is never given away, meaning that the visitors don't have rights to obtain the source code.

In the case of wordpress.com bloggers, when they add custom CSS, they are extending WordPress Mu as a developer, not the theme... since they are hooking into WordPress and WordPress is handling the mix of extended code (Themes, plugins, and custom code). They there for have no rights to the source code for WordPress Mu, or the Theme their blog is using.

The only time a distributer is required by the GPL to give you the source code, is when they give you a compiled program of that source code, which will never apply in the case of WordPress or WordPress Themes, because the compiled program of WordPress only exists for miliseconds at a time and only in Apache. Meaning people are either distributing the source code its self or giving the ouput through a webpage in a browser.

--- ----
Going back to what I quoted... Can someone copy your HTML and CSS? Yes. Does that give them rights to any of your GPL source code? No. Can they use that HTML and CSS without your permission? No, it's the output of your site and under the same copyright as your content. Even though the CSS could be exactly the same as the CSS in the source code.

Remember to distinguish between the source code, the compiled program, and the program ouput.

[chipbennett]

The point of the GPL is to make the source code available to anyone who receives a complied program. With programs such as GIMP or Firefox, this is easy to understand, because there is a clear difference between the source code, the program, and the output. However, with WordPress and WordPress Themes, it becomes harder for people not familiar with the GPL concept to understand.

The HTML and CSS viewed through the browser is the output of a compiled program. In this case, the PHP, HTML, and CSS source code are compiled for each visitor upon request for a webpage. In the case of WordPress Mu and its themes, the compiled program is never given away, meaning that the visitors don't have rights to obtain the source code.

In the case of wordpress.com bloggers, when they add custom CSS, they are extending WordPress Mu as a developer, not the theme... since they are hooking into WordPress and WordPress is handling the mix of extended code (Themes, plugins, and custom code). They there for have no rights to the source code for WordPress Mu, or the Theme their blog is using.

The only time a distributer is required by the GPL to give you the source code, is when they give you a compiled program of that source code, which will never apply in the case of WordPress or WordPress Themes, because the compiled program of WordPress only exists for miliseconds at a time and only in Apache. Meaning people are either distributing the source code its self or giving the ouput through a webpage in a browser.

--- ----
Going back to what I quoted... Can someone copy your HTML and CSS? Yes. Does that give them rights to any of your GPL source code? No. Can they use that HTML and CSS without your permission? No, it's the output of your site and under the same copyright as your content. Even though the CSS could be exactly the same as the CSS in the source code.

Remember to distinguish between the source code, the compiled program, and the program ouput.

I'm not a programmer, so please forgive my ignorance.

Aren't compiled and interpreted code entirely different? Isn't PHP interpreted code?

And, I don't see how any of the above counters the question regarding whether serving PHP/HTML/CSS enables others to copy the underlying PHP/HTML/CSS (with primary emphasis on the CSS).

(Again, please forgive my non-programmer ignorance.)

[hallsofmontezuma]

chipbennett, I'll try and explain what dancole was saying in a different way.

Much of the GPL has code that's been compiled in mind. Say you download the GIMP or CentOS to use. You're not actually downloading the source code. However, since they're licensed under open source licenses, they are required to make the source code available for download if you desire it. Here's an example: Red Hat Enterprise Linux is a commercial linux operative system released under (among others) the GPL. When you download it to install on your computer, you're downloading the compiled program, not the original source code. However, as open source software, they do allow you access to the source code if you wish, and since the source code is GPL, you can modify and redistribute the program, provided that you allow others access to your source code as well, since it also has to be GPL. CentOS is a slightly modified version of Red Hat Enterprise Linux. When you install it, you're not downloading the source, but of course they have to make the source code available if you want it. You can modify and redistribute that as well, as long as you make the source code of your new program available.

Things get a little more complicated with PHP. In general, regardless of license, when you download a PHP program, you're downloading the actual source code. This of course makes it very easy to copy, modify, redistribute, etc at will, even if its license doesn't permit you to.
Obviously, your HTML, Javascript, images, and CSS on your website are viewable by anyone, however you may still not want them copied for their own use or redistribution. On many websites you will see a copyright/terms of use/license section letting visitors know that all materials on their site are copyrighted and can't be reused without their permission.

For example, say you purchase a PHP application from me that I sell that is not GPL or any other open source license. Even though you'll of course have the source code, due to the nature of PHP, I still may require in the license that you not modify, redistribute, etc the code, since it's not under an free (as in free speech) license.

Just to be clear, you can't see someone's PHP code unless they send it to you or you have access to their server. Unlike HTML, CSS, and Javascript, you only see the output of PHP.

[chipbennett]

chipbennett, I'll try and explain what dancole was saying in a different way.

Much of the GPL has code that's been compiled in mind. Say you download the GIMP or CentOS to use. You're not actually downloading the source code. However, since they're licensed under open source licenses, they are required to make the source code available for download if you desire it. Here's an example: Red Hat Enterprise Linux is a commercial linux operative system released under (among others) the GPL. When you download it to install on your computer, you're downloading the compiled program, not the original source code. However, as open source software, they do allow you access to the source code if you wish, and since the source code is GPL, you can modify and redistribute the program, provided that you allow others access to your source code as well, since it also has to be GPL. CentOS is a slightly modified version of Red Hat Enterprise Linux. When you install it, you're not downloading the source, but of course they have to make the source code available if you want it. You can modify and redistribute that as well, as long as you make the source code of your new program available.

Things get a little more complicated with PHP. In general, regardless of license, when you download a PHP program, you're downloading the actual source code. This of course makes it very easy to copy, modify, redistribute, etc at will, even if its license doesn't permit you to.
Obviously, your HTML, Javascript, images, and CSS on your website are viewable by anyone, however you may still not want them copied for their own use or redistribution. On many websites you will see a copyright/terms of use/license section letting visitors know that all materials on their site are copyrighted and can't be reused without their permission.

For example, say you purchase a PHP application from me that I sell that is not GPL or any other open source license. Even though you'll of course have the source code, due to the nature of PHP, I still may require in the license that you not modify, redistribute, etc the code, since it's not under an free (as in free speech) license.

Just to be clear, you can't see someone's PHP code unless they send it to you or you have access to their server. Unlike HTML, CSS, and Javascript, you only see the output of PHP.

Thanks, Halls.

I see two issues (which were the origin of my question):

1) This statement by dancole:

The point of the GPL is to make the source code available to anyone who receives a complied program.

I guess I must assume that this statement is an over-simplification, in that the GPL doesn't only apply to compiled code. Otherwise, the GPL would not be a valid license for PHP, and WordPress wouldn't have a valid open-source/copyleft license.

2) The GPL's definition of convey, and its application to a web server serving a CSS file.

According to your previous post:

To “convey” a work means any kind of propagation that enables other parties to make or receive copies. Mere interaction with a user through a computer network, with no transfer of a copy, is not conveying.

A web server serving an HTML/CSS resource would certainly be a "kind of propagation that enables other parties to... receive copies," since anyone served such a resource can make a copy of both the HTML file and any linked CSS files.

(I shouldn't have included PHP files in my original statement, since the PHP is interpreted and only the resultant HTML is served.)

It is this second point that is most important to my original question.

[hallsofmontezuma]

1) Correct

2) Are you asking if the HTML you see in your browser which is generated by a GPL program, such as WordPress, is licensed under the GPL, therefore allowing you the freedom to do what you want to with it? Nothing is being distributed to you when you go to a website. You're just using/interacting with it. Mere interaction with a user through a computer network, with no transfer of a copy, is not conveying. The nature of how the web works requires that the actual code of the HTML, CSS, Javascript be openly viewable by your browser (and you), but that isn't the same as someone giving you the code.
If you pay me to look at a PHP application you're writing, GPL or otherwise, you haven't actually distributed it to me, therefore I can't turn around and use, modify, redistribute it, just based on the fact that I'm seeing it.

[chipbennett]

1) Correct

2) Are you asking if the HTML you see in your browser which is generated by a GPL program, such as WordPress, is licensed under the GPL, therefore allowing you the freedom to do what you want to with it? Nothing is being distributed to you when you go to a website. You're just using/interacting with it. Mere interaction with a user through a computer network, with no transfer of a copy, is not conveying. The nature of how the web works requires that the actual code of the HTML, CSS, Javascript be openly viewable by your browser (and you), but that isn't the same as someone giving you the code.
If you pay me to look at a PHP application you're writing, GPL or otherwise, you haven't actually distributed it to me, therefore I can't turn around and use, modify, redistribute it, just based on the fact that I'm seeing it.

I'm purposefully avoiding use of the term "distribute".

Mere interaction certainly does not transfer a copy, to be sure. However, a copy can be transferred, in that a copy of a CSS file can be downloaded by the person viewing the web site.

So, I don't understand how, if a user does download a copy of a CSS file served, that said CSS file hasn't been conveyed.

[dancole]

The point of the GPL is to make the source code available to anyone who receives a complied program.

Going into more detail about this statement might help answer your second question. The GPL mainly applies to and was designed for programs which are compiled, but it also works for source code that is interpreted, as well as physical objects.

In all cases, the GPL gives customers the right to reproduce what they have received and this is why the source code must be available to those customers. In the case of compiled programs, this requires the source code to be provided. In the case of a physical object, blue prints need to be available. In the case of interpreted code, nothing else needs to be available.

If you get a hold of part a GPL program, such as the CSS, you are not granted rights to get the rest of the source code. The only time anyone has the rights to a copy of the whole source code is when they have received the GPL product through a purchase, which implies the consent of both parties. In the case of free downloads, you are purchasing the product for $0.00. Maybe "purchase" is not the right word or the technical word...

Well I just noticed that chipbennett has posted, but this took me so long to write this that I'm not undoing it. I guess I'm just talking to the wall now.

--- ----
Compiled code vs. interpreted code

First off, I'm (almost) an Electrical Engineer. Which means I know how to create computer hardware and write low level software (which is processed by the CPU). Anyway, computers can only understand machine code, but to make this code humanly readable it is directly replaced with assembly code. Assembly code is a low level language. In order for high level languages (C++, Python, PHP) to be converted into low level languages, they must be compiled. In the case of PHP, this compiling is done upon request of the end user. Because PHP is not compiled once by the developer and is done so often, it is called interpreted code.
All source code gets compiled into a program, then the program is run and the output producted. Compiled code is technically a "program", while interpreted code is source code that is used in a process of compiling code, running the compiled code, and producing an output, without anything major steps inbetween, in a very short amount of time.