Before we launch the Wiki publicly we need to decide how to handle copyrights and licenses. I can think of several alternatives and I would appreciate your feedback on this. Here comes my thoughts:

What is clear is that by adding material to the Wiki each author must grant CFD Online a perpetual license to use this material in the Wiki. Allowing people to withdraw material from the Wiki would create a lot of trouble to separate things later on.

Since the Wiki is a free resource created by volunteers without any payment I would prefer to have a fairly free copyright policy. Or what do you think? Would you like to limit use and redistribution of the Wiki material? There are a few alternative "free-distribution" licenses:

We could have a copyright similar to the GNU GPL license for software. This would allow anyone to use and modify the content as long as they give credit to its origin and also release their derived works under the same license. Gnu calls this license the Gnu FDL (Free Document License). This license was originally written mainly for manuals, but it has been used in many Wikis also.

An alternative license, which has become popular lately, is the creative commons license, which exists in a few different flavours, all of which allow free distribution but depending on which flavour of the license you use it can also require one or several of:

* Attribution to original author/source * Noncommercial reuse (disallowing others to use the material commecially) * No derivate works (prohibiting others to make derivate works) * Share alike (requiring derivative works to be under the same license)

Here are a few links with more information about these various licensing options:

http://creativecommons.org/ http://www.fsf.org/licensing/licenses/fdl.html http://en.wikipedia.org/wiki/Wikipedia:Copyrights

What is your opinion about the most suitable license to use for material in the Wiki?

Would you like to allow others to take material from the wiki, change and modify it and then redistrubute it?

Would you like to allow others to take material from the wiki and use it commercialy (like selling it etc.)?

With the progress we have made with the wiki over the last few weeks I think that it is time to start discussing this license issue a bit. It might seem boring but it can be very important. If you don't belive mee you can read what happened to Eric's Tresure Troves (now called mathworld) a few years ago: http://mathworld.wolfram.com/about/e...ommentary.html

My $0.02:

> What is your opinion about the most suitable license to use for material in the Wiki?

I generally don't like the creative commons stuff, as they strike me as bit on the over-clever side. The GFDL, on the other hand, seems unsatisfactory somehow. I'm not really sure why it seems that way. Maybe because it is software manual centric. However, when it comes down to it, I would vote for the GFDL (see below)

> Would you like to allow others to take material from the wiki, change and modify it and then redistrubute it?

No. Theoretically, we should know what we are talking about and have put some effort into making sure stuff is correct. We can't stop this from happening, though (unless you just take if offline).

> Would you like to allow others to take material from the wiki and use it commercialy (like selling it etc.)?

Absolutely not. What I know about CFD has come at great personal cost. I don't mean to be a jerk, but if anyone profits of what I know, it should be me. At the very least, the authors of the articles deserve credit. But can we stop this if someone does it without permission? Probably not.

As I typed this, the conclusion I have been moving toward is this: I would rather have a fairly permissive license that allows for reuse, etc. while requiring some attribution than have a license that can't be fully enforced. It seems to me that the GFDL gets pretty close to that, even if it is not the "perfect" solution. I look forward to reading what others think.

jason

Hmm, that answer was a bit confusing. If I understand this correctly the GNU FDL allows commercial reuse of the material and also allows people to create derivate works of the material!

You said that you prefered a GNU FDL license but that you did not want people to be able to use the material commercially or make derivate works of it. That sounds like a contradiction to me. If you want to disallow commercial reuse and derivate works then the creative commons license (in the appropriate variant) is the only option I think.

Here is my preference:

Allow free reuse of material with attribution - yes Allow commercial use - No

I guess my opinion is not decisive, but I think following:

1) We shall provide the future of CFD Wiki

2) We shall provide defence of CFD Wiki

3) We have not to allow to repeat mistakes and troubles of Math Wiki

Of course I am against of commercial use of CFD Wiki. I haven't the slightest of how provide this, but I suppose that Jonas know.

My way in CFD - 10 years struggle with personal sucrifices and failures, and cost is very high (I could create family and buy a good car in money that I invested in my study, and despite that I won't stop and continue my work)

And anyway - We are developing it for public use and humankind in general.

I hope that Jonas as neutral Swedish representative can find a compromise solution between a necessity of editors rights defence and its public usefulness (which means its use by plain first-year student, and possibly by some author, who writes a new CFD-book, and possible by post-gradute student).

I suppose we should provide defence of any commercial use.

As for public use - I have nothing against if some student will use some materials in his work. It can save time of sitting in library and spent there an year (as did I)

I mostly agree with jason that we spend lot of time and money in learning it, but i guess there is some material which by its nature can not be restricted, and on the other hand we have some material where we wish some body else should not make commercial use. Would not it be possible to put two different kind of licenses. Just a thought.

I agree about the free reuse with attribution. I would also like the license to have a "share-alike" statement which requires any derived works to also be available under the same license (ie free). I'm a bit hesitant about commercial re-use. Clearly it isn't fun if someone just takes your work, prints it and sells it as his own. But disallowing commercial re-use has some other implications also. Imagine for example that you write a book or a paper and you cut-and-paste some of the material, say a section on governing equations, from the Wiki. With a license which restricts commercial re-use this would not be allowed since most books and journals are published by commercial companies (publishers).

Two licenses sounds a bit difficult to maintain. How would we handle that? If someone starts an article with one license and then someone else want to add something to the article under a different license ... it would be a mess. I really think everything should be under the same license. If the license can be enforced for all material is another issue.

I doubt anyone is going to be "enforcing" the license anyway. Personally I'm not at all interested in these legal details but I've been asked by several potential new contributors about this so I thought we should have a discussion about it. The primary reason to discuss this is that it might be important for contributors - they might want or they might not want their works to be released under certain licenses and it might affect their willingness to contribute things. Some want it to be free and don't want any restrictions on its use wheras others want it to be restrictive ;-) Many people are also worried about allowing others to modify their texts, but that it an absolute minimum - it is the whole idea of a Wiki.

Thanks for your feedback. See my answer to Praveen concerning commercial reuse. It isn't that easy to have a license which disallows commercial reuse but allows parts of the material to be published by students and scientists in books and papers.

i know two licenses is very messy thing, thats why i wrote its just a thought. I guess we might have two kind of pages, one where we have CFD topics like we have in text book, this page is written by lot of people and it can have mainly the things for which we might need license. There is another kind of content, where people put their work, and they can restrict people from using the matter from commercial purposes. I guess, in such cases author might be contributing the full article. So in that case he can put the legal license some where in article.

So if this two license thing has to be implemented, it then have to be in this form. A general text, which is free for all the people to read and use. Second some ones work, which could be separate article and the author shall put the license then there.

Sorry, didn't mean to be confusing. In spite of the fact that I would rather not allow modification and redistribution (commercial or otherwise), I don't believe that it is something that we can really control. Also, In spite of the fact that I would rather that I profit if anyone does, I would at the minimum like an equal chance to profit. Therefore, I would choose a license (like the GFDL) that explicitly opens things up (with some conditions, most importantly the credit due authors). The GFDL allows *anyone* to commercially distribute it (so if I wished to, I could). I am not a lawyer, but it appears to me that with a creative commons license, the rights holder can still profit commercially. That is (in my opinion) a rather large loophole through which Mathworld-like hijinks could enter. It gives the rights holder a special status that the GFDL avoids (again, not a lawyer).

Does that clear up what I meant? It appears that soem real discussion is happening on this, which is a good thing in my opinion.

It is my turn to be confused. The reuse you describe is either:

1) fair use - Say I write a fantastic article on doing CFD with playstations (sorry I couldn't come up with a better example). As long as they quote my article, give proper attribution, and the quotation is "short," then there is no problem. It would not matter what license is used.

2) plagiarism - If someone publishes words, etc. that they did not write as their own it is plagiarism. The publishers I am familiar with want authors to sign over the rights. If one doesn't have the rights to begin with, one can't sign them over. If someone takes my article on playstation CFD and tries to publish it, the only entity that legally has anything to say about the rights would be cfd-online (no matter what license).

3) strange - If I try to publish my article on playstation CFD, I can't because I no longer own it. Theoretically, I should have known that before contributing it.

Am I missing something? I must be.

jason

Neither the GFDL license nor the creative commons license is exclusive. That means that as the original author you can also publish your work under a different license if you want. For example, you could write an article or a book which you publish and sell commercially under a restrictive license using some of the same material as you also added to the wiki under a free license.

Or in other words, neither the GFDL license nor any version of the Creative Commons license puts any restriction on commercial use of the material from the original author. There is one version of the CC license which puts restrictions on commercial use of someone elses material though.

The attribution + share-alike (allowing commercial reuse) version of the creative commons license is in fact quite similar to the GFDL license. The major difference I think is that the GFDL license requires the whole Gnu license to be distributed with every reuse of any material and it also allows the possibility to specify things that people are not allowed to change.

I don't know if that made anything clearer. I agree that it is good to discuss this now so that everyone has the same view on this. Once we go publich with the Wiki it will also be difficult to change the license.

I'm not sure I understand what you mean. Let me comment on your variants:

1: Fair-use - Fair-use is not affected by any copyrights and whatever license the wiki material has people will always have the right to make "fair-use" of it, like in the way you describe it - quoting a small part with a proper reference.

2: Plagiarism - Only the original author has the right to license something also under a different license. Note that cfd-online will not own any copyrights at all to the wiki material. Each author owns the copyright to the things which he wrote, but by submitting it to the wiki each author also licenses the material under the license which we eventually agree upon (GFDL, CC or whatever) and thereby gives cfd-online the right to use the material in the wiki. The example with using material from the wiki in an article is perhaps a bit confusing since as you say many publishers require the author to sign over copyrights to them. Another example could be that say a linux-cluster vendor selects a few benchmark-cases from the wiki and uses them to demonstrate how well their clusters performs and then distributes the cases + results from the wiki for other clusters etc. as promotion - this would be a type of reuse of material which a non-commercial reuse only license would forbid.

3. If you are the original author you own the copyright to the things you wrote and thereby you can publish it in whatever way you want. The only scenario I can think of where you might have problems is if the publisher does not accept that some of the material has also been licensed under a free license (in the wiki). I think this is a fairly theortical case though - if you decide to write a book or an article based on material which you already submitted to the wiki you would most likely rework and add a lot of things to it and thereby it would most likely not be a problem for the publisher. I mean most scientific books and papers are based on previous material from older papers, internal reports etc. which have already been published and noone questions this.

Having read your responses to my messages, I think I understand now where you are coming from. My wrong assumption was that cfd-online would be the rights holder. As for your response about plagairism, I'm not sure that the situation you describe is not fair use.

Here's a more concrete situation: In my lid-driven cavity article, I refer to a reference with tabulated results - I have the tabulated results in electronic form (I scanned them and then ran the scan through OCR software). I have held off on actually putting up the numbers (I have a plot there with some of the data) because I am not sure that it is fair use, and it would be better if others got it directly from the source (right?). What do you think? From the most technical and picky standpoint, I probably should not even include the plot that I have.

This is probably a pretty important topic for the test cases, because if we start including grids as part of the available content (which is logical) there needs to be a consensus on what constitutes acceptable use.

jason

I agree that we need to think carefully about this when it comes to test-cases. The safest way is of course to contact the original authors and ask for their permission to include their data in the wiki (and thereby license it under a free license).

In the case of the lid-driven cavity case that you have written about - why don't you email Ghia and ask them if it is okay to upload their data to the Wiki. I think their hompage is here: http://cfdrl.uc.edu/ If you would prefer that I email them about it just let me know.