Creative Commons Public Domain Mark

[Marius Kempe]

Hi,

Creative Commons has announced the release of the Public Domain Mark tool today:

http://creativecommons.org/weblog/entry/23830

This could be an excellent tool for IMSLP to use, as it would help make the public domain status of the uploaded clear, and would allow computers and search engines to understand whether or not each work was in the public domain, and thus to automate the categorizing of works.

What do you think?


Best,
Marius

[KGill]

I'm not certain why this would be any better than what we have now. The copyright status of each file is clearly marked - check the copyright tag. V/V/V would signify PD in all countries we look at (Canada, the US, and the EU). If something is under copyright in one of those areas, red text appears in the 'Copyright' field (e.g., 'Non-PD EU'). Also, a large infobox usually appears at the top of such a page informing the user of the fact that it's under copyright in some places. Finally, there are categories containing works broken down by their copyright status. Aside from the fact of our own system, how would we implement it? Is it intended as a watermark for PDF files?

[Boccaccio]

I also am a bit sceptical about this sign... to me it sounds like an invention of the copyright mafia, as a new tool to block access to certain things one day...but maybe I'm a bit paranoid when it comes to these things.

[Marius Kempe]

The main benefit is that this will be a widely used mark, and so it will be much easier to combine IMSLP resources with other resources from museums, libraries, etc., etc., in the future. Ideally, we'll be able to simply go on Google and search for scores which are public domain, and Google will able to reliably tell which ones are public domain because of the mark.

In other words, it's main benefit is in making the copyright status of a work machine-readable, and in standardizing the machine-readability across websites - otherwise search engines will have to learn the different homemade codes of each and every website, like the V/V/V we have here...

[KGill]

How does it work, though? (I can't find any specific information on it.) Would each PDF file have to include the image, or be tagged with specific metadata? Or would it simply be a searchable text field on the page?

[Marius Kempe]

It's very simple; you just include some HTML in your page, which looks like this:

Code: Select all

<p xmlns:dct="http://purl.org/dc/terms/">
<a rel="license" href="http://creativecommons.org/publicdomain/mark/1.0/">
<img src="http://i.creativecommons.org/p/mark/1.0/88x31.png"
     style="border-style: none;" alt="Public Domain Mark" />
</a>
<br />
This work (<span property="dct:title">1st Symphony</span>, by <a href="http://imslp.org/Composers/Beethoven" rel="dct:creator"><span property="dct:title">Ludwig van Beethoven</span></a>), identified by <a href="http://imslp.org" rel="dct:publisher"><span property="dct:title">IMSLP</span></a>, is free of known copyright restrictions.
</p>

Obviously we can customize the visible text however we like. The great thing about it is that the HTML tags are semantic, so search engines like Google and the Creative Commons search engines can detect the copyright status of the work automatically.

I believe it's currently only applicable to works that are in the public domain everywhere.

I'd highly encourage IMSLP to consider adopting it - standardization and searchability are good things.

[Hobbypianist]

I also share the concern that IMSLP should remain independent and not link the definition of public domain to third parties. We already have a established and well working tagging system which indicates the status of every single file for the 3 main regions. If a user wants to download a file he/she must see the copyrights tags before to find out whether it's free in his/her region or not. I don't know how it'll work but if Google just shows the link to the file users may download it directly with disclaimer but without seeing if it's permissible or not.

If the HMTL code is included only on work pages then it's not feasible because the work itself might be in public domain but certain editions may not. If you have a look at Beethovens 1st symphony there are some new typesets (which can't be public domain). And what if a new typeset is added or edition is added which is not public domain in USA/EU? Then we'd have to remove again the html code.
On the other hand if including of the code in file pages is possible we'd probably have to give extra info and not only for example "2nd Piano Sonata" because often we have several editions of the same work. Checking 86000 existing file pages and adding the copyright status is alomst impossible and above all would mean double work.

[steltz]

It seems to me that even if the code is typed into the section of a page that is specific to one edition, a search that finds that code will still bring up the normal work page with all editions available. This might imply that all the editions are public domain. At the moment there is no way to show one edition on an individual page when the work has multiple editions.

This could be very misleading. More clarity is needed on exactly where this code would go and what it brings up in the search.

[haydenmuhl]

I see this being of use more in the documents themselves than on IMSLP. For example, many competitions do not allow loose leaf music because of copyright concerns. A mark like this on the document might be a good way for such organizations to identify legally copied sheet music. Similar to the way CD Sheet Music puts their logo on all that music as a way to signal that it's legit.

There are several obvious problems to be worked out to make this feasible, though. A universal public domain symbol will not work for documents under copyright in some countries, but not others. There should be some way of identifying on whose authority the document was marked public domain. The notation of all this will need to become standard, either explicitly or de facto, before that can become a reality.