Public Domain (PD) License for Metadata

open-data

#1

Background:
Metadata for data (open data) or other content (e.g. bibtex entries) must be considered an independent work. Like written source code it contains information and needs time and effort to be gathered. Thus it contains an investment and is protected under copyright regulations.
UPDATE: This sentence is inaccurate and/or false. See comment below!

For the OEP we gathered a large amount of metadata and want everybody to use it!

  • Just ignore it and trust that nobody will sue anybody? A good option, but we take the OPEN serious.
  • So why not simply license it? Because “normal” open licenses have an attribution (BY) obligation which needs to be followed and create additional work.
  • So we use a Public Domain license! YES, but which one?

I found this presentation about licensing of metadata:

These two options have been found:

  1. CC0 1.0 Universal
  2. ODC Public Domain Dedication and Licence (PDDL)

Since I have never used a PD license before, do you have any experience or suggestions?


#2

Hello Ludwig. Try Kreutzer (2011) who analyses the CC0 dedication. His report is useful for servers located in Germany because he quotes German law. Generally the CC0 dedication is recommended over the PDDL dedication for software for reasons I don’t have at hand.

By the way, your opening premise is not quite correct. Copyright protects originality not effort, albeit at a rather low threshold. For instance, the German Copyright Act (UrhG §2(2)) says (emphasis added) “Only the author’s own intellectual creations constitute works within the meaning of this Act.” Other jurisdictions use similar but not identical criteria. The European database right protects effort irrespective of creativity, but that effort must be substantial. Whether either right applies to the metadata associated with an energy dataset depends on the details. HTH, Robbie.

Update

On reflection, it seems to me that a BibTeX entry would not meet the threshold for copyright. Indeed, any number of people should, when presented with the same document, produce, more or less, the same record. But a collection of BibTeX entries which required substantial effort to assemble would attract a European database right. Note that a database right can be waived with a simple statement to that effect.

Another little known Creative Commons instrument is the Creative Commons Public Domain Mark (CC PDM), used to mark works that are already in the public domain. This instrument is to allow such works to be more easily discovered and recognized as such. It should not be used for waiving copyright: the CC0 is provided for that purpose.

The Open Data Commons PDDL dedication is worded in database terms. It might therefore be more appropriate for databases. As noted earlier, the CC0 is strongly preferred for software.

References

Kreutzer, Till (2011). Validity of the Creative Commons Zero 1.0 universal public domain dedication and its usability for bibliographic metadata from the perspective of German copyright law. Berlin, Germany: Büro für Informationsrechtliche Expertise.


#3

Hello again Ludwig. Aliprandi and Piana (2013) makes a strong case for the use of CC0 for data released by public administrations. But they do not traverse the licensing of metadata. Robbie.

References

Aliprandi, Simone and Carlo Piana (28 March 2013). “FOSS in the Italian public administration: fundamental law principles”. International Free and Open Source Software Law Review. 5 (1): 43–50. ISSN 1877-6922. doi:10.5033/ifosslr.v5i1.84.


#4

I agree with that. The document you quoted (Kreutzer) states:

Bibliographic metadata are generally not copyright protected under German copyright law. Primary and secondary metadata (as defined in 3.1.1 above) that consist of mere information are never copyright protected. However complementary “metadata” like abstracts, forewords, reviews, covers and other works can and will in many cases be copyright protected.
-> Conclusion

That’s the point I want to discuss. The matadata includes a description of the data set and descriptions of each column. In future versions we will also include quality criteria and we work towards an open peer-review of the data of the OEP.

This option is also mentioned in the linked presentation and the document you provided. It is not suggested for Germany, I read it somewhere but now I have a source for it (thanks to you @robbie.morrison).

Notwithstanding there are good reasons to publish bibliographic metadata under the CC0 rather than using the Creative Commons “Public Domain Mark”. Since the Mark is no contractual declaration, especially no offer to grant a license to use actually copyright protected works, its use makes sense only if the worldwide legal status of the respective intellectual good as “free of rights” is definite.
-> Conclusion

This conclusion is quite important and supports my effort to think about the license of the metadata:

Applying the CC0 to actually unprotected metadata is unproblematic since it does not affect the legal situation (if the content is free, it stays free, the CC0 has no legal effect). Furthermore the CC0 can – even if the particular content is not protected – be useful since it informs the user that the affirmer wants his content to be used without restriction, hence will (and can) in case of doubt not enforce potential rights. Additionally the license text of the CC0 helps the users to understand what the meaning of public domain actually is.
-> Conclusion

That would be interesting now. The only one (but a good one) is that the CC0 was investigated by a German lawyer and we have a nice document we can quote :wink:

Big thanks to you @robbie.morrison for another fast and valuable contribution!


#5

Hi Ludwig. In my view, one of the most valuable things that our community could do is develop data and metadata standards for the energy sector information we use. That is ultimately a long road, ending in a systematic treatment (an ontology) of all core information. But the development path can be tackled incrementally, starting with basic metadata. If we could agree a basic metadata standard, then we could start adding metadata to the datasets we circulate. And then start adapting or writing tools that can process those datasets. Unfortunately you appear to be the only person working on the problem at the moment. Robbie.


#6

Just (1.1.18) started another 3 years in a BMWi funded project “SzenarienDB”. One AP is to create (begin) such ontology :wink: First step on the road…

We already agreed on that standard for the OEP! It’s implemented and can be updated if necessary.
Since our community is (should be) aware of this standard and there is no counter proposal it is de facto accepted.

Unfortunately you are right with that but I’m persistent :wink:
In our recent projects we all use the templates that we developed in last years.


Do-a-thon: Draft for 'energy' datapackage standard (?)
#7

There is some movement in the metadata world.
DCAT-AP will be the new metadata standard for governmental data.
Most German data centres will be implementing it.
There is a European Joinup with lots of info and a GitHub repo for feedback to the German version.

An adaptation as the OEP metadata is supposable.

To come back to the original topic: The DCAT-AP is licensed under ISA Open Metadata Licence v1.1 .


#8

Hi @ludwig.huelk Just to note that the ISA Open Metadata License is not currently listed by SPDX (mostly software licenses) or the Open Definition (mostly data and content licenses). Robbie


#9

Thanks for the hint. I opened an issue here.

My first impression of the license is that it will meet the open criteria. I still prefer a PD license for metadata. These obligations look silly:

In accordance to Commission Decision 2011/833/EU, You are herewith granted a worldwide, royalty-free, perpetual, non-exclusive licence to Use and re-Use the Works and any modifications thereof for any commercial and non-commercial purpose allowed by the law and provided that the following conditions are met:
a) Distributions or communication to the public must retain the above Copyright Notice;
b) Distributions must retain the following “No Warranty” disclaimer;
c) Where possible and practical, distributions or communication to the public will provide a link to the Joinup platform;
d) Acts directed to mislead others or misrepresent the Work, its content or source are prohibited;
e) You will not use the name of the European Commission and that of its Contributor(s) to endorse or promote products and services derived from the Use of the Work without specific prior written permission.

[1] ISA Open Metadata Licence v1.1 https://joinup.ec.europa.eu/licence/isa-open-metadata-licence-v11 Copyright © European Union