Dear Adam,

thanks for the pointer. The paper gives an overview of how to design a wiki-based data curation platform for a specific target community. Some of the insights could also apply to Wikidata, while other won't transfer (e.g., you cannot invite the Wikidata community for a mini-workshop to gather requirements).

What I did not find in the paper are numbers of any kind. How do you know that they manage petabytes of data? I also did not figure out how many users they cater for (e.g., they write: 'A small number of testers we called the “seed community” were involved in the testing and experimentation phase. This community generated the initial wiki contents that could then be used to solicit further contributions from a larger community of users' -- but I cannot find how big this small community and this larger community were; this would be important to understand how similar their scenario is to ours).

Anyway, good to know about this recent work. I will send them an email to make them aware of this thread (and of Wikidata).

Cheers,

Markus

On 05/08/13 09:35, Adam Wight wrote:

Dear comrades, I just learned of a system based on MediaWiki which shares many of the same objectives as Wikidata: collaborative data storage and analysis, tracking of provenance, and facilitating citations, to name a few. I'd like to encourage a dialogue with these scientists, I do not think they are aware of your initiative, and they definitely have valuable practical experience after seeing the real-world use of their system. Currently they are managing several petabytes of data.

Research paper by the site creators: http://opensym.org/wsos2013/proceedings/p0301-sowe.pdf

Sorry I cannot link to their site itself-it might require an account...

-Adam Wight

---

Wikidata-l mailing list Wikidata-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-l