[08:24:39] hi all. I'm back and will be around for couple of days before I disappear again for the weekend and couple of days off. [13:18:34] harej: I put in a ticket to make a librarybase project on Phabricator. Thought it might help us keep on top of everything [13:19:56] Sounds good. (I had just been using the Source-Metadata, now WikiCite, tag.) I have some other tasks I would assign to the Librarybase tag [14:19:41] I've still not made the properties; will try to get that done after work today [15:19:18] halfak: I remember seeing your dataset of citation identifiers of Zenodo but can't find it anymore [15:32:29] halfak: anyway I just uploaded https://zenodo.org/deposit/121152/ [15:32:41] and I thought it would be good to link back to yours [15:33:29] o/ pintoch [15:33:44] "Deposition does not exists." [15:33:50] oh sorry [15:34:05] https://zenodo.org/record/55004 [15:34:34] and some stats here: https://github.com/mediawiki-utilities/python-mwcites/issues/10#issuecomment-223993499 [15:37:38] Cool. So the parser you are using works specifically with the cite templates? [15:37:57] yes, it parses anything that uses CS1 [15:38:14] (it's a wrapper of CS1's Lua code) [15:39:33] Here's the DOI link for the citation ID extractor datasets: http://dx.doi.org/10.6084/m9.figshare.1299540 [15:39:55] ah it was on figshare, ok ^^ [15:39:58] I've been planning to work backwards from DOI/ISBNs to metadata by calling APIs like crossref and oclc. [15:40:16] pintoch, figshare is nice because it gives you a DOI for your dataset. [15:40:22] You might consider listing it there :) [15:40:24] Zenodo does too :-) [15:40:30] Oh! Cool :) [15:42:50] pintoch, would be nice to merge those datasets some time soon and get them into librarybase :) [15:43:06] * halfak needs to finalize his metadata extractor scripts. [15:47:12] yeah, it would make sense [17:42:57] tarrow: if I am using content mine for keyword extraction, is there anything I need to do to populate the dictionary so that it knows what to look for?