[openbiblio-dev] Some updates

[Jim Pitman]

Mark MacGillivray <mark.macgillivray at okfn.org> wrote:

> I would be happy to bring the discussion onto the list, if Jim wants to move it from the google doc.

I'm fine with the discussion on the list. But we need also a collaborative editing environment to develop 
something like a formal spec of the metadata standards we adopt for bibjson.  I have added all contributors to the discussion to date to 
have edit permission on the google doc "BibJSON Metadata Issues" which is now publicly visible at http://docs.google.com/View?id=dg263gpx_153dfvhqmhm
Just email me with your preferred email address for google docs if you are willing to contribute to editing it.

I dont insist on Google DOcs for this. A suitable Wiki would be OK. But someone has to manage the doc space and the contributors. 
I find Google Docs easy to manage for this purpose, and have been less impressed with alternatives. OKFN etherpads are OK for very rough work only, 
and seem to get out of control as soon as the doc needs an index.

> On Tue, Oct 4, 2011 at 8:55 AM, William Waites <ww at styx.org> wrote:
> > Also I wonder if you guys have been following the JSON-LD work and if it
> > is still a goal of BibJSON to keep it compatible.
> We are not presently following JSON-LD as the decision earlier in the Summer was just to make bibjson as simple as possible.

Right, I expect BibJSON to remain so simple for the next year or so that it should be easy to embed it in a more advanced effort like JSON-LD
when some benefit from doing so arises. If there are interesting biblio data sets already available in JSON-LD, lets have a look at them
and see what it takes to post them to bibsoup. But without datasets, there is little to discuss.

> > On ordering constraints, you are right that JSON specifies an unordered
> > set of keys. However nothing stops things like BibJSON from recommending
> > a certain ordering, and there are serialisers and parsers that will
> > preserve this ordering. See parallel discussion at:
> >   http://lists.w3.org/Archives/Public/public-linked-json/2011Oct/0014.html
>
> Yes, this is one of the points I made on the google doc - the object
> in a file can appear in any order one wishes to write it into the file. If we so desire, we can specify that a bibjson file should
> contain an object with keys in a certain order.

Yes, but this is not JSONic.  The JSON way to specify an order is to use an array. I would prefer that the main elements of
BibJSON should be easily expressible in JSONic terms and not require any further artifice or conventions, e.g. using
JSON Schema http://docs.google.com/View?id=dg263gpx_153dfvhqmhm

--Jim