[mdx] What next with MDX

[Ian Young]

On 17 Sep 2012, at 17:40, Andy Dale <dalea at oclc.org> wrote:

> From what I understand the primary ‘open issue’ is the extended syntax proposed by Leif. From reviewing the MDX thread on this I am a little lost and would love a basic explanation of the problem and solution.

Leif might give some idea of the specific motivation.  It's obviously possible to come up with various scenarios in which more complex queries might be useful; the question is really whether it's necessary and whether adding things like this before we even have experience with the basic use cases is the right way to be going.

> There is mention on the list that the spec has ‘expired’. Presumably that is an IETF spec status thing.

That's exactly right, and that's all it is.  Chad didn't feel that there was enough active interest at the time to make it worth just renewing the internet-draft every six months.  If there is now some likelihood of people coming forward and attempting implementations, maybe that has changed.  Having it be a current as opposed to an expired internet-draft isn't that big a deal, in my view.  Having said which, it's relatively easy to do.

> Do we as a group want to set a goal to bring the IETF spec into any particular state of formality?

It wasn't headed for the standards track anyway, I'd note.

> I believe that Chad was the sole editor of the original spec and that he is no longer able to contribute at that level (or maybe I’m wrong about that).

That's true, the nature of Chad's work has changed such that he won't be able to participate going forward.

> Is there anyone else who wants to edit? I would be willing to try, I have contributed to several specs but never been ‘the editor’ before.

Chad has transferred the repository to my github account ("iay") for the moment:

	https://github.com/iay/md-query

This process has preserved the open issues, etc.

If anyone feels that I shouldn't be the one to pick the spec up from Chad, they should let me know either here or privately.  Or just clone the repository and make a fork, of course.  I think my views were reasonably aligned with his, but I don't think he thinks of this as a bequest ;-)

> Are there other use cases, extensions, clarifications that we want to add into the mix?

There may be, particularly on the clarification side (having new eyes on the spec will tend to do that).  My own tendency is going to be to want to see real justifications, though, for any extensions.  There's a danger when adding stuff that "might be useful one day" that we turn a very simple access protocol into something a lot more heavyweight, with very different performance characteristics.  That's fine if it's necessary and well thought through; not such a great idea if we're just doing it because adding things is fun.

> (I will send a separate email with my specific problem and what I was thinking about a solution before I became aware of MDX).

Looking forward to it.

Cheers,

	-- Ian



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.iay.org.uk/pipermail/mdx-iay.org.uk/attachments/20120917/0997725e/attachment-0002.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4813 bytes
Desc: not available
URL: <http://lists.iay.org.uk/pipermail/mdx-iay.org.uk/attachments/20120917/0997725e/attachment-0002.bin>