Draft VOEvent 2.1

[Roy Williams]

Dave Morris wrote

> Software for
> implementing the science use cases should be developed based on higher
> level libraries such as the FourPiSky voevent-parse toolkit, which
> represents event messages as Python objects, not as XML or JSON.

Dave
This sound like a fabulous idea. 

The software you mention is from Tim Staley, and is based in the lxml toolkit:
    http://voevent-parse.readthedocs.io/

The STC schema is here (it is 3,600 lines):
    http://www.ivoa.net/xml/STC/stc-v1.31.xsd

Note that the VOEvent 2.0 schema is 398 lines, 
    http://www.ivoa.net/xml/VOEvent/VOEvent-v2.0.xsd

The proposal here to multiply the size of the schema by a factor of 9, to multiply the complexity by 9, to make it 9 times more difficult to rebuild VOEvent in anything other than XML.

To properly understand the argument above, I would suggest the following should be provided by a willing volunteer:
(1) A prototype VOEvent 2.1 that links to the STC schema and uses its elements, and
(2) An extension of the above software that can parse it.
(3) It would also be great to parse another way, eg Java, python etree, etc -- all of which are possible with the much simpler 2.0 schema.

Years ago, I recall a significant challenge from the inclusion of STC. I look forward to seeing the above achieved. Then we can have this discussion with our eyes open.

Roy

---
Royal Observatory Edinburgh
roy at roe.ac.uk
07542 869986