search

scienceai / scholarly.vernacular.io

A vernacular of HTML for scholarly publishing
http://scholarly.vernacular.io/
Apache License 2.0
38 stars 10 forks source link

Details of discourse #30

Open kjetilk opened 8 years ago

kjetilk commented 8 years ago

I belong to the school that prefers to be free to choose my ontologies, but I can see that the world doesn't really follow me there. Anyway, where I feel that scientific discourse really has something significant to gain is when you enable discourse itself to be expressed in structured form. There is a tiny start in my Linked Research paper: http://folk.uio.no/kjekje/2015/noise-essay.html if you look in the source code, you'll find stuff like cito:repliesTo, cito:disputes, etc. The CITO ontology is pretty nice, that is something I would like to see adopted.

darobin commented 8 years ago

I think that we need to distinguish between a baselines interoperability framework that basically means that I can write a tool that produces SH and you can write a tool that consumes it, and they'll work together with no magic. That means being pretty strict in sticking to relatively limited semantics that are guaranteed to be understood. It also means that relying on indirection as may be provided through schemata is problematic.

It does not however mean that people can't use extensions on top! I see no reason why you couldn't use CITO in SH. It just means that my consuming tool might not understand it — but that's okay.

We've started a group to talk about this if you're interested: https://www.w3.org/community/scholarlyhtml/

kjetilk commented 8 years ago

Right, and I see that you may need to be restrictive to find adoption. And I trust your experience in this respect. However, I'm just impatient, what is the purpose of all this? Is it something that is guaranteed to be understood by a consumer that wants to produce proceedings on dead trees? Something that can port LaTeX-type scholarly discourse to the Web? Then, excuse me while I go yawn :-P I feel strongly that there has to be something more in it for the researchers to make them consider not doing it the way it has been done hitherto [sic!]