Authoring versus Interchange

Were I in a philosophical mood I might suggest that science creates
simplicity whereas engineering makes complexity manageable. For the most
part, we're here to do the latter.

This matters because scholarly information can be messy and complex, and
Web technology can be messy and complex, so we need to decide whose life
we are to simplify and whose rug the complexity gets swept under.

Of the formats and solutions on offer out there, some focus on
simplifying authoring (e.g. Authorea, Overleaf) while others try to
establish interchange (e.g. JATS).

Of course, when you can have both there is no reason not to. For
instance that is why there is no reason to use XML for SH: it would make
authoring harder without making interchange any more reliable. But when
authoring and interchange are at odds, we need to pick which wins. In my
experience this drives many core decisions that then follow.

My position is strongly that what is needed first and foremost is an
interchange format. There are several reasons for this.

The primary one is that the most conservative part of the ecosystem —
and that is saying a lot in the scholarly publishing world — is the
authors. The numbers I've seen point to ~90% of them using Word (of
course, in some domains that's 100%), with most of the rest on LaTeX
(and some of the alternatives are even scarier).

Don't get me wrong, there is a small but growing community of people who
want to author papers in HTML, and they should be supported. But the
easiest way to help them is to open the door to experimentation. They
should be allowed to invent authoring formats that suit their personal
tastes, and target a commonly accepted format for submission. One
example of such a tool would be ReSpec[0], which is what a lot of W3C
people use to go from a nicely editable format to a stricter output.

An interchange format can foster innovation not just for authoring
tools, but all around. Once you have that pivot point, you can write
(and share) tools that import from archaic formats, that process it
usefully, reuse them in new projects, etc. Eventually you can enrich
more of the ecosystem.

Authoring formats also tend to be a lot about taste. Some people
actually like Word. Of those who like to author in HTML, some don't mind
weird embedded microsyntaxes while others only swear by sticking to the
core constructs of HTML. Some people even like Markdown! With an
interchange format, they can all happily coexist (so long as they don't
need to co-author too much ;). We want an ecosystem, not a straight-jacket.

[0] https://github.com/w3c/respec

-- 
• Robin Berjon - http://berjon.com/ - @robinberjon
• http://science.ai/ — intelligent science publishing
•

Received on Tuesday, 1 December 2015 04:14:05 UTC