Re: [jats-list] PLOS blog about relative merits of XML (JATS) and HTML in workflows

[Wendell Piez <wapiez@xxxxxxxxxxxxxxx>]

Hi,

I love this blog post and this question. It is an excellent question
and very well-posed.

As an XML/XSLT practitioner, one might expect me (I might expect
myself) to be threatened by the suggestion that XML (and by
implication XSLT with it) are soon to obsolesce next to
HTML5/CSS/Javascript/Markdown. But the more I think about it, the more
I feel like Bre'r Rabbit in the story. "Don't throw me into that briar
patch!" The thing is, I don't really do XML/XSLT. They're just tools I
use. What I really do is data modeling and transformation.

In the world Molly Sharp describes -- markdown facing authors, HTML5
to clients -- there will be just as much data modeling and
transformation as ever. It's true there may not be as many boring
transformations to code and maintain. I actually welcome this, having
written enough stylesheets turning my 'para' element into your 'p'
element. But this by itself doesn't go very far to address the
long-term challenges of design and development. Most especially, it
doesn't, in itself, take care of all the things that HTML5 (powerful
and fun though it may be) doesn't take care of out of the box, the
things now being modeled (in the web space) using microformats, RDF,
home-grown semantic applications of CSS, what have you. Designing,
building and managing the systems that exploit these capabilities, not
only to enable new networked applications (for which stable
interchange formats are a necessity) but also newly differentiating
features, is a growth industry -- Ms Sharp is only pointing the way
here, she didn't make it up. Yet this work -- one might call it the
work of aligning the processing semantics of the machine with the
formal semantics of human purposes and concepts, making the networked
computer more than a platform for exchange of information scraped from
elsewhere, indeed an expressive instrument in its own right -- is
exactly what developers of XML systems and SGML systems before them
have been doing all along. Not all of us, to be sure, and much
interesting work in this area has happened outside XML as well. But in
my experience, this is what most XML people do and want to do, and XML
has been only a serviceable means to this end, not an end in itself,
for us. We *like* CSS, and even HTML of whatever flavor, when it is
done right and serves the purpose, and we always have.
(Notwithstanding what sorts of grumbling you may have heard.)

This work will still go on and indeed we can hope that the next
generation of web technologies will accelerate, not hinder it. There
will be many opportunities for fun, interesting, meaningful and
profitable information modeling, mapping, transformation and
reflection. My own suspicion is that even if XML-based systems are not
visibly at the forefront of all this, they will continue to be used,
at least by the most creative and innovative, in back. I already know
more than one outfit for whom an XML-based information processing
chain is something of a secret weapon, which they are just as happy
not publicizing, and don't see a replacement for in web technologies.

So rather than answer the question, exactly, I would like to reframe
it a bit. To an extent I think the conclusion is based on the premise.
"If XML is being used as an interchange format only" then by
definition it has no other use. I would also ask:

* What are you interested in accomplishing and how might it go beyond
publishing and interchanging data?
* In particular, are you concerned (or interested) in information
retrieval and organization, in long-term archiving, in experimenting
with the media or the processes, or in interpretation and analysis? Or
are you willing to let others take care of these for you (since it's
intellectually challenging and potentially expensive work, which can
be outsourced to experts)?
* Do you plan to be innovative, and where will your innovation focus?

If your answers are "Not much", "No, it's fine for me to contract
others to save my work for me and to lead development of the next
generation of processes, tools and techniques", and "No, I'm not an
innovator or early adopter, I'd rather buy the solution at the store",
then maybe you'll soon be moving into HTML5 (with a back end built for
it) and eventually to whatever succeeds it. (If you're lucky you won't
even much spend time on XML and its extra overhead. When necessary,
you'll contract others to make it for you.) There's a place for
Wordpress and everything like it.

However, if your answers are that do you have other needs and
interests, you look forward to innovating, and you also feel it's
right for you to invest in capabilities aiming at longer-term
strategic objectives, and not only focus on the immediate horizon,
then I expect you'll have uses for XML along with all these other
things for the foreseeable future. At the very least, you'll
understand and respect it, its capabilities and principles, and what
it teaches about abstraction and layering. (Or you may find your lunch
being eaten by someone who does.)

Like any other powerful thing, XML can be badly designed or poorly
fitted. There aren't easy answers here and there never have been. You
might think of an XML-based format (or at any rate one you didn't
design yourself) as being like pre-fab housing: at best utilitarian,
and rarely attractive, and I might agree that depending on what you
are doing with it, you might be right. But to me, XML and its
supporting technologies are more like power tools for making any kind
of information you like.

Cheers,
Wendell

http://americanfolklore.net/folklore/2010/07/brer_rabbit_meets_a_tar_baby.htm
l

On Thu, Jun 6, 2013 at 10:24 AM, Sheila M. Morrissey
<Sheila.Morrissey@xxxxxxxxxx> wrote:
> Apologies for cross-posting, but I think this might be of interest to both
of these lists.
>
> Molly Sharp, newly appointed product manager for content management at PLOS,
has this post on the newly-created PLOS technology blog.
>
>
http://blogs.plos.org/tech/structured-documents-for-science-jats-xml-as-canon
ical-content-format/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%
3A+plos%2Fblogs%2Fmain+%28Blogs+-+Main%29
>
> Her query
>                If XML is being used as an interchange format only, what do
we gain from moving the XML piece of the workflow any further upstream from
final delivery?
>
> Would be very interested in thoughts from members of these two lists.
>
> Sheila M. Morrissey
> Senior Research Developer
> ITHAKA
> 100 Campus Drive
> Suite 100
> Princeton NJ 08540
> 609-986-2221
> sheila.morrissey@xxxxxxxxxx
>
> ITHAKA (www.ithaka.org) is a not-for-profit organization that helps the
academic community use digital technologies to preserve the scholarly record
and to advance research and teaching in sustainable ways.  We provide
innovative services that benefit higher education, including Ithaka S+R,
JSTOR, and Portico.
>



--
Wendell Piez | http://www.wendellpiez.com
XML | XSLT | electronic publishing
Eat Your Vegetables
_____oo_________o_o___ooooo____ooooooo_^