Re: [jats-list] 4. The reason(s) you think JATS works

["Imsieke, Gerrit, le-tex gerrit.imsieke@xxxxxxxxx" <jats-list-service@xxxxxxxxxxxxxxxxxxxxxx>]

Hi Denise,

You are right in both regards: Sasha's language was condescending, and 
stable tag set conventions within JATS are valuable.

In Sasha Schwartzman's (whom I consider my friend) defense I want to say 
that in Russia, where he's from, as well as in other countries such as 
Germany and the Netherlands, conversations are often more direct than in 
the US. And generally on the Internet they are, too, one might add.

But the issue isn't that Sasha's reply was too robust or direct, it was 
arrogant. Just like the bike hobbyist who will tell you: The gear ratios 
are too low? Just mount a larger sprocket and extend the chain a bit. 
This is probably a thing that anyone can do with the right tooling, 
using standard components. As opposed to forging new sprockets from bare 
metal, which probably is the equivalent of conceiving a new tag set from 
scratch.

But for people like me who only want to ride their bikes, assembling 
standard components with standard tooling might already be a stretch.

Apart from that, other people don't care whether my bike uses factory 
configuration, whether I customized it using off-the-shelf components, 
or whether I forged the sprockets myself.

This is different for content. Other people understand, ingest, and 
process content better that adheres to a standard at a predefined level.

So unlike bike configurations, content configurations aren't meant to be 
people's own business.

TEI, like JATS, is also a tagging standard for marking up content, and 
there are certain predefined subsets. But TEI people, unlike JATS 
people, will look down on you if you didn't customize the tag set 
according to your needs, even if your needs are "I want to use an 
interoperable off-the-shelf subset". So Sasha's reasoning was similar to 
what I experienced it in TEI land, without any arrogance intended from 
their part: Customization is easy and anyone can do it. Which isn't the 
point, as you said.

The JATS community is different. There is value in having 
broadly-adopted predefined subsets, and there is value in some degree of 
stability within a given subset. Otherwise frictionless interchange over 
time will be severely hampered and you don't ever stop creating 
Schematrons and ad-hoc XSLTs (which isn't a bad thing per se, but if it 
can be avoided, it should).

--Gerrit


On 01.05.2020 18:15, Denise French dfrenchlibrary@xxxxxxxxx wrote:
> Mr. Schwartzman wrote:
>
> 3) Finally, the built-in customization design principle making it easy
> to create a tag subset or a tag superset was absolutely brilliant. I
> find the whole Swiftian Big-Endian/Little-Endian controversy about
> Greenification of Blue, JATS Lite, etc. quite amusing -- one can
> always create a subset that suits one's need.
>
> Don't sneer at people, it's rude. And you are missing the point of 
> agreed-upon tag sets. It is FAR easier to get a community to agree to 
> all use something standard than to write a custom anything and get many 
> people to use it. The minute you suggest agreeing on a customization 
> everyone wants to change it a bit more, or to have an exception for 
> their convenience, and you don't have interchange any more. THAT'S why 
> we need JATS to be a standard in the first place.
>
>
>
> On Wed, Apr 29, 2020 at 8:29 PM Alexander Schwarzman 
> aschwarzman@xxxxxxxxx <mailto:aschwarzman@xxxxxxxxx> 
> <jats-list-service@xxxxxxxxxxxxxxxxxxxxxx 
> <mailto:jats-list-service@xxxxxxxxxxxxxxxxxxxxxx>> wrote:
>
>     To add to what's already been mentioned...
>     1) It seems to me, one of the less appreciated and apparent reasons
>     that JATS works (or, in my opinion, is more successful than, let's
>     say, TEI or DOCBOOK) is that in the STEM community, compared to the
>     Social Sciences' or Humanities' communities, there have been (a) more
>     consensus and (b) more support/accountability from funders .
>
>     2) Another -- more technical -- reason for JATS success is that it is
>     seamlessly compatible with many technologies; most importantly, with
>     the XML family of technologies, i.e., XSLT, Schematron, and XQuery.
>
>     3) Finally, the built-in customization design principle making it easy
>     to create a tag subset or a tag superset was absolutely brilliant. I
>     find the whole Swiftian Big-Endian/Little-Endian controversy about
>     Greenification of Blue, JATS Lite, etc. quite amusing -- one can
>     always create a subset that suits one's need.
>
>     --Sasha
>
>     Alexander ('Sasha') Schwarzman, Content Technology Architect
>     phone: +1.202.416.1979 | e-mail: aschwarzman@xxxxxxx
>     <mailto:aschwarzman@xxxxxxx>
>     The Optical Society (OSA)
>     2010 Massachusetts Ave., NW
>     Washington, DC 20036 USA
>     www.osa.org <http://www.osa.org>
>
> JATS-List info and archive <http://www.mulberrytech.com/JATS/JATS-List/>
> EasyUnsubscribe <http://lists.mulberrytech.com/unsub/jats-list/225679> 
> (by email <>)