[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
RE: Designing Information for Eternity
- From: cbullard@hiwaay.net
- To: xml-dev@lists.xml.org
- Date: Tue, 19 Nov 2013 11:41:25 -0600
Robert Costello asks: "XML has been around for 15 years. Are there any
XML documents that are 15 years old and are still useful and being
actively used? Will they be used 50 years from now?"
Yes. Most of my projects have been for information that old or older.
It is a matter of the lifecycle of the systems being documented. XML
does not have a standalone lifecycle where used for documentation in
contrast say to messaging. It is coupled to the system being
documented. The first quality for markup way back when (say
Tunnicliffe, gencoding and GML) was transportability of information
among systems (moving files to be printed among publication houses).
The second quality emphasized some years later at the dawn of SGML was
preservation among publications systems when the system being
documented has a longer lifecycle than the publication system (think
Interleaf and movement, repurposing and reproduction among vendors,
contractors, manufacturers, enterprises, etc.).
The lifecycle and transportability qualities have been proven numerous
times as others can attest. This is not reliant on the XML systems,
say XQuery or XSLT, and in fact, works even if these die. The basis
for this is the syntax basis for the specification and the implied
nodeness (well-structured). IOW, as bags o' tags go, it is durable,
flexible and ingestible. Even a simple ASCII editor will do.
Everything past that is for the convenience of the humans even where
they are using tools. XML Doesn't Care. That's the beauty of it.
The project I am currently working is the first time I've had to
consider this from the front end of the life cycle in many years. I
find myself at the end of my career repeating the beginning of my
career. As I noted, I'm glad I was around for the early years of
markup because now it is not only sensible (concepts have been proven)
but easier because the markup technologies have endured. I don't need
logical proofs. I have pragmatic first person reports.
For the two qualities noted above, XML Works Because It Endures. I
cannot say that about a single line of code or processing system I've
used except possibly SGMLS which still lives in the IADS app even if
seldom used.
len
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]