Is XML for the peons or the gods? (was What is XML for?)

[David Megginson]

Tyler Baker writes:

[on CORBA]

 > All of the technology pundits were preaching that it could be used
 > for just about everything, pretty much the same list of items that
 > people are preaching XML to be used for.  In the end I scrapped use
 > of CORBA for use as a dumb messaging layer in an application I had
 > (using CORBA in the first place wasn't one of the most intelligent
 > decisions I have made in my programming life but I was told by all
 > of these "experts" that is was super hihg-performance and all of
 > this other great stuff).  That is not to say CORBA is a bad
 > technology, but I was using it for all of the wrong reasons.  

This is a very good example of the "Golden Hammer" antipattern in
software system design.  CORBA showed (and still shows) great promise
for solving a large set of problems, but CORBA specialists suddenly
started to see *every* problem in terms of CORBA even when it wasn't
the best fit.

Let's not do this with XML (that's all we old grannies are asking).

[snip]

 > CORBA now for all intensive purposes is dead in terms of momentum
 > and most people I know of have totally lost interest in it
 > altogether.

Nah, CORBA's still alive in the enterprise (it's very successfully
used by a customer of mine who happens to be in the Fortune 50, for
example), and it's certainly showing more of a pulse than DCOM as far
as I can tell.  On the light-weight side, Gnu's Gnome desktop
environment is heavily CORBA-based, using a tiny, fast ORB called
"Orbit" (I'm running Gnome right now and it isn't causing any
problems).

CORBA's a nice idea, if a rather clunky implementation in the specs
(not specifying the communication between the stubs and the ORB was a
brain-dead choice, since compiled CORBA clients were, as a result, not
binary-compatible across ORB vendors -- I heard they fixed that for
Java).

[snip]

 > So it basically boils down to is XML for a few really complicated
 > tasks that require "gods" to implement, or is XML for a large set
 > of general tasks that even "peons" can implement.

Yes and yes.  XML scales nicely, and I can do a lot of useful things
with under 100 lines of Perl code (including keeping all of my
company's books to meet the nasty reporting requirements of a
Federally-incorporated Canadian company).  It happens that we
understand the simple stuff pretty well now, so we're starting to try
to figure out how to do the hard stuff -- please don't take that as a
slight against the simple stuff, which is really the backbone of XML
(just as handwritten HTML pages and Perl-based CGIs are still the
backbone of the Web).

<aside> 
 By the way, on a separate note, I find it disturbing that people like
 Paul Prescod, Eliot Kimber, and I have morphed from the young Turks in
 the SGML world a few months ago to the old grannies in the XML world
 now.  It feels like the Star Trek episode where everyone aged rapidly.
 I was in Kindergarten when the Beatles broke up, dammit.
</aside>


All the best,


David

-- 
David Megginson                 david at megginson.com
           http://www.megginson.com/

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)