> tools/environments/products support XML, then the need for knowing XML
itself gets less important.
I am comparing xml to assembler. IT staff generally use higher level
languages and not get worried by assembler itself.
Interesting analogy, one way to answer this question is to think of what
the IT persons will be doing with the XML, as well as the industry trends
and market forces that will shape the future of XML. To help identify
the issues, consider the following "LEVELS" of XML use:
### LEVELS OF XML USE
LEVEL1
XML AS DATA INTERCHANGE FORMAT (aka XML compared to CSV)
If you are familiar with CSV files, you know that this is a widely used
format. CSV is not 'hierarchical' like XML, and has fewer built-in
'delimiters', but serves a similar role. In this category, IT people
use the syntax format simply to get information out of one application
into another application (e.g., out of a spreadsheet and into a payroll
database).
LEVEL2
XML AS SYNTAX FORMAT FOR PROGRAMMING LANGUAGES (aka XML for SVG, XSLT etc)
Since the XML syntax format is 'hierarchical', it allows for
more elaborate expressions, and some people have taken this expressiveness
to create new programming languages. Given this 'level' of use, you can
expect to see all the dynamics in the XML arena that you see in other
'level2' arenas (e.g., programming IDEs, books and tutorials, religious
wars, splintering of functionality and semantics into 'cliques', constant
'improvements' to the language, etc. etc. ad nauseum)
LEVEL3
XML AS BASIS FOR 'SEMANTIC INFRASTRUCTURE' (aka XML as basis for SEMANTIC WEB)
On this level, XML is supposedly used as a way of 'tagging' various
information resources to make it easier for a computer to 'read' text, just
like human beings. I say "supposedly" because there is considerable debate
about whether this level of XML use will ever happen the way its principle
advocates say it will. (see
http://www.well.com/~doctorow/metacrap.htm)
### THE NEED TO KNOW XML
If your IT staff anticipates mostly "LEVEL1" usage, then XML itself will
not be very important to learn.
If more "LEVEL2" usage is in your future,
then indeed the very most minute details of XML should be well within your
grasp, as you will be spending a considerable amount of time wrestling with:
1) DELIMITER COLLISIONS: (e.g. how do I use a 'greater than sign' without
the XML parser thinking its part of an XML tag?; how do I output an
entity reference without XML interpreting it as an instruction and
converting it to something else?)
2) MEMORY JOGGING: (e.g. what is that keyword for specifying background
again? what is that tag for specifying a loop?)
3) KNOWLEDGE CARRY-OVER: (e.g. I know how to make a link in HTML, how do
I do that in SVG, is it even possible?; I know how to define a variable
in PASCAL, how do I do it in XSLT, is it even possible?; I know how to
do a SELECT WHERE clause in SQL, how do I do that in XPATH??)
4) LEAKY ABSTRACTIONS: This is a corrolary to KNOWLEDGE CARRY-OVER in the
case where the 'carry over' seems sensible, but actually breaks down
and causes more problems than it solves.
(see
http://www.joelonsoftware.com/articl...tractions.html)
If more "LEVEL 3" usage is in your future, you will probably be spending
most of your time trying to convince people that your vision is worthwhile,
practical, feasible than you will be in learning the nuts and bolts of
XML.
### MARKET FORCES
LEVEL 1 uses will predominate where practicality and quick turnaround
are paramount. The need to know XML is the least on this level since
people are not as interested in how the data got exchanged, just long
as it did so successfully. Also, market forces will cause LEVEL 1 people
to use whatever tool gets the job done right and cost-effectively, which
may or may not be XML, depending on a case-by-case basis.
LEVEL 2 uses will crop up and be numerous, because LEVEL 2 is a
favorable level for writing books, selling software, developing "expertise"
and having stuff to talk about in usenet groups. This means a need for
intimate knowledge of several of the in's and out's. You can buy my book
for more details.
LEVEL 3 forces will tend to be the stuff of PhD dissertations and
academia. It will inspire mostly the devotees of XML, those who
love XML because it's XML, not because they want *any* old tool, but
because it's the XML revolution.
