[
Lists Home |
Date Index |
Thread Index
]
HI Uche,
Uche said:
This is an important point. I was a huge enthusiast of DOM at the
beginning.
I thought they got it exactly right using IDL for language-agnostic
specification. But at that time my Zen of XML was pretty thin. As I've
understood more deeply how XML is more than yet another data format for
programmers to use, I've realized that the XML should inform the
programming
idiom, not the other way around. Given that I use Python for XML
processing,
and that Python is, regardless of any other value, a language rich in
programming idioms, I realized that there were many very rich ways to
process
XML, and that DOM acted as something of a jail cell restricting me to
one
approach.
Didier replies:
I also have been trapped in the DOM prison whishing to get more
appropriate and cost efficient ways to process XML documents. And I
totally agree that Python is a rich language that would allows to
include useful and productive constructs to process XML documents at a
level beyond the simple syntaxic level. Something more at the problem
solution level. When we read Python programs we should have a good clue
of the type of problems we try to resolve. I think that what has been
forgot is that high level programming languages are for human beings and
therefore readability and efficiency to give info about the resolved
problem are important factors. If that wasn't the case, machine language
would be sufficient (even assembler is an attempt to make it easier for
human beings).
Most of the time, the DOM is simply noise and not relevant to the
resolved problem. However, I can say that the DOM was a first step but
it should not the last. I represents an early stage attempt to resolve
the interface problem between the processing of XML document and the XML
documents. But like I said, it is only a first and immature attempt.
There is better to be done.
Cheers
Didier PH Martin
|
