Conceptual flaw in pxdom?

Paul Boddie paul at boddie.org.uk
Tue May 19 17:44:30 EDT 2009


On 19 Mai, 18:16, "Diez B. Roggisch" <de... at nospam.web.de> wrote:
>
> Sorry to say so, but that's nonsense. DOM is not complicated because it
> contains anything superior - the reason (if any) is that it is formulated
> as language-agnostic as possible, with the unfortunate result it is rather
> clumsy to use in all languages.

Although I presume that people really mean the core standards when
they talk about "the DOM", not all the other ones related to those
core standards, the API is not to everyone's taste because, amongst
other things, it uses functions and methods when some people would
rather use properties (which actually appear in various places in the
standards, so it isn't as if the W3C haven't heard of such things),
and for lots of other subjective reasons: some I can agree with, some
I put at the same level as a lot of the API-posturing in numerous
domains where Python code gets written, where such code jostles above
all other concerns for the coveted "Pythonic" label.

However, when people are actually choosing to use DOM-related
technologies, and when those technologies do not necessarily have
equivalents in whatever other technology stack that could be
suggested, can we not just take it as read that they actually know
that the DOM isn't very nice (or that other people don't think that
it's very nice) and that there are alternatives to the core stuff
(especially when the inquirer has actually indicated his familiarity
with those alternatives) and that reminding everyone for the nth time
about how bad the DOM is (for whatever tangential purpose only
partially related to the topic under discussion) adds very little if
anything in the way of advice? It's like someone saying that they're
going to fly the Atlantic in a 747 only to be told that they should
drive a Lexus because "Boeing make terrible cars".

Feel free to replace "DOM" in the above with whatever else fits,
because this kind of thing comes up all the time.

Paul



More information about the Python-list mailing list