xml-dev - Re: Scripting and XML

[simeons@allaire.com (Simeon Simeonov)]

>Simeon Simeonov wrote:
>> This is a good direction. However, I see a potential inconvenience for
>> scripts that directly modify the document. Embedded scripts implicitly
>> identify the part of the document they operate on with their position.
>> External scripts will have to explicitly specify the part they operate
on.
>
>More likely, I expect the opposite. The parts of the scripts that need
>to be "operated upon" will identify themselves just as they do for
>presentation.

I absolutely agree with this point: markup will have to be added to aid
external scripts. But that could be inconvenient. Yes, it is more powerful.
Yes, it provides more information for processing and, therefore, can enable
tasks previously impossible. But it does require more effort in creating the
markup. My point (which I did not make clear in my short post) was that for
some simple tasks, especially for human-writable languages, this would be
one extra thing that humans would rather not do.

> Let's say you have a form field that validates sin numbers
> (in Canada, we report sins to the government at least once a year). You
> shouldn't have to say this:
> <INPUT TYPE="TEXT" OnChange="blah;blah;blah;blah;blah">
> nor even this:
> <INPUT TYPE="TEXT" OnChange="CallBlah()">
> But rather this:
> <INPUT TYPE="TEXT" CLASS="SIN_NUMBER_VALIDATOR">
> or even this:
> <SIN_NUMBER_VALIDATOR>
> To me this seems blindingly analogous to the move from this: <I> to
> this: <EM STYLE="Italics"> to this: <EM CLASS="WARNING"> to this:
> <WARNING>.

Again, in principle I agree. However, I don't think that this approach is
scalable for some applications. I can easily think of more than a 100
classes of textual input that I would like to validate. I would find it
confusing to have to work with more than a 100 tags just for input
validation. For not very common input types, I would prefer your third
example to the fourth one. Essentially, I am arguing against tag number
explosion.

That brings me to a question that I'd really like to know the answer to:
what are the heuristics of good XML application design? How do you think
they vary between applications aimed to be human-writable vs.
machine-generated? How about between human vs. machine readable apps?

Regards,

Simeon Simeonov





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