Re: Attaching functions to events, from external JS files

On Tue, 02 Aug 2005 13:58:35 +0200, knocte
<knocte@xxxxxxxxxxxxxxxxxxxxxxxx> wrote:

>It seems that you don't know how to append Document nodes to
>pre-existing HTML content using JavaScript DOM Level 1. Functions for
>adding this type of contents doesn't require to write hardcoded HTML
>code on Javascript scripts,

It doesn't make any difference if you have

x.innerHTML="<h1>chicken</h1>";
or
x.appendChild(x.createElement('h1').appendChild(document.createTextNode('chicken'))));

Both are still HTML content in your script, just because one is in a
verbose, but predictable format, doesn't change the fact.

> it's one inconsistency
>that is solved by the "Loading..." method,

You've yet to discuss how a loading method can work unobtrusively.

>or by a future new attribute
>like the one I proposed (onnodeload, or onchildload).

These events already exist, but I'm not trying to change the future
situations, I'm saying your advocacy of onload is flawed, and your
overly strict interpretation of seperation does nothing but harm the
user, for little overall benefit to you.

>> IE6, Opera 8.01, Mozilla, Safari? You know of course that CSS is
>
>What?? You're encouraging me to leave this discussion if you are saying
>that Mozilla doesn't support CSS.

Certainly, all of them can _disable_ CSS, and users do do it, just
like they disable script.

>> optional, and all of them allow you to disable Author CSS (with
>> varying amounts of difficulty, you may need to create a complicated
>> user style*** in some UA's)
>
>Well, what is the amount of people that customize CSS?

If you're falling back on "no-one does that so I don't need to care
about it", then your entire arguments are flawed, unobtrusive
javascript is only worthwhile for the minority user, the majority have
script enabled...

Jim.
.