Re: XTHML 1.0 Strict validation of noscript

>> I tried simply changing the doctype of these pages to 1.1 and the  
>> validator doesn't complain about the <noscript> being there (and  
>> everything else validates, and the page looks fine).
>
> Well, I _still_ think that the ultimate problem remains - you are using  
> document.write, since if you weren't, the problem probably wouldn't  
> arise.

I and another 500 million people. In fact, document.write is probably the  
main reason why JavaScript is used at all. :)

Ultimately, what matters to me is that the code in these pages will render  
correctly in all the main browsers, and will be parsed by the validator  
(so that I can spot errors easily during development - sometimes a single  
error can "mask" others, that's why I want it to validate). I'm not doing  
it to gain karma points with the god of standards (it's not like I agree  
100% with the current XHTML or CSS standards anyway).

So, if the code validates and if it works in all the main browsers, I'm  
pretty confident that the site will be able to do its job during its  
lifetime. It's extremely unlikely, as I'm sure you'll agree, that any  
major web browser will stop supporting document.write (or even innerHTML,  
which is much more of a "hack" - and I almost never use - and yet works  
fine in all the major browsers) during the next decade.

Some kinds of things that are possible (and even easy) to do  
"programatically" are not easy to regulate or verify at a "static" level.  
That doesn't mean we should limit design or "presentational" aspects just  
so we can comply with the static standards. For essential information,  
I'll generally care more about following the standards exactly (and will  
never rely on the availability of JavaScript, for example, because many  
people disable it). But for stuff like this (personalising the page,  
printing non-essential information, adding interactivity, etc.), the only  
thing that really matters is that it works, and works identically in all  
browsers. Standards are a tool, not a goal.

RMN
~~~

Received on Monday, 16 October 2006 15:47:12 UTC