Re: predefined class names (was Getting beyond the ping pong match)

At 15:54 +1000 UTC, on 2007-05-06, Lachlan Hunt wrote:

[supposedly Jukka wrote:]
>>>> (For example, in my page about intellectual rights, I may well have
>>>> marked parts _discussing_ copyright issues with such an attribute,

[...]

> you presented a completely hypothetical example without any evidence
> to show that it actually happens in reality.
>
> It would really help if the people who are objecting to the predefined
> class names could find real world evidence to support their claims of
> clashes, and explain precisely what real practical problems can occur
> from it.

I looked at the first 20 results of
<http://www.google.com/search?q=%22class=copyright%22>. Three of them
contained actual results: <http://www.csmrobotics.com/copyright.html> and
<http://cars.uk.msn.com/Photo_Gallery/photogallery.aspx?cp-documentid=477037>
seem to use it more or less as Web Apps 1.0 currently defines it. But
<http://www.munchfonts.com/HTML/additional.html> appears to be exactly what
Jukka suggested might be out there.

However, given that the Web allegedly consists of billions of dopcuments,
what do a few such cases mean? Surely this isn't thorough enough material to
base a spec on, no matter which side of this particular sub-debate you're on.

If we are to back up arguments with representative data, which seems a
sensible idea to me, what source/method do we have available? It's not
particular easy (unless I'm overlooking something) to search within the
source of HTML documents using Google. If there is no such source available,
can we require people to back up their arguments with this sort of data?

(Btw, this is not to say I'm for or against predefined class names yet. Still
pondering :))


-- 
Sander Tekelenburg
The Web Repair Initiative: <http://webrepair.org/>

Received on Sunday, 6 May 2007 07:52:16 UTC