[whatwg] HTML4's profile="" attribute's absence in HTML5

[Ian Hickson]

On Wed, 7 May 2008, Mike Schinkel wrote:
>
> Your logic is filled with fallacy.  It pressumes that because people are 
> not doing something that would otherwise cause a problem because that 
> problem currently doesn't exist.
> 
> That would be like you saying "But people in this region don't need 
> bridges to cross the river because no body travels to the other side" 
> when there are currently no bridges close enough make it realistic. 
> Build a bridge locally and things will change drastically.
> 
> By the same point, people avoid this problem because there is no safe 
> way to address it. If there were a safe way, usage will blossom.
> 
> In summary, you are using far too much confirmation bias in your 
> decisions.

Generally speaking, on the Web, if there is a need people will work around 
the problem in obvious ways. To continue your analogy, if people wanted to 
get to the other side, they would use boats, they would swim, they would 
climb on the back of aligators, they would dig tunnels, etc.

In this particular area -- disambiguation of class names and other 
metadata fields -- we see very little in the way of people avoiding name 
clashes. Indeed, the rare times where a clash becomes a problem, the names 
are simply changed to avoid the problem.

It's possible that this will become more of a problem in future, but then 
that's when we should fix it -- when we know what the needs are. At the 
moment, any fix here would be based on theoretical concerns.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'