Re: Requirements for research (Was: Dropping <input usemap="">)

On Aug 16, 2007, at 9:47 AM, Philip Taylor wrote:

>
> Robert Burns wrote:
>> On Aug 16, 2007, at 8:58 AM, James Graham wrote:
>>> Robert Burns wrote:
>>>>
>>>> Which results are you referring to here. Ian looked for misused  
>>>> <input usemap> elements. Philip looked for <blink> and <element- 
>>>> I just-mad-up> on another data set.  The couldn't reproduce the  
>>>> same results because they were looking for different things.
>>>
>>> I meant, more generally, that the two different surveys they did  
>>> have produced similar results for questions like "what fraction  
>>> of pages use the <foo> element?" rather than questions  
>>> specifically related to <input usemap>. IIRC Phillip didn't find  
>>> any use of <input usemap> (which you would predict from the  the  
>>> small usage found by Hixie and the relative number of pages in  
>>> the two surveys).
>> Niether Hixe nor Philip looked for use of <input usemap> . Hixie  
>> looked for misuse (or esoteric use) of <input usemap>. Philip  
>> looked for other stuff (and unsurprisingly didn't find any <input  
>> usemap>).
>
> http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/ 
> index - I looked for all start tags and for all attributes on start  
> tags.
>
> http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/attr/ 
> usemap is the data for <... usemap> - out of 7739 pages, <img  
> usemap> was on 1006, <td usemap> was on 1, and usemap was not found  
> on any other tag. (Given the sample size, that just indicates that  
> any other usemaps are on somewhere below 0.2% of the pages from  
> dmoz.org. I could probably look at a hundred times as many pages  
> without too much trouble, but I'd be happier if there was a better  
> source of 'normal' pages than dmoz.org)

Understood. that's probably the hardest part of this kind of research  
is finding a good listing of the population.

In any event, it sounds like we wouldn't be breaking much content if  
we specced <input usemap> in a slightly different way than the  
current draft (or at least the draft before the changes were made  
that did not reflect the views of the WG).

Take care,
Rob

Received on Thursday, 16 August 2007 14:55:43 UTC