[uf-discuss] FYI: Google web authoring statistics

[Ryan King]

On Jan 25, 2006, at 7:15 PM, Mark Rickerby wrote:

> Sorry if some of you have already seen this, if not, I think it's
> quite interesting...
>
> Google's massive survey of text/html documents on the web:
> http://code.google.com/webstats/index.html

Blogged, as well: http://microformats.org/blog/2006/01/25/google- 
releases-web-authoring-statistics/

> Of specific relevance to this list:
>
> http://code.google.com/webstats/2005-12/classes.html
>
> class="title" is the third most commonly used class, and
> class="content" the 6th most, which could be significant in terms of
> some of the recent hatom/hcard/hreview disambiguation discussions.

Certainly.

> http://code.google.com/webstats/2005-12/element-a.html
>
> It looks like only a tiny % of hyperlinks use the rel attribute, but
> of those that do, that data is also interesting.
>
> rel="nofollow" is *far* more common than anything else, but
> rel="license" is surprisingly widespread too. rel="tag" is catching up
> to rel="bookmark", though their commentary on this is a little
> disparaging.

I assume you're refering to: "Web 2.0 collaborative remixability  
social tagging" ? Yeah, a bit snarky, but I've seen worse. :-D

> Though I do find their phrase "rampant bad UI" quite amusing, and
> unfortunately all too true.
>
> see also...
> http://code.google.com/webstats/2005-12/linkrels.html

-ryan
--
Ryan King
ryan at technorati.com