[whatwg] <summary> tag to help avoid redundancy of meta description tag!?

Ashley Sheridan ash at ashleysheridan.co.uk
Fri Mar 19 07:17:39 PDT 2010

On Fri, 2010-03-19 at 13:43 +0100, Roger Hågensen wrote:

> On 2010-03-18 10:04, Ashley Sheridan wrote: 
> > The main problem with that would be that parsers would then need to
> > read into the <body> of the page to produce a description of your
> > site. This might not produce much of an overhead on a one-off basis,
> > but imagine a parser that is grabbing the description from hundreds
> > or thousands of pages, then this could become a bit of a problem.
> I do not see how that is any more or less of an problem than today
> with pages that have meta description missing,
> what do those parsers do then? Do they stop at </head> ? What do they
> use as description instead? The first paragraph?
> The parsers used by all major search engines certainly do not halt,
> they break down the entire page right?
> As for delays, that is not an issue for consumers, I can not recall
> any browser ever showing me the meta description unless I explicitly
> view page properties.
> I can imagine that the seeing impaired community would love something
> like this, as it would basically tell screenreaders that "this" is the
> first paragraph/summary/description/teaser of the page,
> allowing blind people to more rapidly jump from page to page.
> Currently the meta description is not always good content, would be
> interesting to see a Google analysis of how the meta description is
> used,
> i.e. how many are basically repeating page content (like I do) and how
> many just dump keywords in there, how many pages on a site have a site
> wide identical description? And so on.
> Roger.
> -- 
> Roger "Rescator" Hågensen.
> Freelancer - http://EmSai.net/

Search engines and people are not the only content parsers. Sure, you
would expect a parser to maybe look further into the content if the
description meta tag was missing, but imagine if a parser had to do this
for all the content it looked at? There are still overheads to consider.

Why not just use server-side code to output the first paragraph of
content as the description for the page also?

I just feel that the <head> and <body> areas of a page have two distinct
uses, and unnecessary crossovers shouldn't occur if it's avoidable.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.whatwg.org/pipermail/whatwg-whatwg.org/attachments/20100319/bc7cb66d/attachment-0002.htm>

More information about the whatwg mailing list