W3C home > Mailing lists > Public > www-talk@w3.org > March to April 1997

Re: We need a META schema registry!

From: Benjamin Franz <snowhare@netimages.com>
Date: Tue, 18 Mar 1997 05:44:27 -0800 (PST)
To: meta2@mrrl.lut.ac.uk, www-talk@w3.org, se-dev@franz.ww.tu-berlin.de
Message-ID: <Pine.LNX.3.95.970318052136.13817A-100000@ns.viet.net>
On Tue, 18 Mar 1997, Andreas Koenig wrote:

> Even with the best and most diligent registration authority, the META
> tag is a two edged sword. It transports stuff nobody is asking for
> explicitly. It steals the net bandwidth. Please read
> http://www.kulturbox.de/aid/pics-se/ for a proposal how the net can be
> freed from this exacerbating bloat today.

I've said it before and will undoubtedly say it again: NO scheme for
meta-infomration significantly more complex that the existing de facto
standard of 'Description' and 'keywords' META tags is going to be
successful with the market as it is right now. Users don't *understand*
them. Anything the *average* user cannot understand will simply be ignored
by them. Even the *original* META scheme that produced 'description' and
'keywords' included two more meta fields regarding data type
(resource-type) and distribution - both of which have vanished from
general usage due to incomprehension of application or need by users. 

PICS derived meta-data? Yah, right. The *only* reason even the original
PICS has any market is due to fear of legal and public relations
repercussions. It is a *programmers* approach to meta data - and damn few
people are programmers.

And you are wrong that no one is asking for META info - at least one major
search engine *displays* meta 'description' for search results. A very
good thing in my opinion since it eliminates the 'Here is my best guess on
a representative sample of text from the page' that some other places
produce - at least one search engine will actually return raw JavaScript
as a page's description if it is present (doh!). 

While PICS-SE has some admirable goals - it is destined to founder on the
rocks of user incomprehension and indifference. 99%+ of users will neither
care nor even be *ABLE* to implement it correctly. 

The *ONLY* way I can see out of this will be if at least one of the top 5
or so search engines made a public announcement that they will search rank
sites using a high level meta-information standard *higher* with that
infomration than sites not using them. This would align the *market*
behind it because they *do* understand 'ranked high is GOOD' and will be
*motivated* to try then. You would then see pressure on the HTML editor
makers to include whatever standard the search engines were using - which
would bypass the general user incomprehension problem. 

-- 
Benjamin Franz
Received on Tuesday, 18 March 1997 08:44:38 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 27 October 2010 18:14:22 GMT