W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > April to June 2001

Re: Frames & Font Sizes (was Re: 10.4 Re: Checkpoints 10.4 and 10.5)

From: David Woolley <david@djwhome.demon.co.uk>
Date: Tue, 5 Jun 2001 07:55:19 +0100 (BST)
Message-Id: <200106050655.f556tK301447@djwhome.demon.co.uk>
To: w3c-wai-ig@w3.org
> If we wanted to break deep links ourselves in order to keep them from being 
> broken in the future when we changed content, then this seems valid.

That's typical commercial design practice, but it goes against the
philosphy of the web.  The right thing to do is to design the partitioning
of the site and the resource names so that, whilst pages may be taken out
of the main navigation structure, they are still available.  (I think
this is one of Jakob Nielsen's gripes; Microsoft are particular bad
at changing their site to break any attempt to deep link it.)

The unintended message from completely removing old material is that
you are not interested in customers once they have made the purchase,
or that you are afraid to be associated with it.  Frustrating deep
linking loses you the benefit of free advertising from other sites.
A 404 response, or even a redirect to the home page/ a search page are
likely to result in the back button being pressed, rather than going to
the trouble of working out where the reference really pointed.

> Isn't this the responsibility of Web TV? How long have frames been in the 
> HTML specs? Should one poorly designed UA keep everyone else from using a 

They have been in for about four years, always with an indication that
they are a stop gap measure to be eliminated from sites as soon as CSS 
implementation permits.  Most people never read the specs, though.
I believe W3C has never liked, them, even though it has had to codify them.

> specific technology?

It's actually poorly designed mainstream browsers that are causing
frames to last longer than they should.  Web TV has to work within
constraints that don't apply to ordinary browsers.

> Again, isn't this the responsibility of Lynx developers?

Lynx has had to cope with frames because people don't use the fallback
mechanisms provided (with some exceptions, the W3C HTML specifications
have been designed so that well written pages will degrade gracefully).
Implementing something that looks vaguely like a GUI frames implementation
on an 80x24 character cell display just uses too much screen real estate.

> Yet most of these "first line" web sites never had frames to begin with. 
> Search engines have no use for frames. Microsoft does use frames when 
> necessary (ie. MSDN documentation libraries).

I'm pretty sure that Microsoft once used them on their main page.  Netscape
almost certainly did, as they invented them.  MSDN is unlikely to be
designed by their primary web designers and, in any case, is really an
application for GUI browsers rather than a true web resource.

> We're a bit leery about using GIFs on a commercial website due to the 

I think there is an element of paranoia here.  It's getting too far
off topic, but there are legal ways of using GIF and other alternatives
that are better than JPEG.
Received on Tuesday, 5 June 2001 02:55:29 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:13:55 GMT