RE: Anchor Names -- a P4 item?

This can get quite complex. It is valuable to be able to pass a URI on a
napkin (although more valuable to pass it via email, since then you don't
have to try and click the napkin, or read through the food on it). And for
local maintenance it is often helpful to have something that vaguely makes
sense.

On the other hand, one of the biggest reasnos why finding things on the web
today is so hard is becuase people change the names of things to reflect
their new idea about how they are organised. This breaks everybody else's
links to those things, from search engines to advertising, and people have to
try and find where the new version is. Or run the risk that what they pointed
to yesterday is a different thing today.

Giving things names that are very meaningful means that you can deal with
them easily, as can anyone who understands how you named things. But it also
runs the risk of running out of names. There are societies where people are
used to things changing names regularly, but they tend to be small, and the
changes tend to be according to well-defined and understood rules. The web is
not such a society.

My personal view is that it is more important to make sure the names stay the
same, so links don't all break, than that they are memorable. But there is no
intrinsic reason for them not to be memorable just becuase. (It is a bad idea
to try and read too much into a name, or to expect someone to have invented a
scheme they can maintain for ever.)

cheers

Charles McCN


On Thu, 1 Jun 2000, Al Gilman wrote:

  At 05:39 PM 2000-06-01 -0700, Kynn Bartlett wrote:
  >At 6:15 PM -0400 6/1/00, Bruce Bailey wrote:
  >>Or is this a general design issue, and therefore not in the domain of the
  >>WCAG?  "Use sensible file names for your HTML documents" is not in the WCAG
  >>either.  Can anyone point me to a reference (with face validity) that
  >>includes such basics?
  >
  >Using sensible file names is not a requirement nor should it be.
  >The URI scheme is not meant to convey information or data -- it is
  >meant to be an almost-entirely arbitrary system (once past the
  >protocol and hostname identifiers) that associate arbitrary content
  >with a unique identifier.  There is no requirement that the
  >unique identifier be human readable, human readable, human
  >parsable, or convey any information at all apart from uniqueness.
  >
  
  If these objects (the files) were always accessed by clicking on links, we
  could say that safely.  But that is not true.  Mnemonic file names are
  valuable because the resource is accessed through the local file system as
  well as through URIs.
  
  And a basic requirement on URIs is that they be able to pass via a cocktail
  napkin to the "go to location" manual input of the broser User Interface.
  Read the RFC.
  
  Al
  
  >-- 
  >--
  >Kynn Bartlett <kynn@idyllmtn.com>
  >http://www.kynn.com/
  > 
  

--
Charles McCathieNevile    mailto:charles@w3.org    phone: +61 (0) 409 134 136
W3C Web Accessibility Initiative                      http://www.w3.org/WAI
Location: I-cubed, 110 Victoria Street, Carlton VIC 3053
Postal: GPO Box 2476V, Melbourne 3001,  Australia

Received on Thursday, 1 June 2000 22:34:26 UTC