W3C home > Mailing lists > Public > www-style@w3.org > February 2004

[css3-hyperlinks] Comments about the New Draft

From: Lachlan Hunt <lachlan.hunt@iinet.net.au>
Date: Fri, 27 Feb 2004 00:33:32 +1100
Message-ID: <403DF5AC.4060000@iinet.net.au>
To: www-style@w3.org

   Here are some of my comments on the first css3-hyperlinks draft.  I 
think it's quite good, although there are a few issues about some of the 
properties, and some features missing.

The ‘target-name’ property
   Except for ‘current’, the other properties may cause unexpected 
actions such as opening a new window/tab and replacing a frameset. 
Therefore, I think it should be stated that the UA SHOULD/MUST make the 
  action known to the user, through a UA specific means, prior to 

The ‘target-new’ property
* window
   Personally, I hate it when websites cause links to open new windows,
   currently either by the @target or some javascript, or whatever.
   If as I mentioned previously, this was known to the user, the user
   could then take action to prevent the new window, open a new tab
   instead, etc.
* tab
   Same response as window, except the options are reversed.
   (ie. prevent new tab, new window instead, etc.)
   If tabs are not supported, the UA should treat this as either window,
   or as if the target-name property was ‘root’
* none
   What is this for?
   “No new destination is created. The target is not displayed.”
   That essentially means that when the user activates the link, nothing
   will happen.  This is a big usability problem, because when any user
   clicks a link, they expect the link to work! otherwise it's
   essentially a dead link, which is quite frustrating.

Missing feature
   A feature that

   The link-behaviour property, which was described in an old css3-box 
draft [1] had the value ‘confirm’.  This looked quite useful, because it 
allows the author to give more information about the resource and ensure 
that the user actually wants to retrieve it.  The user is then able to 
make a more informed decision before continuing.  eg. file download may 
be too large, content may be considered innappropriate to some, etc. 
Although, I wasn't so happy with it's implementation.  Confirm would 
have been better implemented as a property:

Name:	confirm
Value:	[<string> | attr(<identifier>)]+ | none | inherit (maybe?)
Initial:	inherit or none?
Applies to:	link elements
Inherited:	yes?
Percentages:	N/A
Media:	all
Computed value  specified value (except for 'inherit')

[1] http://www.w3.org/TR/2001/WD-css3-box-20010726/#L4112

Lachlan Hunt
Received on Thursday, 26 February 2004 08:33:41 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:27:11 UTC