W3C home > Mailing lists > Public > www-style@w3.org > July 2001

Re: activel inks?

From: AMollah <am@freephone.fsnet.co.uk>
Date: Tue, 3 Jul 2001 13:14:57 -0400 (EDT)
Message-ID: <009001c103e4$6decd930$0401a8c0@localnet>
To: "Manos Batsis" <m.batsis@bsnet.gr>
Cc: <www-style@w3.org>
Yes, I think I was being a bit optimistic to believe that people will go
straight to the chosen link without going over other links (especially in
vertical navbars), but still for the majority of the time the "current" link
would be displayed properly, and the current link would stick out.

It is true that for my intended purpose it would be better to have no :hover
for the :current property (to make the link less like a link), but on the
whole the :hover property just indicates that the mouse is over something,
and it masks the nature of what it is over. So it is just as corrupting when
:hover goes over a :visited link. How are you to tell, when the mouse is
over a link (and :hover is being used) whether the link is normal or
visited? I think you are attempting to overcome a problem with :hover in
order to insert :current, which is not necessary.

As for :sub-domain, :sup-domain, (perhaps :external, :sub-external to
emphasise external as opposed to internal) that would be a great idea to
allow people to choose between the two (they needn't use both). I think
again here you are looking too hard for problems. I mean, anyone could set
:hover to have the same attributes as :link or :visited, why should this be
any more or less of a problem with :*domain? (There are enough colours to go
round -if I had my way everyone would use the same set of (colour at least)
attributes for the various link states, it would be alot easier for
novices - but I suppose would spoil the colour scheme of just about every
site out there). Of course it would be nice to have an extra pseudoclass for
":*domain-visited" and needless to say ":*domain-current" being a logical
inconsistency would not be required.

Maybe for :domain or :external property you could have instead of or as well
as a different property for subdomains, a number that specified how far from
the current url is to be considered external:

a:external{
dist: 3
colour:green}

So that url's that are more than 3 directory levels up or down are
considered external (or could have - and + values). This would help with
those ISPs that use subdirectories for different sites.

I hear what you are saying about time and deadlines - I only intended to put
a suggestion forward at the start, and we seem to be getting deeper and
deeper into this (and I'm not even on workgroup). This has probably been
discussed in depth already somewhere if I look for it. If you don't want to
answer any more that's OK, and thanks for the time so far.

Best Wishes,

Dave




----- Original Message -----
From: "Manos Batsis" <m.batsis@bsnet.gr>
To: "AMollah" <am@freephone.fsnet.co.uk>
Cc: <www-style@w3.org>
Sent: Tuesday, July 03, 2001 2:19 PM
Subject: RE: activel inks?




> -----Original Message-----
> From: AMollah [mailto:am@freephone.fsnet.co.uk]

> but I see that the same problem would apply in that when the
> mouse is moved
> over the "current" link the colour would change and the
> underline would reappear.

Exactly :-)


>  I don't know what the answer is but this issue
> also affects the
> :visited pseudoclass too though I think, so your points 1 to
> 3 should apply
> to that also, using the same logic (maybe even :link if you
> think about it -

Um... I don't really think so. A visited link is still a destination
outside the current document so :hover effects should apply to indicate
that.


>  My intention with using the
> colours above is
> that people aren't attracted to click the "current" link
> because they know
> this is the page they are on already, so they wouldn't hover
> over it anyway.

Sorry, I have to disagree. Usually, links are very close together,
forming 'navigation panels/areas' in web page design. In short, I
usually hover above 3 or 4 links before clicking the one I want. So, it
is easy for non technical people to get confused.



> You are absolutely correct about blurry nature of the
> discussion of the
> :external links issue, I think that is why I think this would
> be better as a
> html element.


href is just an href. If you are thinking of adding linktype='external'
or something in the <a> tag, it is still better to use css classes to
style the links indicating they lead outside the current domain.



> html. Any urls
> outside subdomains might be OK, so in your example the second
> url would be
> external. This would at least account for all the different
> .tripod (and
> similar free isp) sites, but might cause some large corporate
> sites some
> trouble (although alot have distinctly different sections in
> subdomains) and
> there would still be the problem with ISPs that use just
> directories for
> different sites.


As you can see, it is difficult to make something that will suit
everyone. I still think that letting developers make a solution of their
own, may it be css classes, or scripting, or behaviors, or whatever is
ok. That way spec editors will have more answers ready when they'll try
to make something to include in the css spec by gathering input.
OR we can make something like

:sub-domain

:super-domain

And the problems begin. If you think that the presentational overlap
between :hover and :current is a problem, wait till you see the above in
action. Example of problematic style selector authoring:

a{
font-weight:normal
color:blue;
}

a:current{
font-weight:bold;
color:black;
}

a:hover{
color:red;
font-weight : bold;
}

a:visited{
color : #c0c0c0;
}

a:sup-domain{
color : red;
font-weight:bold;
}

a:sub-domain{
color : blue;
font-weight:bold;
}


Ok. We try solving the problem with :hover and :current adding something
like the following attribute:

a:current{
hover-effect:false;
/* OR */
hover-effect:true;
}

Something more difficult now. How about :external that are also :visited
?

Um... I got no time for this right now, it may take a while and that
deadline is knocking my door but you get my point.
What do you think?

Kindest regards,

Manos
Received on Wednesday, 4 July 2001 11:01:43 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:10 GMT