W3C home > Mailing lists > Public > www-style@w3.org > July 2005

Re: New layout language.

From: Orion Adrian <orion.adrian@gmail.com>
Date: Thu, 7 Jul 2005 17:36:02 -0400
Message-ID: <abd6c801050707143674378a4f@mail.gmail.com>
To: www-style@w3.org

On 7/7/05, Kris@meridian-ds.com <Kris@meridian-ds.com> wrote:
> 
> Orion Wrote:
> 
> >There is a difference between criticism for refinement and criticism
> >for the sake of criticism. To discover whether something is useful or
> >not requires exploring it, not dismissing it. Criticism should happen
> >after understanding, not beforehand. Criticising before you understand
> >something is pointless is it not?
> 
> Come on man.  First off, you're human like the rest of us and are probably
> a nice person IRL.  I certainly didn't critisize you for the sake of it.  I
> did it because the system you claim is better isn't. (we'll get into that)
> And your statement of when Criticism should take place could CERTAINLY
> apply to yourself concerning CSS.  Your entire proposal was based on a
> criticism of CSS which was false.  I can't help but feeling you still don't
> fully understand what CSS is capable of, and that is a bit disconcerting
> considering your stance on Criticism.

For which I criticised myself. And yes I'm a pretty nice person in
real life. My pointedness against the behavoir I'm talking about
wasn't against you. I've just seen people treated rudely accross
multiple groups.

> As far as your system goes, let's be honest, there's nothing differen about
> it from absolute positioning.  Sure you have a couple of concepts that you
> like about your system, but ultimately, we're stating the same things in
> the same ways minus your shorthanding. In fact, I've gone to the trouble to
> prove it to you.  You'll find a style sheet at the end of this e-mail
> stated in css syntax.  That style sheet is the same layout as before, but
> I've fixed the issues inherent in the last one I sent you.  It was a simple
> matter really, I just needed to think about it some since I seldom use
> absolute positions.

The one difference between absolute positioning and what I'm proposing
is that it doesn't interact with the rest of the system. It's cordoned
off. There's also the benefit (though argued) that it provides one
solution for one problem (i.e. it's deterministic).

> As far as seperation of Layout and Formatting goes, look man, whatever,
> I'll let the other folks argue it out with you.  The CSS is as follows:
> 
> body {
>       margin:0px; /* since there's margin automatically on the body */
> }
> 
> .canvas { /* this was added for 2 reasons.  1 you needed it for your model,
> and 2 I needed it to stop the degradation of my model.  With this we no
> longer get the overlap of different regions on each other.  Apparently I
> don't need calc(), I just need some fore-thought*/
>       position:absolute;
>       width:100%;
>       height:100%;
>       min-width:620px;
>       min-height:420px;
> }
> 
> .left {
>       position:absolute;
>       top:10px;
>       left:10px;
>       width:200px;
>       bottom:10px;
>       border:1px solid #000000;
> }
> 
> .right {
>       position:absolute;
>       top:10px;
>       right:10px;
>       width:200px;
>       bottom:10px;
>       border:1px solid #000000;
> }
> 
> .midleft {
>       position:absolute;
>       top:10px;
>       bottom:20%;
>       left:222px;
>       right:50%;
>       margin-right:5px;
>       border:1px solid #000000;
> }
> 
> .midright {
>       position:absolute;
>       top:10px;
>       bottom:20%;
>       right:222px;
>       left:50%;
>       margin-left:5px;
>       border:1px solid #000000;
> }
> 
> .bottom {
>       position:absolute;
>       top:80%;
>       bottom:10px;
>       right:222px;
>       left:222px;
>       margin-top:10px;
>       border:1px solid #000000;
> }
> 
> Minimums were taken out of all elements since the canvas took care of that
> problem.  Since absolute elements are absolute to the object they're in...
> I just defined minimus for that object instead of each individual region.
> Made more sense to me ulimately.  I'd also like to point out that I got to
> looking at your code for your system and I think you've got a couple flaws.
> Specifically:
> 
> bottom {
>       top: 340px + 80%;
>       left: 630px;
>       height: 70px + 20%
>       widht: 410px + 100%;
> }

This value was incorrect, but was correct in the C# demo code.

> Secondly, I'd like to point out that since you've made no
> right or right margin, this box, while knowing where to begin, does not
> know where to end... thus it should overlay the right most box's lower
> quadrant.  Assuming that I understand your system.  (which I think I do at
> this point) these are flaws I see.  Again, I apologize up front if I'm
> incorrect in my observations.

Where it ends is (left + width) where left and width are the
appropriate equations.

There is a consistent model for converting one into another and for
the time being I'm going to use it for pages directed at Firefox.

If you're curious it's

Canvas - (Left + Width) = CSS right

So if you had any three you should be able to get the last value.

Now what calc does get around is the need for margin. If you could get
rid of them you'ld have safer layouts. Though it's neat that you found
a non-degrading layout.

I appreciate the time you've taken to talk to me about this. Most
people haven't. I appreciate the time all people have taken to
understand what's been presented.

What I don't understand is the use of classes in the examples you're
giving. Wouldn't ID's be more appropriate since each div is unique in
a page? A class should be able to be used many times, should it not?

-- 

Orion Adrian
Received on Thursday, 7 July 2005 21:36:07 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:39 GMT