W3C home > Mailing lists > Public > www-style@w3.org > January 2008

Re: flowing around both sides of a float

From: James Elmore <James.Elmore@cox.net>
Date: Wed, 2 Jan 2008 10:54:26 -0800
Message-Id: <C1922E54-9178-4544-9421-5E9C8D0C7FBF@cox.net>
Cc: Brad Kemper <brkemper@comcast.net>, CSS <www-style@w3.org>
To: Philip TAYLOR <Philip-and-LeKhanh@Royal-Tunbridge-Wells.Org>

On Jan 2, 2008, at 10:10 AM, Philip TAYLOR wrote:
> Brad Kemper wrote:
>> I don't want the user agent to design my page for me, or to second  
>> guess my intentions. I will not use anything that removes the  
>> control I currently have over the design.
> What "control" do you believe you have ?
> HTML & CSS allow you to /influence/ the
> design, but if you want to /control/ it
> you would do better to adopt a page-
> description language such as PDF.
>>> When writing a web page, I usually don't actually care whether it's
>>> floated left or right, and in general I don't know whether it  
>>> will look
>>> best floated left or right without knowing the font size and window
>>> width etc.
>> Maybe you don't, but I do.
> Were it not for my New Year's resolution to be kinder
> to all men, I would be sorely tempted to suggest
> <arrogance> tags around the last sentence.
> Philip TAYLOR

Different people want different things from HTML and CSS. Some want  
the pages to look beautiful, according to their vision of beauty.  
This may not mean the same thing to Brad as it does to Philip. That  
is why HTML and CSS provide for different levels of control. People  
who "don't actually care" can use straight HTML. People who do care,  
can use CSS with HTML. People who have to be in complete control can  
use a page description language, or even capture the 'perfect' with a  
gif or other image file.

Why are we arguing about the 'ideal' page design? I want to improve  
CSS so I can make my web pages more closely match my ideals, and  
recognize that any tool imposes some limitations on me. I just want  
to suggest things for CSS which, for me, would make my designs better  
(simpler, clearer, more accessible, whatever better means to me). If  
most of the CSS group thinks that my suggestions are not suitable for  
CSS, that is fine too. The questions I want to ask myself about any  
proposal are:
	Does it fill a need, providing the tool (CSS) with better control  
and more power?

	Does it make the tool easier to use, especially in the area of  
understanding how to use it? (This is one reason many people complain  
about HTML+CSS -- things happen which they don't understand and have  
a hard time controlling to make their pages look the way they want.  
If my suggestions make the tool easier to understand, to document, to  
use, that is the main reason I am suggesting them. If they make it  
harder, I gladly will let them be voted down or even ignored.)

	Is it possible to implement? (I program computers, I know some  
things are impossible.)

	Would any implementation make the layout take significantly longer  
to display? (People hate to wait.)

	Would the current *AND FUTURE* constraints of viewing hardware make  
this control moot? (If I really want pixel-level control -- I don't  
-- would some guy looking at my web page in 640x480 or HiDef see some  
very stupid things because of that control. Also, what will happen  
next year when iPod nanos have 1020x800 (very small) pixel screens,  
or when standard displays become 36 inches or more with a minimum  
screen size of 10mega pixels?)

Because of the way my mind works (my wife says I often think in a  
tangent to the rest of the world), I see some things which are  
missing from CSS (like this ability to separate positioning from text  
wrap), or some things which CSS can do, but which require hacks (such  
as using tables for layout). Please don't instantly jump to the  
conclusion that a suggestion (from me or from anyone) is seeking  
'perfect' control. Maybe we see some things differently than you do,  
and would like to combine our visions to make CSS a better tool.

James Elmore
Received on Wednesday, 2 January 2008 18:54:54 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:27:32 UTC