Re: [css3-text-layout] New editor's draft - margin-before/after/start/end etc.

On Jun 7, 2010, at 1:48 AM, MURATA Makoto (FAMILY Given) wrote:

>> 
>> Fallback can be handled by other proposals, too. E.g. [1]
>> 
>> [1] http://lists.w3.org/Archives/Public/www-style/2010Jun/0190.html
> 
> It is not impossible.  However, such pseudo selectors or media queries 
> basically require authors to create two sets of margin/padding/border properties: 
> one for vertical writing and the other for horizontal writing.  Authoring of such two 
> sets is certainly painful.  Murakami-san thinks it is prohibitively painful.  At least,
> one could argue that such design is not fair.

I would argue that for complex designs it will still take a lot of care and attention to get it right, and it is easier to do that when then two designs (one horizontal, the other vertical) can be isolated into two blocks or style sheets.

> At least two implementations already decided to abuse existing physical properties 
> as logical properties.  They are: http://www.lovelyreader.com/ and 
> http://blog.print.cssj.jp/?itemid=197 (I learned them from Murakami-san.)
> Apparently, their implementors believe that the omission of logical properties 
> in CSS is a serious flaw.

Or they just see a market opportunity by not waiting around for something better (for which I cannot fault them). 

But epub books seem to typically have very simple designs that are more easily adaptable to such a simplistic solution. Many Web sites and Web apps are far more complex, and may not produce good results from the same strategy. As a single simple example, consider those that use 'overflow-x:hidden; overflow-y:auto' in order to create an area that scrolls in the direction of a long list, and use 'white-space:nowrap; text-overflow:ellipsis' to indicate when there is hidden text in the X direction. I saw no mention of overflow in that second link you provided, which had a list of the properties it did transmute.

Received on Monday, 7 June 2010 16:40:07 UTC