W3C home > Mailing lists > Public > www-style@w3.org > March 2011

RE: [css3-text] Proposed pruning & scoping of hyphenation properties

From: Christian Stockwell <cstock@microsoft.com>
Date: Thu, 31 Mar 2011 22:20:19 +0000
To: fantasai <fantasai.lists@inkedblade.net>
CC: "www-style@w3.org" <www-style@w3.org>
Message-ID: <2B4FAE26EDC5A14CAD4376398DC6DFBE64990FF7@TK5EX14MBXC122.redmond.corp.microsoft.com>

> -----Original Message-----
> From: fantasai [mailto:fantasai.lists@inkedblade.net]
> Sent: Tuesday, March 29, 2011 11:37 PM
> To: Christian Stockwell
> Cc: www-style@w3.org
> Subject: Re: [css3-text] Proposed pruning & scoping of hyphenation
> properties
> On 03/29/2011 09:21 PM, Christian Stockwell wrote:
> >
> > I don't follow. What is the hyphenation dictionary supposed to take
> > care of here? Perhaps it would help to describe an expected use case
> > for "word-wrap: hyphenate" and how it would relate to usage of "hyphens".
> The use case is that hyphenation is turned off. We don't really want
> hyphenation. But there's this really big word that doesn't fit. We have three
> options here, represented by the "emergency-wrap" ('word-wrap')
> property:
>    - overflow the containing block ("word-wrap: normal")
>    - break arbitrarily ("word-wrap: break-word")
>    - fire up the hyphenation engine and wrap the word via hyphenation
>      ("word-wrap: hyphenate")
> > (Adding one point that I just noticed) 7. 5 of the "control"
> > properties are specified as optional:
> > The following author controls are not required to be supported for the
> > UA to claim conformance to CSS Text Level 3:
> > *'hyphenate-limit-zone'
> > *'hyphenate-limit-chars'
> > *'hyphenate-limit-lines'
> > *'hyphenate-resources'
> > *'@hyphenate-resource'
> >
> > It seems like hyphenate-limit-last should be included in that list,
> > but is not. Was it omitted in error?
> Nope, that was intentional. There was some discussion in the F2F that a lot of
> the hyphenation controls become not so necessary if you support Tex-style
> paragraph breaking: there's not so much need for tweaking the hyphenation
> limits to get good behavior. That's why I put those limit properties in the
> optional list.
> 'hyphenate-limit-last' doesn't fall into that category: whether hyphenation is
> allowed on the last line of a page or spread is more stylistic choice than
> optimal breaking lever. So it made sense to leave it out of the list.
> ~fantasai

Can you elaborate on how you are distinguishing "optimal breaking lever" from "stylistic choice"? To me this distinction appears to be completely arbitrary. In particular, perhaps it would be helpful to explain why "hyphenate-limit-lines" is "a lever" rather than a "stylistic" property since it seems to be most closely related to hyphenate-limit-last.

Given that CSS does not require UAs to implement TEX-style paragraph breaking I think it's important that we make a strong statement: Either basic hyphenation control is important for CSS or it is not. If it is, these basic control properties should be required for conformance (with the exception of hyphenate-resource, which is not yet sufficiently defined). If basic hyphenation control is not important for CSS, then all of these properties should be optional. Our mixed message here doesn't make much sense to me.

On an unrelated note, I think we need to properly define (or remove) the "spread" value. As far as I can tell it is not sufficiently defined anywhere, including in the print module where I'd expect to see it described. My preference is to remove the value since "page" should be sufficient for this use case and does not require us to define a completely new concept that hasn't been tackled elsewhere in CSS.
Received on Thursday, 31 March 2011 22:21:15 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:38:44 UTC