W3C home > Mailing lists > Public > www-style@w3.org > January 2015

Re: [css-text] Preventing typographic orphans

From: Florian Rivoal <florian@rivoal.net>
Date: Sun, 4 Jan 2015 12:50:38 +0100
Cc: www-style list <www-style@w3.org>
Message-Id: <6249F59D-8626-4BBB-80B8-2E827B90861B@rivoal.net>
To: Michał Gołębiowski <m.goleb@gmail.com>

> On 03 Jan 2015, at 23:58, Michał Gołębiowski <m.goleb@gmail.com> wrote:
> 
> I'd b very happy if such a thing existed in CSS. In Poland it's considered an error to put single-letter words like 'w' ('in'), 'z' ('with') etc. at the end of the line. Clients often expect this to be fixed; in one project we had to read the text from the backend and manually replace all spaces after such words with non-breaking ones via JS on the client side. Such a thing doesn't influence performance in a good way so I'd be glad if we could just do it with plain CSS.

This is somewhat similar to the practice called kinsoku-shori in Japanese, where certain letters are forbidden to be at the end of the line (for example opening [ ( { etc) even though naïve line breaking would put them there.

As far as I can tell, css-3-text gives enough flexibility to UAs to do the proper behavior when the language is identified, but does not require it.

This should probably fall under the list of things that are affected by "line-break: auto | loose | normal | strict". Maybe calling it out as an example there would give a hint to UAs that this is a desirable behavior for Polish, giving some basis to justify an implementation.

 - Florian
Received on Sunday, 4 January 2015 11:51:03 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:50 UTC