- From: Peter Moulder <pjrm@mail.internode.on.net>
- Date: Sun, 8 Feb 2015 08:24:48 +1100
- To: www-style@w3.org
- Cc: Alan Stearns <stearns@adobe.com>
On Fri, Feb 06, 2015 at 05:23:56AM +0000, Alan Stearns wrote: > [Perhaps the not-fully-balanced case might be described as] > 2. Try to fix a short last line without changing the “normal” line measure > too much I've already clarified that I don't think that "without changing measure too much part" is really part of it (in fact I think we agree that these use cases are more able to change their measure when necessary). Here I'll comment on the "fix a short last line" part. If we consider the case of a higher end UA such as used in print, I don't think we need a property to say "avoid short last line", because such a UA should always *try* to avoid a short last line, even in body text. The distinction between body text and the sorts of paragraphs we've been discussing is not that a short last line looks worse[*1], but that these paragraphs are more able to reduce their visible measure than normal body text is. For a web-browser-like UA, it might be that the main effect is to change the effective line-breaking width to make a short last line less likely. Whereas for a UA more like what Adobe or I would be working on, this freedom can also be used to avoid other badnesses, whether in rag shape or in avoiding weakly dispreferred break opportunities. (Even a web browser might consider avoiding weakly dispreferred breaks if the resulting paragraph still uses only two lines.) pjrm. [*1]: Indeed, these sorts of paragraphs don't have to contend with text-indent, and one might be more willing to prefer a short last line over a hyphenation for a pull quote than in body text.
Received on Saturday, 7 February 2015 21:25:26 UTC