[css3-multicol] Feedback

[Sylvain Galineau]

In section 4, Column gaps and rules [1]:

"Column rules are only drawn between columns that have content."

Assuming column 2's entire content did not fit e.g. an image and needs to overflow to the next page, I assume column 2 is considered to have content in this case.


[1] http://www.w3.org/TR/2009/WD-css3-multicol-20090630/#column-gaps-and-rules

[Håkon Wium Lie]

Sylvain Galineau wrote, on 14 Jul (!) 2009:

 > In section 4, Column gaps and rules [1]:
 >
 > "Column rules are only drawn between columns that have content."
 >
 > Assuming column 2's entire content did not fit e.g. an image and
 > needs to overflow to the next page, I assume column 2 is considered
 > to have content in this case.
 >
 > [1] http://www.w3.org/TR/2009/WD-css3-multicol-20090630/#column-gaps-and-rules

It's a good question. Two more general questions, which the spec also
should answer, are:

  1) what are the endpoints of a column rule?
  2) between which columns should a rule be drawn?

It seems that we currently have three implementations: Gecko, Prince,
AntennaHouse. All three answer question 1) in the same way: the
endpoints are positioned on the content edge. However, I don't think
this is optimal. Consider this screen shot:

  http://people.opera.com/howcome/2009/tests/column-rule-ss.png

>From a visual/optical point of view, the column rules stick up too
high. This is due to the line-height being bigger than the height of
the glyphs. A better (albeit more complex) rule would be to base
computations on the font in the top/bottom lines, rather than the
content box.

For question 2), implementations differ. Gecko and AntennaHouse do not
display column rules if there isn't content on both sides. For
example, if an element has three columns, but only one word, there
will only be content in the first column. In this case, Gecko and
AntennaHouse will not display column rules at all. Prince, however,
will always display all column rules.

In all three implementations, all column rules of an element always
have the same length. This seems like a principle worth keeping.

Here's a test case:

  http://people.opera.com/howcome/2009/tests/column-rule.html

Comments welcome.

-h&kon
              Håkon Wium Lie                          CTO °þe®ª
howcome@...                  http://people.opera.com/howcome

[Håkon Wium Lie]

Also sprach Sylvain Galineau:

 > "Column rules are only drawn between columns that have content."
 >
 > Assuming column 2's entire content did not fit e.g. an image and
 > needs to overflow to the next page, I assume column 2 is considered
 > to have content in this case.
 >
 > [1] http://www.w3.org/TR/2009/WD-css3-multicol-20090630/#column-gaps-and-rules

After thinking about this for some time, I've reached the opposite
position: if some content doesn't appear in a column -- for whatever
reason -- it shouldn't be considered to be there for the purpose of
deciding if column rules should be drawn or not.

The primary reason is that column rules are visual distinctions
between content in columns; it the content has been moved, the visual
distinction isn't necessary any more.

This view, however, isn't strongly held and I can live with the
opposite conclusion.

I've added this example to the editor's draft to illustrate:

  If a tall image is moved to a column on the next page to find room for
  it, its natural column may be left empty. If so, the column is
  considered empty for the purpose of deciding if the column rule should
  be drawn or not.

  http://dev.w3.org/csswg/css3-multicol/

-h&kon
              Håkon Wium Lie                          CTO °þe®ª
howcome@...                  http://people.opera.com/howcome

[Sylvain Galineau]

Thanks, I am OK with this.

> -----Original Message-----
> From: Håkon Wium Lie [mailto:howcome@...]
> Sent: Sunday, October 18, 2009 12:06 PM
> To: Sylvain Galineau
> Cc: www-style@...
> Subject: Re: [css3-multicol] Feedback
>
> Also sprach Sylvain Galineau:
>
>  > "Column rules are only drawn between columns that have content."
>  >
>  > Assuming column 2's entire content did not fit e.g. an image and
>  > needs to overflow to the next page, I assume column 2 is considered
>  > to have content in this case.
>  >
>  > [1] http://www.w3.org/TR/2009/WD-css3-multicol-20090630/#column-
> gaps-and-rules
>
> After thinking about this for some time, I've reached the opposite
> position: if some content doesn't appear in a column -- for whatever
> reason -- it shouldn't be considered to be there for the purpose of
> deciding if column rules should be drawn or not.
>
> The primary reason is that column rules are visual distinctions
> between content in columns; it the content has been moved, the visual
> distinction isn't necessary any more.
>
> This view, however, isn't strongly held and I can live with the
> opposite conclusion.
>
> I've added this example to the editor's draft to illustrate:
>
>   If a tall image is moved to a column on the next page to find room
> for
>   it, its natural column may be left empty. If so, the column is
>   considered empty for the purpose of deciding if the column rule
> should
>   be drawn or not.
>
>   http://dev.w3.org/csswg/css3-multicol/
>
> -h&kon
>               Håkon Wium Lie                          CTO °þe®ª
> howcome@...                  http://people.opera.com/howcome

[Håkon Wium Lie]

Also sprach Sylvain Galineau:

 > > After thinking about this for some time, I've reached the opposite
 > > position: if some content doesn't appear in a column -- for whatever
 > > reason -- it shouldn't be considered to be there for the purpose of
 > > deciding if column rules should be drawn or not.

 > Thanks, I am OK with this.

Fantasai had a preference for your original proposal. I don't care
that much, and your original proposal is simpler. So, I've reverted
the draft.

That's the last issue, I believe -- the draft will advance to CR
shortly.

-h&kon
              Håkon Wium Lie                          CTO °þe®ª
howcome@...                  http://people.opera.com/howcome