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

Re: [css3-tables] Optimize per column whitespace due to line wrapping

From: Shelby Moore <shelby@coolpage.com>
Date: Sun, 2 Jan 2011 12:03:34 -0500
Message-ID: <5e0b3b55bb19bc27c59eaf40bb039798.squirrel@sm.webmail.pair.com>
To: "timeless" <timeless@gmail.com>
Cc: www-style@w3.org
> I'm confused.

Thanks for taking the time to explain and get clarification.

> I don't see a problem, the author asked for something mildly stupid by
> overdesigning a table, and got what was requested.

I am the author.

What overdesign? It was "designed" (quickly slapped together) to make sure
the equations don't split and become unreadable (as if they were rendered
equation images). It was also intended to make the logic of the complex
set expression easier to follow with horizontal alignment or multiple
lines of logic, with comments that apply to each line to the right.

Off topic, there may be cleaner ways to encode that layout objective, e.g.
nowrap style instead of so many '&nbsp;', but that is irrelevant to the
point of this thread. I didn't have time to get into those intricacies and
browser version issues with more complicated nowrap.

> first column:
> Fi(w | w ? W) =
> Fi(n | n ? N) =
>
> second column:
> ?
> ?
> ?
> ?
> where Recurse is (Fi(n) \ ?) ? Fi(tail(w))
> fold(map(hash(G, n), Fi), ?, ?)

The last 2 rows have a colspan.

> third column:
> { ?
> { t
> ( Fi(n)
> { ?
> ( Recurse
>
> The third column is clearly stretching to fit the last cell "( Recurse"
> The relevant source is:
> <td>(&nbsp;Recurse</td>
> -- Which clearly indicates that the author doesn't want any wrapping.

Correct.

> fourth column:
> |
> |
> |
> |
> |
>
> The "where Recurse" bit from the second column is left aligned (like
> everything else) and is undersized relative to the row above it:
>
> ? ( Recurse | Recurse ? ?, ? ? Fi(n), n ? N, n = head(w) )
> In my browser, I have that split as:
>
> ? ( Recurse |                    Recurse ? ?,
>               ? ? Fi(n), n ? N, n = head(w) )
>
> where Recurse is (Fi(n) \ ?) ? Fi(tail(w))

Apparently you have the window width that just luckily makes it appear
there isn't a problem.

On my wide screen, I see:

? ( Recurse |                              Recurse ? ?,  First-set...
                        ? ? Fi(n), n ? N, n = head(w) )  First-set...
where Recurse is (Fi(n) \ ?) ? Fi(tail(w))               First-set...

The excess gap can be as large as the width of the "Recurse ? ?,".

Or more generally, the excess gap can be as large as the shortest wrapped
word.

The adverse effect is not limited to wide screen, if the middle words are
also long on every word-wrapped, then the effect will occur at various
window widths.


Generally this an optimization problem that applies to word-wrap and
column inline sizing.  The optimization becomes more needed (noticeable)
as the words that are wrapped become longer.
Received on Sunday, 2 January 2011 17:04:52 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:35 GMT