W3C home > Mailing lists > Public > public-html@w3.org > March 2011

Re: ISSUE-155 table-border: Chairs Solicit Proposals

From: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>
Date: Fri, 25 Mar 2011 21:41:40 +0100
To: Aryeh Gregor <Simetrical+w3c@gmail.com>
Cc: Sam Ruby <rubys@intertwingly.net>, HTML WG <public-html@w3.org>
Message-ID: <20110325214140277243.c406983e@xn--mlform-iua.no>
Aryeh Gregor, Fri, 25 Mar 2011 12:57:32 -0400:
> On Wed, Mar 23, 2011 at 8:22 PM, Leif Halvard Silli wrote:
>> UPDATE: It turns out that HTML5 requires UAs to treat @border as a
>> boolean:
> You missed the prose underneath:


The CSS rules above the prose affects cell borders. But I was also 
aware, if not of the prose, then at least of the of how @border affects 
table element itself. Making @border officially a boolean attribute, 
thus result in a 1px border around the table.

Do you have any view on the pros/cons of changing @border=value into 
@border=border (AKA a boolean attribute)?

These are the pros and cons that I see:

 * Pros: 
  - more obvious that it is a 'fallback' styling measure
  - better match of the the defined CSS rules.
  - help authors to avoid border="\digit", which doesn't
    affect border width (except on <table> itself) anyway.
  - simpler to for vendors to understand its effect if it
    is works as a boolean. (Currently there are some bugs,
    most prominently in Webkit - see bug report.[1])
 * Cons: 
  - some current UAs have bugs which means that they may not 
    treat <table border="border"> equal to <table border="1">
    (Most prominently Webkit - see my bug report. [1])

In my view, the pros outweighs the cons.

[1] https://bugs.webkit.org/show_bug.cgi?id=56982
leif halvard silli
Received on Friday, 25 March 2011 20:42:17 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 29 October 2015 10:16:11 UTC