[whatwg] Default (informal) Style Sheet

Mon Apr 2 00:59:50 PDT 2007

At 19:30 +0200 UTC, on 2007-04-01, Anne van Kesteren wrote:

> On Sun, 01 Apr 2007 20:16:12 +0200, Sander Tekelenburg <st at isoc.nl> wrote:
>> Who are we (as spec definers) to decide that x is the only correct
>> behaviour or presentation? And why should we want to stifle innovation
>> by requiring some specific presentation?
>
> Defining default rendering for certain constructs such as that the <body>
> element has a default margin of 8px (iirc) is important for
> interoperability reasons

I'm not sure I understand. Exactly what interoperability are you referring to
here? Surely we're not trying to ensure that a Web page is presented the same
in every browsing environment? What would be the use of that?

> and for new UAs trying to enter the market (saves
> them reverse engineering other UAs).

Hm... That might indeed be a problem looking for a solution. But I'm not at
all convinced that requiring body {margin:8px} is the proper solution. Even
if it were the ony possible solution, I'm not convinced the benefits outweigh
the objections I raised.

> A lot of web pages rely on default
> rendering of elements. For instance, that an element such as <p> has a
> default style of "display:block" as opposed to "display:none".

Defining preseantation up to *that* level is no problem IMO. The current
(HTML 4) spec already does so, and in fact this is no more than a translation
of HTML's distinction between block and inline level elements to CSS
terminology.

I didn't get the impression from the OP though that the aim was to restrict
specifying of presentational defaults to this level. (The OP said "informal"
and "within limits", but didn't define that.)

> It is very important that UAs falling within the same conformance class
> agree on these basic principles so authoring against those UAs becomes
> more predictable

As I asked before: how does an author provided 'CSS zapper' not do that? How
in fact does requiring default presentations remove the need for authors to
provide 'CSS zappers'?

-- 
Sander Tekelenburg
The Web Repair Initiative: <http://webrepair.org/>