W3C home > Mailing lists > Public > public-html@w3.org > May 2009

Re: Design Principles

From: Sam Ruby <rubys@intertwingly.net>
Date: Tue, 26 May 2009 08:26:14 -0400
Message-ID: <4A1BDFE6.7000704@intertwingly.net>
To: Ian Hickson <ian@hixie.ch>
CC: Julian Reschke <julian.reschke@gmx.de>, HTML WG <public-html@w3.org>
Ian Hickson wrote:
> On Tue, 26 May 2009, Julian Reschke wrote:
>> Ian Hickson wrote:
>>> ...
>>> This is the approach I have taken, and intend to continue taking, in editing
>>> the HTML5 specification, so long as this working group continues to allow me
>>> to edit it. I believe technical soundness and practical usefulness is more
>>> important than theoretical purity and consistency with other specifications.
>> Can we please have a vote on on the last part?
> 
> I would definitely support having a vote on this. Sam?

Phrasing it as "theoretical purity" would make such a vote a virtual
push poll, but leaving that aside for the moment...

What would be practical implications of this choice be?

Time to have a little fun (read: please don't take the below too seriously):

   Taking both to the equivalent extreme, and using the excellent
   terminology that Mark Pilgrim introduced[1], isn't this essentially
   being a choice between being a sociopath ("argue that the [other] spec
   is ambiguous, or misleading in some way, or ignoreable because nobody
   else implements it, or simply wrong") and and asshole ("write code
   that is meticulously spec-compliant, but useless. If someone yells at
   them for writing useless software, they smugly point to the sentence
   in the [other] spec that clearly spells out how their horribly broken
   software is technically correct").

   Of course, the right answer is not to take either position to such an
   extreme.  My experience is that nobody here has done so, though many
   at times act as if others others have done so.

   I believe that "technical soundness" and "theoretical purity" are both
   in the eye of the beholder, essentially reduce down to the same thing,
   but of course the former is more often used to refer to ideas the
   person making the statement favors, and the latter to refer to all
   other ideas.  In any case there is a rather low upper bound on how
   well you can achieve either based on "don't break the web".

   "Practical usefulness" is motherhood and apple pie.  Is there anyone
   here opposed to "practical usefulness"?  Come on, raise your hands if
   you do!

   "Consistency with other specifications" in practice means "make every
   effort to work with authors of specs when there is a conflict, but
   when impasses occur, don't let that stop you".  A concrete example of
   this can be found in the HTML 4.01 spec[2].  I know of no effort to
   "improve" the SGML spec in this manner.

As Maciej has apparently picked up developing the Design Principles, I'd
suggest that one or both of you simply propose adjustments or additions
to this document.  Meanwhile, I prefer dealing with specifics.  Ian, the 
answer you previously gave on feed sniffing was most helpful.

- Sam Ruby

[1] http://diveintomark.org/archives/2004/08/16/specs

[2] http://www.w3.org/TR/REC-html40/appendix/notes.html#h-B.3.7
Received on Tuesday, 26 May 2009 12:26:57 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:16:37 GMT