Ian Hickson wrote:
On Sun, 18 Jan 2009, Shelley Powers wrote:
The more use cases there are, the better informed the results will be.
The point isn't to provide use cases. The point is to highlight a
serious problem with this working group--there is a mindset of what the
future of HTML will look like, and the holders of the mindset brook no
challenge, tolerate no disagreement, and continually move to quash any
possibility of asserting perhaps even the faintest difference of
opinion.
I'm certainly sad that this is the impression I have given. I'd like to
clarify for everyone's sake that this mailing list is definitely open to
any proposals, any opinions, any disagreement. The only thing I ask is
that people use rational debate, back up their opinions with logical
arguments, present research to justify their claims, and derive proposals
from user needs.
I've been especially critical of you, which isn't fair. At the same
time, as you have said yourself, you are a "benevolent dictator", which
seems to me to not be the best strategy for an inclusive HTML for the
future.
I know I'm not comfortable with the concept. But I'm also late to this
group, and shouldn't disrupt if the strategy works.
Regardless, I got the point in the comment. That, combined with this
email from Ian, tells us that it doesn't matter how our arguments run,
the logic of our debate, the rightness of our cause--he is the final
arbiter, and he does not want RDFa.
For the record, I am as open to us including a feature like RDFa as I am
to us including a feature like MathML, SVG, or indeed anything else. While
I may present a devil's advocate position to stimulate critical
consideration of proposals, this does not mean that my mind is made up. If
my mind was made up, I wouldn't be asking for use cases, and I wouldn't
be planning to investigate the issue further in April.
There is a fine difference between being the devil's advocate, and the
devil's front door made of thick oak, with heavy brass fittings.
How does one know if one has provided a use case in a format that is
more likely to meet a successful outcome, than not. Is the criteria
documented somewhere? It's difficult to provide use cases with the
twenty questions approach.
What are the criteria by which a possible solution to a problem is
judged? Is there a consistent set of questions asked? Tests made? A
certain number of implementations? Again, is this documented somewhere?
I am not paid by Google, or Mozilla, or IBM to continue throwing away my
time, arguing for naught.
It may be worth pointing out that, many of our most active participants
are volunteers, not paid by anyone to participate. Indeed I myself spent
many years contributing to the standards community while unemployed or
while a student. I am sorry you feel that you need to be compensated for
your participation in the standards community, and wish you the best of
luck in finding a suitable employer.
The point I was trying to make, and forgive me if the my writing was too
subtle, is that it's not the fact that the work will time, but whether
the time will be well spent.
Operating in the dark and tossing use cases in hopes they stick against
the wall, without understanding criteria is not a particularly good use
of time. However, having specific tasks that meet a given goal, and
knowing that the goal is stable, and not a moving target, goes a long
way to ensuring that the time spent has value.
Knowing that one can, with diligence, ensure that the best result occurs
is a good use of time.
Spitting into the wind, at the whim and whimsy of a benevolent dictator,
is not a good use of time.
As far as Google goes, we have no corporate opinion either way on the
topic of RDFa in HTML5. We do, however, encourage the continued practice
of basing decisions on data rather than hopes.
Bully for Google.
Shelley
