On Feb 21, 2008, at 5:49 PM, Maciej Stachowiak wrote:
On Feb 21, 2008, at 5:03 PM, Graydon Hoare wrote:
So presenting yourself as a participant with neither of those
supports in place, you're sort of walking into a room, kicking the
legs out from under a table and asking why it's suddenly on the
floor. We need to revise the strategy a bit to help for your case,
if we can't rely on those. Please try to be patient if this takes
some time; none of us had put "write pre-impl docs" in our current
work-assignment schedules. We'll need to make room to do it. I can
probably dedicate a month of nearly-full-time energy to this
starting in the first week of march. Soon enough?
I'm totally willing to wait for documentation that is more clear
and up to date. What concerned me was that it wasn't part of the
plan at all, and that Brendan at least seemed unreceptive to the
request.
You keep misrepresenting things. I never said anything about not
writing up-to-date specs. I said specs should be co-evolved with
implementations to be up to date, and to the extent that they can be
based on extracted RI code, they should be developed that way, so as
to be mechanically checkable.
This is a serious point. It deserves more than what I regard as
flippant attitude about "ask Graydon" combined with overstatement of
the value of all-prose specs. If you have to look at four files to
understand let as implemented in the RI, so what? ES1-3 require
multiple readings of sections 8-15 just to settle simple questions.
The overview and evolutionary programming tutorial take more top-down
approaches to presenting (not specifying) features.
Programming languages with the kind of type system, binding rules,
and compatibility constraints that ES4 has absolutely need executable
semantic specifications, not merely all-prose specs.
That it takes time is natural.
Thanks for understanding. Now with that in mind, please re-read
Jeff's post and mine. We are talking about working intensively in the
next three months on both specs and implementation. Now is the time
to step up. Apple was hors de combat for a long time in Ecma TC39.
Kicking the legs out from under the table and pointing at the floor
is not good citizenship in my book, whatever our (real) failings in
keeping proto-specs up to date.
Ultimately for ES4 to be a meaningful standard, it has to have a
spec that is comprehensible without special insider information.
That's true if you mean by "comprehensible" "things only in people's
head". I claim it's false if you mean "things specified only in prose".
My expectation was that at least parts of it would start to
approach that point before they reached the implementation phase.
I'm surprised that this wasn't the plan already, but I'm glad you
are willing to be flexible.
I'm not thrilled about people detouring into recapitulating in
uncheckable, duplicative, buggy prose what is in the RI. I'd rather
we focus on where the final spec will want to use prose anyway, not
extracted, lowered RI code. Probably we will have to compromise and
do both prose and RI work. How willing to be flexible are you?
/be
_______________________________________________
Es4-discuss mailing list
[email protected]
https://mail.mozilla.org/listinfo/es4-discuss
