Re xml-dev Digest V1 #348

[Oren Ben-Kiki]

Frank Boumphrey wrote:

<Oren Ben-Kiki>
It seems as though you seriously suggest that the lack of documentation of
the technical process of creating the spec (specifically, the issues, the
alternatives, and the rationale for the adopted solutions) is "OK" since one
can simple E-mail some WG member and ask him to explain it for you. After
all, it works so well for you!
</Oren Ben-Kiki>

<Frank Boumphrey>
There is usually a requirements document, followed by several public working
drafts working drafts, followed by a propoaed recommendation, followed by a
recommendation.
</Frank Bounphrey>

Alas, a working draft/recommendation is not accompanied by a summary of the
internal W3C issue/alternatives/resolution documentation. As the XHTML case
demonstrates, this is problematic.

<Frank Bounphrey>
Public input is invited at every stage of this process.
</Frank Bounphrey>

The problem is not the ability to send recommendations to the WG, it is
having to second guess the reasons for the decisions it makes.

<Oren Ben-Kiki>
At any rate, I don't have the E-mail addresses of the XHTML WG members. I
can't officially get these addresses AFAIK.
</Oren Ben-Kiki>

<Frank Boumphrey>
The addresses of the editors are usually on the document.
</Frank Boumphrey>

True. But this address is not intended for asking questions about the
rationale for the draft - it is intended for submiting possible improvements
for it. I personally would not feel comfortable in using it in order to
obtain "rationale", though I'm certain the editors would provide it on
request. Every one of them I've been in contact with has been very helpfull.
However, this can't be the standard way to obtain it, because, as I've said:

<Oren Ben-Kiki>
 Even if I could, it would not be
practical for them to answer my questions - because they'd be swamped with
the questions of every other interested reader of the draft.
</Oren Ben-Kiki>

<Frank Boumphrey>
I have never failed to get a response when I've written, and the times any
one has called me, I have always replied.
</Frank Boumphrey>

I fully appreciate the efforts that W3C members - editors included :-) - are
making to convey the rationale to the public. But I feel that answering
private E-mails should not be the prefered way to do so.

<Oren Ben-Kiki>
 Obviously there is a reason why proper process
documentation is not being provided. The problem is that the simplest reason
is "to hide any shady politics between member companies".
</Oren Ben-Kiki>

<Frank Boumphrey>
I think you've been reading too many conspiracy novels!:>)
</Frank Boumphrey>

I didn't say I believed this to be the reason, I said it is the simplest
one. There are several advantages for providing a clear rationale:

- The quality of the drafts would increase. It turns out that having to
spell out the reasons for a decisions forces better decision making.

- The quality of outside input to the WG would improve. People won't be
rehashing old arguments - at least not as much.

- Developers would find it easier to accept decisions they disagree with.
The fact that the favored alternative was considered and found less
satisfactory for clearly specified reasons does wonders against the feeling
of being "left out of the process".

So the reason for _not_ providing a rationale has to be good enough to
override these three advantages. A pretty tall order, which explains why
there are  "conspiracy theorists". I'm _not_ one - as I've repeatedly said,
I feel the W3C is doing a pretty good job overall. The ill feelings
manifested in this list towards the W3C can be pretty easily dispersed by a
bit of PR effort - and by releasing technical rationale documentation.

<Frank Boumphrey>
The w3c works by consensus!

The documentation that is not publically available is for the most part not
worth reading! Many of the internal mailing lists make this list look like
the embodiment of decorum and lucidity!
</Frank Boumphrey>

I didn't say anything about a general release of all internal documentation
and not even a release of the meeting protocols. I agree with you that this
would be pretty much useless. What I would like to see is some rationale
documentation as an integral part of each working draft and recommendation.
This would be _based_ on the existing internal documentation already
required by the W3C process.

<Oren Ben-Kiki>
Another reason is
"because it would harm the quality of the resulting recommendation" -
obviously absurd, or maybe "it would slow things down too much"
</Oren Ben-Kiki>

<Frank Boumphrey>
It's very difficult to get  16 people to agree, yet alone a whole mailing
list
</Frank Boumphrey>

Nobody said that the whole world should agree with every decision made by
the WG. As I've said above, a rationale is a very good way to make
developers accept reasonable decisions they don't agree with.

Share & Enjoy,

    Oren Ben-Kiki



xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)