W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > May to August 1995

Complaints regarding Draft Status and Author

From: Roy Fielding <fielding@beach.w3.org>
Date: Wed, 09 Aug 1995 14:03:55 -0400
Message-Id: <199508091803.OAA03854@beach.w3.org>
To: Rich Salz <rsalz@osf.org>
Cc: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
>This attitude is offensive.  We are not the huddled tribes waiting for
>stone tablets to come down from above.  The members of this mailing list
>are those people who expect to be involved in the definition of the HTTP
>protocol.  At least now I understand why a 500-line detailed commentary on
>v10-spec that I sent to this list at the end of May got no reaction.

I am sorry that you feel that way, but you are incorrect on all counts.

First, I have never refused to make a change to any of the drafts I have
edited once there became rough consensus to do so.  I have the highest
regard for the members of this mailing list, and the people whom I *do*
ask about specific design changes are all members of this list.  The reason
I ask them first and not the list-at-large is because THEY HAVE ALREADY
IMPLEMENTED THE ITEM BEING DISCUSSED or because I can talk to them
face-to-face at an IETF meeting, at UCI, at MIT, or at any one of a
dozen other meetings/conferences I attended over the past four months.

Second, being "involved in the definition of the HTTP protocol" means
exactly what you are doing now.  Read the draft.  Send in comments.
The WG consensus will be polled and, if desired, the changes will be
made to the draft.  That is why it is called a DRAFT.  If you want to
propose the exact wording of the item in question, than do so.  If not,
then I will write what I believe to be a suitable draft based on those
comments, and then invite additional input (just as I am doing now).

Third, not all comments I receive on the spec are sent via the WG mailing
list.  For every one public commentary, I receive three or four private
messages.  I must take all into consideration, though all final decisions
are made by the WG.

Fourth, the reason your commentary received no response was because

   1) It was 500 lines long, making it difficult to "discuss"

   2) Most of the comments were accurate and agreeable; these changes
      were all made to the draft [BTW, it took me 16 hours of editing
      to do so, because most of your suggestions were wholesale changes
      to the structure of the draft and style issues like replace all
      parentheses with commas, use must instead of required to, etc.].
      Did you check the draft for your suggested changes?

   3) It was mailed while I was on vacation, and thus I was not available
      to respond to it when it was fresh.  Sorry, but I now get 4-6 hours
      worth of mail to read and respond to *every day* -- I cannot respond
      to everything that comes my way, particularly not when I agree with
      what someone else has already said.

>This is not the way the IETF works.  At the very least, an apology is
>called for.  More likely, a major procedure reset probably needs to be done
>by the Area Director (e.g., replace the current editor).

I am sorry that my statement has been misunderstood as being a pronouncement
from behind some exalted status.  Please keep in mind that this is E-MAIL
and thus is a lousy medium for communicating intentions.

We are proceeding exactly as "the IETF works" with the singular exception
that I was unable to edit the draft for two months because I was
disconnected from my authoring environment.  Yes, that's bad, but there
wasn't a damn thing I could do about it (I *tried*, honestly I did, but
its difficult to edit a document when you are on a plane, driving a car,
looking for an apartment, and trying to deal with a computer that crashes
every 30 minutes).

Finally, note that I am a volunteer who has invested 10 months of my
*free* time to edit the specification.  To do so, I have delayed
my Ph.D. progress by one year and have not released a new version of my
three publically available software products in a year.  Aside from the
fact that I'm listed as an author, I did not get paid for this work
until I arrived at W3C, and even now it is only one of my duties.

Even so, I consider both RFC 1808 and the HTTP/1.0 draft to be among
the best written specifications produced by the IETF, as is clear
from the comments I have received outside the mailing list.  I invite
you to find a way in which it can be improved.

 ....Roy T. Fielding  Department of ICS, University of California, Irvine USA
                      Visiting Scholar, MIT/LCS + World-Wide Web Consortium
                      (fielding@w3.org)                (fielding@ics.uci.edu)

p.s. I just spent 2 hours composing this response.  On most days, that would
     be the sum total of my available free time.  Fortunately, I have
     arranged to be in a situation where I can spend my full day in what
     I would normally consider "free time".  Please don't waste what has
     caused me a great deal of pain, expense, and time to arrange.
Received on Wednesday, 9 August 1995 11:09:07 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:31:24 EDT