Re: Comments on Section 3 of the Requirements Document

Yaron Goland wrote:
> 
> Requirements documents are first and foremost goal posts. They defined
> what success is, they do not define how to succeed. So while I think the
> question of what we are going to do w/HTTP is critical, I do not believe
> that discussion should be included in the requirements doc. The
> requirements doc should state that we must use HTTP as our base but
> where we take it is the whole point of this group.

I don't even think the requirements document needs to state that "we
must use
HTTP as our base", either.

The requirements document is not the only constraint on the group's
output,
of course.

I think it would be a viable option to define a DAV protocol that didn't
use HTTP at all.

> As for when to discuss what we are doing to HTTP, I would strongly
> suggest that we discuss this on a case by case basis. Simply stating
> "free for all" or "no alterations" only guarantees that original ideas
> will be killed.

Well, there's a fundamental choice, really, that isn't to be made 
case-by-case, which is "are we layering a completely new distributed
object protocol on top of HTTP". If you're going to do that a little,
you might as well do it all the way. If you're not, then you shouldn't
do it in little parts.

This isn't a matter of killing original ideas, it's one of coming up
with
a spec that's self-consistent.

I don't think "self-consistency" needs to be in the requirements
document,
even though it's a requirement.

Larry

Received on Thursday, 13 February 1997 00:03:54 UTC