ietf-nntp Issue: empty groups

[Jack De Winter]

At 07:57 PM 12/31/96 -0500, Rich Salz wrote:
>>What I am trying to say though, is that it wasn't defined in the spec,
>>so the implementors decided to do something that appealed to them.  I
>>do not want to get into a holy war about which of them were right and
>>wrong, but I do want to see an explanation for that section that makes
>>sense.  To me, first=last=count=0 only makes sense if there never have
>>been any articles in that group.
>
>Not quite right.  At this point, if it's common practice that is not
>documented in RFC 977, then all this group could/should do is say "this
>is how to do it."  A note can explain that it's kindaa stupid, that
>the original reason was lost, etc., etc.
>
>It doesn't matter if you don't like it, it doesn't make sense, or if
>you have to add extra code.  If the common practice for some undefined
>behavior is "x", then "x" is what you must do.

I see your point, but don't 100% agree with it.  Our job, as
members of the NNTP-EXT WG is to document common practice and
to fix known problems with the existing specification.  The
'empty group return value' rathole is an example of where it
was not specified and there are people who have a lot of problems
with the common practice.  Others simply want to fix the bug
of not having it documented in the original spec.

When it boils down to it, it is the WG chair who will say if
something is applicable or not.  But we have the charter to
go by, and we need to make sure that we fix problems and document
common practice, just not one or the other.  Our goal is to
enable Stan Barber to take our information and to derive a
concise specification from it.

regards,
Jack

p.s. Happy new year to all... T-3:30 to go here, and everyone
in my house has the flu.
-------------------------------------------------
Jack De Winter - Wildbear Consulting, Inc.
(519) 576-3873		http://www.wildbear.on.ca/

Author of SLMail(95/NT) (http://www.seattlelab.com/) and other great products.