Message-Id: <v03100606ae947125f6dc@[205.214.160.36]>
In-Reply-To: <v0310060bae9424e8cfdb@resnick1.isdn.uiuc.edu>
References: <EMZN514237D460@MPA15AB> Your message of "23 Oct 1996 11:47:42
 -0500" <v03100603ae93f3913621@resnick1.isdn.uiuc.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
Date: Wed, 23 Oct 1996 18:12:20 -0700
To: drums@cs.utk.edu
From: Dave Crocker <dcrocker@brandenburg.com>
Subject: Re: draft-ietf-drums-abnf-01.txt

At 12:48 PM -0700 10/23/96, Pete Resnick wrote:
>You have a very sick gateway that needs fixing. This To: address is broken.

	Apparently, Unisys is quite proud of its broken gateway.  It's been
broken for some years and Unisys has been given very considerable,
ummmmmmm, feedback about it.  It's astonishing that they remain insensitive
to the embarassment that continues to accrue to the company (and its
employees) in this matter.

>>0 -   It is called "...abnf-01.txt" on the internic host
>I believe that is correct. Wasn't the last one "00"?

oops, sorry.

>>3.3-  Wouldn't it be better to take a lesson from the C experience
>>      and put the modifier character before the equal sign rather
>>      than after? Thus, "=/" -> "/=".
>I think this makes the parsing harder. The way it is now, you look for the
>equal sign and then figure out what's to the left (the rule name) and

That's exactly why I did it =/.  I prefer /= but decided to keep the
parsing simpler.

>>3.4-  The second example implies that alternative is higher precedence
>>      than concatenation.  This can't be right!

At the least, it should be listed explicitly and now it is.

>>3.11- I think the order (highest to lowest?) of these should be stated.
>>      I also think that an example would help immensely, since people
>>      always seem to have clear but conflicting ideas of what higher
>>      precedence versus lower means.
>
>Yes, absolutely.

feed me one and I'll include it.

>I don't think concatenation has ever been highest precedence. It should be
>in the list, though, whereever it goes.

turns out that it seems to be LOWEST priority, I think.

>>      Some mention of the convention of defining some very low level
>>      items as:
>>
>>            "<" <a description of what we all know anyway> ">"

done. (Ned got to me first.)

>I don't think we should be encouraging this practice at all. I would prefer
>to see no standard ever use this ugliness.

	It's a necessary evil, to be avoided whereever possible, not not
always possible.  As Ned observed, I had two of them in the version you
folks got.

d/

--------------------
Dave Crocker                                             +1 408 246 8253
Brandenburg Consulting                              fax: +1 408 249 6205
675 Spruce Dr.                                  dcrocker@brandenburg.com
Sunnyvale CA 94086 USA                        http://www.brandenburg.com

Internet Mail Consortium                http://www.imc.org, info@imc.org