[John Moreno <phenix@interpath.com>] Re: GNKSA and Gnus

[John Moreno <phenix@interpath.com>] Re: GNKSA and Gnus

[Lars Magne Ingebrigtsen]

Here's the latest GNKSA evaluation report.  16c has been fixed, I
think, but 7c will not be fixed.

The GNKSA has evolved from a sensible bare-bones minimum thang into
something, uh, else, and I think the whole think has become
pointless.  Gnus will therefore from now on not strive for GNKSA
compliance.  

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@ifi.uio.no * Lars Magne Ingebrigtsen
------- Start of forwarded message -------
Message-Id: <199712280107.UAA02498@mail.interpath.net>
Subject: Re: GNKSA and Gnus
Date: Sat, 27 Dec 97 20:07:56 -0500
From: John Moreno <phenix@interpath.com>
To: "Karl-Johan Noren" <k-j-nore@dsv.su.se>,
        "Lars Magne Ingebrigtsen" <larsi@gnus.org>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: 8bit

Karl-Johan Noren wrote:

[...]

>> Also did you send a copy of the review to the author?
>
>Nope - but I guess I should have, and I don't have the review
>available on the computer I use right now (I'm not sure it's
>still on my Unix account either). Gnus have a homepage at
>
>   http://www.gnus.org/
>
>though, I think.

Well here it is:

============================================================================

                                          GNKSA   Evaluation  Form  v2.0


Product Name           : Gnus
Product Version        : 5.3
Tested on platform     : Unix/GNU Emacs 19.34
Available for platforms: Emacs 19
Evaluation date        : December 10, 1997
Evaluated by           : Karl-Johan Norén


Summary
=======

Regarding the strict GNKSA requirements (MUSTs), this software:

[ ] PASSES, hence can from now on proudly carry the Seal.
[X] FAILS on the following points:
    7c   Does not restrict references sensibly
    16c  Does not warn when posting quoted text only


Additionally, the software violates the `soft' GNKSA requirements
(SHOULDs) in the following way:
    10d  Does not omit (proper) signatures from quoted text
    10e  Does not let the user indicate which part to followup to
    10g  Attribution line lacks Message-ID of original article
    16b  Does not refuse posting an empty article
    16d  Does not refuse to post quoted text only
    16f  Does not try to prevent posting multiple copies entirely


The software sports the following nice net-keeping features unmentioned
by the GNKSA:

Fully customisable via elisp.

Support for NoCeM (I haven't tested this, though).

Asks if posts are to be sent to unknown groups (ie groups not
present on the newsserver).

Good easily customisable threading, and it's easy to fetch
older articles in the thread.

Allows colouring of quoted text, signatures, new text, headers etc.


Additional Comments:

I haven't been able to test the 998-character limit on references
lines, it does however wrap the reference line.


Checklist
=========                                                       (M)UST /
                                                                (S)HOULD
1) Displays all essential header information
   Software clearly displays:
   [Y] a) Article's author (From)                                      M
   [Y] b) Article's Subject                                            M
   [Y] c) List of groups posted to (Newsgroups)                        M
   [Y] d) Where (and how) to direct followups (Followup-To)            M
   [Y] e) Where to reply to if not the From-address (Reply-To)         M
[Y] PASS: Satisfies all MUSTs                                          M

2) Provides clear, separate commands for new  posting, followup, and
   e-mail reply
   [Y] a) for posting a new article                                    M
   [Y] b) for posting a followup article                               M
   [Y] c) for replying by e-mail                                       M
   [Y] d) Uses standard terminology                                    S
   [Y] e) Avoids ambiguous terminology                                 S
[Y] PASS: Satisfies all MUSTs                                          M

3) Provides cross-posting functionality
   [Y] a) Allows specifying multiple groups                            M
   [Y] b) Warns about, or prevents, posting to large numbers of groups S
   [Y] c) Strongly encourages setting Followup-To: on large crossposts S
          (`Y' if large crosspostings are disallowed)
[Y] PASS: Satisfies all MUSTs                                          M

4) Allows users to change essential headers
   [Y] a) Allows editing Subject at all times during composition       M
   [Y] b) Allows specifying new Subject of at least 70 characters      M
   [Y] c) Allows setting "Followup-To: poster"                         M
[Y] PASS: Satisfies all MUSTs                                          M

5) Ensures followups and e-mail replies contain a correct Subject
   [Y] a) Prepends "Re: " if (and only if) not already present         M
   [Y] b) Preserves entire original Subject (modulo minor repairs)     M
[Y] PASS: Satisfies all MUSTs                                          M

6) Directs followups to the correct newsgroups
   [Y] a) Posts to groups in Followup-To if present                    M
   [Y] b) Initiates e-mail reply rather than a followup posting on
          "Followup-To: poster", clearly informing the user            M
[Y] PASS: Satisfies all MUSTs                                          M

7) Make sure followups contain valid References
   [Y] a) Creates References header with Message-ID of original article
          as the last element                                          M
   [Y] b) Includes last three References from original                 M
   [N] c) Ensures References will fit in 998 characters                M
   [Y] d) Keep as many References from original as fit                 S
   [Y] e) Does not propagate broken Message-IDs in original References S
[N] PASS: Satisfies all MUSTs                                          M

8) Direct e-mail replies to the correct address
   [Y] a) Uses Reply-To if present                                     M
   [Y] b) Uses From address otherwise                                   
[Y] PASS: Satisfies all MUSTs                                          M

9) Allow the user to change her mind about whether to post or mail (or
   do both) and behave if doing both
   [Y] a) Allows users to change their mind and mail rather than
          post after having initiated a followup message               M
   [Y] b) Allows users to change their mind and post rather than
          mail after having initiated a reply message                  M
   [Y] c) Does not offer both posting and mailing as default behaviour M
   [Y] d) Inserts a notification that the message was posted as well
          as mailed in the e-mail copy when both posting and mailing
          a followup article                                           S
[Y] PASS: Satisfies all MUSTs                                          M

10) Provide adequate quotation and attribution facilities
    [Y] a) Allows including quoted original                            M
    [Y] b) Clearly distinguishes quoted material                       M
    [Y] c) Prefixes quoted material with `>'/`> '                      S
    [N] d) Omits correctly delimited signatures from quoted material   S
    [N] e) Provides a means of indicating which part(s) to followup to S
    [Y] f) Attribution line containing original author precedes quotes M
    [N] g) Attribution contains Message-ID of quoted article           S
[Y] PASS: Satisfies all MUSTs                                          M

11) Provide a user-specified "Subject: " header
    [Y] a) Requires non-empty, user-specified Subject for new articles M
    [Y] b) Refuses posting articles without, or with an empty, Subject M
    [Y] c) Does not provide default Subject if user did not set one    M
    [Y] d) Allows changing the Subject at any time while editing       M
[Y] PASS: Satisfies all MUSTs                                          M

12) Provide a valid "From: " header
    [Y] a) Sets "From: " header to syntactically valid e-mail address  M
    [Y] b) Refuses posting articles without a syntactically valid
           "From: " header                                             M
    [Y] c) Uses correct e-mail addresses (valid and belonging to the
           user) only, as far as it can possibly know                  S
[Y] PASS: Satisfies all MUSTs                                          M

13) Allow users to both cancel and supersede their own articles (and
    _no_ others!)
    [Y] a) Allows cancelling articles                                  S
    [Y] b) Allows superseding articles                                 S
    [Y] c) As far as possible, does not allow cancelling or superseding
           other peoples' articles                                     M
    [Y] d) Uses standard terminology                                   S
[Y] PASS: Satisfies all MUSTs                                          M

14) Try to respect the 80-character line-length convention
    [Y] a) Articles are posted as edited, with linebreaking intact     S
    [Y] b) Warns about lines over 80 characters                        S
    [Y] c) Does not refuse to post articles containing long lines      S
    [Y] d) Allows rewrapping quoted text                               S
    [Y] e) Enforces formatting requirements on article after external
           editing (`Y' if there is no support for external editors)   S
[Y] PASS: Satisfies all MUSTs                                          M

15) Separate signatures correctly, and don't use excessive ones
    [Y] a) Uses (and enforces) standard signature delimiter            S
    [Y] b) Warns against or refuses to use excessive signatures        S
[Y] PASS: Satisfies all MUSTs                                          M

16) Try to prevent obvious user errors
    [Y] a) Warns when attempting to post empty articles                M
    [N] b) Refuses posting empty articles                              S
    [N] c) Warns when post articles containing quoted material only    M
    [N] d) Refuses posting quoted-text-only articles                   S
    [Y] e) Warns against posting multiple copies (if possible at all)  M
    [N] f) Prevents multiple posting entirely                          S
[N] PASS: Satisfies all MUSTs                                          M

17) Post human-readable articles unless ordered otherwise
    [Y] Does not (and can not) encode or encrypt articles unless
        on explicit user demand                                        M
[Y] PASS: Satisfies all MUSTs                                          M

18) Provide self-protection
    [Y] Allows filtering out annoying articles (killing)               S
[Y] PASS: Satisfies all MUSTs                                          M

19) Be kind to servers, leave room for others
    [Y] a) Does not unnecessarily open multiple connections            M
    [Y] b) Does not generate excessive server load otherwise           M
[Y] PASS: Satisfies all MUSTs                                          M





------- End of forwarded message -------

Re: [John Moreno <phenix@interpath.com>] Re: GNKSA and Gnus

[Russ Allbery]

Lars Magne Ingebrigtsen <larsi@ifi.uio.no> writes:

> Here's the latest GNKSA evaluation report.  16c has been fixed, I
> think, but 7c will not be fixed.

[snip]

| [X] FAILS on the following points:
|     7c   Does not restrict references sensibly
|     16c  Does not warn when posting quoted text only

[snip]

| 7) Make sure followups contain valid References
|    [Y] a) Creates References header with Message-ID of original article
|           as the last element                                          M
|    [Y] b) Includes last three References from original                 M
|    [N] c) Ensures References will fit in 998 characters                M

7c is complete and utter hogwash.  News readers have absolutely no
business arbitrarily truncating References headers and destroying
threading information, and nothing in any news standard requires this.

And on top of that, there are absurdities like:

|     10e  Does not let the user indicate which part to followup to

This is nonsensical.  In the entire time that I've been reading and
posting to Usenet, I've never had a need for any such thing, or even have
any coherent idea how one would *do* such a thing.  It sounds to me like a
broken attempt to encode etiqutte that requires human intervention and
forethought (namely, snipping unneeded quoted material) into software.
Efforts of that sort are invariably doomed to failure.

|     10g  Attribution line lacks Message-ID of original article

And this is absurd and utterly wasted space.  Inclusion of message IDs in
attribution lines serves absolutely no purpose other than to drop hash not
intended for human consumption at the beginning of human-readable text.
This is the purpose served by the References header, not the attribution
line.

> The GNKSA has evolved from a sensible bare-bones minimum thang into
> something, uh, else, and I think the whole think has become pointless.
> Gnus will therefore from now on not strive for GNKSA compliance.

Thank you.

The GNKSA started as a very reasonable and interesting endeavor; it's
really a shame that it's turned into something this useless and misguided.

-- 
Russ Allbery (rra@stanford.edu)         <URL:http://www.eyrie.org/~eagle/>

Re: [John Moreno <phenix@interpath.com>] Re: GNKSA and Gnus

[Karl Kleinpaste]

Lars Magne Ingebrigtsen <larsi@ifi.uio.no> writes:
>> The GNKSA has evolved from a sensible bare-bones minimum thang into
>> something, uh, else, and I think the whole think has become pointless.

Russ Allbery <rra@stanford.edu> writes:
> The GNKSA started as a very reasonable and interesting endeavor; it's
> really a shame that it's turned into something this useless and misguided.

The real horror is that GNKSA started out as nothing more challenging
than codifying the behavior of rn (for the most part), and yet I
recall such innumerable complaints about all the supposedly heavy
requirements being forced on all those suffering software authors,
most of whom just somehow couldn't get it right.

Re: [John Moreno <phenix@interpath.com>] Re: GNKSA and Gnus

[Kai Grossjohann]

>>>>> On 04 Jan 1998, Russ Allbery said:

  GNKSA>     10e  Does not let the user indicate which part to followup to

  Russ> This is nonsensical.  In the entire time that I've been
  Russ> reading and posting to Usenet, I've never had a need for any
  Russ> such thing, or even have any coherent idea how one would *do*
  Russ> such a thing.  It sounds to me like a broken attempt to encode
  Russ> etiqutte that requires human intervention and forethought
  Russ> (namely, snipping unneeded quoted material) into software.
  Russ> Efforts of that sort are invariably doomed to failure.

Maybe they're talking about digests?  I found the feature of TM to be
able to hit "a" (or was it "f") on the header line of a submessage
quite useful.  It was a MIME multipart message with a message/rfc822
part, I think, where I found that useful.

As the evaluation form doesn't really say, and I like to see the good
things in people... :-)

I would like to see the signature omitting stuff, though.

-- 
Kai Grossjohann, Informatik VI        grossjohann@ls6.cs.uni-dortmund.de
Uni Dortmund, D-44221 Dortmund        http://ls6-www.cs.uni-dortmund.de/
                                      Vox +49 231 755 5670, Fax -2405
I like both kinds of music.

Re: [John Moreno <phenix@interpath.com>] Re: GNKSA and Gnus

[Russ Allbery]

Kai Grossjohann <grossjohann@charly.cs.uni-dortmund.de> writes:

> Maybe they're talking about digests?  I found the feature of TM to be
> able to hit "a" (or was it "f") on the header line of a submessage quite
> useful.  It was a MIME multipart message with a message/rfc822 part, I
> think, where I found that useful.

Hmm.  Okay.  I can buy this, although I encounter very, very few digests.
They're common in some varieties of mailing list, but pretty uncommon on
Usenet overall (although admittedly I have seen them).  Most of what I see
on Usenet in digest format are FAQs.

> As the evaluation form doesn't really say, and I like to see the good
> things in people... :-)

Right.  :)

> I would like to see the signature omitting stuff, though.

So would I, although I don't really need it.  Following is a bit of elisp
code I wrote that has been referred to by at least one friend of mine as
"the best reason to switch to using emacs to post and send mail that I've
seen."  :)

;; Delete anything between the current point and the signature (defined as
;; the block starting with "-- " at the beginning of a line.  *Extremely*
;; useful for killing quoted text.
(define-key message-mode-map "\C-cz" 'rra-kill-to-signature)
(defun rra-kill-to-signature ()
  "Deletes all text up to the signature."
  (interactive)
  (save-excursion
    (push-mark (point) t t)
    (re-search-forward "^-- ")
    (forward-line -1)
    (kill-region (mark) (point))))

-- 
Russ Allbery (rra@stanford.edu)         <URL:http://www.eyrie.org/~eagle/>

Re: [John Moreno <phenix@interpath.com>] Re: GNKSA and Gnus

[Lars Magne Ingebrigtsen]

Russ Allbery <rra@stanford.edu> writes:

> ;; Delete anything between the current point and the signature (defined as
> ;; the block starting with "-- " at the beginning of a line.  *Extremely*
> ;; useful for killing quoted text.
> (define-key message-mode-map "\C-cz" 'rra-kill-to-signature)
> (defun rra-kill-to-signature ()
>   "Deletes all text up to the signature."
>   (interactive)
>   (save-excursion
>     (push-mark (point) t t)
>     (re-search-forward "^-- ")
>     (forward-line -1)
>     (kill-region (mark) (point))))

Seems like a useful command to me.  What keystroke should it be bound
to?  The `C-c C-LETTER' keymap is starting to get rather full...

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

kill-to-signature (was GNKSA and Gnus)

[Russ Allbery]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Seems like a useful command to me.  What keystroke should it be bound
> to?  The `C-c C-LETTER' keymap is starting to get rather full...

Well, the original choice of C-c z (in the space dedicated to user
bindings, of course) was for the mnemonic similarity to ESC z.  In my
version of Gnus, C-c C-z still hasn't been taken, but I'm still using a
fairly old version.

-- 
Russ Allbery (rra@stanford.edu)         <URL:http://www.eyrie.org/~eagle/>

Re: kill-to-signature (was GNKSA and Gnus)

[Lars Magne Ingebrigtsen]

Russ Allbery <rra@stanford.edu> writes:

> Well, the original choice of C-c z (in the space dedicated to user
> bindings, of course) was for the mnemonic similarity to ESC z.  In my
> version of Gnus, C-c C-z still hasn't been taken, but I'm still using a
> fairly old version.

Ok; I've put it on `C-c C-z'.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen