From: Charles Lindsey (chl@clw.cs.man.ac.uk)
Date: Fri Nov 05 1999 - 06:12:37 CST
In <19991104122952.57985@main.templetons.com> Brad Templeton <brad@main.templetons.com> writes:
>The syntax I would propose would be simple. The body would consist of
>lines. One message-id per line. Anything after a semicolon is MIME-style
>optional parameters. I had proposed a "replaced-by" parameter as a
>step until Replaces is well implemented, but if we use version numbers it's
>not that important.
>Parameters could include:
> group=x (specify newsgroup)
> score=y (BI or other number)
> type=hide (is there any other kind?)
> reason=z (some other reason coding)
> groupdex=a,b,c (Integers saying which group from
> the newsgroups line this message was
> in.)
OK, here is a proposal, which I could write up properly if needed:
Newsgroups: alt.binaries.nasty-pictures,alt.binaries.nastier-pictures,...
From: Cosmo <wherever@cosmo.is>
Subject: spam cancel
Approved: Cosmo <wherever@cosmo.is>
Signed: if/when we have authentication
Control: bulkcancel
Content-Type: application/news-cancel
Message-ID: <123456@cosmo.is>
Distribution: ALT
<123456@spamhaus.com>; BI=23.6; type=hide
<654321@spamhaus.com>; BI=23.6; type=hide
...
But I am not sure what you gain from those parameters. They all have the
same BI (they are all the same spam). What other than 'hide' do you want?
And can someone please show me how the same example would appear in
accepted NOCEM syntax (or point me at the NOCEM RFC).
And who else is in favour of such a scheme?
>For mvgroup? The syntax does not need to be specified here. Elsewhere
>in the specification we will lay out the extensible header syntax, or simply
>point to the MIME rfcs where it is documented. Then all we need say is that
>all newly defined headers and control messages have their initial parameter,
>and everything after the first unquoted semicolon is additional parameters
>in MIME syntax that may be ignored if not understood. There is no need
>(or desire) to define a syntax for individual items.
The syntax you want is already in 4.1. But it also says (4.2.1):
The possibility of allowing header-parameters to appear in all
headers is provided mainly for the purpose of allowing future
extensions to existing headers, since only a very few USENET-
header-parameters are actually defined in this standard. Observe
that such header-parameters do not, in general, occur in headers
defined in other standards, except for the Mime standards [RFC
2045] et seq. and their extensions. Nevertheless, compliant
software MUST accept all such header-parameters in headers defined
by this standard and its extensions (ignoring them if their meaning
is unknown) and SHOULD accept (and ignore) them in all headers.
The implication there is that we have not specified any usage of
header-parameters in existing headers, notably in the Control header (for
reasons of compatibility with existing software). You will find the syntax
of the Control header in section 6.6. Anything done specially for mvgroup
has to be done within that syntax, or else that syntax will have to change
(in which case you need to show that existing implementations will not
barf).
So if you want some different syntax for mvgroup (and you are the only one
asking for it), then please make some specific proposals consistent with
the above limitations.
>Mapping corporate subnets, plus new subnets and distributions. With my
>new proposed rule that a control message should not, perhaps must not be
>executed if it contains a distribution header and the site is not an
>explicit member of the distribution, it becomes possible to use these
>control messages to map new subnets and see how they are growing. No, one
>would not likely map the whole net.
Well we have no such Distribution rule yet. Does anyone else want it (I
could write it into the Control message rules quite easily)?
I would have expected corporate subnets to know their own topology without
the use of 'sendsys' messages. The only usage I could think of would be for
finding possible sources of a Usenet2 feed (that is the only wide-ranging
subnet currently in existence).
But I do not hear others calling for retention of these messages (except a
lukewarm desire for 'version' from Russ).
-- Charles H. Lindsey ---------At Home, doing my own thing------------------------ Email: chl@clw.cs.man.ac.uk Web: http://www.cs.man.ac.uk/~chl Voice/Fax: +44 161 437 4506 Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K. PGP: 2C15F1A9 Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5