1997-01-04 - Re: Experiments on Mailing Lists

Header Data

From: Adam Back <aba@dcs.ex.ac.uk>
To: ph@netcom.com
Message Hash: eb9e516c255dd99692c9edb379a68d0dc25b22106eb1f6d2876eedce8f73d2db
Message ID: <199701031310.NAA00485@server.test.net>
Reply To: <v02140b01aef46e21565d@[192.0.2.1]>
UTC Datetime: 1997-01-04 23:37:10 UTC
Raw Date: Sat, 4 Jan 1997 15:37:10 -0800 (PST)

Raw message

From: Adam Back <aba@dcs.ex.ac.uk>
Date: Sat, 4 Jan 1997 15:37:10 -0800 (PST)
To: ph@netcom.com
Subject: Re: Experiments on Mailing Lists
In-Reply-To: <v02140b01aef46e21565d@[192.0.2.1]>
Message-ID: <199701031310.NAA00485@server.test.net>
MIME-Version: 1.0
Content-Type: text/plain



Peter Hendrickson wrote:
> At 11:02 AM 1/4/1997, Timothy C. May wrote:
> > All of these schemes--some of them pretty clever--for posting tokens,
> > reputation-based killfiles, buying and selling reputation futures, etc.,
> > are almost certainly far too complicated to deploy on a list like ours.
> 
> > ...[Interesting Extropian history deleted.]...
> 
> > So, rather than do similar tinkering with the Cypherpunks list--not that
> > either Hugh Daniel or John Gilmore have given any hint they are willing to
> > do such tinkering--I suggest those who want to try token-based posting, or
> > information markets in reputation capital, or herd-consensus killfiles,
> > etc., set up a separate mailing list and implement whatever they wish.
> 
> If the point of these schemes is to filter content (*), there is no
> reason why they have to be implemented for the entire list.  They
> can always be converted to a tag on the message which readers can
> use, but are not required to.

I think this filtering/censorship stuff is getting way too complicated
to be practical.  It is in my opinion inappropriate to moderate or
automatically filter the list itself, because that is a form of
censorship.  Expecting people to install and use new software is
asking too much.

The way I see it, you have several simple options:

1) read everything, and hit d when you decide a post is noise

2) filter at your own machine - there's lots of software.  Presumably
those requesting list server user controlled filtering are saying that
they find it too much hassle to install or learn how to use such
software, or that they don't want to pay for downloading garbage

3) start up a mailing list mirror, subscribe the mirror to
cypherpunks@toad.com, and implement user controlled filtering in
majordomo on the mirror.  This frees up some of toad.com's current
load, and puts all filtering load on the mirror.  If there's a demand
for such a service, someone can do it.

4) subscribe to a filtered version of cypherpunks, and run the risk
that the owner filters material which would be interesting to you.

5) setup a filtered version of cypherpunks which explicitly filters
just commercial spam, and mail loop errors.

Currently I use method 1.  A couple of people lately have so
consistently posted garbage that I am considering it may be worth the
effort of switching to method 2.  If someone did 5 and the delay was
reasonably short, I would probably subscribe to it.

The only thing I would be happy to see happen in the way of list based
filtering, is anything to cut out pure commercial, non crypto related
spam.  Spammers seem to have discovered mailing lists as efficient
distribution methods in addition to direct mass mailing lately.
Unfortunately this is difficult to filter automatically, and no one
has the time to do it in close to real time, and time lags hinder
discussion.

Adam
--
print pack"C*",split/\D+/,`echo "16iII*o\U@{$/=$z;[(pop,pop,unpack"H*",<>
)]}\EsMsKsN0[lN*1lK[d2%Sa2/d0<X+d*lMLa^*lN%0]dsXx++lMlN/dsM0<J]dsJxp"|dc`





Thread