// COMPLETE THREAD

Re: Experiments on Mailing Lists

7 expanded posts ยท every known parent and child

NODE 7f0db760Re: Experiments on Mailing Lists
At 01:10 PM 1/3/97 GMT, Adam Back wrote:
>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.

Some evildoer has been posting messages to Usenet purporting to be from
"cypherpunks@toad.com"; some of the messages posted have been to newsgroups
frequented by the make-spam-fast crowd, so now we've apparently been
identified as within an especially gullible market segment. The messages
sent to Usenet are labelled "SPAM BAIT" (or something like that) -
apparently the sender of the messages thinks they're doing something useful. 

>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.

For me, real time access to Cypherpunks is unproductive; I find that I get
the most value out of the list if I read my accumulated messages once every
day or two instead of once an hour.

Also, at a "macro" perspective, too much feedback can be as harmful or
inefficient as too little feedback.

I think that a "3 posts per person per day" rule might produce interesting
results; at least from my perspective, people who send many messages (> 5,
or so) per day usually don't have anything of substance to say and I
frequently skip all of their posts. It would also encourage people to avoid
the "Me, too" or "I think you're an idiot" messages which can just as
easily be sent privately or not at all. Implementing such a rule would be
disproportionately burdensome technically and politically, so I'm not
seriously suggesting that we implement it, but I do think it's useful to
think of "fewer, better" posts as a goal. 

--
Greg Broiles                | US crypto export control policy in a nutshell:
gbroiles@netbox.com         | 
http://www.io.com/~gbroiles | Export jobs, not crypto.
                            |
NODE 9b76777eRe: Experiments on Mailing Lists
At 6:38 PM -0800 1/4/97, Greg Broiles wrote:

>I think that a "3 posts per person per day" rule might produce interesting
>results; at least from my perspective, people who send many messages (> 5,
>or so) per day usually don't have anything of substance to say and I
>frequently skip all of their posts. It would also encourage people to avoid
>the "Me, too" or "I think you're an idiot" messages which can just as
>easily be sent privately or not at all. Implementing such a rule would be
>disproportionately burdensome technically and politically, so I'm not
>seriously suggesting that we implement it, but I do think it's useful to
>think of "fewer, better" posts as a goal.

Is this 3 per day _on average_, or _peak_? (I was quiet for Xmas and NYE,
so I built up a "credit" of about 21 posts, of which I still have some left
:-))

Seriously, this rule was also tried by the Extropians, with little useful
effect.

While the goal of "fewer, but better" posts is a noble one, mechanical
rules such as are implied by quotas are a bad idea. For various reasons.

Most of my posts are reasonable long ones (is this also a violation of a
policy Greg would support?), and I try to completely avoid short, "me too,"
rejoinders or the typical net.repartee that so pollutes other lists.

(If you think our list is bad, you ought to see lists which are completely
dominated by one-line witticisms and inside jokes....)

Frankly, some of us have more time and interest in posting to this list
than some others have. As I like to say, "for various reasons." Clearly all
1200+ subscribers cannot post 3 messages a day and have the list survive;
that some posters post many more than 3, some post 3 or fewer, and the vast
majority post none at all is completely unsurprising. Attempts to
mechanically limit the number of peak posts to some arbitrary limit,
without taking into account other factors, will backfire badly.

--Tim May


Just say "No" to "Big Brother Inside"
We got computers, we're tapping phone lines, I know that that ain't allowed.
---------:---------:---------:---------:---------:---------:---------:----
Timothy C. May              | Crypto Anarchy: encryption, digital money,
tcmay@got.net  408-728-0152 | anonymous networks, digital pseudonyms, zero
W.A.S.T.E.: Corralitos, CA  | knowledge, reputations, information markets,
Higher Power: 2^1398269     | black markets, collapse of governments.
"National borders aren't even speed bumps on the information superhighway."
NODE 523db469Re: Experiments on Mailing Lists
Timothy C. May wrote:

> (If you think our list is bad, you ought to see lists which are completely
> dominated by one-line witticisms and inside jokes....)

  Or 10,000 messages saying only, "Merry Christmas", followed a few days
later by another 10,000 messages saying, "Happy New Year".
  Anyone who needs 'proof' that I am speaking the truth, here, (and who
has a 10Gb hard drive), can just send me their email address, and I
would be glad to forward them.

Toto
NODE 1887a995Re: Experiments on Mailing Lists
Hi,

Thought that I would add a few statistics on cypherpunks list filtering.

Since this afternoon I received 70 messages from the mailing list.

Of those 70 28 went into my twit folder (40%).

Of those 28:

- 2 anonymous posts:
    1 was moved back to my cypherpunks folder
    1 was Tim May slander message
- Remaining 26 belonged to 4 users on my twit list (non spam)

42 of 70 made it past my filter into my cypherpunks folder (60%).

Of those 42:

- 4  were spam that got past my filter. <sigh> I only filter repeat offenders. - 12 were
on Controlling spam on the list
- 6  were on ITAR/EAR
- 3  were on SSL Spoofing
- 3  were on Tim May/Cryptoanarchy
- 3  were on Ecash
- 2  were on OCR
- 2  were on California Inet Law
- 1  was on Secret Sharing
- 6  were on misc topics


I have found that by filtering out a handfull of users plus anonymous posts the list
becomes quite manageable. As far as spam goes I don't think there is any solution other
than the delete key. A good % of the time one can tell from the message topic that it is
spam and can delete the message without opening it.

As far as the issue of paying to download unwanted messages. One a user gets a good feel
for who the want to filter out they should be able to create a
killfile on their ISP's server. 

--
-----------------------------------------------------------
William H. Geiger III  http://www.amaranth.com/~whgiii
Geiger Consulting    WebExplorer & Java Enhanced!!!
Cooking With Warp 4.0

Author of E-Secure - PGP Front End for MR/2 Ice

Look for MR/2 Tips & Rexx Scripts
Get Work Place Shell for Windows!!
PGP & MR/2 the only way for secure e-mail.
                          
Finger whgiii@amaranth.com for PGP Key and other info
-----------------------------------------------------------
NODE 899ae4d8Re: Experiments on Mailing Lists
At 9:38 pm -0500 1/4/97, Greg Broiles wrote:
>Some evildoer has been posting messages to Usenet purporting to be from
>"cypherpunks@toad.com"; some of the messages posted have been to newsgroups
>frequented by the make-spam-fast crowd, so now we've apparently been
>identified as within an especially gullible market segment.

Wow. Is it really true that all we need is a cancelbot?

It can't really be that easy...


Cheers,
Bob Hettinga

-----------------
Robert Hettinga (rah@shipwright.com), Philodox,
e$, 44 Farquhar Street, Boston, MA 02131 USA
"The cost of anything is the foregone alternative" -- Walter Johnson
The e$ Home Page: http://www.vmeng.com/rah/
FC97: Anguilla, anyone? http://offshore.com.ai/fc97/
NODE 39e4e51eRe: Experiments on Mailing Lists
Robert Hettinga wrote:
> 
> At 9:38 pm -0500 1/4/97, Greg Broiles wrote:
> >Some evildoer has been posting messages to Usenet purporting to be from
> >"cypherpunks@toad.com"; some of the messages posted have been to newsgroups
> >frequented by the make-spam-fast crowd, so now we've apparently been
> >identified as within an especially gullible market segment.
> 
> Wow. Is it really true that all we need is a cancelbot?
> 
> It can't really be that easy...

... And we have the cancelbot. Very easy to run, can be set up
to robocancel everything from cypherpunks@toad.com, on the spot.

	- Igor.
NODE 658fac3aRe: Experiments on Mailing Lists
Robert Hettinga <rah@shipwright.com> writes:

> At 9:38 pm -0500 1/4/97, Greg Broiles wrote:
> >Some evildoer has been posting messages to Usenet purporting to be from
> >"cypherpunks@toad.com"; some of the messages posted have been to newsgroups
> >frequented by the make-spam-fast crowd, so now we've apparently been
> >identified as within an especially gullible market segment.
>
> Wow. Is it really true that all we need is a cancelbot?

Use mine. It's in Phrack.

---

Dr.Dimitri Vulis KOTM
Brighton Beach Boardwalk BBS, Forest Hills, N.Y.: +1-718-261-2013, 14.4Kbps