SPAMS are killing bionet

Wolfgang Schechinger wgschech at med.uni-tuebingen.de
Thu Jul 24 07:20:08 EST 1997


Hi all, 

I don't think that spams will kill bionet.
Bionet is too valuable to me (and hopefully to others, too)
the methods group has become an instrument in my daily lab work I do 
do not want to miss.

Of course spams annoying and I agree with the statement that they 
are theft because they take bandwith and produce extra costs for 
those having to pay for news or email services by time, volume or 
number. Thus, everything possible should be done to  minimize spam.

The growing number of spams had forced me to manipulate my email 
address thus that someone willing to reply had to remove characters 
from it to get the message delivered. I have stopped this since it 
caused too many problems to everyone. I have decided to live with 
spam since it's easier for me just to delete the unwanted messages.

Many mail and news programs have the ability to filter incoming 
messages, thus it's possible to trash all messages dealing with 
"nude", "sex" or "money fast" automatically. But it also will trash 
messages accidentally containing these words. 

Maybe it might be possible to implement such a filter in the 
newsgroups distribution machinery.
Since almost all spammers are using invalid sender/reply-to addresses
(they probably want to avoid getting mailbombed by my 32 MByte 
386spart.par virtual memory file ;-) a mail notifying the sender that 
his/her message has been killed could avoid unwanted cancellations.

Another way could be using a strategy similar to that used by the 
"deja news" posting service: You send your message to a server that 
asks you for confirmation of your message by an email dialog. This 
should only work when correct email addresses are used.

A third idea was allowing only "registered users" to post to the 
group. The email adresses of the users are known to the bionet sever 
and in normal newsgroup dialg including replies, individual codes a 
re used instead of email adresses in order to identify users. This 
could be a means to avoid bulk mailers extracting email adresses from 
messages. Of course this will cause an enourmous amount of 
adrimistration. If automated by the listserver, this at least would 
cancel out al spammings by senders using invalid email addresses.

All suggested strategies of course will cause some kind of annoyance 
and need somebody to set up the control stages and caring for their 
maintainance.

In addition, a "law" could be added to the newsgroups charter stating 
that every spammer (if ever revealed) agrees to pay e.g. 2500 bucks 
to the bionet as advertising costs by submitting his spam to the 
group. 

Maybe this all looks a bit like 1984, doesn't it? 

WS
-----
                                     
Wolfgang Schechinger         
University of Tuebingen
email: wgschech at med.uni-tuebingen.de


http://www.medizin.uni-tuebingen.de/~wgschech/research.htm




More information about the Methods mailing list