My munging gives a repliable address and states the terms. Mail
without a leading "Re:" in the subject is segrgated and I may
opt for bounce if I decide it is working well. My munging is also
only done for news and mailing lists.
>from one of you who did _not_ do that. My mailer, elm, does not put the
>headers in the editor and did not call any special attention to the
>headers. When I receive a message by _mail_, I expect to be able to
>reply to that mail without a single thought toward whether the address
I have gotten mail like that. Very annoying. Apparently the sender
(in my case) was just relying on Netscape's put-your-email-address-
here preference. Since Netscape does not use different preferences
for news and mail....
>And now we return to the situation where people want to merge mail and
>news into a single function. Now you won't be able to tell whether
>you're sending bogus headers by mail or by news, and we're told this is
>a good thing?
Nor is it a good thing when I notice an increasing frequency of
"unsubscribe" messages in Usenet.
>Perhaps someone should unleash a sanitybot which supersedes all messages
>with syntactically invalid From: headers with their equivalent valid
>headers.
Too hard to do reliably. remove.com is a real hostname, for instance.
I submitted a Q&A to Brad Templeton for Dear Emily Postnews about
the problem, his opinion is that it is not wide spread enough to be
included yet.
Elijah
------
gets really peeved at the ones who totally remove their addresses
>>I also am against net abuse. In particular the net abuse practised by the
>>spammers. They use fake addresses to you know. They dont give their real
>>one either like I do. Yet you seem to be directing your anger more towards
>>me than to the spammers. I am telling you in my post what my real address
>I agree. Get the car thieves, not the people who install car alarms.
Changing your address is not installing car alarms. It is sticking
extra numbers on your license plate or in some cases (like you Terry)
driving without license plates, putting the plates on the inside of the
car.
I pay attention to the headers. But the fact that people respond to
trolls that send their messages crossposted into several unrelated
groups shows that many people do not. I've occasionally made that same
mistake. It is possible I could miss someone's invalid address and the
bounce messages will affect me, not the damned forger who will screw up
a working system because he can't be bothered to send REMOVE messages
back to spammers. (I've only ever gotten 3 spams that did not have that
option, and I still got them to stop sending me spam. The only real
problems I have now is junk mail that is actually approved by the
chancellor of the university for e-mailing to 50,000+ users, consuming
gigabytes of mail spool space.)
Of you people who use munged or forged headers, how many of you remember
to strip the extraneous information yourself when you send an e-mail
reply from your newsreader, or Cc: someone in e-mail? I got a message
from one of you who did _not_ do that. My mailer, elm, does not put the
headers in the editor and did not call any special attention to the
headers. When I receive a message by _mail_, I expect to be able to
reply to that mail without a single thought toward whether the address
of the sender is valid (unless it is UCE, and UCE is easily
recognizable).
And now we return to the situation where people want to merge mail and
news into a single function. Now you won't be able to tell whether
you're sending bogus headers by mail or by news, and we're told this is
a good thing?
Perhaps someone should unleash a sanitybot which supersedes all messages
with syntactically invalid From: headers with their equivalent valid
headers.
>This whole thing reminds me of the caller ID debate. Some people
>were against blocking, as though everyone you call has a right to know
>your phone number. Others argue that the right to privacy is more
>important.
I particularly enjoyed the people who said they would use blocking on
all of their outgoing calls and not accept any calls that were blocked.
Still, the Caller-ID comparison is not entirely appropriate. You're not
blocking your number from being shown, you're adding numbers to it and
expecting people to decode it to find out what the real number is.
If you have such a problem with spammers that you can't just put in a
mail filter or tell them to permanently remove you from their list, then
hire a damn secretary to read your mail first and delete the spam for
you before you see it. If you have a problem, it is _your_
responsibility to deal with it; don't foist another problem on others
for your own selfish, irresponsible benefit.
>I say, make people pay a penny per email. If someone sends, say, 10
>messages a day, no big deal. A spammer sending 10,000 will have more
>of a disincentive.
I don't think a commercial spammer would blink twice at a bill of $100 a
day. A mailing list maintainer though would rather shut down the list.
Just contact your local government offices (whatever government (or
definition of "local") that may be) and tell them to make junk e-mail an
illegal business practice. That is the only thing that will stop them.
--
_-<#)-=# http://cse.unl.edu/~gberigan/War-of-the-Worlds.html
___/___
_-~_--<###) "Imagine having one of them lovely things with its Heat
<~c>' __--< Ray wide and free! We'd turn it on Martians, we'd turn
\_--=____#) it on men! We'd bring everybody down to their knees!"
Greg Berigan <gber...@cse.unl.edu> wrote:
[snip for space]
+ I pay attention to the headers. But the fact that people respond to
+ trolls that send their messages crossposted into several unrelated
+ groups shows that many people do not. I've occasionally made that same
+ mistake. It is possible I could miss someone's invalid address and the
+ bounce messages will affect me, not the damned forger who will screw up
+ a working system because he can't be bothered to send REMOVE messages
+ back to spammers.
1) The burned hand teaches best: now that you've (in theory) made that
mistake of not seeing the headers, you'll know to look. /One/ bounced
reply will not kill your mail server.
2) We have proof that REMOVE requests are being used to verify that a
real person is at the other end; your address is then placed on /all/ of
the lists. This is being done mainly by the worst of the spammers in
the first place; sending a REMOVE only makes things worse for yourself.
IMHO, contacting the spammers is a lost cause; if you have the time and
determination to respond, contacting the site admins is the only way to
go.
[snip]
+ Of you people who use munged or forged headers, how many of you remember
+ to strip the extraneous information yourself when you send an e-mail
+ reply from your newsreader, or Cc: someone in e-mail?
I don't have to worry about either case, and I do both frequently.
Anything approaching a real newsreader doesn't handle the actual sending
of mail (replies /or/ cc's), which means that the generation of From:
lines on the mail message is handled by the mailing software -- which
hasn't been munged. It's the all-in-one newsreaders that do too much
causing your problem there; programs at least on Unix will usually
follow the modular it's-already-been-written-why-duplicate-work-left-hand-
doesn't-know-what-the-right-hand-is-doing-and-doesn't-care approach.
+ And now we return to the situation where people want to merge mail and
+ news into a single function. Now you won't be able to tell whether
+ you're sending bogus headers by mail or by news, and we're told this is
+ a good thing?
I junked the "mail and news and why they're the same thing" thread as
soon as it appeared; IMHO whoever wrote that is either an idiot, is
insane, or both. Survival of the fittest will ensure that morons like
that won't last long enough on the net for their ideas to take root. Or
something like that. :-)
+ Perhaps someone should unleash a sanitybot which supersedes all messages
+ with syntactically invalid From: headers with their equivalent valid
+ headers.
You're going to rewrite my mail in transit? I think not. That's right
up there with putting up posters in the town square in big bright
letters, proclaiming the real identity of that Federal Witness
Protection recipient down the road. You have no right to affect my
email unless -- and until -- it's delivered to you.
[That was for the benefit of the audience, I know you're too intelligent
to mean that seriously.]
[someone made a caller-id analogy here]
+ Still, the Caller-ID comparison is not entirely appropriate. You're not
+ blocking your number from being shown, you're adding numbers to it and
+ expecting people to decode it to find out what the real number is.
We're not adding numbers. We're adding letters and punctuation and big
flashing icons to our 'phone numbers. The munging is obvious to anyone
who cares to look. ("Uh, yeah, I /am/ posting from the top-level .toast
domain....")
+ If you have such a problem with spammers that you can't just put in a
+ mail filter
And the filter is triggered on /what/? From: lines of *.com? There
went all the kind people at dec.com. Maybe Subject: lines containing a
dollar sign? The VMS administrators will wonder why mail from them
rarely gets answered or read.... The spammers don't add an "X-Spam: yep,
this is some" header, you know.
+ or tell them to permanently remove you from their list,
Tried that. Doesn't work. They have no reason to do so, and now even
more of a reason to /not/ remove me: they know I'm reading my mail.
+ then
+ hire a damn secretary to read your mail first and delete the spam for
+ you before you see it. If you have a problem, it is _your_
+ responsibility to deal with it; don't foist another problem on others
+ for your own selfish, irresponsible benefit.
You'll first have to convince me that I /want/ to hear from people who
either can't change the To: line in a reply, or are too stupid to
notice, or are too lazy to find the backspace key. If you don't care
enough to delete a few characters (fifteen seconds work?), then you must
not really care whether I get your message or not.
"Selfish"? Possibly. Self-defense is more like it. "Irresponsible?"
If it was irresponsible, we wouldn't be writing THIS ADDRESS IS SLIGHTLY
MUNGED, PLEASE FOLLOW THE FOLLOWING INSTRUCTIONS in bright lights, now
would we? No, we'd just toss out our posts and not give a rat's ass
whether you knew it was munged or not.
+ Just contact your local government offices (whatever government (or
+ definition of "local") that may be) and tell them to make junk e-mail an
+ illegal business practice. That is the only thing that will stop them.
*choke* The reason I munge my address is because I don't have /time/ to
deal with the spam! You think I have the free time to deal with government
bureaucracy??!?
[And before you try, "oh, but you have time to read usenet?" -- I read
and post late in the evening, when my work is done. The government
shuts down at five o'clock, when I'm still working. And I obviously
can't just say, "don't send any spam during the day when I'm depending
on readable, working email to do my job; only send it at night when I
have time to deal with it instead of news."]
Luck++;
Phil
--
http://www.cs.wright.edu/people/students/pedwards/ The gods do not
Address is deliberately munged. You must edit protect fools. Fools are
the headers before replying -- although it can protected by more capable
be entertaining not to, if you're really bored. fools. -Larry Niven
I've only ever gotten about 3 spams the DID have that option. The
first two, I tried it; I still receive spam from them.
: ....................................................... The only real
: problems I have now is junk mail that is actually approved by the
: chancellor of the university for e-mailing to 50,000+ users, consuming
: gigabytes of mail spool space.)
That's the main problem I have with the practice. There is only a
limited amount of disk space to store incoming email. When that
space is filled, the mail server rejects any new mail until more
space is available. As long as all the users on the system I'm on
check their mail regularly and delete the spam, it doesn't affect
things much; but, if the mail disk is filled, every piece of spam
to any user on the system is contributing to the loss of any
desired new mail I might want to receive. I have no control over
how soon those other users will check their email and delete the
spam they received.
: Of you people who use munged or forged headers, how many of you remember
: to strip the extraneous information yourself when you send an e-mail
: reply from your newsreader, or Cc: someone in e-mail? I got a message
: from one of you who did _not_ do that. My mailer, elm, does not put the
: headers in the editor and did not call any special attention to the
: headers.
My mailer, elm, DOES put the headers in the editor. Try editing your
elmrc file. The option is available.
: ......... When I receive a message by _mail_, I expect to be able to
: reply to that mail without a single thought toward whether the address
: of the sender is valid (unless it is UCE, and UCE is easily
: recognizable).
Too bad; as long as we're getting spammed with UCE, especially when
they are culling our addresses from our posts, people are going to
munge their address in an attempt to prevent that. If you want the
address munging stopped, stop the spammers.
[snip]
: ttr...@world.std.com (Terry Traub) wrote:
: >This whole thing reminds me of the caller ID debate. Some people
: >were against blocking, as though everyone you call has a right to know
: >your phone number. Others argue that the right to privacy is more
: >important.
I'm not sure which is more important, but I don't think I should be
the one to pay to get my phone number changed (and even more to keep
it unlisted) just because the phone company can't block the calls of
the guy harrassing me and my wife because his phone company blocks
his caller ID info.
: Still, the Caller-ID comparison is not entirely appropriate. You're not
: blocking your number from being shown, you're adding numbers to it and
: expecting people to decode it to find out what the real number is.
Do you find it easier to reply to someone when you cannot get their
number, or when you have to decode it?
: If you have such a problem with spammers that you can't just put in a
: mail filter or tell them to permanently remove you from their list, then
: hire a damn secretary to read your mail first and delete the spam for
: you before you see it. If you have a problem, it is _your_
: responsibility to deal with it; don't foist another problem on others
: for your own selfish, irresponsible benefit.
1. Telling them to remove you from their list DOES NOT WORK!
2. A mail filter only works once they have sent you UCE already.
3. While it might be my problem, I do not have the ability to
deal with it. I cannot go around deleting UCE spam sent to
each user on the system I am on; since I cannot do so, the
spam wastes a limited resource.
: ttr...@world.std.com (Terry Traub) wrote:
: >I say, make people pay a penny per email. If someone sends, say, 10
: >messages a day, no big deal. A spammer sending 10,000 will have more
: >of a disincentive.
: I don't think a commercial spammer would blink twice at a bill of $100 a
: day. A mailing list maintainer though would rather shut down the list.
I'm sure a mailing list mainter can get an exception made once they
prove it is a valid mailing list (i.e. requires registration *before*
it sends someone any mail).
: Just contact your local government offices (whatever government (or
: definition of "local") that may be) and tell them to make junk e-mail an
: illegal business practice. That is the only thing that will stop them.
Local government can't do much when it comes from outside the state.
Federal government has to deal with it then.
--/Edward Burr/-------------------+----------------------------------
| Senior, Science Education, OU | Use of my e-mail address or URL |
| egb...@wildstar.net | is restricted. Visit my home |
| http://www.wildstar.net/~egburr | page for details. |
\ / \ /
\ Unless otherwise noted, all opinions here are solely my own./
The "From" line contains the electronic mailing address
of the person who sent the message, in the full Internet
syntax.
This is a required header, and people that intentionally
munge it are violating RFC 1036.
Ben Polk
Netscape News Server Engineering
> From RFC 1036:
>
> The "From" line contains the electronic mailing address
> of the person who sent the message, in the full Internet
> syntax.
>
> This is a required header, and people that intentionally
> munge it are violating RFC 1036.
I'll have to go look at that again. I thought it only applied to e-mail,
and not to the From: field on messages passed via UseNet.
Of course, the spammers who are intentionally munging their From: header
in order to get around the various filters are also violating the same
RFC.
--
PGP Public Key available at http://www.neta.com/~caradoc
Use finger for additional verification
Some info about forged spams also on the WWW page
<HTML><H1><BLINK>Don't read news with Netscape!</BLINK></H1></HTML>
"In the wiring closet, no one can hear you scream!"
> From RFC 1036:
>
> The "From" line contains the electronic mailing address
> of the person who sent the message, in the full Internet
> syntax.
>
> This is a required header, and people that intentionally
> munge it are violating RFC 1036.
Yup.
And sendmail violates RFC822 by rewriting addresses.
Your point is?
> Ben Polk
> Netscape News Server Engineering
Incidentally, is the current Netscape news server still using old buggy INN
code?
> The "From" line contains the electronic mailing address
> of the person who sent the message, in the full Internet
> syntax.
>This is a required header, and people that intentionally
>munge it are violating RFC 1036.
Technically yes. My own anti-junk-mail From: header *is* valid. But
still, I (and perhaps others) will take you more seriously when Netscape
software stops misusing the Newsgroup: header in email and also stops
gratuituously leaving html droppings in random places.
--
Rahul Dhesi <dh...@spams.r.us.com>
a2i communications, a quality ISP with sophisticated anti-junkmail features
*** Now featuring Strategy C for junk-mail-proof News postings ***
|| "please ignore Dhesi" -- Mark Crispin <m...@CAC.Washington.EDU> ||
> > This is a required header, and people that intentionally
> > munge it are violating RFC 1036.
> I'll have to go look at that again. I thought it only applied to e-mail,
> and not to the From: field on messages passed via UseNet.
RFC 1036 describes the article format for USENET messages.
> Of course, the spammers who are intentionally munging their From: header
> in order to get around the various filters are also violating the same
> RFC.
Well, yah, but spammers are evil net abusing dogs. What's these other
people's excuse?
>> Perhaps someone should unleash a sanitybot which supersedes all messages
>> with syntactically invalid From: headers with their equivalent valid
>> headers.
> You're going to rewrite my mail in transit? I think not.
When did the term "supersedes" start applying to mail? Obviously I'm
talking about news.
>We're adding letters and punctuation and big
>flashing icons to our 'phone numbers.
And when someone uses some old software that capitalizes one or more
elements in the address (some put the .edu domain as .EDU), people won't
get confused?
>> If you have such a problem with spammers that you can't just put in a
>> mail filter
> And the filter is triggered on /what/? From: lines of *.com?
In mail, you do have the message body available. A useful criteria for
refusal is the presence of the word "REMOVE" in the message.
Now if the people munging their addresses get blocked for using the same
word, why should I care? If I did, I could code up a special patched
mail reader that would display the 2 lines before and after that word
and I can determine if it is spam or a munger or normal use of the word
"remove".
> You think I have the free time to deal with government
> bureaucracy??!?
That's what representatives are for. If you aren't willing to push to
get the law changed, I can only assume you don't really care. You just
want an easy out, and you don't care what else you screw up as long as
it isn't your problem anymore. That is net abuse.
>>Of you people who use munged or forged headers, how many of you remember
>>to strip the extraneous information yourself when you send an e-mail
>>reply from your newsreader, or Cc: someone in e-mail? I got a message
>My munging gives a repliable address and states the terms.
I have no problem when the used address is reachable.
>Mail
>without a leading "Re:" in the subject is segrgated and I may
>opt for bounce if I decide it is working well.
Hmm, I guess I shouldn't have changed the subject (using "(was: ...)") a
couple days ago.
>>Perhaps someone should unleash a sanitybot which supersedes all messages
>>with syntactically invalid From: headers with their equivalent valid
>>headers.
>Too hard to do reliably. remove.com is a real hostname, for instance.
A configurable special case.
>I submitted a Q&A to Brad Templeton for Dear Emily Postnews about
>the problem, his opinion is that it is not wide spread enough to be
>included yet.
A strange opinion. How widespread must it be?
Simply put, the "excuse" is the fact that spammers are evil net
abusing dogs. I haven't taken the step of using an altered From:
header yet, but I can certainly understand those who do. I can't
imagine that any of them really care that they are violating an RFC,
nor do I. There is at least a reasonable argument that the harm is
far worse than the cure, particularly for those who insert dummy
characters in the domain field but leave their reply address otherwise
intact.
Mike - Griffin, GA
Mike....@pobox.com
Does your news server still violate RFC 977? (eg., lines
with 2 periods at the beginning terminating a post.)
--
|======================================================================|
| Seth Kroger "If God made us in His image we |
| skr...@slonet.org have certainly returned the |
| http://www.slonet.org/~skroger compliment." -Voltaire
>: I don't think a commercial spammer would blink twice at a bill of $100 a
>: day. A mailing list maintainer though would rather shut down the list.
>I'm sure a mailing list mainter can get an exception made once they
>prove it is a valid mailing list (i.e. requires registration *before*
>it sends someone any mail).
This is silly. Who would collect the money? What authority would
they have to make anyone pay?
--
Richard Braakman
It's certainly not silly. The way I think it will work is that my
(augmented) smtp server won't accept incoming mail (except perhaps for
that signed by authorized senders and/or bearing an authorization signed
by me) without collecting the required fee. The "authority" is that if
you don't pay the fee then your mail won't be accepted. Of course, this
won't happen until the net has a workable system for efficiently
making such small payments.
David desJardins
--
Copyright 1997 David desJardins. Unlimited permission is granted to quote
from this posting for non-commercial use as long as attribution is given.
Please do share.
--
to reply via email, remove "oink-" from my return address
> Does your news server still violate RFC 977? (eg., lines
> with 2 periods at the beginning terminating a post.)
No. That was a bug that was fixed a year or so ago. If
you are aware of any way in which the News Server is not
conforming to standards let me know and I'll try to fix
them.
>> screw up
>> a working system because he can't be bothered to send REMOVE messages
>> back to spammers. (I've only ever gotten 3 spams that did not have that
>> option, and I still got them to stop sending me spam.
> I've only ever gotten about 3 spams the DID have that option. The
> first two, I tried it; I still receive spam from them.
I've had 100% success. Perhaps you're doing something wrong. Maybe the
spam is being forwarded through another account to you, and the removal
has the wrong address as compared to their database.
> My mailer, elm, DOES put the headers in the editor. Try editing your
> elmrc file. The option is available.
Thank you. I shall investigate whether this option is available on the
installed version here.
Still, if people are going to post with invalid addresses, they can at
least ensure that their software -- when sending mail messages --
provides the correct address, or when using software that acts as both
newsreader and mailer, that they make that change on a per message
basis. Take some responsibility for your actions people.
There is no reason to use a faked address in a person-to-person e-mail
communication.
> I don't think I should be
> the one to pay to get my phone number changed (and even more to keep
> it unlisted) just because the phone company can't block the calls of
> the guy harrassing me and my wife because his phone company blocks
> his caller ID info.
Have you tried *57 (last-call trace)? Works even on blocked calls.
>> If you have such a problem with spammers that you can't just put in a
>> mail filter or tell them to permanently remove you from their list, then
>> hire a damn secretary to read your mail first and delete the spam for
>> you before you see it. If you have a problem, it is _your_
>> responsibility to deal with it; don't foist another problem on others
>> for your own selfish, irresponsible benefit.
> 1. Telling them to remove you from their list DOES NOT WORK!
Maybe the address from which the removal requests come doesn't match the
address to which they are sending the UCE. Maybe your outgoing mail has
munged addresses.
> 2. A mail filter only works once they have sent you UCE already.
The spammers aren't the only ones that can maintain lists of e-mail
addresses. Get one of the blacklists of known spammers and spammer
sites.
> 3. While it might be my problem, I do not have the ability to
> deal with it. I cannot go around deleting UCE spam sent to
> each user on the system I am on; since I cannot do so, the
> spam wastes a limited resource.
If you're running the system, consider setting up a system-wide filter.
(To be safe, also add a pyramid scheme detector on outgoing news and
mail.)
>> I don't think a commercial spammer would blink twice at a bill of $100 a
>> day. A mailing list maintainer though would rather shut down the list.
> I'm sure a mailing list mainter can get an exception made once they
> prove it is a valid mailing list (i.e. requires registration *before*
> it sends someone any mail).
Spammer starts a mailing lists, applies for exception, then spams from
the same address.
I'd rather e-mail remain flat-rate for unlimited use and just outlaw the
spammers outright.
>> Just contact your local government offices (whatever government (or
>> definition of "local") that may be) and tell them to make junk e-mail an
>> illegal business practice. That is the only thing that will stop them.
> Local government can't do much when it comes from outside the state.
> Federal government has to deal with it then.
Which is why I made the definition of "local" floating.
Meanwhile, there are now efforts to expand the federal junk fax law to
include junk e-mail. For the international UCE, how about making junk
e-mail an extremely high tariff per item import? One state is also
putting forth a law that would apply within its borders. Also, other
state laws have been applied to people who do not live in that state
when interstate communication is used.
I happen to be able to follow your advice, but not all people have that
option, i.e. they may not be able to get/install aliases or/and not able
to get/install screening filters.
Personally I find human-fixable munging acceptable, provided that the
mungers accept the consequences of their munging, i.e. also the
consequences for themselves. Since many mungers don't realize the
consequences, it is unclear if they would accept them.
Greg Berigan <gber...@cse.unl.edu> wrote:
+ In news.software.readers, pedw...@cs.wright.edu (Phil Edwards) wrote:
+ >Greg Berigan <gber...@cse.unl.edu> wrote:
+
+ >> Perhaps someone should unleash a sanitybot which supersedes all messages
+ >> with syntactically invalid From: headers with their equivalent valid
+ >> headers.
+
+ > You're going to rewrite my mail in transit? I think not.
+
+ When did the term "supersedes" start applying to mail? Obviously I'm
+ talking about news.
Sorry, sorry.... s/mail/news/ in my last sentence. And while I'm
allowed to supersede my own posts, you certainly aren't. And vice
versa. The obvious exception is if I dump nine thousand copies of MMF
on you, in which case I really hope that someone will hose my spam
before it goes anywhere.
+ >We're adding letters and punctuation and big
+ >flashing icons to our 'phone numbers.
+
+ And when someone uses some old software that capitalizes one or more
+ elements in the address (some put the .edu domain as .EDU), people won't
+ get confused?
*gape* Dear Lord... this means people will have to learn to READ!
I have litt- no, make that zero sympathy for people who don't bother to
read headers, even just the basic ones.
+ >> If you have such a problem with spammers that you can't just put in a
+ >> mail filter
+
+ > And the filter is triggered on /what/? From: lines of *.com?
+
+ In mail, you do have the message body available. A useful criteria for
+ refusal is the presence of the word "REMOVE" in the message.
Agreed. Unfortunately, that's the same criteria Spamford and Co. use to
/confirm/ your address as a valid target.
+ Now if the people munging their addresses get blocked for using the same
+ word, why should I care? If I did, I could code up a special patched
+ mail reader that would display the 2 lines before and after that word
+ and I can determine if it is spam or a munger or normal use of the word
+ "remove".
I would advise that you /don't/ care, that you should just block both
the spammers and us mungers from your newsreader. That way, you won't
have to see the spam, and you won't have to put up with us evil mungers
who are so obviously responsible for single-handedly destroying the
otherwise-perfect Usenet.
+ > You think I have the free time to deal with government
+ > bureaucracy??!?
+
+ That's what representatives are for.
I'll ask my question again... you think I have time to deal with that
kind of bureaucracy? The reps waste just as much of my time.
+ If you aren't willing to push to
+ get the law changed, I can only assume you don't really care.
Is this an "I volunteer" that I'm hearing? Fine; /you/ be our
representative. If you chase after the problem as heatedly as you chase
after us, nobody will have room to complain.
+ You just
+ want an easy out, and you don't care what else you screw up as long as
+ it isn't your problem anymore. That is net abuse.
You accuse a Unix person of wanting an easy out? Funny, we were always
accused of looking for the most difficult solution before.
Sorry, Greg, but you're flat wrong here. I am willing to put in the
hours -- I /have/ put in the hours, maybe you just /might/ be looking in
the wrong place, eh? -- and I would welcome a working solution even if
it was difficult. But there isn't a solution that I, as a full-time
student, part-time worker, and off-time human, with no money, no
reputation, and no free time, can effect.
For the last time: I am not praising nor advocating munging. I'm
saying that it works. Unlike any other stopgap measures available, it
WORKS. Once you, with all of your free time and ferocious words, have
solved the problem, I will honestly be happy to go back to posting in
the clear.
In the meantime, there's little point to continuing. I'm removing
news.*.{!policy} from my .newsrc, since it is just as devoid of
information as the last time I looked. (Somewhere, I swear, there is a
mailing list consisting of Chris, Seth, Tim, and about five other
people, where work is actually being done, since it obviously isn't
here. Forget any Cabal[TINC] theories, I'm just talking about daily
struggling to stay afloat.)
Which option is that? Is it in the ELM guides?
Sven
: Which option is that? Is it in the ELM guides?
: Sven
never read the elm guides, but...
somewhere in your home directory, probably in a dir called '.elm',
you should find a filed named 'elmrc'. Edit the file and look for
lines similar to:
# when messages are copied into the outbound buffer, don't include headers?
noheader = ON
and set the value to OFF.
If you can't find the elmrc file, it might not exist yet (if you
never made any config changes to elm, it probably doesn't exist). To
get it created, start elm, choose 'o' for options, then '>' for save.
This will save the current settings to the elmrc file, which you can
now edit as described above.
hope this helps,
Edward Burr
--in some areas, not all. This varies by phone company, and I believe by
political jurisdiction as well.
>> 1. Telling them to remove you from their list DOES NOT WORK!
>
>Maybe the address from which the removal requests come doesn't match the
>address to which they are sending the UCE. Maybe your outgoing mail has
>munged addresses.
Or maybe the spammer ignores all remove requests, and just offers the
option as a cynical attempt to confuse the issue.
>I'd rather e-mail remain flat-rate for unlimited use and just outlaw the
>spammers outright.
Here we agree.
It seems to me that junk email is wrong for the same reason that junk
faxes are wrong: they force the recipient to pay for unsolicited and
unwanted mail. Junk faxes are currently illegal. Though the law is
carelessly enough draw that an argument _could_ be made that it applies
to junk email, it would be far better to have explicit legislation that
covers the latter.
--
Stan Brown, Oak Road Systems, Cleveland, Ohio USA
email: stb...@nacs.net Web: http://www.nacs.net/~stbrown/
USD 500.00 charge for proofreading unsolicited commercial emails.
Would then be a simple matter to post spam from off-shore.
--
-----------------------------------------------------------------
Ghengus Khan kh...@phoenix.net :Opinions and Commentery are free
:That is what I charge and that is
http://www.phoenix.net/~khan :what they are worth.
-----------------------------------------------------------------
: Would then be a simple matter to post spam from off-shore.
True, but at least it would get rid of the US junk mail, which is the
vast majority of what I receive.
They'll soon learn; it's not a problem.
>It doesn't matter whether newsreader software lets you edit the address
>if you don't notice that you need to. And the whole idea of having the R
>key (or other simple "reply command) is to make it unnecessary to deal
>with the address of the person who posted.
Access to that R button has for too many become an autonomic reflex, with
the arm reaching for it even when there is no message coming from the brain.
So, where the buttonpush is not restrained by reason or responsibility, it
can usefully be restrained by effort.
>So what happens? Someone sends you a reply. It bounces. Now what do they
>do?
If the message was just a hasty one-line knee-jerk response tacked onto
a quoted copy of my entire posting then the respondant will hopefully
have come to the realization (unless he is totally clueless) that his
hasty unconsidered riposte was really of little worth, so he now won't
bother making that further tiny extra effort to foist it onto me. And with
any luck, the experience might make him less inclined in future to fire
off mail without giving due thought and care in forming his message. In
short, it will have helped guide him along the path to becoming a
responsible net user, so in the long run it's to the benefit of us all.
On the other hand, if he is determined to convey something of importance,
something that by its nature he is certain I will value and appreciate,
he will make that small effort to do so (editing the header and possibly
even take this opportunity to add extra polish to his message - I wish!)
and will make a mental note that in future he should pay closer attention
to the article he is replying to, and his care and consideration will be
appreciated by all who are genuine in their desire to keep the net a
useful and affordable discussion medium.
It could be, too, that upon reflection he will be suitably impressed by
the filter effect of a munged address, and might adopt the idea himself.
Then, if most posters were to follow this lead, it would start to put
some of the spammers through hoops.
>And some of them may, one day, be answering a question that you asked.
In which case they will see the message in big letters at the foot of
my posting: YOU WILL NEED TO EDIT MY ADDRESS IF REPLYING BY MAIL OR ...
Yea. So?
Do you have any idea how _few_ people actually _conform_ to RFC 822 and/or
RFC 1036?
Not many.
These seem to be standards which are honored more in the breach than in
the observance.
--
-- Ron Guilmette, Roseville, CA -------- Infinite Monkeys & Co. ------------
---- E-mail: rfg(at)monkeys(dot)com ----------------------------------------
------ Copyright (c) 1997 by Ronald F. Guilmette; All rights reserved. -----
And as long as we are taking about the sins of Netscape...
Why is it that every piece of mail which has ever been sent to me
from anyone at Netscape appears to be forged? The From: and envelope
From_ headers always say so-a...@netscape.com, but the Received
headers indicate that it never even entered any such domain! Rather,
all these mails seem to have originated at some entirely different
place... someplace called `mcom.com'.
An official company-wide mail forging policy perhaps? Or just ordinary
system admin incompetence?
You be the judge.
Insufficient information on the part of the poster. mcom.com is a domain
owned by Netscape Communications, and I believe it was in use before
netscape.com was. I imagine the "M" probably stands for Mosaic :)
Incidentally, going to home.mcom.com takes you to Netscape's homepage.
--
Steve Sobol | New Age Consulting Service, Inc./Internet Service Provider
Technical Support| Modem: 216 619-2015 (Cuyahoga Co.), 951-8266 (Lake Co.)
sjs...@nacs.net | Outstanding Customer Support: 216 524-8414/FAX 524-6699
sjs...@nstc.com | 815 Superior Avenue, #425 Cleveland OH 44114
>>Well, it's not a problem for would-be repliers *IF* they happen to notice
>>the address and its "obvious" need for editing.
>They'll soon learn; it's not a problem.
So on one hand you have the design goals of making the function
completely transparent to the user, to give him an intuitive interface
which requires the user not have to see or deal with the niceties of
the communication medium, and on the other you have people insisting
that the user be exposed to the function out of their own selfish wants.
>It could be, too, that upon reflection he will be suitably impressed by
>the filter effect of a munged address, and might adopt the idea himself.
Which is why this behavior should be stopped now. Anyone have any
ideas on how these bad messages can be detected and dropped on the
floor so they don't get distributed anywhere? If people are going to
write messages that violate the RFCs, then they should have no
expectation that their messages will go anywhere.
There's a very simple way to avoid spam: don't post. If you want to
respond to someone's message, reply by e-mail only.
(And that elmrc option to include headers is not what I had in mind.)
>There's a very simple way to avoid spam: don't post. If you want to
>respond to someone's message, reply by e-mail only.
If nobody posts, what messages will there be to reply to?
I hate to use foul language, but that's the stupidest fucking proposal
I've heard in an awfully long time.
As of recently I'm posting with a bogus address. As penance for all those
posts from days gone by when I used my real name, I've got procmail
between the outside world and my inbox, and monitoring this group helps me
keep the "killfile" current. When spam does get through, I've got perl
code to generate a spam report to various postmasters. Ultimately, I've
got notices of account termination to prove that it frequently gets the
job done.
It works reasonably well for me; if EVERYONE (not just the hard core spam
fighters that post here) would use a technique like this, we'd see the
spam problem reduce quickly. Unfortunately, too many people are content
to just through the 'remove' hoops and/or delete the incoming mail without
really doing anything to stop the problem.
Today's trivia: pwrnet gets connectivity through autonet.net -
I'm not sure if autonet.net just has no TOS or they simply refuse to do
any enforcement, but they are kind enough to provide a phone number for
pwrnet's spam victim's to voice their displeasure: (800) 829-2206, as
evidenced at www.autonet.net
Have at it, folks.
--
-- My address has been forged to deflect unsolicited email from people who
collect lists of victims from usenet archives. Don't bother replying...
>>There's a very simple way to avoid spam: don't post. If you want to
>>respond to someone's message, reply by e-mail only.
>If nobody posts, what messages will there be to reply to?
Messages from people who are either not bothered by spam or have better
ways to avoid spam than posting with false addresses.
It is clear that there are two disparate views of spam, the view of any
particular netuser being determined simply by his ISP's charges. Those
who pay a flat monthly charge for email fail to see what all the fuss is.
Surely, you say, it is easy enough to simply delete unwanted mail? What's
the big deal about a few unwanted intrusions into your mailbox each day?
But those of us who are charged for email traffic are finding it more and
more galling to be confronted by the conceited attitude that spammers
should continue untrammelled draining our accounts at a rate in my case
of x10 to x20 times the rate it would be dropping by normal personal mail.
(I receive about 2-3 spam mailings per day; the record for a single spam
being about 480 lines, IIRC.)
That most email spam originates from the US, where we all know that net
access charges are comparatively low and where few ISPs charge according
to the amount of traffic, is a point not lost on those of us outside of
North America.
And I am playing the Devil's Advocate by taking the side of those who
suffer in silence, rather than cause ructions by pointing an accusatory
finger at any one country. Implementing procmail to handle spam is out of
the question when you are paying $$ for email; far preferable to bounce
it _FROM_THE_SPAMMER'S_SERVER_ by using a munged address than have the
suffering recipient pay to deal with it. (By way of example, my brother
who lives in a very remote region pays 25c for each kilobyte of email.)
In the short term, the only solution is to use a munged address. Perhaps
a compromise in the longer term will be for those outside the US to obtain
a US account and forward filtered mail from that to the real account. Far
better, too, to munge the domain rather than the username, because the
latter still reaches the recipient's server -- from where it is then
bounced because there is no known user.
And what, you may ask, do we say to someone posting from his free .edu
account and who has the audacity to sneer at those who don't get free
net access and who put a good case for needing to implement a spam filter
in their posting addresses?
Nothing. Just leave him clueless.
>So on one hand you have the [newsreader] design goals of making ...
big money ... $$$$
inroads into market share
dependency upon a particular ISP
dependency upon a particular platform
unabashed self-promotion on a world stage
a tilt at domination by one giant monopoly
compact, net-efficient newsreaders redundant, while making glitzy, bloated,
inefficient readers the norm
... &tc.
I think your bedazzled view of newsreader designers is misplaced. They
are more and more treating the internet with contempt. Witness the
profligate distain shown towards "bandwidth" and its efficient use:
Want to reply? just click here
Want to include a mime attachment? just click here
Want to duplicate your post by email? just click here
Want to duplicate your post as html? default or just click here
Want to crosspost to every group under the sun? just click here [well, almost]
Want to quote the whole article you're replying to? just click here
Want to edit the quoted article? Nah! Send it all.
Want to edit the address? Good grief! Actually type something?!
Got anything to say? doesn't matter
>>It could be, too, that upon reflection he will be suitably impressed by
>>the filter effect of a munged address, and might adopt the idea himself.
>
> Anyone have any
>ideas on how these bad messages can be detected and dropped on the
>floor so they don't get distributed anywhere?
Yup. You'll nab every one of 'em with killfile From: *
>If people are going to
>write messages that violate the RFCs, then they should have no
>expectation that their messages will go anywhere.
You're off on an Ego Trip? See and let us know when you get back, now.
Pah. I can beat that. Every user at my site, and in fact, almost
everyone I know with a ".uk" E-mail address got spammed with a crock
which had 4,000 odd addresses in the CC line alone...
That was a UK chap so I could afford to spend a fortnight getting him
investigated by everyone I could think of ( Office of
Telecommunications, Dept of Trade & Industry, police, Office of Fair
Trading, his suppliers, trading standards... )
I currently estimate that 90% of the e-mail though my site is spam and
I know of several sites which get 95% spam contents.
>That most email spam originates from the US, where we all know that net
>access charges are comparatively low and where few ISPs charge according
>to the amount of traffic, is a point not lost on those of us outside of
>North America.
Oh this'll get people claiming it's all coming from Europe again. Or
that Chris Lewis is sending it all out...
The big problem is that complaining directly to the Americans is
expensive -- and mailing them produces a "Yeah, yeah" from their
autoresponders and apparently nothing else.
>In the short term, the only solution is to use a munged address. Perhaps
>a compromise in the longer term will be for those outside the US to obtain
>a US account and forward filtered mail from that to the real account. Far
>better, too, to munge the domain rather than the username, because the
>latter still reaches the recipient's server -- from where it is then
>bounced because there is no known user.
The problems with this are we'd have to actually pay in dollars and
they'd have to cope with the idea of people, actually being, OUTSIDE
the US ??!?!?
[ You should have heard the reaction from MCI when I called thenm in
the US to complain and the first question was "can I have your phone
number with dialling code" and I started with "plus forty-four".. ]
I think the future is an on-going battle between semi-intelligent
auto-mailing systems and semi-intelligent mail filters. The world will
get divided into people who use whacking great filters and those who
live with the crap. Kind of like the way small lumps of the
population watch everything off video so we can fast-forward the
adverts, and everyone else puts up with 25% of their viewing time
being used to sell them washing powder.
-----------------------------------------------------------------------------
"It's not a personality.. it's a bulldozer" sill...@excession.demon.co.uk
Current project: Computer wargaming's next generation...
------------------------------------------------------------------------------
Commercial emaillers to this site should be prepared to accept our complaints
by email and phone, complaints to their ISP, notifications of possible tax
frauds, help requests to the appropriate authorities and maybe kernel images.
-----------------------------------------------------------------------------
[followups set to nana.email only]
No Spam <No....@holy.cow.net> wrote:
> It is clear that there are two disparate views of spam, the view of any
> particular netuser being determined simply by his ISP's charges.
While the discussion that follows is quite a good one, I would
take issue with this particular claim, arguing that there are
at least _three_ different views of spam.
One view is that of the person who must pay for connect time in
one way or another. For such a person, spam imposes real,
immediate costs, and is generally despised for (at least) that
reason.
Another is the view of someone who pays no metered charges for
internet access and uses email wholly or largely for entertainment,
communication with friends, and so on. Spam may be annoying for
such a person, but any costs imposed are minimal (the time it
takes to delete a message), and such a person may or may not view
spam as a serious problem.
A third group consists of those who, regardless of whether they
pay any metered charges for access, use email for something of
some importance, such as their business. For such persons, the
fact that spam does not impose any immediate monetary costs may
be less important than the other costs imposed.
Suppose you are in the middle of some project, and the bell
announces that some new mail has arrived. Is this an important
message from your client, or is it just Yuri with more exciting
news for Minoxidl users? How much time to you take out of your
project to check, and how much productive time is lost by
dropping what you are doing, just to "hit delete".
Certainly one can set up filters of various kinds, but good
filters take some work to keep up (Did you forget to add that
new client to your filters? Oops, you won't see his/her mail
until you sort through the junk tomorrow.) and still let you
get hit with the first salvo from every new spammer.
--
gregory byshenk No one I know is sensible enough
gbys...@tezcat.com to allow me to speak for them -
gbys...@prairienet.org wise though such a thing might
chicago, illinois, usa be...
I do realize that even local calls get charged by the minute in the UK, but
surely it's cheaper to complain in email to the spammer's ISP (Sprint, for
example) than it is to either telephone them (As you did) or even snailmail
them.
Getting the people who work at Sprint/MCI/etc. to understand that there are
places that aren't in the US is another whole ball of wax, unfortunately.
In any case, I'd have to say your response to Greg Berigan seems to have
completely missed the mark - you'll note that, in my address, the spam block
is almost at the very end. It's not totally at the end because I want the
spammers to TRY to mail me - but it's close to the end so that the absolute
minimum of servers will be involved in bouncing their crap back to them.
----------------------http://www.teleport.com/~atari-----------------------
ATARI COMPUTING - the new 60 PAGE printed magazine for all Atari users
NEWS - GOSSIP - FEATURES - REVIEWS - HUMOUR - HOW TO DO IT - Q&A
email for subscription details and further information
Europe: mker...@cix.compulink.co.uk USA: at...@teleport.com
--------------------ftp://ftp.teleport.com/users/atari---------------------
>I do realize that even local calls get charged by the minute in the UK, but
>surely it's cheaper to complain in email to the spammer's ISP (Sprint, for
>example) than it is to either telephone them (As you did) or even snailmail
>them.
Freephone numbers that bounce the call to the US are your FRIENDS !!
And besides which, MCI US in particular just seemes to ignore
complaints unless you make an effort. I asked them by e-mail to have
someone call me to discuss multiple problems with their internet
connectivity provision a while back and nothing so far.
>Getting the people who work at Sprint/MCI/etc. to understand that there are
>places that aren't in the US is another whole ball of wax, unfortunately.
I think half the problem is that I only ever seem to reach the
frontline droids who probably aren't bright by anyone's standards.
>In any case, I'd have to say your response to Greg Berigan seems to have
>completely missed the mark - you'll note that, in my address, the spam block
>is almost at the very end. It's not totally at the end because I want the
>spammers to TRY to mail me - but it's close to the end so that the absolute
>minimum of servers will be involved in bouncing their crap back to them.
That hadn't occurred to me -- I'll fix mine.
-----------------------------------------------------------------------------
"It's not a personality.. it's a bulldozer" | Current project: Computer
sillywiz at excession dot demon dot co dot uk | wargaming's next generation...
------------------------------------------------------------------------------
Commercial emaillers to this site should be prepared to accept our complaints
by email and phone, complaints to their ISP, notifications of possible tax
frauds, help requests to the appropriate authorities and maybe kernel images.
-----------------------------------------------------------------------------
IQ test in my email address. Humans can solve it to reply, spambots can't (yet)
-----------------------------------------------------------------------------