NODE 9c5faffaFIDOnet encryption (or lack thereof)
Just-in-time terraforming 30-Sep-1993 1008 <yerazunis@aidev.enet.dec.com>Thu, 30 Sep 93 07:11:54 PDT
>Mike, Please advise the sender of this message that I DO NOT allow
>encrypted mail to pass thru this system. I expect folks to abide by this
>rule voluntarily...
>
>I would hate to have to block all messages from this source becuase
>someone wishes to violate my policy :)
>
>BTW, the debate about "encrypted" mail with me is MOOT... I will not vary
>from my position... (just thought I'd let you know in case you wanted to
>try to convince me it is OK to allow encrypted mail...) please have the
>other person send encrypted mail directly to your machine...
Heh. OK. Well, if one behaves "ethically", then I guess *that* closes
the issue. It's his machine and he gets to make the rules. (this is
my personally-adhered-to point of view)
On the other hand, he doesn't seem to have protected himself against
steganographic users (though the low bandwidth of steganography compared
to obvious encryption may make the steg channel less useful). Others
may choose to take this point of view- but it's your karma.
-Bill
NODE 46da936dRe: FIDOnet encryption (or lack thereof)
Mike Godwin <mnemonic@eff.org>Thu, 30 Sep 93 08:12:00 PDT
Bill writes:
> Heh. OK. Well, if one behaves "ethically", then I guess *that* closes
> the issue. It's his machine and he gets to make the rules. (this is
> my personally-adhered-to point of view)
My question is this: how does he know that the mail is encrypted if he's
not examining the mail that passes through his system? If he *is*
examining the mail that passes through his system, it seems likely that he
is violating the Electronic Communications Privacy Act.
--Mike
NODE 947ddf2bRe: FIDOnet encryption (or lack thereof)
Al Billings <mimir@u.washington.edu>Fri, 1 Oct 93 13:13:39 PDT
On Thu, 30 Sep 1993, Mike Godwin wrote:
> Bill writes:
>
> > Heh. OK. Well, if one behaves "ethically", then I guess *that* closes
> > the issue. It's his machine and he gets to make the rules. (this is
> > my personally-adhered-to point of view)
>
> My question is this: how does he know that the mail is encrypted if he's
> not examining the mail that passes through his system? If he *is*
> examining the mail that passes through his system, it seems likely that he
> is violating the Electronic Communications Privacy Act.
Only if he has stated that he allows private mail. Most sysops have
specifically worded policy statements for their systems that say that the
sysop can read any and all messages on the system and may do so at any
time. Bulletin boards do not normally offer truely private mail because of
some of the legal implications.
NODE ca9729f1Re: FIDOnet encryption (or lack thereof)
Mike Godwin <mnemonic@eff.org>Fri, 1 Oct 93 13:43:40 PDT
Al Billings writes:
> On Thu, 30 Sep 1993, Mike Godwin wrote:
> >
> > My question is this: how does he know that the mail is encrypted if he's
> > not examining the mail that passes through his system? If he *is*
> > examining the mail that passes through his system, it seems likely that he
> > is violating the Electronic Communications Privacy Act.
>
> Only if he has stated that he allows private mail. Most sysops have
> specifically worded policy statements for their systems that say that the
> sysop can read any and all messages on the system and may do so at any
> time.
That's all very nice, but it doesn't enable a FIDO sysop to intercept
messages from people who are not users of his or her particular system.
Those people did not waive their rights to privacy under the ECPA.
> Bulletin boards do not normally offer truely private mail because of
> some of the legal implications.
This is a common myth. First of all, there are many BBSs that do
offer truly private mail, or whose sysops, as a matter of policy, do not
read others' private mail. Secondly, there's no legal liability associated
with allowing e-mail privacy. Third, federal law (the ECPA) bars
sysops from examining mail except under some very precisely defined
circumstances.
I suggest that you inform sysops who tell you otherwise that they can
contact me at the Legal Services Department of EFF. You've got my e-mail
address already--my phone number is 202-347-5400.
--Mike
NODE c4b50abaRe: FIDOnet encryption (or lack thereof)
Al Billings <mimir@u.washington.edu>Fri, 1 Oct 93 14:28:11 PDT
On Fri, 1 Oct 1993, Mike Godwin wrote:
> Al Billings writes:
>
> > Only if he has stated that he allows private mail. Most sysops have
> > specifically worded policy statements for their systems that say that the
> > sysop can read any and all messages on the system and may do so at any
> > time.
>
> That's all very nice, but it doesn't enable a FIDO sysop to intercept
> messages from people who are not users of his or her particular system.
> Those people did not waive their rights to privacy under the ECPA.
As has already been shown from Fidonet policy, Fidonet does not guarantee
private mail in any from and, in fact, advises that mail will be going
through many sites and can be read along the way.
> > Bulletin boards do not normally offer truely private mail because of
> > some of the legal implications.
>
> This is a common myth. First of all, there are many BBSs that do
> offer truly private mail, or whose sysops, as a matter of policy, do not
> read others' private mail. Secondly, there's no legal liability associated
> with allowing e-mail privacy. Third, federal law (the ECPA) bars
> sysops from examining mail except under some very precisely defined
> circumstances.
The third point does not apply if the sysops offer no private mail in the
first place.
> I suggest that you inform sysops who tell you otherwise that they can
> contact me at the Legal Services Department of EFF. You've got my e-mail
> address already--my phone number is 202-347-5400.
I don't need sysops to tell me otherwise. I've been running my own BBS
for over three years. My system has a very clear policy statement that
refers to the ECPA and states VERY clearly who can read the messages
posted on my system in different areas. As I'm not a Fido hub (and barely
participate in that network at all), I don't have to worry about passing
other mail through my system.
Wassail,
Al Billings,
Sysop of The Sacred Grove 1:343/56
NODE 0a4d10d3Re: FIDOnet encryption (or lack thereof)
Mike Godwin <mnemonic@eff.org>Fri, 1 Oct 93 14:43:39 PDT
Al Billings writes:
> As has already been shown from Fidonet policy, Fidonet does not guarantee
> private mail in any from and, in fact, advises that mail will be going
> through many sites and can be read along the way.
You could be extrapolating from Fidonet's refusal to *guarantee* e-mail
privacy (after all, how could Fidonet *enforce* it?) that all users
of every Fido BBS everywhere have waived their rights under ECPA.
My understanding is that Fidonet policy was drafted not in order to comply
with ECPA, but to acknowledge that, in this decentralized network, there
was no authority a user a could appeal to if his e-mail was not kept
private. But I'd be interested in seeing a direct quote of the policy
provision you're alluding to here.
And what about me? I don't post from a Fido BBS, so even if there's a
Fidonet-wide waiver of ECPA rights, it's not a waiver *I* have agreed to.
What if mail from me passes through a Fido node on its way to a non-Fido
destination?
> The third point does not apply if the sysops offer no private mail in the
> first place.
Certainly, if they offer no mail at all, they're not liable, since no mail
passes through their systems.
But the interesting case is this: let's assume that you're right that all
Fido users everywhere have agreed to waive their ECPA rights. Then are the
sysops who reserve their right to read e-mail reading *all* e-mail that
passes through their systems?
If not, this puts the lie to the claim that they're limiting their
liability by reserving their right to read e-mail. After all, the
criminally significant communications may be the ones they're skipping.
In general, criminal liability depends on *knowledge*--you normally can't
be held criminally liable for acts and communications you didn't know
about.
I know of no case in which a sysop has been held *civilly* liable for
failing to read all e-mail on his system. So, in terms of classic risk
analysis, what does that statistic tell you?
> > I suggest that you inform sysops who tell you otherwise that they can
> > contact me at the Legal Services Department of EFF. You've got my e-mail
> > address already--my phone number is 202-347-5400.
>
> I don't need sysops to tell me otherwise.
I wasn't referring you to any sysops.
> I've been running my own BBS
> for over three years. My system has a very clear policy statement that
> refers to the ECPA and states VERY clearly who can read the messages
> posted on my system in different areas. As I'm not a Fido hub (and barely
> participate in that network at all), I don't have to worry about passing
> other mail through my system.
If all your users have agreed to waive their e-mail privacy rights, and
you're not dealing with any mail that does not either originate or
terminate on your system, then you're not in violation of ECPA.
--Mike
NODE bfbe6d82Re: FIDOnet encryption (or lack thereof)
J. Michael Diehl <mdiehl@triton.unm.edu>Thu, 30 Sep 93 08:36:38 PDT
According to Mike Godwin:
> Bill writes:
> > Heh. OK. Well, if one behaves "ethically", then I guess *that* closes
> > the issue. It's his machine and he gets to make the rules. (this is
> > my personally-adhered-to point of view)
>
> My question is this: how does he know that the mail is encrypted if he's
> not examining the mail that passes through his system? If he *is*
> examining the mail that passes through his system, it seems likely that he
> is violating the Electronic Communications Privacy Act.
That was my first question. Then it occured to me that I have seen bbs's which
have disclaimers wrt email privacy. That is the loophole he is exploiting.
J. Michael Diehl ;^) |*The 2nd Amendment is there in case the
mdiehl@triton.unm.edu | Government forgets about the 1st! <RL>
Mike.Diehl@f29.n301.z1 |*God is a good Physicist, and an even
.fidonet.org | better Mathematician. <Me>
al945@cwns9.ins.cwru.edu|*I'm just looking for the opportunity to
(505) 299-2282 (voice) | be Politically Incorrect! <Me>
Can we impeach him yet? |*Protected by 18 USC 2511 and 18 USC 2703.
PGP Key = 7C06F1 = A6 27 E1 1D 5F B2 F2 F1 12 E7 53 2D 85 A2 10 5D
NODE b0740714Re: FIDOnet encryption (or lack thereof)
Mike Godwin <mnemonic@eff.org>Thu, 30 Sep 93 08:46:54 PDT
J. Michael Diehl writes:
> That was my first question. Then it occured to me that I have seen bbs's which
> have disclaimers wrt email privacy. That is the loophole he is exploiting.
Well, there's no doubt that users of his system can agree to allow
the sysop to read their mail. But what about people whose mail passes
*through* his system on the way to somewhere else?
He has no agreement with them.
--Mike
NODE 1c30d610Re: FIDOnet encryption (or lack thereof)
jet@netcom.com (J. Eric Townsend)Thu, 30 Sep 93 11:41:55 PDT
Mike Godwin writes:
> My question is this: how does he know that the mail is encrypted if he's
> not examining the mail that passes through his system? If he *is*
> examining the mail that passes through his system, it seems likely that he
> is violating the Electronic Communications Privacy Act.
With UNIX it's quite simple to grep for "-----BEGIN PGP
MESSAGE-----"... and ditch messages that match.
I guess one could also run the incoming mail through a spell-checker
and reject messages with greater than %99 failure rate.
Neither of these require actual examination of the message by a human,
neither reveal content of a message to a human.