From: Kee Hinckley <nazgul(_at_)somewhere(_dot_)com>
At 3:22 AM -0600 4/10/03, John Fenley wrote:
Administrator maintained public "choicelist"(functions as a whitelist, and
a blacklist) database:
Who administers this? And who can read it?
Each entry in the database is maintained by the creator of that entry.
anyone can read it because it is a public database.
Database curator:
Again. What's the scope of this job? Per-ISP?
2. protect the security of the database against unauthorized changes.
How are changes authorized?
By password, then by password reset, then by an actual person contacting the
creator using the contact info they give.
4. give id# of the list's entry to subscribers so they can recieve the
list, or use a system where they send the first message to subscribe to
the list.
These ID #'s are unique world-wide? Who guarantees that?
Yes they are unique. there is one central database that that generates a
unique number when an entry is created.
1. enter # given when you sign up for a list if your mail service supports
it, or send a subscribe message to the list.
Enter # (or list?) where?
into your Mail User Agent so that your MUA can contact the database to
maintain your whitelist.
1. allow user to enter id #s.
Into what?
into the whitelist manager
When a new user joins the system:
1. They choose a user name, and a password.
2. they enter contact info, to prvent hijacking.
How does that prevent hijacking?
Fenley, Dagmar
1985 N 360 East
PROVO, UT 84604
In the event of a lost password, contact info would be used to give them
control of their account.
#1 Each time your system recieves an email message the senders address is
checked against a whitelist.
Where is "your system" in this model. MTA, MUA?
The Mail User Agent.
If the name IS on the list deliver the message. end
If the name IS NOT on the list procede to #2.
#2 Check the choicelist database, and compare the numbers returned against
the users choicelist. If no number is returned go to #3 If the number is
in the users choicelist deliver the message. end
If the number is not in the users choicelist delete the message. end
More critically. You've described a complex whitelisting system using
unique numbers (not sure why, the email address is also unique, and the
mapping between the two appears to be public, although I'm not certain from
the description). What does that gain over just whitelisting the address.
And how does it do any more to prevent forgery than just whitelisting the
address?
Hypothetical situation:
I sign up at amazon.com.
they give me a #
I enter that # into my MUA
ANY address that Amazon.com now decides to send me mail from is now
whitelisted.
Amazon.com is responsible for adding a new address to the public database.
When they add an address to the public database, it is added to my
whitelist.
All I have to do is enter the # once.
even if someone creates a fraudulent entry in the main database with
"admin(_at_)amazon(_dot_)com" in it, That does not help them send me spam, because my
system only reads the entrys that I have authorized it to read, and I probly
didn't authorize that number. Even If i did, I could just remove a single
line from my whitelist to never hear from them again.
John Fenley
_________________________________________________________________
MSN 8 with e-mail virus protection service: 2 months FREE*
http://join.msn.com/?page=features/virus
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg