There is no 100% foolproof automatic mechanism. Consider, if you will, that
the emailer that accepts your your request, might actually store and forward
this to another host, or pass it through a firewall and onto some corporate
email system. sendmail was designed to do this, and to send back / bounce
email. Some sites may hold your email for days before bouncing it.
If you want clean and Polite! email (avoiding spam and nuisance attacks)
then you should send an email to that account asking them to reply for it to
be really setup. I'm sure you'll remember seeing something like that when
you registered for this list *g*. It serves 2 very nice purposes (and many
more but ...) 1. verify the email account; 2. verify it was actually you
that signed up for this service and not your neighbours kid signing you up
for all sorts of spam mail.
The last part is my opinion *g*
Thor HW
----- Original Message -----
From: Rod McChesney <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, August 17, 1999 2:31 PM
Subject: Re: Help on Email Validation
> Unfortunately, VRFY usually doesn't work either -- it can be turned
> off and often is. For instance, O'Reilly's sendmail book recommends
> setting PrivacyOptions to goaway to disable VRFY and EXPN commands.
> Sometimes VRFY returns a 250 or 252 when all it's actually checking is
> that the name looks something like foo@bar:
>
> VRFY asdf
> 550 asdf... User unknown
> VRFY foo@bar
> 252 <foo@bar>
> VRFY xyz@abc
> 252 <xyz@abc>
>
> Our site sends email to people who register, and travel agents need to
> reply to requests via email or phone, so it's important to us to have
> clean emails. But it appears to be an intractable problem in general.
> If someone knows of a way that mostly works, I'd love to hear
> about it...
>
> Rod McChesney, Korobra
>
>
> Craig R. McClanahan wrote:
> >
> > Rod McChesney wrote:
> >
> > > Bear in mind that whois often returns false negatives, because it's a
> > > huge security breach to allow random outsiders to skim your list of
> > > email accounts. For instance, I believe most corporate email systems
> > > refuse whois requests. You should probably try whois out on a sample
> > > of your actual user base to see how many could have been validated
> > > this way, and whether it's worth doing.
> > >
> > > Rod McChesney, Korobra
> > >
> >
> > The better way to validate users's EMAIL addresses would be to use the
verify
> > (VRFY) command already present in the SMTP protocol used to transport
Internet
> > email. The protocol is pretty simple, and is outlined in RFC 821. It's
probably
> > a 50-liner to do the remote validation using sockets, given an email
address as
> > an argument.
> >
> > Craig McClanahan
> >
> >
___________________________________________________________________________
> > To unsubscribe, send email to [EMAIL PROTECTED] and include in the
body
> > of the message "signoff SERVLET-INTEREST".
> >
> > Archives: http://archives.java.sun.com/archives/servlet-interest.html
> > Resources: http://java.sun.com/products/servlet/external-resources.html
> > LISTSERV Help: http://www.lsoft.com/manuals/user/user.html
>
>
___________________________________________________________________________
> To unsubscribe, send email to [EMAIL PROTECTED] and include in the
body
> of the message "signoff SERVLET-INTEREST".
>
> Archives: http://archives.java.sun.com/archives/servlet-interest.html
> Resources: http://java.sun.com/products/servlet/external-resources.html
> LISTSERV Help: http://www.lsoft.com/manuals/user/user.html
>
___________________________________________________________________________
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff SERVLET-INTEREST".
Archives: http://archives.java.sun.com/archives/servlet-interest.html
Resources: http://java.sun.com/products/servlet/external-resources.html
LISTSERV Help: http://www.lsoft.com/manuals/user/user.html
