[sitefinder-tech-discuss] Pseudo code please

[Kee Hinckley <nazgul@somewhere.com>]

I'd like to see some pseudo code.

1. Given a host name in .com/.net, determine whether the host really exists.
2. Given a domain name in .com/.net, determine whether the domain 
really exists, or whether it has been configured to *look* like a 
Verisign wildcard domain.
3. Given a host/domain, determine whether one should be able to send 
mail to it.

In all of the above examples, I want to know how long I can 
legitimately cache the results of any *.TLD lookups and expect them 
to remain unchanged.

If the code assumes that *.TLD will always return one and only one A 
record, then I'd like to see that assumption defended in light of the 
realities of load balancing and DDoS attacks.

Prior to any SiteFinder release I would expect real libraries in (at 
least) C and Perl.  Those libraries should handle caching and timeout 
of the *.TLD results.  And they should not require duplicate lookups 
(which means that either you need to duplicate the entire set of DNS 
APIs, or you need to have a library that takes the normal results of 
a given lookup and returns the answer).  In addition they should 
allow for easy expansion to include logic for other wildcard TLDs 
from other TLD contractors.  I don't know how you're going to easily 
implement all of that, but it's obviously cheaper for you to do it 
once than for thousands of companies to do it separately.


None of this consists of an acceptance of the wildcard.  We've spent 
thousands of dollars dealing with the Verisign wildcard, I just want 
to minimize our future costs.
--
Kee Hinckley
http://www.messagefire.com/         Next Generation Spam Defense
http://commons.somewhere.com/buzz/  Writings on Technology and Society

I'm not sure which upsets me more: that people are so unwilling to accept
responsibility for their own actions, or that they are so eager to regulate
everyone else's.

----------------------------------------------------------------
To subscribe or unsubscribe from this elist use the subscription
manager: <http://lists.elistx.com/unsubscribe>