Re: Internet security code said vulnerable to hackers

[Eric Murray <ericm@lne.com>]

Adam Shostack writes:
> 
> Robert Hettinga wrote:
> 
> |   	 ATLANTA, April 9 (Reuter) - The new security protocol for
> 
> |   	 Steve Mott, senior vice president of electronic commerce
> |   and new ventures for MasterCard International, said it could
> |   take hackers as little as a year to break the industry's
> |   standard encryption code, which is supposed to render
> |   credit-card numbers unreadable to outsiders on the Internet's
> |   World Wide Web.
> 
> 	The security problem with SET is not its crypto, but its
> complexity, which makes it impossible to determine if the thing is
> secure or not.  Its also a nightmare to implement, and was supposed to
> be ready six months ago.


"Security through incomprehensibility".


Set's problem, or one of them anyhow, is that it uses ASN.1.  ASN.1
is useful for some things, but it really sucks as a description of
a protocol.  It is incredibly complex and figuring out the actual
contents of a given message is very difficult.   There is little description
in the SET documents of the protocol itself; the definition is pretty much
left up to the ASN.1.
The ASN.1 by itself often fails to describe what is in an object, i.e.
a gkThumb is an object of type CertThumb, which is defined in ASN.1 as:

CertThumb ::= SEQUENCE {
 digestAlgorithm   DAlgorithmIdentifier -- (sha1)--,
 thumbprint        Digest
 }

But the ASN.1 doesn't say what data is hashed in the Digest.
So you have to flip back to the text and hope there's a description
of what is hashed.   Sometimes there is, sometimes there isn't.


BTW, Set was the name of an old Egyptian god, the one who slew Osirius.
Brewers' Dictionary of Phrase and Fable says under the entry for Set that he
"came to be regarded as the incarnation of evil".

-- 
   Eric Murray  ericm@lne.com  Network security and encryption consulting.
PGP keyid:E03F65E5 fingerprint:50 B0 A2 4C 7D 86 FC 03  92 E8 AC E6 7E 27 29 AF