ref:
http://www.garlic.com/~lynn/2005t.html#27  RSA SecurID product

part of the issue with authentication techniques is there pervasive
use in electronic environments.

one of the issues with "something you know" shared-secrets, is that a
unique shared-secret was required for every different security domain
... as a countermeasure to entities in one security domain
compromising a different security domain (aka the students hired for a
local neighborhood garage ISP accessing the online bank accounts of
ISP customers).

as electronic environments proliferated ... the number of different
security domains proliferated and some people now have requirement for
scores of shared-secret passwords. the early guidelines also were that
shared-secret passwords be impossible to quess and memorize and also
changed frequently. an old april 1st password corporate directive from
the early 80s
http://www.garlic.com/~lynn/2001d.html#51 OT Re: A beautiful morning in AFM.
http://www.garlic.com/~lynn/2001d.html#52 OT Re: A beautiful morning in AFM.
http://www.garlic.com/~lynn/2001d.html#53 April Fools Day

the inability of humans to keep track of scores of impossible to guess
and memorize shared-secret passwords that change frequently led them
to recording passwords in various ways ... which opens up new
vulnerabilities (the combination of password proliferation and
difficult, "secure" passwords results in the creation of new
vulnerabilities).

another way of looking at hardware tokens ... is that they are a
physical object that acts as a countermeasure to simply guessing (or
capturing recorded) passwords (nominally you have to capture the
physical object).

however, some of the token technology in the past just produced static
data as proof of their possession. however, static-data possession
proof is vulnerability to evesdropping/skimming. furthermore, in
two-factor authentication, the basic premise is that the different
factors have unique and different vulnerabilities. unfornately all
static-data authentication technologies may have common
evesdropping/skimming vulnerabilities, aka a token that generates
static-data as proof of its possession and the entering of a
shared-secret password might have a common evesdropping/skimming
vulnerability (an attacker inserts a skimmer and captures both token
static data as well as an entered pin/password). even shared-secret
biometrics may be vulnerability to evesdropping/skimming threat.

the following is discussion of form of evesdropping/skimming attack on
one-time-password standard specificed by internet standard RFC 2289
http://www.garlic.com/~lynn/2003m.html#50 public key vs passwd authentication?
http://www.garlic.com/~lynn/2003n.html#0 public key vs passwd authentication?
http://www.garlic.com/~lynn/2003n.html#1 public key vs passwd authentication?
http://www.garlic.com/~lynn/2003n.html#2 public key vs passwd authentication?
http://www.garlic.com/~lynn/2003n.html#3 public key vs passwd authentication?

which includes discussion of a form of man-in-the-middle attacks.
misc. collected postings on mitm-attacks
http://www.garlic.com/~lynn/subpubkey.html#mitm

part of hardware token authentication should be technology that
generates unique values for each authentication operation ....  as
countermeasure to evesdropping/skimming attacks.

and as previously noted, hardware token technology may also be used to
move from shared-secret paradigm to a secret paradigm ... where the
person effectively authenticates themselves to the token (pin/password
or biometrics) and the token is certified to only operate properly
when provided with the proper authentication. The relying party when
presented with token authentication, then may assume two-factor
authentication based on previously obtaining details of the token
certification.

if the token authentication technology for generating unique
authentication is sufficiently robust ... then there is the
possibility that you could actually have a single token (aka
person-centric) that is acceptable for multi-factor authentication in
multiple different security domains (since there is no knowledge in
the possession of one security domain that could be used for attacking
another security domain).

misc. past postings on the subject of person-centric vis-a-vis
institutional-centric authentication technologies
http://www.garlic.com/~lynn/2005g.html#57 Security via hardware?
http://www.garlic.com/~lynn/2005m.html#37 public key authentication
http://www.garlic.com/~lynn/2005p.html#6 Innovative password security
http://www.garlic.com/~lynn/2005p.html#25 Hi-tech no panacea for ID theft woes

--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/

Reply via email to