[PGP-USERS] Re: ADK Bug: Statement from cert.org.

[Simpson, Sam]

[PS: Sorry to ramble but I'm pissed off]

> -----Original Message-----
> From: Kai Raven [mailto:kai.raven@ob.kamp.net]
> Sent: 25 August 2000 10:30
> To: pgp-users@cryptorights.org
> Subject: [PGP-USERS] Re: ADK Bug: Statement from cert.org.

<SNIP>

> *That is not enough*. 
> As we learned, many users do passphrase chaching, many users don't
> know something about ADK, although ADK is implemented since 
> the V.5 series of PGP.

Indeed.  If we ever want PGP to be used by more than geeks and paranoid's
then this kind of explanation is woefully inadequate.

Patching key servers IS NOT ENOUGH!

> PGP and NAI have nothing changed in the default configuaration (adding
> the ADK column as the *first* column in PGPkeys) since PGP 5 to PGP
> 6.5.3.
> The only adequate solution in my point of view is the release of
> Freeware, Personal Privacy and Desktop Security versions *without* any
> ADK functionality.

I agree.  Any chance of an ADK/GAK free CKT build?

> And the argumentation, that ADK protects against key recovery law
> plans is bullshit - There is no evidence, that any government  can
> force key recovery and programs, that have key recovery 
> mechanisms have no chance on the market today.
> It is not a solution to change key recovery with message recovery and
> patch and fix again and again.

The anti-christ of security is complexity.  ADK is big and complex and NAI,
PGP and you and me as end users have paid the price.

Amusingly I've spent a great deal of time replying to people who either
state "PGP is broken" or asking "Is PGP broken?", hell I've even written a
250Kb FAQ that's distributed from an official NAI site and included in CKT
builds but I really believe we need some "good faith" response from NAI
(e.g. releasing a new free upgrade version (e.g. 6.6 or the soon to be
released v7) that is verifiably GAK-secure for FREE!).

> Kai Raven (who don't use PGP 5/6 any longer until a really secure PGP
> 6... version exists without any ADK-Bullshit)

GPG already won't encrypt to ADK keys so it's a good alternative and I have
every confidence that Werner Koche is currently working on a version of GPG
that is at anti-ADK as possible. This won't help existing NAI/PGP users
though.

Maybe the release of DSS-2 / the new SHA, AES and the lapse of the RSA
patent in the next month or two will give an ideal opportunity to revisit
the implementations of OpenPGP and RFC2440 and get rid of the
security-sapping "baggage" called ADK?

One has to wonder when it makes sense to "cut our losses" now and define &
implement a clean, simple and non complex mail security standard rather than
producing a v5 specification that forces implementers to jump through more
hoops for the sake of backwards compatibility.  Werner: what's your take on
the best way forward?


Rgds,

Sam

-- 
Archive is at http://lists.gnupg.org - Unsubscribe by sending mail
with a subject of  "unsubscribe"  to gnupg-users-request@gnupg.org