[linux-elitists] GPG luser rant
Fri Apr 13 13:16:58 PDT 2001
begin Karsten M. Self quotation:
> There is a distributed public keyserver network. This seems to work
> reasonably well from a data distribution standpoint. I'd be interested
> in knowing what specific problems exist with it.
I know I'm lame (again) in saying this, but Schneier's coverage of the
appallingly depressing overall picture and its numerous sub-disasters is
your best reference on this.
I've now read Joey's post, which I had not when I sent my own earlier
reply. Consider the example of Joey's correspondent, whose extreme (and
extremely inconvenient) precautions to protect his private key _might_
just be adequate: Can he place a similar level of trust in the chain of
authentication required to use a public keyserver? Remember, you can't
just concentrate on one aspect of communications security; you need to
scrutinise all of them, at once, all the time. If your level of
paranoia is excessive in any one of those areas, you'll waste vast
amounts of time. If it's _inadequate_ in any of them, you lose entirely.
> S&L is by my side right now, as is AC. That would be Chapter 15,
> "Certificates and Credentials"?
_Applied Cryptography_, if memory serves, describes how it works when
everything goes right. _Secrets and Lies_ maps out the myriad ways it
can and does fail.
> See the LSec Wiki node: SecurityIsAProcess
I will. Thanks.
> Exporting the display of a protected device exports the security threat.
That depends on what's exported. A crypto-signing security fob need not
display much at all, to do its work. Possibly nothing at all; certainly
nothing security-sensitive. (Electronic-signal output would total little
more than just the signed text.)
Cheers, Right to keep and bear
Rick Moen Haiku shall not be abridged
firstname.lastname@example.org Or denied. So there.
More information about the linux-elitists