Re: secure plist store

[Ted Zlatanov]

On Wed, 29 Jun 2011 17:07:57 +0900 Daiki Ueno <address@hidden> wrote: 

DU> I didn't notice that the field encryption code is already checked in.
DU> However, it does not work for me at all and looks too complicated

I would appreciate it if you were more specific about "does not work" by
either filing a bug or explaining.  Obviously it works for me and Lars,
who tested it.  Similarly, since you wrote EPA/EPG, your advice on
reducing complexity when we use it is greatly appreciated but you don't
give us any here.  Should we dig through plstore.el and figure out how
to use the pieces in auth-source.el?  I don't think that was your
intent...

DU> I don't want to see that the Gnus password-caching feature becomes
DU> harder and harder to use daily...

I don't think we've done anything that makes it harder to use.  The GPG
token functionality is off by default right now.

DU> Yeah, I'm reluctant to repeat the same discussion - here is a
DU> proof-of-concept implementation of searchable, partially encrypted,
DU> persistent plist store, called plstore.el.
...
DU> As you see, secret properties are prefixed with ":secret-" and the value
DU> is hidden, and the real properties are encrypted together in the GPG
DU> data at the end.

So it's not line-based.  I think that's a minus: people expect to be
able to copy a line out of the netrc file, and it makes managing such
files easier.

Ted