| 1 |
> first off, fix your e-mail client. this long line crap is ridiculous. |
| 2 |
|
| 3 |
:) ever heard of flowed text? absolutely no need to get aggressive... |
| 4 |
|
| 5 |
> second, anyone can add/remove e-mail addresses. we arent verifying |
| 6 |
> e-mail addresses, we're verifying keys. |
| 7 |
|
| 8 |
Unfortunately you are misunderstanding the GnuPG trust model here. As a third |
| 9 |
party you are not signing someone's key, but someone's userid associated with |
| 10 |
that key. |
| 11 |
|
| 12 |
> the *only* thing that matters |
| 13 |
> is that the key we have on file (0xabcd) is the one that was used to |
| 14 |
> sign. |
| 15 |
|
| 16 |
That's a policy decision. Basically there are several ways to go by |
| 17 |
implementing our own trust model. |
| 18 |
|
| 19 |
1) Rely on an existing list of keys somewhere distributed in portage, and |
| 20 |
automatically trust all keys in that list. |
| 21 |
VERY BAD, because if someone manipulates the portage tree he/she can |
| 22 |
manipulate that list as well. I'm pretty confident however you actually meant |
| 23 |
option 2) or 3): |
| 24 |
|
| 25 |
2) Rely on an existing keyring somewhere distributed in portage; the file (not |
| 26 |
the keys themselves) is signed with a master key. |
| 27 |
Is a very clumsy workaround. |
| 28 |
Pros: you can exactly decide what keys are used and trusted, without thinking |
| 29 |
about GnuPG's inner workings. |
| 30 |
Cons: People tend to modify their keys. Add user ids. Add new subkeys. Expire |
| 31 |
or revoke subkeys. Revoke userids. (My photo in the key is pretty old by now. |
| 32 |
:o) Whenever anything of this happens, the key file changes, needs to be re- |
| 33 |
signed by infra and re-uploaded. |
| 34 |
|
| 35 |
3) Rely on an existing key list somewhere distributed in portage; the list |
| 36 |
file with the key id's (not the keys themselves) is signed with a master key. |
| 37 |
Is a mediocre and potentially insecure workaround. |
| 38 |
Pros: you can exactly decide what keys are used and trusted, without thinking |
| 39 |
about GnuPG's inner workings. A user can edit his key and the key remains |
| 40 |
trusted. |
| 41 |
Cons: Mainly that the key id is a pretty short hash afaik.(Any better-informed |
| 42 |
people around?) |
| 43 |
|
| 44 |
4) Rely on an existing list of keys somewhere distributed in portage and |
| 45 |
possibly somewhere else (keyservers); a key userid is signed with a master |
| 46 |
key. Work with GnuPG's well-tested and well-thought-out trust relationships. |
| 47 |
Back to start, time to re-read the entire thread... :) |
| 48 |
|
| 49 |
Am I missing something? |
| 50 |
|
| 51 |
-- |
| 52 |
|
| 53 |
Andreas K. Huettel |
| 54 |
Gentoo Linux developer |
| 55 |
dilfridge@g.o |
| 56 |
http://www.akhuettel.de/ |
Archives