| 1 |
On Fri, Aug 4, 2017 at 12:09 AM, Matthew Thode |
| 2 |
<prometheanfire@g.o> wrote: |
| 3 |
> On 17-08-04 05:51:38, Andreas K. Huettel wrote: |
| 4 |
>> > > >Apparently, the Foundation only has a list of PGP key IDs in |
| 5 |
>> > > >https://wiki.gentoo.org/wiki/Foundation:Member_List. Even worse, most |
| 6 |
>> > > >IDs listed there are only 32 bit IDs, providing no security at all. |
| 7 |
>> > > > |
| 8 |
>> > > >I would like to ask the Foundation to keep a list with the (160 bit) |
| 9 |
>> > > >PGP fingerprints of its members. (For developers, this information |
| 10 |
>> > > >should be readily available in LDAP.) |
| 11 |
>> > > > |
| 12 |
>> > > >Ulrich |
| 13 |
>> > |
| 14 |
>> > What do we need to prove? |
| 15 |
>> > |
| 16 |
>> > That the the key belongs to a given individual or just that the key on the |
| 17 |
>> > vote is the same as the key used for the membership application.? |
| 18 |
>> > |
| 19 |
>> |
| 20 |
>> That the key on the vote is the same as the key used for the membership |
| 21 |
>> application. |
| 22 |
>> |
| 23 |
>> This is impossible without the full fingerprint. |
| 24 |
>> And with only the short keyid it's trivial to hack. |
| 25 |
>> |
| 26 |
>> -- |
| 27 |
>> Andreas K. Hüttel |
| 28 |
>> dilfridge@g.o |
| 29 |
>> Gentoo Linux developer (council, perl, libreoffice) |
| 30 |
>> |
| 31 |
> |
| 32 |
> I think we should just record the full key id, would we still need to |
| 33 |
> also have the fingerprint in that case? |
| 34 |
> |
| 35 |
> -- |
| 36 |
> Matthew Thode (prometheanfire) |
| 37 |
|
| 38 |
64 bit key IDs can still be duplicated. It's not as easy, but it is |
| 39 |
doable. K_F has a nice blog post on why you really should be checking |
| 40 |
full fingerprint: |
| 41 |
https://blog.sumptuouscapital.com/2016/08/openpgp-duplicate-keyids-short-vs-long/ |
| 42 |
|
| 43 |
-Doug |
| 44 |
dwfreed |
Archives