Re: gnunet-guile reboot & guix (take two)

[Ricardo Wurmus <rekado@elephly.net>]

Hi,

Ludovic Courtès <ludo@gnu.org> writes:

> Hi amirouche,
>
> amirouche <amirouche@hypermove.net> skribis:
>
>> After discussing gnunet & guix at fosdem with gnunet
>> people I have better picture of where things can go.
>>
>> The short story is:
>>
>> 1) There is no way to know the gnunet hash aka. gnunet uri
>>   of a substitute before the build.
>>
>> 2) There is no way to associate gnunet hash and guix hash
>>   in a secure/trusted manner over gnunet. Except maybe
>>   if we use GNS to publish guix hash as subdomains of
>>   substitute-server.guix.gnu?
>
> Before jumping to conclusions, did you check the narinfo format that we
> use on HTTP?  A narinfo is essentially a statement that a given store
> item, resulting from a .drv build, has a specific content hash.

This may have been due to my unclear communication at the Guix Meeting
on Friday.  What I said there is that we cannot possibly know the hash
of the output *ahead of time*, which is why it cannot be recorded as
part of the package definition.

The file format of the narinfo files accompanying substitutes, however,
(as provided by files served through “guix publish”) *does* contain the
hash of the contents.

> Lookup by content hash over GNUnet’s AFS is all we’d need.

Indeed.

For items that are *not* reproducible (or that are marked as not
substitutable) a different mechanism would be needed, but it is much
less important to have it.  For those items a keyword lookup (where the
keyword is the name of the derivation or the name of the item’s path in
the store) with filtering by GNS trust would be possible; but
considering that this case is less important and harder to implement I
don’t think we need to bother with it.

--
Ricardo

GPG: BCA6 89B6 3655 3801 C3C6  2150 197A 5888 235F ACAC
https://elephly.net



_______________________________________________
GNUnet-developers mailing list
GNUnet-developers@gnu.org
https://lists.gnu.org/mailman/listinfo/gnunet-developers