[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Monotone-devel] Re: Old Feature Revisited?
From: |
hendrik |
Subject: |
Re: [Monotone-devel] Re: Old Feature Revisited? |
Date: |
Wed, 28 Feb 2007 11:31:09 -0500 |
User-agent: |
Mutt/1.5.9i |
On Wed, Feb 28, 2007 at 04:02:29PM +0000, Bruce Stephens wrote:
> Ulf Ochsenfahrt <address@hidden> writes:
>
> [...]
>
> > A friend of mine accidently created a new branch and synced it to my
> > server, where he has write access. The result is not good.
> >
> > The previous discussion hasn't brought up a solution, but how about
> > this: - a server does not accept branch certificates for branches
> > for which the client does not have write access
>
> That would involve netsync throwing away potentially useful
> information. I think the intention is that netsync ought to take the
> certs (as it does), and then you choose to accept them or not later
> (using get_revision_cert_trust).
>
> I don't think that's ideal (I don't think anyone things the existing
> hooks are ideal), but in outline I think it's the right way to
> approach things.
I always worry about netsync receiving huge amounts of spam. None of
ever gets certified or authorised, but it ends up taking space.
If the size of a network of syncing monotones were to grow like, say,
usenet, this might become a real problem.
-- hendrik