Eric S. Raymond wrote on Sat, Dec 01, 2012 at 01:03:28 -0500:
> kmra...@rockwellcollins.com <kmra...@rockwellcollins.com>:
> > Possibly I'm naive, but a client provided email address is far
> > from being a GUID.  In fact, I can pretty much set my email address
> > to anything in most DVCS tools.  Who is to say I haven't used
> > your email address when committing?
> 
> Technically, nothing.  The underlying assumption is that you trust
> your contributors not to *want* to spoof each other.
> 
> Sure, it would be nice to have better authentication than that, but
> if you think for a bit you'll see that this is a very hard problem.  
> The cost of solving it would so high that DVCSes have decided they have
> to ignore the spoofing case and hope everybody behaves well.
> 

Haven't a few projects decided to require PGP-signed revisions instead?

> So far, this has worked.
> -- 
>               <a href="http://www.catb.org/~esr/";>Eric S. Raymond</a>

Reply via email to