On Wed, Sep 17, 2008 at 07:41:04PM +0100, Eric Kow wrote:
> Hi David,
> 
> On Wed, Sep 17, 2008 at 11:14:20 -0400, David Roundy wrote:
> > Here's a quick fix for issue27.  I've been resistent on this, but it
> > seems that since so many people don't know how to use darcs properly
> > (particularly when they're first starting by converting old
> > repositories using tailor), it'll be simples to just make darcs more
> > robust.
> 
> Hooray for defensive programming!
> 
> Now, I realise that I'm being a pain here by quibbling over details, but
> since we are adding junk anyway, I still think we should prefer
> deterministic junk (e.g. hash of patch-and-creation-context) over
> non-deterministic junk (random numbers) :-D
> 
> To justify this, I could invoke a belief in Murphy's Law, i.e. that if
> anything could get Tailor users up in arms, it will get Tailor users up
> in arms.  (For the interested, the "anything" in question is that
> tailoring the same repository twice will now produce two subtly
> different repositories)...
>
> But I suppose the real reason is not so much the Tailor users, but
> my own natural squeamishness against introducing randomness when it
> is possible to avoid doing so.
> 
> What do you think?

I think deterministic junk is considerably harder to produce, and
requires proof that it is generated correctly, while non-deterministic
is easy to produce.  We can always change it to be deterministic
later.

David

Attachment: signature.asc
Description: Digital signature

_______________________________________________
darcs-users mailing list
[email protected]
http://lists.osuosl.org/mailman/listinfo/darcs-users

Reply via email to