[Monotone-devel] Re: dumb servers

[Zbynek Winkler]

graydon hoare wrote:
[snip]

> that's sort of the nice thing about hashtree synchronization: it's 
> really just set union, so if I sync with a server then you sync with 
> the same server, you have all my changes. if I sync with the server 
> again later, I have all your changes. it would replace probably 10 
> different commands in monotone with just one: "monotone sync <url>". 
> things which reduce the amount of code so dramatically are very 
> attractive.

Hmm. What exactly would be involved in implementing such a command? I've 
never done anything related so I am trying to learn here. Could we say 
that the postcondition of the sync is that the hashes of all head 
manifests are contained in both databases? If so, could the 
synchronization start by exchanging the top level hashes, if they are 
different send the other a copy and go to a parent version(s), 
continuing up to a point when both hosts find that they already have the 
hashs in the their databases (find the closest common parent). Then the 
things to send to the other host should be fairly easy to identify by 
examining the contents of the manifests in question. If this is the case 
it could be relatively easy to do and it could even be layered on top of 
http (no new holes to those firewalls).
Zbynek

--
http://zw.matfyz.cz/     http://robotika.cz/
Faculty of Mathematics and Physics, Charles University, Prague, Czech Republic