On Mar 17, 2004, at 5:44 PM, Andreas Raab wrote:

> But if you start to think about an evolutionary model where you'd 
> really
> take updates from lots of different places it throws various of the 
> basic
> assumptions that we use today right out of the window. Maybe up to a 
> point
> where it gets questionable even to think about maintaining a system 
> with any
> kind of "files" whatsoever - wouldn't it be much more useful if a smart
> client and a smart server could negotiate what server has, the client 
> could
> say what it wants and they both figure out how to get it done? Hm....

As I understand it, that's precisely what Squat does.  Or did you 
already know that?

I think "files" are a bit of a red herring though.  The Monticello 
model works very nicely in a "pure objects" world - there's no reason 
you have to do a bulk transfer of 300k worth of serialized definitions, 
you could have a proxy to a remote Version object instead.  The more 
relevant question is what Versions consist of, and what kinds of 
operators you have for them - what are the rules for comparing, 
diffing, merging?  Monticello's rules are quite simple, which I think 
is probably a good thing, but I'm certainly curious what operators 
would be possible for more complex notions of version and package.

I'd also be curious to hear more about which assumptions you think go 
out the window, and why...

Avi