On Sunday, September 21, 2003, at 11:49 PM, Colin Putney wrote:

> I think one of the side effects of letting VersionInfo live 
> independent of it's Version is that won't have to be duplicated quite 
> so much. I imagine Monticello could build a single VersionHistory with 
> information gathered from several sources. So that would present a 
> convenient, central place to make corrections. (Sure, you'd still have 
> a "central" tree per image or maybe per MC user, but that's still much 
> less duplication than per Version.)

I don't like this at all.  It doesn't matter how little or how much 
they're duplicated - they're still immutable.  I *really* don't think 
you want to change that.  You have no idea how many .mcz files are 
sitting around on disk with the old ancestry information.

Fix problems by adding nodes to the tree, not modifying existing ones.  
For example, if you had

A <- B <- C

and you really, really needed it to look like

Q <- A <- B <- C

then create a parallel set of versions that adopt Q into each of A, B, 
C:

      A <- B <- C
    |     |     |
Q <- A' <- B' <- C'

If someone later has something derived from, say, B, and needs to merge 
in something derived from Q, they just need to merge in B' first (which 
shouldn't affect anything except for the ancestry tree).