[Rebuild] Deadline for undecided?

[Dermot McNally]

On 17 February 2012 15:36, ant <antofosm at gmail.com> wrote:

> I like Dermot's idea, and actually it is precisely what I was thinking of.
> (My point being only that if an undecided mapper accepts last minute while
> the gradual rebuild has already started, things get messed up.)

Well, my thoughts on the model suggested were that this would not be a
problem - so yes, we'd start processing objects in ways that would
mark some objects dirty (that is, to be dirty on changeover). But if
the mapper concerned then agreed, we'd reprocess the objects. This
would happen at the latest on April 1st, but probably periodically
anyway, since we would (I suggest) want to use the evolving DB to
drive our remapping tools such as the various reports, Bad/Cleanmap
etc.

In like fashion, if such a mapper agreed in, say, May, we would have
the option of deciding to re-admit data from his older versions,
though this would have to be done with extreme care because of the
possibility of conflicts where remapping has occurred.

Dermot

-- 
--------------------------------------
Igaühel on siin oma laul
ja ma oma ei leiagi üles