[Rebuild] Getting underway before 1st April

[Simon Poole]

Am 31.01.2012 17:32, schrieb Michael Collinson:
...
> - (for me to continue) to persuade the community that the WTFE algo 
> should be the basis of the transition
Yes.
> - that split/merged ways is the only point of contention that could 
> make any potential real difference
Well yes and if we do actually try to process them, how far we go in our 
analysis (as in potentially synthesizing new objects).
> - that the WTFE "rules" could indeed by mapped into scripts that 
> actually altered the database.
I'm not sure why we would want to run the risk of experimenting on the 
live database and doing changes in situ. In doing -anything- like this I 
would always require that we have the option of doing a complete roll 
back of any change, essentially at the flip of a switch.

The most obvious way to do this would be to rebuild on a "new" machine 
and keep the current DB running till the time we switch over for good. 
This probably also implies not changing object IDs (if we still want to 
be able to apply diffs from the "new" DB to the old for a while.
>
> - and that the resulting map would look almost exactly the same as the 
> CleanMap Simon is publishing now?

With the exception of the  geometry of objects that is..

Simon