[OSM-dev] 0.6 Replication

[Brett Henderson]

Andy Allan wrote:
> On Wed, Sep 24, 2008 at 12:30 PM, Brett Henderson <brett at bretth.com> wrote:
>
>   
>> Perhaps I completely misunderstood you.  I was worried that you might do
>> something like this:
>> 1. way 12345 is currently at version 1.
>> 2. Via potlatch you edit way 12345 adding a new tag and create version 2.
>> 3a. Via potlatch you immediately add a second tag and update version 2 to
>> include the extra tag.
>> 3b. Via potlatch you immediately add a second tag and create version 3
>> including the extra tag.
>>     
>
> 3b. "In this situation a given changeset will increment the version
> number of an object more than once."
> http://wiki.openstreetmap.org/index.php/OSM_Protocol_Version_0.6#Version_numbers
>   
Cool.  I had a chat to Richard on IRC and he's also confirmed this.  I 
was just nervous that Potlatch might be doing something funky that the 
main API doesn't allow.  I should have more faith :-)

>> I'm happy with whatever label is applied, so long as I understand how it
>> works :-)
>>     
>
> Bah, the labelling is the easy bit and potentially the only bit that
> I'll be any good at :-)
>   
:-)
>> It might make more sense to discuss the changesets implementation.  Here are
>> several ways I could see them being implemented:
>>     
> [...]
>   
>> I suspect we're using scheme 3.
>>     
>
> From my understanding, that's correct. Individual modifications become
> ACIDy, but changesets are a loose grouping of changes for labelling,
> comments and so on.
>   
Yep, okay.  This all means that a replicated database (ie. osmosis 
replication, not true database replication) will not have a copy of the 
changeset information.  Hopefully this won't be an issue.

Cheers,
Brett