[OSM-talk] Request for Romano-British features

Lester Caine lester at lsces.co.uk
Mon Jan 16 14:57:21 GMT 2012


John Sturdy wrote:
> On Tue, Jan 3, 2012 at 3:12 PM, Nathan Edgars II<neroute2 at gmail.com>  wrote:
>
>> For something like this, where there is very limited overlap between past
>> and present, it makes sense to use a separate database. But in cases where
>> most of the features still exist, such as railways or Roman roads, it's
>> silly to duplicate the effort between databases (or somehow require everyone
>> improving a way in one to upload it to the other and fix all intersections).
>
> Agreed.
>
> As long as the tagging used is such that things that no longer exist
> are not normally rendered (and only show as thin outlines on standard
> editors) I think including historic data shouldn't be a problem.
> Compared with the amount of modern ("current") data, there's not
> really that much of it, anyway, so its effect on the storage
> requirements is going to be fairly small; and we still meet the
> requirement of the most accurate map of what is current.

This has been my argument all along. The majority of the historic mapping 
locally is simply add the date when a road was constructed, but there are very 
small sections of roads that have been re-routed and it does seem ridiculous to 
have to go to a second database for just a few extra lines of data. We simply 
need to apply the end_date tag consistently rather than looking at ways to move 
that data to a second database?

There is data that is better supported by a secondary database, such as perhaps 
the movements and location of troops during an engagement, but the historic 
development of ground features is swamped by start_date information. This does 
of cause leave a grey area with administrative boundaries. Something that 
changes quite regularly in the UK at least. The history changes here are best 
managed as secondary data, in which case, the 'current' view would just be 
synchronised from that database and changes of country names, boundaries, and 
other 'political' data would be edited in the secondary database first? I'm sure 
that other examples would also make sense?

-- 
Lester Caine - G8HFL
-----------------------------
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php



More information about the talk mailing list