On 2018-11-12 22:00, Warin wrote:
> On 13/11/18 01:07, Allan Mustard wrote:
>
>> Not contrived at all in these days of tight budgets. I see no reason the
>> inverse would not work. I'll add it.
>
> I think there are too many things in the proposal. Keep it simple. Yes the
> 'extras' might sound nice but they add complexity and each one is a point
> that can lead to someone objecting to that specific thing and leading to
> enough no votes that it fails.
At moments like this I like to invoke one of my heroes: Albert Einstein.
One famous saying attributed to him is: As simple as possible, but no
simpler.
If you simplify complex realities too much, you lose valuable detail. If
it's complex, it's complex. If you want to leave out a level of detail,
such as being able to distinguish between the different types of
services provided on behalf of multiple "tenant" countries in a
diplomatic mission, then so be it, but let's discuss whether it is
desirable to leave that out, and whether the resultant ambiguity is
acceptable. Data modelling means constructing an approximation to
reality, and is all about what details to keep in and what to leave out.
Once it is left out, it cannot be reconstructed from the rest of the
data. (If it can, your data model is not properly normalised.)
If OSM is being limited to being suboptimal because of politics and the
inability to reach consensus, I would rather the system was fixed
instead of condemning the whole business to eternal mediocrity.
_______________________________________________
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging
