Am 06.01.2011 16:44, schrieb Simone Saviolo:
2011/1/6 Steve Bennett <stevag...@gmail.com <mailto:stevag...@gmail.com>>
Putting in place a serious process for tag migration will be
difficult. I suggest that a first step will be definition of an
actual schema, with version number. For example, define an actual
list of several hundred tags, with semantics, that correspond to
"OSM core 1.0". Then, we could have votes on changes to the
schema, with advance notice given: "On November 1, 2011, the main
database will be updated to OSM core 1.1. Please have your editor
and renderer patches ready for this date."
I would even go further. I would like to see such a schema become a
sort of "OSM certification", to be awarded to consumers that fully
support it.
What's the benefit of that?
What beside of this - I fear, stupid - "certification" is the benefit
for a hiking map in supporting e.g. maxspeed of motorways as part of the
OSM core being the decision basis to get the certification?
To make a better example: Garmin AiO for Europe is getting too large for
many devices currently - so the core definition you propose would
require to include buildings in the map, no matter of their size and the
drawbacks of excluding most old devices by including the building layer?
An alternative to "this product supports OSM Core n.n completely" as a
(the?) requirement for the certification would be "this product does not
interpret attributes from OSM conflicting to the Core definition".
This for me fit's better, but nevertheless not completely, as it does
not prevent devices to support other variants as well probably
conflicting in parts.
Additionally there would have to be an organization/council/something to
give the certification to the application (in wide interpretation)
developer/vendor.
But: even that does not automatically lead to high quality of the
product: A map without legend is worse than a map with missing
housenumber display. A public bus information system without support of
railway is complete in it's self defined domain - but lacks of the
railway network and therefore does not support the complete core (I guess).
I thought about some kind of layered-extracts a while to get kind of a
stable, well defined architecture for application vendors:
An API (not THE api) could export a "car routing layer", a "building
shape layer", a "footway layer" and so on.
That could be an enhancement of the XAPI concept by shortcuts for
complex queries, too.
It probably could provide a bridge of the wiki principle of OSM
including the free-to-invent-new-tags to the application requirement of
a as much as possible stable base for data updates.
Last, but not least it could be one solution to define renames for
"deprecated" tag usages of some kind, e.g. new invented subtags.
To use the tree example discussed a few month ago (a tree versus an
important or lone tree) the "tree" layer could be split after that
discussion to an "tree" layer and an "important tree" layer considering
the new subtags and only providing the trees explicitly tagged as important.
I'm neither sure yet if that's a good idea, nor do I know if it is clear
what I wanted to say - but well - everybody can ask me.
regards
Peter
_______________________________________________
Tagging mailing list
Tagging@openstreetmap.org
http://lists.openstreetmap.org/listinfo/tagging
