> is harder to parse than simply parsing Map Features as it is now ;-)
well, at least the ugly stuff could be left to an XML parser ;-) - and the raw
export would be transformed into [1] which is certainly
not more complex than JOSMs preset file (I even think it could replace the JOSM
preset file in the future ...)
> I'm also unsure about additional complexities introduced by the semantic wiki
> approach; if that were implemented, would it still
> be possible for non-IT people to manage the information - would it still be a
> true Wiki in the sense that everyone can edit it?
Users ("wiki gardeners") don't get in touch with the Semantic Wiki. They still
use more or less the same wiki templates
they have been using so far. And the complex template collection used in the
german wiki (one wiki template per feature and property?) would be obsolete.
Writing the text of map feature pages is more challenging than filling in ~10
property values in a template and SMW has no impact on the former task. I
agree, however, that overview pages like [[Map Features]] would probably have
to be maintained by IT-people.
On the other hand, there would be less need to edit overview pages. If
individual pages were modified or added, changes would automatically be
propagated to them. But note, that using semantic queries in overview pages is
not mandatory. You can still trade ease of copy(paste against data redundancy
without putting the main goal at risk: generating a machine-readable list of
map features.
There are just two rules users should follow (it doesn't "break" the wiki if
they don't it's just more work for others to get it fixed in order to generate
a *consistent and complete* machine-readable map feature list):
1) if you document a key, create a page Key:key and insert a template {{KeyDef
...}} at the beginning (for a translated page, call the page lang:Key:key)
2) if you document a tag, create a page Tag:key/value and insert a template
{{ValueDef ...}} at the beginning (for a translated page, call the page
lang:Tag:key/value)
Everything else is negotiable.
My concern is stability and responsiveness of the wiki, also in the light of a
recent thread about insufficient (?) server infrastructure in the OSM backend.
--Karl
[1]
http://wiki.openstreetmap.org/wiki/Machine-readable_Map_Feature_list#File_format
-----Ursprüngliche Nachricht-----
Von: Frederik Ramm [mailto:frede...@remote.org]
Gesendet: Mittwoch, 21. Januar 2009 19:54
An: karl.guggisb...@guggis.ch
Cc: talk@openstreetmap.org; 'Denny Vrandecic'; 'Yaron Koren'
Betreff: Re: [OSM-talk] OSM-WIKI: reducing redundancy, less redundant and
standardized data exchange
Hi,
Karl Guggisberg wrote:
> I opened a discussion on Talk:Wiki [3] on whether and how this approach could
> be moved to wiki.openstreetmap.org.
Well to me it looks as if
http://dev.openstreetmap.org/~edgemaster/semwiki/index.php/Sample_Export_for_Key:highway
is harder to parse than simply parsing Map Features as it is now ;-)
I'm also unsure about additional complexities introduced by the semantic wiki
approach; if that were implemented, would it still be possible for non-IT
people to manage the information - would it still be a true Wiki in the sense
that everyone can edit it?
Bye
Frederik
--
Frederik Ramm ## eMail frede...@remote.org ## N49 00'09" E008 23'33"
_______________________________________________
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
