[OSM-talk] [OSM-dev] GSoC'10

Karl Guggisberg karl.guggisberg at guggis.ch
Thu Mar 11 23:16:18 GMT 2010


Hi,

 > 4. conversion of the wiki into OWL, so that the rules and relations 
of the OSM are documented formally.
see also
   
http://wiki.openstreetmap.org/wiki/Machine-readable_Map_Feature_list/OWL_Semantic_Wiki_and_more
   http://wiki.openstreetmap.org/wiki/Machine-readable_Map_Feature_list
an attempt to maintain and harvest a machine-readable list of map 
features on and from the OSM wiki using
OWL, RDF, SemanticMediaWiki.

Note that this is more than a year old, and no progress has been made. 
JOSMs tageditor plugin now adopted
a more pragmatic approach. I manually maintain a list of 
machine-readable map features (yet another list, in
addition to JOSM presets, and other lists).

 > 4. conversion of the wiki into OWL, so that the rules and relations 
of the OSM are documented formally.
That's the key issue. And the most difficult part. SemanticMediaWiki 
could help here, because people working
with the wiki and its templates today would still work with the very 
same wiki and the more or less same
templates. We had a prototype up and running a year ago. Emphasis on 
"prototype", though. Unfortunately, it never
made into the production wiki (lack of hardware ressources at the time, 
also lack of interest in general).

"so that the rules and relations of the OSM are documented formally" 
describes the major obstacle here: a significant
share (if not the majority) of OSM contributors doesn't see value in 
raising the level of formality

Regards
Karl

Am 11.03.2010 23:39, schrieb jamesmikedupont at googlemail.com:
> Description :
>
> Integration of the java swoop ontology editor into JOSM. A JOSM 
> ontology plugin.
>
> Work :
>
> 1. create a live mapping from OSM into RDF , so that the swoop can 
> access the data in OSM without conversion.
> 2. be able to have the changes in the rdf be reflected back into the OSM.
> 4. conversion of the wiki into OWL, so that the rules and relations of 
> the OSM are documented formally.
> 5. running of the pellet inference engine and all the other reasoning 
> tools from swoop to infer new facts and validate the data.
> 6. encoding of the rules of the JOSM validator into OWL rules, maybe 
> we will have to include new derived geometric things like if two ways 
> intersect and also ways to only process data in a certain radius.
>
> mike
>
> On Thu, Mar 11, 2010 at 11:16 PM, Graham Jones 
> <grahamjones139 at googlemail.com <mailto:grahamjones139 at googlemail.com>> 
> wrote:
>
>     Hi,
>     I agree that improving documentation would be  a really useful
>     contribution to OSM, but Google are quite explicit that this is
>     outside of the scope of GSoC
>     (http://socghop.appspot.com/document/show/gsoc_program/google/gsoc2010/faqs#doc_proposals).
>
>     A project along the lines of 'context sensitive help in OSM
>     editors' could be a possibility though.  I have never worked on
>     one of the main editors, so I have no idea how hard this would be
>     from that point of view.  The Artifical Intelligence aspects are
>     quite difficult in themselves though - it would have to try to
>     guess what you would like to do from what you have just done -
>     quite a challenge, but please add it to the list if you can manage
>     to describe it!
>
>     Regards
>
>
>     Graham.
>
>
>     On 11 March 2010 10:59, jamesmikedupont at googlemail.com
>     <mailto:jamesmikedupont at googlemail.com>
>     <jamesmikedupont at googlemail.com
>     <mailto:jamesmikedupont at googlemail.com>> wrote:
>
>         lets put it in a different perspective :
>
>         Make the documentation as part of the program!
>
>         I would like to see for example a help system that is
>         integrated to the wiki,
>         Click on a tag, have it pull up the wiki entry, be able to add
>         new unknown tags or rename them.
>         We could even have an OWL Ontology created with a reasoning
>         engine to specify rules for tagging.
>
>         Really, the software should be so good that you dont need docs
>         or videos.
>
>         mike
>
>
>         On Thu, Mar 11, 2010 at 11:49 AM, Peteris Krisjanis
>         <pecisk at gmail.com <mailto:pecisk at gmail.com>> wrote:
>
>             > GSoc
>             > student pool is a very talented one - it would be good
>             to use them for more
>             > critical things. Of course, OSM unlike other projects is
>             basically a
>             > collection of tools maintained by various people, so
>             difficult to achieve a
>             > consensus.
>             >>
>
>             Having one place of knowhow of mapping is quite critical
>             for project
>             like OSM. And believe me, creating good documentation
>             requires quite a
>             skill and isn't easiest job in IT world as we would like
>             to see.
>
>             Cheers,
>             Peter.
>
>             _______________________________________________
>             talk mailing list
>             talk at openstreetmap.org <mailto:talk at openstreetmap.org>
>             http://lists.openstreetmap.org/listinfo/talk
>
>
>
>         _______________________________________________
>         talk mailing list
>         talk at openstreetmap.org <mailto:talk at openstreetmap.org>
>         http://lists.openstreetmap.org/listinfo/talk
>
>
>
>
>     -- 
>     Dr. Graham Jones
>     Hartlepool, UK
>     email: grahamjones139 at gmail.com <mailto:grahamjones139 at gmail.com>
>
>
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk
>    
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20100312/f30d615e/attachment.html>


More information about the talk mailing list