No subject

sine (som våre byggjer på) på eit overordna europeisk nivå. Dei gjer
sikkert validering på verdsnivå òg, utan at eg har sett rapportar på
dette. Valideringane som blir gjort, viser at prognosane er svært
treffsikre (90-95%) dei fyrste dagane, og at det fell ned mot 60% for
dag 10 i prognosen."

That is, the EVWMF validates their own weather forecasts on a large
scale level and the hit rate for the first few days is around 90-95%.
I'm not sure what they are measuring. Later that goes down to a 60% hit
rate for day 10.

I'm not sure about how the rest of the world are modeled, the grid
resolution, etc.

John

John at Darkstar wrote:
> Wikimedia Norway have had a meeting with Meteorologisk Institutt
> (http://met.no) in Norway about using the dataset published by the
> institute. Today the dataset is used for the service Yr.no
> (http://yr.no), a joint venture between them and Norwegian Broadcasting
> Corporation (http://nrk.no). Met said it would be pleased if the data
> could be utilized somehow, later this was confirmed by NRK.
> 
> The dataset is localized to within 4km in Norway and updated in about 4
> hour cycles. There are available lookup for 7.5 million places around
> the world, and 700 000 places in Norway. Base data are the HIRLAM model
> (High Resolution Limited Area Model) and the EC model from The European
> Weathercenter ECWMF. HIRLAM has 10 km resolution of the grid. When
> requesting data for named places the data from the grid is interpolated
> to build the new weather forecast.
> 
> The dataset is made available with an open license, they request
> attribution and "truthful reproduction", basically you can't say it is
> from Met and falsify the dataset. You can't add a few degrees to make
> the weather look better at a certain place and still claim its sourced
> from Met. This is analogous to editing a news photo at Commons so its
> not a truthful reproduction of the event anymore. There are one
> disclaimer for sites with pornographic or racist content which may be
> troublesome for a CC-by-sa -licensed site. They do although allow use on
> commercial sites. Other than that the licensing should be given due
> consideration.
> 
> My personal opinion is that they should be kindly asked to make the
> dataset available according to CC-by-sa to make reuse easier.
> 
> Our proposal to Met/NRK was to make an extension that could be localized
> for various languages, based on the XML-data streamed from their site,
> and to use something like this for Wikinews. Unfortunately we have not
> been able to get sufficient funding for the project. We also need some
> backing from the community if this should be considered a viable project.
> 
> Short information in english about Yr.no
> http://www.yr.no/english/1.2025949
> 
> Page about licensing (in Norwegian)
> http://www.yr.no/verdata/1.3321307
> 
> Page about services available (in Norwegian)
> http://www.yr.no/verdata/
> 
> Some locations
> http://www.yr.no/place/Norway/Oslo/Oslo/
> http://www.yr.no/place/Bouvet_Island/
> http://www.yr.no/place/Norway/Finnmark/B%C3%A5tsfjord/Makkaur/
> http://www.yr.no/place/United_States/New_York/New_York~5128581/
> http://www.yr.no/place/China/Xinjiang/%C3%9Cr%C3%BCmqi/
> Page about how the weather forecast is made (in Norwegian)
> http://www.yr.no/informasjon/1.3687572
> 
> John Erling Blad
> Wikimedia Norway
> 
> Tris Thomas wrote:
>> Dear All,
>> I don't know whether this has been discussed before, apologies if it has.
>>
>> I'm interested in people's thoughts on a new Wikimedia project-maybe 
>> WikiWeather, which basically would do what it says on the tin.  Along 
>> with importing national weather from other sources(especially to begin 
>> with), contributors could then put their weather where they are.  This 
>> could evolve into many contributors giving very localised weather 
>> forecasts worldwide, which could be used by many of the other projects 
>> and anybody else.
>>
>> Would people be interested in this proposal/have any thoughts on it?
>>
>> Thanks!
>>
>> Wikinews User Page <http://en.wikinews.org/wiki/User:Tristan%20Thomas>
>> _______________________________________________
>> foundation-l mailing list
>> foundation-l at lists.wikimedia.org
>> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>>
> 
> _______________________________________________
> foundation-l mailing list
> foundation-l at lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>