Joined: 26 Oct 04
BBC UK Meteo Department should do its own BOINC Weather Forcasting ... with only limited involvement from climateprediction.net or the Met Office.
This means a different BOINC project, run somewhere else in the UK.
If one is just doing 10 day weather forcasts [and forward looking seasonal predictions] (model run days less than 90) there is less a need for Met Office dependancy.
With near real time weather forcasting, one can use datasets from higher resolution weather monitoring networks -- just like the Met Office currently does.
Using BitTorrent Files embedded into an RSS feed, the multimegabyte models can be distributed globally for nearly free.
-- The update cycle for 10 day weather forcasts is 8 hours, and has been for 20 or so years. With a 5.5 to 6.2 hour work unit deadline this is not an issue, noting that the seasonal look ahead models should have 40 day deadlines.
-- Most models would have computed in 3 modes : 1 degree + 1 gradiant +r 1/2 gradient in order to average out resolutional variations.
-- Even with 15 minute time resolution, a 7 day global model will not be huge. Even at 1/4 Gradiant resolution and 10 minute steps the file will not become explosivly huge (100 mb+).
-- Note that there is an assumption that Ensembles of 100 to 1000 models would be integrated into the released data sets.
SO LONG AS THE ADS ARE NOT OBTRUSIVE (NOR CULTURALLY DISTASTFUL) AND FOR UK MANUFACTURED GOODS ONLY
-- Having some room for adverts in the screen saver would be OK.
ALL OF THE CLIMATE MODELS (and data sets where applicable) SHOULD BE OPEN SOURCE -- THAT IS TO SAY THE GENENRAL PUBLIC SHOULD BE ABLE TO INSPECT THEM.
THE BBC (as well as most other UK broadcasters and media outlets) need to embrace the open source movement as the UK has become impovrished to such an extent that closed source systems and services are no longer affordable for most of the population.
Joined: 5 Sep 04
That's something that you should take up with the BBC.