> What we need is carefully detailed and thought out information. 

In an ideal world, yes, but given the millions of dollars *not* going into bee research, isn't it something? Plus, controlled experiments have their own selection criteria are not necessarily truly randomized when it comes to location. Frequently, it's the availability and willingness of beekeepers to use their yards for experiments.

I've just volunteered for Cornell's research and was told that they need to limit their reach in NY to locations where the expense will be worth their visit. There weren't other volunteers from Dutchess County. This is perfectly understandable, but the locations for this research will be limited, just like broodminder.io's data is limited. I'd think that researches know how to separate the noise from good data. At the very least it can confirm trends.

For the research project that our team did for Prof.Bromenshenk's MB class at UMT we ran a survey and found out that the biggest influence on new beekeeper's management style is the person who mentored them. 1,500 self-selected responses from 48 states and 20 countries. Is it noise? Does it confirm what some of the experts we've interviewed were suspecting? We're waiting for the word to submit it for publication (hint, hint, prof. B, :-)), but one can imagine a lot more useful correlations could be extracted from extremely large (10,000 samples) and frequent (default is every 1h) data collection that Broodminder is aiming for.

Przemek

P.S.
After almost a week of data, the weight graph has stabilized. They didn't swarm. About 6-7 pound drop in the morning, then recovery as foragers come back. 

             ***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software.  For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html