A few friends and I are thinking of working on an open source everything  project over IAP (January).
Some research into people working on similar projects:
Many links from: http://news.ycombinator.com/item?id=3276596
Twine — Cambridge, MA
~$100, automation oriented web app, 3 sensors, wifi connected. (oh hai mit media lab people o.o)
whoa sticker sensors! eeeeee!
ethernet base station, 250+ feet range, probably in $100+ closer to $200 range
tedx towatch http://www.youtube.com/watch?v=evKoXorLhqI
http://postscapes.com/companies IoT company listings
http://ifttt.com/wtf software only
http://java.net/projects/spots/pages/Home older SPOTS, not commercial (educational)
“small, Java-based, wireless devices developed at Oracle Labs. The SPOTs project is the place on java.net for development of open source code relating to Sun SPOTs. This includes system code, application frameworks, demonstrations, and applications.”
Rascal — Somerville, MA
$100+ (close to $200) open source hardware FTW! ^^ From the frontpage:
|Basically takes care of talking to internet, web GUI, and data logging for your arduino-connected sensor|
Can I steal the design?
I’m giving you the design, so no, technically you can’t steal it. If you want to make your own version of the Rascal under a different name, or use it in a commercial product, go ahead. Seriously, do it. I’ll never saturate the market for devices like the Rascal, and I’m already working on the next design anyway. asdfjhadf yay oshw
https://code.google.com/p/pubsubhubbub/ idk what are this. pushing atom/rss updates instead of making scripts constantly poll?
http://mbed.org/handbook/Tour — makes experimenting with microcontrollers easier
educational, <$50 range for sensors (requires Arduino, not designed with connecting to Internet in mind?)
Grove Easy Prototyping Series
None of the sensor platform-y ones are open source hardware, though. Or designed with the intent of enabling people to modify and extend the sets of sensors. >__> bwahahaha
Our idea doesn’t actually fall into “internet of things” because (at least initially) we’re aiming for Really Cheap ($10 or less a sensor).
Thoughts: conflict, we want lots of data so we’re aiming for inexpensive sensor boards and part of the cost-savings is not having wireless capability on each board. But that implies we want people to get lots of sensors (yay more data to crunch and visualize!), which means that’s a lot of data to read manually (will people have the patience? Perhaps it’d be like watering plants — not everyone is cut out for it)
My hope is that in two weeks, when we get to wireless networks in MAS.863, I will have the magic. See: http://fab.cba.mit.edu/content/tools/circuits/network_transport/
More on the business-y side:
nice infographics: http://www.readwriteweb.com/archives/cisco_50_billion_things_on_the_internet_by_2020.php
nice graphical overview of IoT idea and potential: http://wwwen.zte.com.cn/endata/magazine/ztetechnologies/2010/no5/articles/201005/t20100510_184418.html
Economic-y Research Reports
 (everything as in: hardware (schematic to gerber files), possibly our applications for various funds, software (data visualization), firmware (microcontroller code), businessware (using os shopping cart software like prestashop or zencart), research), documenting everything not just the final product — everything we learned and fixed)