The API will be provided by Pachube.com and every datapoint will be saved forever, accessible at all times in full resolution in CSV, XML, or JSON via REST or real-time sockets. Everyone will own their own data and can control access to it via their Pachube account. The default for everyone's data will be "Public", and so you will be able to use the Pachube API to request every datastream in the world tagged with "AirQualityEgg" or by geography, etc.
For the DIY kits, you'll need basic soldering skills, a soldering iron, and maybe some wire snips. We just did some workshops building these up, so there is some documentation that might be helpful for you as an example here:
We'll certainly build this out more, but you can see the process there.
The sensors will not be calibrated and their precision and sensitivity is mediocre. However, we believe that generating data, even poor data, will not only contribute in a significant manner given the scale and density we can achieve, but we are also giving a community of people who previously had no voice a way to participate. You'll notice that the Air Quality Egg project was not defined as a way to gather precise, scientific air quality data. It was defined as a way to give PEOPLE a way to PARTICIPATE in the conversation! More here: Pachube blog posts: http://blog.pachube.com/search/label/air%20quality%20egg "Crowdsourcing data accuracy" : http://blog.pachube.com/2011/06/crowdsourcing-data-accuracy.html
YES! While the data collection model does draw strength from the density of users, there is no better way to say "I CARE ABOUT THIS ISSUE" and start a movement in your own home area than sharing data about your own air quality where you live. We can also work together to use this project to address specific local issues in specific places. You should tell us your story in the comments.
The nature of Kickstarter is that we are raising funding for a project that hasn't been completely designed yet. So, we will most certainly take into account the need for this thing to exist out in a varied environment. We don't have exact specifications yet, though.
Don't see the answer to your question? Ask the project creator directly.