Week 4: Android client: work out OAUTH signup procedure.
The icecondor client has significant functionality without using the server. It displays the current location on a map (gmaps app does this too) and has the lat/long at the bottom (hard to read). It is a georss reader - give it a list of RSS feeds and it will poll them every X minutes. Using the geo:rss tags, it will put a marker for the most recent entry from each feed.
Once the position updating is turned on, then the openid kicks in. This is delayed in the user experience because a) icecondor client should by default not be reporting position, and b) reporting position means authenticating with the server which is a multistep process. Its this process that I've been working on during week 2 and 3. The server side has very basic OpenID and OAUTH functionality now. I'm moving back to the android side development to implement the other half of authentication.
Week 5: Client graphics update
The graphics in the android client need an overhaul. The app logo and a status logo with about 6 variations on the status logo.
Week 6 Add subscriptions / Amazon FPS integration
a subscription covers a period of time. it is created by the reception of a payment received message from FPS. during
the time period, features are enabled. the current feature idea is long-term history for $2/month. free option is 12 hours
of history (good enough for many needs).
Week 7: Announce. Silicon Florist, etc.
The server capacity will be an issue here. I would like to have the launch stay at around 100 position-reporting users, max. though i don't
like the idea of limiting access. i would use appengine if it supported ruby.
a subscription covers a period of time. it is created by the reception of a payment received message from FPS. during the time period, features are enabled. the current feature idea is long-term history for $2/month. free option is 12 hours of history (good enough for many needs).
Week 7: Announce. Silicon Florist, etc.
The server capacity will be an issue here. I would like to have the launch stay at around 100 position-reporting users, max. though i don't like the idea of limiting access. i would use appengine if it supported ruby.