The Open Pipe Kit (OPK) is the missing plumbing between sensors and databases that will empower thousands of data journalists and civic hackers to collect data without needing a programmer's assistance or being locked into one data platform from a proprietary turn-key solution.
Until the OPK reaches 1.0, it is intended for developers only. See the OPK Roadmap to 1.0.
OPK isn't a framework, it doesn't have plugins. It's the opinion that we should build drivers for sensors as Command Line Interfaces (cli) in any language we want. Follow a couple of well thought out conventions (like pulling data from a sensor should be the pull command) so other humans (and computers) can follow you. We're going to build amazing things.
Every sensor or database driver is a Command Line Interface. Sensors have a standard pull command, databases have a stand push command. They both have a help command that shows the parameters for push or pull as well as the options for each of those parameters. The help commands are how the GUI is going to be able to display a form to the user that empowers them to configure pipes without having to go down to the command line.
| Name | Type | Status | Link | Description |
|---|---|---|---|---|
| simpleserial | sensor | prototype | https://github.com/open-eio/opk-simple-serial-cli | Pipe sensor driver that gets values between x and x found over serial |
| phant | database | prototype | https://github.com/open-eio/opk-phant-cli | Pipe database driver that sends data to a Phant service like http://data.sparkfun.com. Phant project page here |
| temper1 | sensor | prototype | http://github.com/open-eio/opk-temper1-cli | For the temper1 USB temperature sensor |
OPK will run on Linux-capable devices. So far, most of our development has focused on the Raspberry Pi; but the approach should work on a variety of Linux platforms (including: your laptop, the Beagle Bone Black, etc) with slight modifications.
We're keeping a list of 'recommended hardware' here.