Week 2

Published on 25 January 2015

This is a post about weeknotes.

Spotify and Radiodan

Made a start on a Spotify radio using the Radiodan framework. Dan had already done the hard work swapping out mpd for the mopidy player . It was a bit of an effort installing all the dependencies via homebrew and then installing the mopidy-spotify extension. However, once that was done all it needed was my spotify username and password (requires premium). The plugin doesn't support loading playlists via their spotify URL e.g.spotify:user:andrewnicolaou:playlist:7va6oap6VjoxxlapQq5aCT. You need to do mpc lsplaylist to get a list of playlists with their displayed names and then mlc load "<playlist display name>" which then works. I'm hoping that perhaps through radiodan I might get more detailed objects.

Working towards an idea of a synchronised, virtual office radio. All the radios would share a collaborative playlist that you can manage via Spotify. They all start playing at the same time and stay in sync.

Documentation time for Radiodan

Reviewing Libby and Dan's excellent docs.

Team review

First of the year. Making plans for the next quarter. Used my giant planning poster from last week. The layout confused people a little but the plans themselves were fairly well received. Seems like things are going in the right direction.

Shuffle

A prototype that I worked on a while back is still generating quite a bit of interest internally. We used it for a user trial and it still has that utilitarian design. Thinking of design ideas for new landing page that are bolder and more "televisual".

Project Ara

A trip down the road to Google's fancy central London office. Inside the Renzo Piano building that looks like a colourful technical drawing of a building. A live stream of Google's Project Ara Module Developer's Conference.

Regina Dugan gave a great general keynote about power of platforms and how greater levels of abstraction always have performance penalties. But the benefits of greater diversity of ideas outweighs it in the long run.

They're trying to scale up some impressive technology - inductive communications between modules so that there's no wear and tear. Electro-permanent magnets that can be made to switch state between stronger and weaker force using an electric current but that take no power when in either state. UniPro, a packet switched device bus.

They also have impressive timescales and ambitions, a 2 year project to have "demonstrators at convincing scale" including a market trial for selling the phones in Puerto Rico, creating a 2 sided market bringing together module manufacturers and consumers.