Leap Motion launches Skeletal and Bone APIs for tracking, replicating hand motions. Hack for LA faced limitations in data quality, crimping APIs. Plus: DocuSign's first mobile SDK, and Google Chrome phasing out Netscape APIs.
Leap Motion Announces API for Tracking Hand Gestures
Leap Motion, which sells a tiny hand tracking device for just under $80 that can serve as means to control games and other software, has released a skeletal API aimed at spurring integrations. The API allows for full freedom of all fingers, while a new Bone API can extract data once the motions are recorded and project them onscreen. With one, users can control objects on screen. With the bone API, they can see their hands on screen; their online hands following their own, and can manipulate objects on screen, or rather command their digital hands to do that on screen.
As David Conrad points out in I-Programmer, there are some sticky points with how the program handles occlusions, which will no doubt get ironed out. But,
The most important part of the API, however, is that it is getting feedback from early adopters that sounds more like the sort of thing the original LEAP launch should have generated. It seems to work much better than the original tracking API and it seems to be easy to use.
An AirSpace store lists apps and games that take advantage of the Leap Motion technology.
Hack For LA Reveals Limits of Datasets
Your API is only as good as your dataset. That's one lesson coming out of last weekend's LA hackathon focused on using city data. The expanded city data, unveiled just as the hackathon started, wasn't comprehensive enough to be useful, many hackers found.
As Soumya Karlamangla writes in the LA Times, the hackers mostly sidestepped the city data site:
Zach Latta, whose team won first place for its app that connects homeless shelters with volunteer groups and restaurants, said there weren't enough data sets available related to homelessness, and the ones that were available weren't comprehensive enough for what his team needed. They instead relied on crowd-sourcing data to build their app.
This is likely a sign of early days in the civic hackathon world. The runner up did manage to use data on city water to create its API. One problem was that a lot of the data provided was actually summaries of data--bar charts and so on--when what was needed was the actual datasets. To hack a hackneyed phrase, the data is in the details.