Last week, the SAMI team’s Dr. Yujing Wu hosted a local meetup in Sunnyvale, CA for developers who wanted to learn more about SAMI and what makes it special. In this post, we share the slides from the presentation and highlights of the Q&A that followed. Read on!
In her presentation, Yujing explains what we call “silo systems” (devices and data made inaccessible between different vendors) and how SAMI provides true interoperability. The case study walks through SAMI device and application development separately, as an example of how you can bypass building a complex end-to-end system and focus instead on your area of expertise.
Beyond interoperability, with data from diverse sources (e.g., digital health, smart home, smart car) all coming to the same place, SAMI affords service providers a unique opportunity to discover and deliver rich insights about users in an unprecedented way.
After the presentation, we received a number of insightful questions that we would like to answer below for your benefit:
How can I create my own custom analytics?
Your service makes API calls to access the data from SAMI in real-time or in bulk. SAMI currently provides simple analytics such as aggregates and histogram, but a few partners have worked with us to provide real-time analytics services to their users. Privacy is very important to us and developers need to get permission from users to access their data.
Are there limits on data storage?
We do not have storage limits at this point. However, we have rate limits. In addition, message payloads are limited to 10 KB.
How is SAMI different from its competitors?
SAMI is a fully open ecosystem designed to break silos. There are a few open platforms that focus on message/transportation services between diverse IoT devices (e.g., iOS/Android), and others that provide storage or analytics services for IoT devices. However, in these cases the source and destination devices are developed by the same company, and interoperability across vendors is not possible.
Other platforms do provide an ecosystem that allows interoperability across vendors. However, these platforms only focus on one type of data (like fitness data) or one platform (like iOS).
SAMI is agnostic and can take in any kind of data, from any device, in any format. This is the purpose of the Manifest that you define. Likewise, SAMI includes extensive APIs for searching data in real-time and historical modes, as well as tools for managing devices and applications.
Finally, SAMI is a private data bank: users are the sole owners of their data.
What commercial products has SAMI integrated with?
We are actively integrating with various partners and are covering this on the blog—see the integration tag!
In the presentation you showed how to make API calls to get individual messages. Can we get large amounts of data from SAMI?
Yes. Depending on your needs, you should consider WebSockets for high-performance, real-time access to data. SAMI also offers an export API that is ideal for extracting data in bulk for offline analysis, for example using MATLAB or SciPy.
To hear about future meetups, join our mailing list below!