I am new to Spark and the cloud API. How can I use my Spark core to automatically login to a particular website using my credentials and store particular personal or profile information that I request from the website in it’s memory?
The Spark core does HTTP but not HTTPS, so the data you send to or receive from a host over the Internet over HTTP is visible to anyone looking in.
But the Spark core does have a secure connection to the Spark cloud, so depending on what you want to do, that may be the best option. There is a feature in beta right now that lets the Spark cloud server act on your behalf with other Internet hosts over a typical HTTPS connection.
So the flow is your core talks to the Spark cloud which then talks to an website on your behalf. As you might imagine, you might have to give the Spark cloud your credentials to the other website in some manner (OAUTH would nice!) so it can act as your agent.
Maybe you should explain more concretely what you are trying to do so we can suggest a way to do it?
Hi @bko first of all, thanks a lot! That knowledge is really helpful and pertains to what I want to achieve. To be more specific, what I basically want to do is very similar to what certain third party apps do by storing basic user info from servers like google and facebook. I want the Spark to be able to through the cloud, as you suggested, connect to my google plus or facebook and pull out my profile information like name, interests, groups, etc. and store it in the Spark. Let’s call this Spark X.
Suppose I have another such device Y which is connected to a Facebook page I manage. Now consider a situation where the user on X wants to like the page on Y. What I want to be able to do is just by bringing X near Y, the user should be able to automatically like that page.
Can you advise on how I can achieve this and if there are certain resources that would be helpful? Thanks much!
Wow that is a big project! There are lots of security issues with user credentials to consider, plus there is the proximity detection aspect of knowing when core X is “near” core Y.
There has been some very similar academic work at the MIT Media Lab and a spin-off called nTag now apparently part of Alliance Tech. You can read about that here and here: