Tag Archives: context automation

Ideas for context automation at BYU

Posted by on 10 May 2010 14:17 6 comments

  • The book you placed on hold last week at the BYU Library is now in for you to pick up. You receive a notification on your phone just as you enter the atrium. Want to check it out right now?

  • Your favorite stand-up comedy group is doing a show right now next to the Cougareat. Your phone alerts you to such as you walk through the quad just outside the Wilk. Care to take a look?

  • Your Physics 121 homework is in your backpack, all finished. Now that you’re in the MARB, your phone reminds you to run upstairs and turn it in before you forget.

  • You’re still on the other side of campus and class starts in 10 minutes. Since you’re on your work computer, perhaps you’re deep in thought on a project. Don’t forget to leave in time!

  • You’ve been searching Google, Wikipedia, and the Library’s website for 10 minutes with keywords all relating to Rembrandt and van Gogh. You don’t seem to have found anything satisfactory yet. There’s an art history subject librarian online right now. (She even has a specialty in Dutch painters.) Maybe she can help you break this mental block–care to chat?

Location-aware context automation is an extremely powerful and relevant concept. The possibilities are endless.

How do we make this a reality on a college campus?

BYU could use information I supply about my intent to give me relevant, useful information (call it advertising if you must): books I want to read, my favorite groups, academic interests, homework reminders, etc.

Such location-aware context automation is currently a hot topic at Kynetx, industry leader in pioneering the purpose-based, context-aware web of the future.

Kynetx provides the platform to enable the kind of context-aware applications I mentioned earlier.

What if BYU’s network routers were Kynetx endpoints? Then my smartphone (also acting as a Kynetx endpoint, probably through an app) could detect when I’m in range of one of these wireless access points. It can also trivially determine which building or perhaps even which classroom I’m in. That takes care of the location part of the context puzzle.

What if my laptop were a Kynetx endpoint that knew what building I was in and what I was working on–where I’m browsing the web, whether I’m online on my chat client, etc.?

Other data can be used to determine the intent part of my context–books of my Amazon.com wish list, books I’ve checked out or reserved at the library (both currently and in the past), classes in which I’m enrolled, clubs of which I’m a member, homework assignments that are due soon. Insert your favorite piece of data here. You get the idea.

Given that information, BYU could provide me contextually relevant information to make me more productive at my current location or the current time of day. That context automation could stretch across my web browser, my computer, or my smartphone.

That opens the doors for some really powerful applications that aren’t currently possible with the web as we know it.

What ideas do you have?

Kynetx and the spectrum of identity

Posted by on 29 April 2010 2:09 No comments

This article is also cross-posted on Kynetx Code.

This week was the Kynetx Impact Conference 2.0, held in Sandy, Utah. I met a lot of cool people there, and I thoroughly enjoyed the intellectually stimulation of being in such close proximity to so many brilliant innovators.

One topic of particular interest was treated by Chad Engelgau from Acxiom Corporation: the spectrum of identity.

Chad argued that in order for the web to work, we need to have an identity continuum that places anonymity at one end and verified identity at the other, with room for one or more personas in between. In light of Facebook’s recent announcement of the Open Graph Protocol, this is an intriguing idea. Facebook would like to do away with anonymity and personalize everything. But that model is broken.

From Chad’s remarks, I’d like to construct a real-world analogy of Facebook’s new proposal:

Imagine you walk into a grocery store. You have to scan your government-issued ID card before they’ll let you through the door. Once they’re sure you are who you say you are, you’re granted access to the store and are given a cart. It knows what kinds of things you’re looking for (either ones you explicitly declared or ones that are relevant based on your demographics, etc.) Advertising all over the store is modified when your cart rolls by to offer products and services of interest to you. This can make it very easy (but potentially very annoying) to find the things you really want and may or may not speed up your shopping trip.

Such a grocery store would be simply absurd. If you frequent the store, you will know what is being sold and where it is located; you’ll know what you need to buy. Preserving your anonymity in a grocery store is the most efficient and the most relevant way to shop.

Now consider going to a bank to take out a loan for a new car you just purchased. If anyone could just walk in to the bank anonymously and obtain a loan, the banks would soon be in deep financial trouble. They have a need to know who you are and what your financial background is before they will offer you a loan. In this case, anonymity is absurd; a verified identity is necessary.

If the real world works this way–built around spectrum of identity–why shouldn’t the web?

If Facebook’s new method of personalizing sites by using your full, “verified” identity everywhere was ever intended to become mainstream, it is a broken method. It is simply not necessary.

Chad Engelgau postulated that users of the web ought to be able to browse anonymously but still get a personalized experience. In some situations, anonymity is best; in others, one or more personas that may or may not accurately represent the “real you.” And in a few situations (especially where financial transactions or sensitive data are involved), a verified identity is absolutely necessary.

This is where the power of Kynetx and context automation enter the picture.

With Kynetx, users can browse the web anonymously without the need for some intermediary (think Facebook) to store and disperse personal information about you. The user can instead give information about relevant pieces of her context and allow Kynetx apps to leverage that information. If I’m Amazon.com and I want to show my visitors relevant purchase suggestions, I only really need to know what they’re thinking about buying or what they like to buy from me. I don’t care what their email address is or who their Facebook friends are or whether they use Visa or MasterCard. We can figure out the relevant details later when I need to know more about the user’s identity (e.g., when the purchase is actually made).

Facebook wants to kill this spectrum of identity by doing away with anonymity. While that brings some benefits with it, the model is inherently broken. How do we fix it? Kynetx.

Kynetx: This changes everything