The Context Engine
Consider this a bit of science fiction, but don’t leave it at that, because we’re not that far from any of this really working out. The basic premise is this: imagine a whole new “layer” to our human/environment operating system such that we apply “context” to more things. I know it sounds abstract, so let me go into some examples for you to think about this in action.
Coffeeshop
Chris walks into the coffeeshop. His wristwatch glows blue when it detects a WiFi network. Software scans the available networks for “free, allied, or unknown,” in that order. There are no free networks in range, so Chris’s context engine sends a question to his wristwatch (only because it’s likely Chris is still viewing this screen; the same message could go to his phone, tablet, etc). Question: “cost to connect, $3.99 for 2 hours. Yes/no.” Chris clicks “yes” on his watch and the context engine handles the payment.
The same transaction today takes a LOT more hassle, and yet, the above-mentioned isn’t un-doable. I didn’t use anything more than a program and a set of rules and perhaps Bluetooth to accomplish that scenario.
Social Network
Chris’s context engine has 3 favorite orders for this particular coffeeshop chain. He clicks the second option on his phone, and waits to collect it, making smalltalk with the server. Meanwhile, the context engine has noticed that 14 friends within 5 minutes distance of the coffeeshop have revealed status and location information favorable for a visit. The engine offers up a “meetup” option, with checkboxes next to each person’s name. Chris selects 3 of the 5 and invites them by for coffee and a chat.
Why should I have to do all the ping/acknowledge to get that done? Couldn’t we have robust status databases that our devices could ping to determine all this? Sure!
Bookstore Browsing
Chris leaves the coffeeshop after talking with his friends a while, and heads to the bookstore down the street. The context engine reads the storefront’s RFID and launches an auto-wiki with Chris’s latest books of interest, recommendations his friends have blogged or tweeted lately, and then gathers via api comparison shopping data for 3 online stores. The context engine grabs a store map and highlights where the first 5 books Chris might choose are located, just in case.
Nothing stopping these transactions from happening. With more places having databases and RFID, why not permit locative storefront data? Why not make it easier for me? And auto-wikis? Makes sense if they can somehow be human-curated. I’m thinking of Mahalo, in this case, only a little more…spontaneous and flexible.
The Point of All This
We’re not using the power of data and we’re certainly not using devices that are aware of what’s around us. We use GPS for directions, but stop there. We use lots of manual effort where we could have something that makes an “easy” layer on some of our more common transactions.
Without a lot of heavy lifting, but some innovation, we could make some things happen that might really change how we interact with the physical world by using data and networks and context (probably what we call “business intelligence”) software. And in that world, it is still the connectors who will be helpful to the process.
What’s Needed
We’ll need strong security to protect this context data, especially if you consider some of the other neat ways this could be used (giving us real-time awareness of our spending patterns; GPS-location data released; etc). We need lightweight databases ( LDAP?), and we need some kind of context rules engines. We needs better APIs for our user interfaces, and wider adoption of standards to make passing data easier.
What do you think?
What am I missing? What could you add? What scenarios can you see this empowering?
If you enjoyed this post, please consider leaving a comment or subscribing to the feed to receive future articles delivered to your feed reader.
Comments
Chris walks into a bookshop. His “context engine” notices that he’s in a bookshop, and applies a whole bunch of rules to work out what he might want to buy.
It presents the first three choices, showing:
- The location in the store
- The price in the store
- The price on Amazon
It also notifies the store’s engine to say “Customer who’s come in is a regular book buyer - we’re going to recommend him X, Y, and Z - will you price-match Amazon.”
The till flashes this up, and the server clicks “yes”.
5 seconds later, Chris’s watch pings to say “The Store will price match Amazon on Y and Z” and Chris thinks “why the hell not?”
[…] Carlos Lorenzo had some great ideas on this topic.You can read a snippet of the post here.The context engine reads the storefront’s RFID and launches an auto-wiki with Chris’s latest books of interest, recommendations his friends have blogged or tweeted lately, and then gathers via api comparison shopping data for 3 online … […]
Hi Chris,
That’s a fantastic vision (especially the coffee shop view)… Really got me thinking. Perhaps the cafe should understand it’s not just “we sell coffee and biscotti” but “we’re a great gathering place for the community.” I live in the middle of an incredible neighborhood with three Starbucks within view of each other, as well as a small mom-and-pop java establishment a half block from home (the opposite ends of the spectrum.) Sure, they offer WiFi, but the signal’s often weak (only one box, and that’s in the office) and the stores can get really crowded. WiFi banners seem to be used to get people through the doors. The chains are too big to change, and the mom-and-pops are too small (and vulnerable) to devote the time and money into expanding into new realms. At least, that’s the way I see it in my neighborhood - which has a WalkScore of 100 out of 100! http://www.walkscore.com/get-score.php?street=1401+17th+Street%2C+NW&go=Go
My reality check today was to take my laptop to my local wifi enabled coffee bar. Only they charged £10 minimum for a connection. And then the noise of screaming kids with their mums shouting above them made me realise that out of London UK coffee bars don’t even know how to provide for a single market segment (nothing there for the kids to do and no good for business users) let alone have any idea how to extrapolate that to provide a better service.
Maybe Second Life is the place for such innovation. Then when the technology back office systems can support it in RL, a few bright sparks can make it happen. I look forward to that level of interactive service.
[…] our experience of life would be like if the things around us were put into context for us using a Context Engine. The basic premise is this: imagine a whole new “layer” to our human/environment operating […]
[…] a year ago, Chris Brogan wrote some science fiction: Chris’s context engine has 3 favorite orders for this particular coffeeshop chain. He clicks the […]
Great ideas, but the big hurdles we have in the near future is that security, and above that, the public’s comfort level with that security. It’s been a process to get the mass to even “store” their money in an electronic form and give up the old check book. And the idea of tracking via GPS and “somebody” know where you might be all the time scares the bejeezus out of people. It’s all great as long as it works. And our massively context driven device doesn’t get a blue screen of death.