Use cases
From Pelennor
This article will attempt to theorize on a variety of computing scenarios experienced by end users. Focus should be on how Pelennor's envisioned data/information architecture makes them possible. After all, anyone can dream up futuristic ideas. The real work is in bringing them to reality. Pay special attention to Pelennor's guiding philosophies at work.
Some of these ideas will be composed in a rather "stream of consciousness" manner, in order to illustrate the change of thinking required for a successful user experience. When thinking of futuristic technologies, it is easy to jump to faulty conclusions on how they would be optimally used. Just look at the naive assumptions made in 1960's sci-fi movies!
After describing each scenario, we will break down the problem and see how Pelennor technology can help solve it. Since we are foremost concerned with the need for a universal, distributed data architecture, it seems fitting to analyze all of the diverse input data involved with the scenario - often data which is totally inaccessible to today's computers!
Personal computing
Distributed scheduling and social planning
Diet and health management
Automated price shopping and shopping trip optimization
Casual correspondence with family and friends
Electronic entertainment
Interaction with local community groups
Neighborhood community building and collective security
Travel planning and optimization
Automotive navigation
GPS navigators have gone mainstream, but they offer little help when trying to find a store or restaurant that opened a few months ago.. or when major roadwork has completely changed the layout of exits and onramps. But imagine if your navigation data could be updated in real-time while you drove -- not only through a mobile internet connection, but from short-range data "beacons." Now, think way beyond "computerized maps." Imagine being able to ask your navigator, "Where is the cheapest fuel station that serves good coffee, within 15 minutes on my route?" Better yet, imagine if your latte could be prepared and paid for by the time you got there. But wait! We've already violated the principle that technology should be unobtrusive. We should not have to ask about fuel prices or route optimization. These are given concerns which do not change. Neither do we need to specify "good" coffee. Let us simplify the initial user interaction to "I want coffee." Now it is up to the computer to determine the rest: Do you need fuel? Are there travel time restrictions? What is the probability of delays that could offset your schedule? What are your economic preferences? (save time/fuel at expense of favorite coffee?) In the end, the only remaining user interaction is to choose among calculated best scenarios, as prepared by intelligent software agents.
Information needed for this use case to be seamless:
- current location
- final destination
- road maps surrounding the path from current location to final destination
- average traffic rates on these roads
- probability of various delays along this path (accidents, signal lights, etc.)
- current fuel capacity remaining
- fuel economy of the vehicle under different conditions (windspeed, grade of road)
- weather conditions
- identity of driver
- personal coffee brand and type preferences
- personal economic preferences
- hard and soft personal schedule deadlines
Pelennor technology breakdown:
Roadway automation
Today's road systems are quite inefficient at optimizing traffic flow and have little or no ability to handle anomalies. Most traffic signals still operate on timers preset for estimated throughput during the day and at best have proximity sensors. How many of us have stopped at lonely red lights at 1am or waited out a "no turn on red" even though there were no vehicles for a half mile in the opposite direction? How often have police had to manually operate traffic lights after concerts or sporting events? Why do radio announcers still need to provide traffic and accident reports during rush hour?
Suppose that our vehicles could "talk" with the roadways. Imagine if sensor networks were combined with our heading and intended route, as shared by our navigation computer. The amount of time and fuel saved by constant routing optimization would be outstanding. Traffic lights would more often be green when they "obviously should be." Our navigator could tell us to slow down or speed up in response to changing intersection timings or routing changes. (Why race to a red light? Why drive at 65 mph when construction has traffic backed up 2 miles ahead?) Speed limits could become dynamic to account for road conditions, traffic density, and traffic jam avoidance. (Drivers are less stressed by slower speeds than having to stop and wait.)
Automotive safety
Roadway automation in itself would dramatically increase safety by moderating speeds and reducing driver stress. However, we can take this concept much further. Our smart roadways of the future include slickness sensors, visibility sensors, and hazard beacons. Likewise, suppose that vehicles involved in an accident could beacon their state as a warning to others nearby. Of course, they would also notify police if the collision was bad enough or if the vital signs of any occupants were in danger. Not all roads will be outfitted with elaborate sensor networks, but future vehicles can provide plenty of safety data themselves. If onboard accelerometers detect poor road conditions, this data can be uploaded to municipal repositories for retrieval by other drivers and to prioritize work of road maintenance crews, from snow plowing to pothole repair!
The elusive "kitchen computer"
Researchers have been trying to find a way to put a computer in the home kitchen for decades. It seems to make sense, right? Wouldn't everyone desire easy access to recipes, technique references, and nutritional data? But it's not that simple! Past attempts have produced designs that are more hassle than they are worth. Indeed, cooking is more of an art than a science, and today's software is simply too rigid. Real cooks substitute ingredients, experiment, make notes in the margins of their recipe books, adjust temperatures to coordinate timing of multiple dishes, and elicit random help from family and friends. Who wants to bother fussing with a mouse or even a touch-screen while their hands are covered with flour or while preparing a delicate zabaglione? And yet, what if the computer could disappear? What if it could simply become a virtual "master chef" looking over your shoulder and ready to help if called upon.
Imagine a kitchen of networked appliances, measuring tools, RFID scanners, cameras, and conveniently located screens. Suppose that most human interaction is passive, because the system, as a whole, is intelligent enough to provide only the information relevant to current tasks. Any other interaction can be accomplished by simple voice commands or by using a simplistic countertop touch-screen. Lets walk through an example of the user experience:
Personal financial management and planning
Why should taxes be a yearly chore? Why should loan and insurance shopping take more than a few minutes? For that matter, why should anyone have to input data into a personal finance manager or even balance a simple checkbook ledger? All of our personal financial data is currently available, somewhere in the world, in various electronic forms. The problem is that there is no way to automatically aggregate it for processing. We welcome the day when bills, taxes, and other financial hassles waste no more than an hour per year. Such time will be used mostly to make simple policy choices and agree to pre-calculated optimization strategies.
Of course, financial planning goes far beyond keeping accounts and records in order. Budgeting and savings plans will be much easier with all financial data aggregated automatically. For example, software may generate possible scenarios to consider, after presenting current spending patterns. Afterward we might ask to be warned of purchases that will stretch the categories we agreed to cut back on. Long term financial planning will greatly benefit as well, especially in the area of investments. Subscription market data, news, and screening services will be much easier to integrate. And our virtual advisor will be ever-present to inform us when a market decision must be made. (once again, redefining the "human factors" at play in the stock exchanges!) On a daily basis, we will also find ourselves presented with practical wisdom, such as, "Given your average investment returns, that fancy HDTV you're about to purchase carries an opportunity cost of $75,000 by the time you retire. According to expected price trends, that cost will drop to $38,000 if you can wait another year to buy."
