Monday, December 18, 2006

Functional optimization

This will be the first of several posts about information appliance and multi-use product issues.

Place: reading a magazine in a store.

Design problem: Interface designers often face a conundrum when designing products. How do you get a lot of features in while still producing a usable product? The above is a real product review in a fairly well read magazine. They have placed an enormous swiss army knife in their overview of 'cool products'.

Design solution: Products fall into a complexity continuum leading from low to high. Lower complexity products (sometimes called information appliances) typically only do a few related things and can potentially have very simple interfaces. In locations such as a kitchen countertop or a large desk, many individual interfaces can be easily accessed and organized. However when we look at mobile devices, the ability to carry (or wear) multiple interfaces is dramatically reduced and the benefits of combination become more apparent.

The pocket knife above has gone so far down the complexity continuum that it has actually ceased to become easily portable. Furthermore the quality of the experience of using any individual tool has become so degraded that it becomes pointless to have that many tools. The opposite can also happen, such as in the iPod, which has an overly simplified interface (no added complexity to support advanced or long-term use) which is used to access a variety of complex features. All good design is about finding the right balance and optimising for what people need to get done.

Wednesday, December 06, 2006

Paper towel dispensers (continued)

Not to belabor the point of towel dispenser design, but as a followup to the previous post on the topic, I saw this very NICE dispenser design recently. The use of both graphics and text which directly tell the user how to interact with the device is very effective. Could a motion sensor-based design be this clear?

Thursday, November 23, 2006

Digital roads and accident avoidance

Place: freeway to portland

Design problem: Accidents and other unexpected blockages to lanes on freeways can create dangerous situations for approaching traffic. For planned blockages (e.g. road work) trucks with programmable signs are used to redirect traffic out of a blocked lane (see above). Car accidents that block a lane typically attract a police or fire truck which turns on its emergency lights. The problem with these solutions is the proximity to the accident. Given the speed traffic travels at, approaching drivers are very close the actual accident by the time they first know about it and they still lack a good understanding of how to avoid it. It also results in rapid lane changes which can cause additional accidents. All of this results in a greater probability of grid locked traffic occuring before the problem area and less highway efficiency.

Potential solutions: Oncoming drivers need to be informed of a problem long before an accident, informed what to do to avoid it, in day or night, and in a reliable way so the 'crying wolf' phenomenon common in road work signs doesn't occur. What if we had 'smart' roads? Roads that had imbedded led lights in the pavement could animate arrow flashes pointing out of a lane, miles before a reported accident. Lane sensors or cameras could auto detect small numbers of stopped cars and auto trigger avoidance systems and alert the police. Even if this cost prohibitive, road location labels could enable people to call in an accident location to 911 and rapidly have the avoidance lights triggered for the corect location. We need digital roads.

Monday, November 13, 2006

Interacting with an automatic towel dispenser

Place: Public bathrooms

Design issue: In an environment such as a mall or airport with users who may not read english well (kids, foreign travelers) how can a towel dispenser which is movement-activated best communicate to the user how to interact with it?

The first example image above tells the user exactly what to do, which is good, but it may be challenging for those who don't read english well. The second example tells the user how the device works, but not how to interact with it. It also uses a picture more commonly understood by most users and has a clear case which allows the user to see when the paper is empty.

Potential solutions: Combine the two interfaces. Show two hands in the proper orientation with an arrow pointing below. Also remove the time delay commonly found on these units which makes it hard to get two sheets of paper. Also add a light which blinks when a hand is detected (to help the user understand what the device is seeing) and add an out-of-paper blinking light that only comes on when the unit is empty. If this can't be determined reliably than the clear case is a decent solution. Interfaces that don't require touching are great for public areas and could be used more widely than they are.

Sunday, November 05, 2006

Efficient restaurant experiences

Place: san francisco food court

Design issue: Restaurants that don't do table service commonly have to announce upcoming orders to customers. In its more primitive forms this entails shouting numbers from behind the counter, customers trying to find order numbers on their receipts, customers not hearing announcements, and customers needlessly standing waiting for food to be prepared. Not an ideal customer experience.

Solution: Some wise soul figured out that pagers had become dirt cheap and that when integrated with a vibration unit and LED lights could work as a personalized alert system between the restaurant and a hungry customer. Now (in some locations) when you pay for your food you get a "smart" coaster which vibrates and blinks when your food is ready or when you are ready to be seated at your table. It is a simple and elegant solution I have yet to find a problem with.

Tuesday, October 24, 2006

Not the car for gangsters

Place: sequoia nat. Park, ca

Design problem: what if you found yourself inside the trunk of a car? How would you get out? Now, I know, You're asking how someone would get there in the first place... let's just say that maybe you took out a loan from the wrong person.

Solutions: our rental car has a safety pull cord inside the trunk. It shows a drawing of a person leaping out of the trunk on it. I'm not sure you could actually see this in the dark, so possibly a glow in the dark pull cord or possibly some mood lighting for unexpected trunk rides might be in order.

Friday, October 20, 2006

Misuse of multimedia systems

Place: westfield shopping center, san francisco

Design problem: the large floor map of the mall in the information kiosk is a large touch screen. It is placed next to a static listing of businesses and locations. Touching the map removes the mall floorplan and replaces it with a dialog box offering a variety of services such as searching for a brand or store name. However public information kiosks are often used simultaneously by multiple customers. In the current design if one person touches the screen, no other customer is able to get their questions answered until the first has completed their interaction. When I first saw the kiosk there was a crowd of people standing around it - not in amazement - but with irritated looks waiting for an elderly shopper to finish using the device. Most walked off without getting the information they wanted.

Proposed solutions: the potential for mall kiosk systems is high. They can show you how to get to a store from your current location, or let you search for a specific product and see if it in stock or compare prices (this kiosk did not do this). However they must support multiple users at the same time, be at least as efficient as the old paper posters for common tasks (e.g. Browsing stores in a category), and they should ultimately make the shopping experience better - not more frustrating. Multiple smaller kiosks would help, as would side by side paper and digital kiosks, so that the paper could be used while the digital one was busy.

The two photos demonstrate the new unusable touch screen and the old usable printed poster (located in different parts of the mall).

Wednesday, October 11, 2006

Getting off an aircraft

Place: on a recently landed plane in Florida

Design problem: A lot of people need to get off a plane in a short period of time, *every time* a plane lands. The current solution is to make one or two long single-file lines go out *one* door. Is there a worse possible solution to this problem?; only not letting people off comes to mind. The issue is partially how the plane is managed (the administration Southwest airlines) and partially how it is structured (e.g. Boeing).

Potential solutions: Many busy buses have two doors - with good reason. Trains also have multiple doors. Virgin airlines in Australia boarded and emptied planes from both the front and the back of the plane at the same time. This helps, but it only cuts the problem in half. The real problem is that people need to unclip their seatbelt, get into the aisle, find their bags, and check if they have forgotten anything - all while having a line behind them. A passing lane works on a freeway for slow motorists, why not on an airplane? Make the plane a little wider and the passengers a lot happier. And why not use three doors to enter and exit? Sure, it might require redesigning the entry ramps and planes, or deplaning people onto the tarmac, but the customers would get in and out more rapidly (which makes the airlines more timely) and customers would be less frustrated with their flights.

Thinking outside the box: another thing that would help would be to make the plane double decker, with bags placed in lockers in front of each passenger instead of above them. That way people could access their luggage without getting up from their seats during the flight (no more 'your baggage may have shifted warnings'), and you wouldn't have to fight for a free spot in the communal lockers. It also would mean that passengers wouldn't block the communal aisle from affording movement while they are getting their bags ready. In summary: at the time of boarding or deplaning, the aisle is a communal space which is having far too many tasks required of it at the same time; no wonder passengers get annoyed whenever they have to take a flight.

Tuesday, October 10, 2006

Safety lectures on airplanes

Place: southwest airlines flight, seattle

Design problem: Can you imagine getting on your fifth flight of the year and not hearing the same safety lecture for the fifth time? Wouldn't it be nice to just read your book in relative silence before takeoff? Does anyone actually listen to those announcements? I don't think I have ever listened to one, and I travel a fair bit. Furthermore I doubt anyone remembers significant parts of a boring repetitive safety lecture after they've just survived a crash landing. The current system is a significant reduction in the (already greatly diminished) quality of the air travel experience.

Possible solutions: The airline passenger is part of a captive audience who has to stay in a seat for long periods. Their ability to use devices (e.g. Ipods, dvds, games, phones) is highly controlled by airline staff, with little distraction possible at certain times. With so much time being spent bored, sitting looking at the back of the seat in front, why is this space not being used to convey safety information?

Remove the audio safety sermon and put in a simple graphical overview of safety instructions on the back of the seats. Even I would know the safety procedures if I had nothing else to read. Alternatively, have an automated system which verbally announces simple procedures in the event of a crash - just in time information. Plane travel doesn't have to feel like being in boot camp.

Saturday, October 07, 2006

Replacement of car parts

Place: Festival parking lot, Shelton WA

Design problem: GMC Jimmy cars have a plastic grill on the front of the car. A rock can easily fly up and chip or break portions of the grill leaving the car looking bad. Cost of repair is $24 for the letter. But cost of shop labor time to repair is $80. It turns out that replacing the letter requires dismantling the entire front of the car (because the pins holding the letter go far back in), and it requires special screw drivers which most people don't have.

Potential solutions: Make emblem letters that are cheap and dismantle from the front of the car; mold the emblem into the front grill so it can't fall out; use standard parts so people can fix their own cars; allow access to the front grill from inside the engine compartment so repairs can easily be made; lastly make it easy to replace all small parts on the car should something happen to them.