Why Promenade?

An iPad app built for two

Today my second (non-work-related) iOS app went live on the App Store. It’s an iPad app called Promenade. I’m quite proud of this one.

Promenade is an app meant for the family members or caregivers of people suffering from dementia and related illnesses. The idea is that the app can be used as a tool to stimulate conversation, revisit old memories, and make personal visits more engaging. It presents groups of photos based around selected topics, with options to view each photo at full resolution, but it's much more than just a frontend to an image search. The real magic is that it provides suggestions for 1) what to talk about for a given topic, and 2) what to talk about next in a way that segues well from that given topic.

Search for "shoes" and you'll get a list of questions about fashion. Search for "Sean Connery" and you'll get a list of questions about movies or celebrities. Promenade doesn't try to boil the ocean and match each word in the dictionary with a set of questions, but it does its best to figure out what the broad subject is, and provides a word association funnel where meaningful prepared questions can be provided for most searches. In addition, each search will also provide a selection of related topics so the conversation can move on from there. Search for "shoes" and you'll get boots, walking, hiking, sandals, high heels, etc. It’s not perfect though. Right now searching for “cupcake” brings up questions about television, because there’s a bunch of cupcake shows (Cupcake Wars, DC Cupcakes) that take precedence. There are plenty of other examples like this. Part of maintaining the app will entail a steady stream of continuous semantic improvements to keep this from happening as often.

That’s a brief overview of what it does, but the best way to get a sense of how it works is to try it out. Check it out in the App Store.

But I wasn't motivated to write this post just to talk about what it is, this is mainly about why it is. These are the reasons why Promenade exists:

Experience

I have a grandmother who had a stroke over 10 years ago, and has suffered from dementia (as well as other severe health problems) for quite some time, so I’m aware of how awkward and frustrating visits can be. I don’t want to dwell on this point too much, it’s quite personal, and it’s caused the family a lot of pain and grief over more than a decade now, but it’s important as a starting point to the whole thing.

Family

I have a sister who studied Gerontology at university, and now works as a Recreational Therapist at a long-term care home in Hamilton. She was the one that first suggested putting something like this together, and she’s really the main reason I ever found myself in a position to make the app. She talks with people in long term care every day, and she was able to come up with the hundreds of questions that are presented in the app. So let’s just say I have some subject matter expertise in the family.

Non-Technocratic

One of the things I was very cognizant of while working on Promenade was the fear of it being perceived as a technocratic solution. I’m very aware of the sometimes clumsy ways that technology is purported to solve all of society’s problems (teaching the homeless to code?). I liked the idea of Promenade as being software for two people to use together, not a video game or a distraction device to keep someone’s parent “busy”, but something that would enhance and encourage conversation instead of substituting for it.

Hacking Health

When this idea was still in its infancy, I attended the “Hacking Health” weekend conference in Toronto. This was a hackathon that brought together a large roomful of developers, designers, business people and medical professionals all interested in how technology and software could be directed to improve people’s health and Ontario’s health industry in general. It seemed like an ideal place to make a pitch, get a small team of people together with different types of expertise, and see if I could get the project started. Ultimately, it was kind of a disaster for me, and I got pretty much zero interest. I don’t want to sound too bitter about it, (and it probably has more to do with my ability to pitch an idea), but I was disappointed to see the lion’s share of attention going to “sexier” projects around social networking and crowdsourcing. I stood at a table for an hour, fielded a few odd questions, but eventually left the MaRS building with no interest, a bit of a chip on my shoulder, and something to prove.

UI

I’m an iOS developer, not a designer, and even though I have an “I have no idea what I’m doing” moment every time I open up most graphics software, I have some strong opinions about UI design for iOS that I wanted to put into practice. To make a broad, sweeping declaration: I don’t think native iPad apps generally look as nice as iPhone apps. Obviously there are a ton of exceptions to this rule, but in general, I find there’s a clash between the common UIKit elements (buttons, navigation bars, toolbars, search bars, etc) that seem very small and pushed towards the edges of the screen, compared with a relatively large central area for content that’s often quite empty. Apple has been preaching a “content first” approach to design with iOS7, which would seem to celebrate the very issue I see as being a problem. But with so many of the native iPad apps (iTunes, Messages, Notes, etc) this gap between the minimal chrome and the huge amounts of empty space usually found between the toolbars strikes me as awkward. To some, this minimalism may be attractive, but I find it off-putting. In addition, there are the usability issues: I find it hard to hit search bars, cancel buttons and toolbar buttons on an iPad (especially a Mini). If this app was going to be focused more on elderly people and their families and caregivers, I wanted to design a UI that would avoid some of those frustrations.

Design Decisions

Here are a few of the more interesting design decisions that were made during development:

• Only iPads are supported, not iPhones, and only in landscape mode. This application is best suited for a large screen. Two people sharing an iPad in landscape mode seemed to best encourage the use case I had in mind, so it was an attempt to “force” people to use the app a certain way. I also wanted to prevent situations where the screen would be rotating wildly after being handed back and forth between two people (do most people even know rotation lock exists?).
• An earlier version used Flickr as the main image source. Although the photos tended to be very high quality, people tended to upload certain “kinds” of pictures to Flickr, and I found certain things were over-represented, while other things were under-represented. For example, a search for “Greece” on Flickr provided a sensational list of photos that people had taken in the Greek Islands. However, a search for “Gardening” returned pictures of garden spiders. It’s important for Promenade to be good at returning images for more seemingly banal things like knitting, gardening, walking, shopping, etc, the kinds of things that don’t get uploaded to Flickr or other photography sites.
• YouTube presented a similar problem. An earlier version of the app included the ability to switch the media type from images to videos, but YouTube suffered from that problem even worse than Flickr did. Lots of videos of Beyonce, not so many videos of knitting. In the end, I wound up simply providing a link to search results on the YouTube website, to disassociate the app with the types of video results that get provided on the site.
• Everything in the app is large on purpose and requires explicit actions. Large text, large buttons, large hit areas, lots of ALL CAPS. I wanted to make sure that elderly users wouldn’t struggle to get around. Another example: early versions of the app allowed a user to tap a thumbnail to get the full photo, and tap the full photo to return to the thumbnails. However, I found that people often liked to “point” to certain things in the photo, which would dismiss the picture and annoy them. I switched to an explicit close button (the “new search” button gets appropriated and rotates from a + to an x).

Sales

I’ve spent a lot of time thinking and discussing the idea of how to make a more successful app. It’s a frequent topic of conversation with friends and coworkers, and I wrote an article about this earlier this year. It’s mostly a depressing read about how hard I feel it is to succeed in the App Store, and towards the end I mentioned that I was working on something as my best answer to the situation. That’s what Promenade is.

In the time since that article was written, the situation looks to only have worsened. Indeed, it’s almost accepted wisdom now in the iOS developer community that in general (with certain exceptions), charging for apps outright is a bad strategy, and that in-app purchase is the way to go. Whole episodes of the Accidental Tech Podcast have focused on this very issue. Partly as an experiment to go against that line of thinking just to see what would happen, and party because my idea doesn’t lend itself well to in-app purchase, I’ve decided to ignore those trends. I’m charging $4.99. It’s quite possible this is a dumb idea. But here’s what I’m counting on:

My Theory

I believe that consumers might be more willing to spend money when the purchase is directed towards somebody they care about, instead of being for themselves, especially if it’s virtuous/aspirational in nature. I think that aspirational mindset creates an opportunity to convince people of the value of a piece of software like this, since the whole point is to enhance something that people like to imagine themselves doing (going to visit mom or dad). Not the greatest analogy, but think of it kind of like getting a gym membership, something that you like the “idea” of, no matter how often you wind up going (but hopefully you go a lot!).

So this is clearly not for the Bay Area 20-somethings who want an app to get drones to deliver their burritos (or via Uber, if necessary). But there’s a much bigger audience than that. This market just seems a bit less crowded, but certainly growing (but not completely ignored, think Sudoku, “Brain Age” games, etc). Some good background reading about this is The Exotic Underclass from MIT and the Hacker News discussion around it.

All Long Reads Must Come To An End

Right now, I’m mostly happy to just have shipped something. My other iOS application in the App Store is Critic Match, a tool to figure out which movie critics your film tastes align with, which was providing some data that I personally found quite interesting, but that never really amounted to much in terms of downloads (although I did smile when 6 months later, Rotten Tomatoes came out with a feature that pretty much ripped off my idea completely).

Promenade, by contrast, isn’t really meant for everyone, and certainly not a group as wide and vague as “movie lovers”. It’s targeting a niche, but with an aging population and more people destined for long term care and retirement homes, it’s a large and growing one, and one that I feel is relatively underserved by available tools and technology. I’ve had great initial feedback from early testers, but now I’m putting it out there for everyone to see. If you find it helpful or interesting, I’d love to hear about it.