Blog: Josh Lewis

A few years ago I had an idea for a website that would help its users to see which stores had a particular item in stock. I was tired of phone shopping for popular items, and I was even more tired of driving to a store only to find that they don’t have the one thing I’m looking for. It seemed simple: I want Item X for a reasonable price. I am located at these GPS coordinates. List the retailers that have Item X in order of ascending distance from me. However, it was pointed out that this idea would require the cooperation of each store’s inventory system, and it was likely that I wouldn’t ever get such deep cooperation from the biggest, most popular stores. So I ditched the idea.

When the software for what Apple calls iPhone 2.0 was introduced on March 6th, Apple spoke of a framework they were adding to the device called Core Location (optionally see the Core Location documentation on Apple’s Developer Center, registration required). The framework, according to Apple’s documentation, “lets you determine the current latitude and longitude of a device.” That data can then be used in applications on that device. (It is my strong opinion that locational awareness will be the defining feature of killer iPhone applications, just as connectedness and participation were the defining features of all killer desktop computer applications of the last 15 years, just as cheap desktop publishing was the defining feature of the 15 years of computing before that.)

Let me tell you about the iPhone application idea I had in the evening on March 6th. I’ve kept it under wraps until now, but I know now it’s pointless to stay silent for personal reasons I’ll go into later.

Let’s say you’re at the store. You’re going to buy a product. But you want to know if there is a better deal on that product at another nearby store, or somewhere else in town. So you get your iPhone out, open this application, and point the camera at the product’s barcode. It scans the barcode, talks to “the cloud” to see what the product is, and talks to Core Location to figure out approximately where the user is currently standing. If there is more than one store in the reasonable proximity of the user, it asks them which store in the list they are at, and asks them to either enter or confirm the price (if someone else has already entered it) of that particular product at that particular store. The user can then see other stores in the area (perhaps within a preferentially-set distance radius) and the prices they are charging for that product, with a quick click over to Google Maps to call the store and check to see if it’s still in stock, and to get driving directions to the store. The application would pay for itself after a few smart uses, and in this economy, I’d rather be selling aspirin software than vitamin software. This application is definitely an aspirin.

I think stores would start to scan their own stuff, or talk with the owner of the aforementioned “cloud” database for a way to quickly upload their entire inventory with up-to-date prices every so many hours. Users would initially power the system, and eventually help to fill in the gaps. (Don’t tell me relying on users to provide useful information doesn’t work. Look at Wikipedia.) The application has a great sense of participation and anti-”The Man” that I think people would really get into. Another boon: the search and matching algorithms could be updated to match products that are very similar to the one scanned (say, if they scan something purely functional and commoditized like bread or windshield cleanser) so that users wouldn’t have to make extra effort to search outside the brand they’re currently scanning. Perhaps it would eventually expand to items that were difficult to scan, like gasoline prices, and encompass everything a person would buy.

In late March I informally pitched this idea to Wil Shipley for reasons that will be obvious to users of Delicious Library. He liked the idea, but passed on the offer to take almost all of the profits from it if he developed it. I don’t blame him; ideas these days are a dime a dozen, but makers are hard to find. Moreover, talented makers are extremely rare and valuable. That’s why after thinking it through, I decided not to try to pitch the idea to more people on the hopes that I could make a profit from it without doing any work. That’s cheap. I know I can’t create this alone because I’m a “web guy” and I just don’t have the 40+ hours per week of free time this would require of me if the application was going to come out in 2008. I’d have APIs to learn. My Obj-C is extremely rusty. I’ve never really used Xcode seriously. All these things start to sound like cheap excuses though, don’t they? Maybe that’s what they are. I can’t tell.

So, if I want to use this application badly enough, I’ll round up a group of people and we’ll make it happen. Maybe it’ll be open-source and free. I don’t know how one would go about wisely leading such an effort, so if anyone would like to lend me a hand or give me advice, I’m wide open and willing to listen. All that I ask is that I be able to steer this idea from a very high level, and have a lot of input on the user interface and feature list. I think that’s where I could add the most value. I feel like I’ve already used this application I’ve thought about it so much. The only thing I don’t have is a name. I’ve heard many; they all suck. (Sorry, peeps who made suggestions.) Price Finder is too obvious (and probably too taken), and please kill me if I call it “Scannr.”

On a personal note, one of my defining struggles in life over the last few years has been to become the kind of person who would rather make than consume. To be a creator of culture rather than a consumer of it. To be a shaper of the world rather than allowing the world to shape me. I’m far from that goal, but I intend to continue pursuing it. Maybe by the time I’m 50 or 60 I’ll be there. That’s plenty of time to learn about more than just web development.