This is a response to Ashu Joshi's blog about about the Physical Web. Ashu and I actually have quite a bit in common and I'm very much aligned with his overall motivations (e.g. simplicity for the user) However, by slightly misunderstanding what the Physical Web is and how it would work, his concerns are misplaced.
First I want to thank him for taking the time to reply. The whole purpose of the Physical Web project is to be open and have discussions like this so I'm very glad his post is encouraging this debate. Also, I don't expect that my answers here will cover all of his concerns so I hope we can continue this discussion if I don't properly address everything correctly. Let me take his points in order
#1. It is about Category of Devices, not individual things
He makes the point that having an app for every device is absurd, it should be about categories of devices. This is easy, I agree. I didn't mean to imply every device and I'll look into changing the wording. However, even if I have to install 10,000 apps to control a billion, my overall point is the same. Installing apps proactively is a user hassle and a roadblock. I'm a UX designer, and this project was born when I was a Creative Director at frog design where we were doing user studies, getting our clients to app to be installed. It was torture. It was tremendously hard to get clients motivated enough to run the install gauntlet. I hope we can agree that no matter the numbers, having users get to the info/app with a few clicks is better than forcing them to install (and likely uninstall later) This is especially true if in the future, we expect there to 1000s more smart devices that will be entering our lives.
#2. App-less Interaction Complicates the UX
His basic point is the bookmarks are just as complex as app. Again we mostly agree. The physical web isn't meant to be just for the 10 things you'll do every day. It's sweet spot is when you want to use something once (e.g. a mall map, a parking meter, a bus stop in a new city) In these situations we're trying to take the core value of the web, a link, and apply it to situations that would require an app today. By making it that light weight, we expect it to be used in a variety of places. In fact, much like the web is long tail, we hope the Physical Web unlocks a long tail of interaction, every device can now offer a bit of info or interaction if you want. Of course, I'm not saying bookmarks are bad, just not part of the core experience. If you indeed do you 10 things regularly then by all means, bookmark them. It clearly creates clutter, like apps do, I'm not disagreeing there but it's still easier than installing apps as you're already there, it's a simple save action.
#3. Connecting and Discovering is just the start, the real challenge is in the System
Again we agree (noticing a pattern here? ;-) The Physical Web is really just a discovery service, we often say geekily that we just a form of "Proximity DNS" that's all. You actually have to build the service for the URL to point to it. His negative example is a bus stop system. Again, I agree it's a lot of work but people ARE building these systems now and I'd rather access it through a lightweight URL linked to the exact stop I want, than I app I have to navigate to, open, and drill down to the info I need.
#4. Separate the Data from the UI
His basic point here is that using the web implies that "User Interface is embedded in the device" OK, here I'll break with my pattern, he misunderstands the tech here ;-) The device ONLY has a URL, that URL points to a server which is super easy to upgrade.
#5. Use Cases should drive the approach
Here Ashu calls into question why a vending machine would ever use web pages when it could just use Apple Pay/GoogleWallet/etc. It's perfectly fine to quibble with the value of the vending machine, lots of people have told me they much prefer cash, etc. I actually think that using the Physical Web for a candy bar isn't likely the killer app, but it was a good example of walk up and use. As I stated earlier, this IS very much a user driven experience, we're trying to take the pain out of the user journey from "What is that?" to "I'm using it now" The long tail I spoke of earlier is where we feel the real value will be as everything is capable of offering tiny bits of information, not just apps. So yet, please criticize the vending machine, I won't fight you there. My point is that we're trying to do much more.
Finally, Ashu throws Steve Jobs at me, implying that we aren't a simple service. I'll admit, that one hurt a bit ;-) Here I don't think he really understands the user flows before/after. We offer a service that with 2 taps gets you to interaction with what you are standing in front of. All without pestering you or having any alerts annoying you in your pocket. Compare that with the steps necessary to install an app and I think we come out much MUCH simpler. We are very user driven here.
Again, in the spirt of the internet, it's so easy for things to become personal or negative. I'll reiterate that I'm very thankful for Ashu for writing his blog post for the simple reason if he feels this way, so do others. I hope it's clear this is an intellectual debate not a personal one and I fully expect Ashu to have a rebuttal to my points. I'm happy to continue the conversation. The point here isn't to take sides but to explain our sides and hopefully find common ground. Thanks again for your feedback Ashu.