Barry Po (NGRAIN) How Much AR is Enough? Lessons learned from Live Examples of AR in Practice

Editor's Notes

• #2: Show of hands: How many of you in the audience are in product or sales in your organization? The reason I’m asking is because I wanted to share with you an experience I’ve had, and am curious to see how many can relate. From time to time, I’ll get an email in my inbox that reads like the following:
• #3: The email will read something to the effect of: [Read email] And when I get these emails, my first reaction is — “YES! New sales opportunity” — right? Like any true product/sales guy. And then I read it again, and then reality starts to set in. Then it occurs to me: Uh oh… here’s another prospect who’s interested in AR, who saw a great concept video, and now thinks that AR is going to turn him into:
• #4: Tony Stark — I mean, we, as an AR community, have done such an incredible job of selling people on the promise behind augmented reality that there are increasing numbers of people out there who really want to make it happen, whether that’s where they play, or as we want to focus on here, where they work. The problem is that the way AR is often pitched today, whether that’s in concept videos, or the way popular press or media portrays it, sets the stage for massive disappointment.
• #5: The reality is that AR is on a fast track to look more like this. All your content, everywhere, all the time — no way to turn it off — a total living nightmare. And that’s assuming you can even get the technology to work reliably regardless of where you are, and can get past the hump of getting it setup properly.
• #6: By the way, you can’t make this stuff up — there is an article that’s been making the rounds in media circles over the last couple of weeks, and I thought the Gizmodo version captured it quite aptly: This Augmented Reality Future looks like a living hell. “It’s fun for the occasional tech demo, but now we know the reason AR solutions have failed to catch on. The future AR solutions promise will eventually turn into the same nightmare that surfing the web has become." And I think it captures in a nutshell, what it is that we, as a community of people deeply invested in the technology, are on the verge of creating:
• #7: Technology that is all about the technology, rather than technology that is meant for people. Now, I don’t necessarily agree with the idea that the “living hell” is a future we’re destined for, but I think it should give us something to think about: How many of us can really say we really understand the use cases for AR in the enterprise, and even more importantly, the end users who would use it?
• #8: You see, I think it’s easy to buy into the notion that we know our users, we understand our audience, and all we really need to do is build the tech, and magically, the world is a better place. That the future looks like this worker here — we are on the cusp of building technology that is going to propel us into a future worthy of science fiction And that this is the kind of environment we can all look forward to working in
• #9: Well, as unsettling as it is, the reality is that today, many of the beneficiaries of AR in the enterprise world actually look much more like this. It’s dirty, it’s dangerous, and it’s thankless, hard work. I’d like to share with all of you an experience we had at NGRAIN, working to bring AR to an oil & gas company — and some of the things we learned.
• #10: We were first engaged by our client about two years ago now, and the problem statement was seemingly straightforward: They have very expensive equipment crucial to their extraction operations — problem is, because the equipment is so hard to maintain, They could only count on it working 20 weeks out of every 52, and the downtime was absolutely killer.
• #11: The closest analogy would be like any of us buying a car, only to find out that we could use it 3 days a week, because it would spend the other 4 in the shop — every week. It would probably make you very upset, right?
• #12: Well, like any good AR tech provider, we were confident that we could deploy a solution that would make a world of difference And so, we set off to do that: we said, let’s go, meet our end users, show them what we can do, and we’ll be on our way to building an incredible business case for AR And we did. Some of our team at NGRAIN set off on a three week assignment to live and work alongside a maintenance team at a remote location so we could learn what it meant to be a maintainer in the oil & gas industry And we learned a lot.
• #13: We learned that these maintainers would often find themselves working in frigid, sub-zero temperatures, sometimes as low as -22 degrees F (-30 degrees C). Temperatures so low in fact, that they would fill out checklists, work orders, and do all their “computer stuff” in their trucks instead of doing it in front of the equipment they were supposed to be working with, because it was nicer to do it someplace warm.
• #14: We learned that these maintainers worked for extremely long periods of time, and in fact, frequently opted to work even in their off shift hours because it was boring and there was so little to do in their spare time.
• #15: And we learned that these maintainers often found themselves working in dangerous, dirty environments — to the point where it could be considered “routine” or “common” that maintainers would lose fingers, limbs, or even their lives.
• #16: What we got out of that whole experience was something much more revealing: despite our own self-confidence in our technology and the promise for AR to “change the way people work,” we came to the realization that we didn’t really have a clue as to who our end users were, or what they needed, or what an AR solution needed to be.
• #17: Which leads me to the title of this talk, which some of you may have noticed, I haven’t really talked about up to this point… “How much AR is enough?” Well, like all things – the answer is “it depends.” But, it might be fair to say the answer is definitely not “all the time.” You see, in the time we spent with those O&G maintainers in the field, we learned some very important lessons about the true “barriers to entry” for AR, and that challenged us to look beyond some of the naive assumptions we were making about what an AR solution needed to be, to be successful. I want to share three insights based on our experience…
• #18: The first assumption we make as a community is that someday, it will be possible to build an AR solution that is robust enough to be used in practice. This is something I think we can all relate to — “Don’t worry, we know the technology isn’t mature enough today, but it will be one of these days." Well, when your incumbent work environment requires you to deal with ice storms, corrosive, or explosive materials, and random movements of the earth, it would be fair to say it doesn’t matter how mature your tech stack is, it is an extremely tall bar to be mature enough to deal with all of the uncertainty inherent in these work environments. 
• #19: In trying to use AR in the field, we came to realize one thing pretty quickly: it’s dirty out there, and the world is full of unexpected surprises Tracking didn’t always work, not always ideal to use markers, and you can’t even assume they have Wifi (they don’t, but they might be getting it next year)
• #20: Key insight: Assume the worst – be prepared for when AR doesn’t work, because even if it doesn’t, your user still has a job to do
• #21: And in the scenario we found ourselves in, it became evident that “graceful degradation” was a useful quality in an end solution — including things like offering: End users the ability to swap into a virtual 3D experience on demand Animated video content instead of an interactive experience Access to even basic PDFs or images with the most crucial information By offering end users multiple alternatives, we can at least start to build some assurance for the end user that the end solution guarantees a certain level of reliability, even when the environment doesn’t cooperate or things don’t work according to plan
• #22: A second assumption we make as a community is that end users are willing to put up with a certain amount of overhead to setup a tool, so long as the tool has proven its value. And in the context of AR, we often find ourselves in scenarios where some level of setup is necessary: Setting up an environment so it can be tracked Registration/calibration Orienting for initial pose, etc But when you are required to work in an environment that actively works against you, your patience for setting up tools is going to be non-existent — your only priority will be to get in, get the job done, and get out. Any sort of setup is just wasting time that could be spent doing the job.
• #23: In working with users, we learned that there is a zero tolerance policy for things that waste their time Not necessarily technical, nor inclined to want to learn – it’s got to work for them, or be embedded in their process, otherwise they won’t do it And setup, especially when there are so many variables involved in getting the tech to work, is just one of those things that can easily be perceived as a “waste of time” because it’s really hard to make setup of any kind, “easy enough”
• #24: Which led us to the following insight, which is that: If you have to set it, then forget it What’s needed are solutions that…
• #25: To address that, we need to build applications that get end users to work right away: Get started in a “limited” mode/context if there has been no setup Skip setup steps and get straight to work Setup in the background — get started in a limited fashion, and then let the app catch-up as it gets the info needed
• #26: One final assumption rests in the belief that somehow, the focus of the end user experience is on the app itself… that somehow, AR must be the star of the show. Well, as much as we hate to admit it, it’s not – nor will it ever be the case that AR is the center of attention for users in the field
• #27: Some common feedback we got from maintainers in the field was that “it’s just a tool” – which is true – the reality is that AR’s purpose is to help the user, not be the center of attention And that really got us to think – most design methods used in app development today are centered around the assumption that the app is the focus for the user, rather than an adjunct to something else In the field, that’s simply not true – the focus is getting the job done, and the app needs to be at the right place at the right time
• #28: Which led us to the following insight, which is that: It’s not about all the info, all the time – it’s really the right info at the right time Which is to say that users shouldn’t be forced to interact, or pay attention to an app, when it doesn’t help them – you have to get out of their way so they can focus on doing their job
• #29: Some techniques we needed to consider: Methods for “blanking” an app with a single interaction Minimal UI, minimal presentation in use — it’s not about flooding the user with a lot of information, just what they “need to know” to get the job done User choice and freedom — give the user control over how they want information presented instead of forcing them to do things a certain way all the time
• #30: Now – realizing that I have just a few moments here, I have just a couple more thoughts here…
• #31: The first is that we really need to challenge ourselves as a community, to hold ourselves up to a higher standard in terms of what we promise to deliver, and what we actually deliver… While it might seem like a brilliant tech demo is all that’s needed to get usin the door, delivering on the promise of the technology is a much higher bar than perhaps we think it is And if there is one thing I’d like you to consider out of everything I’ve talked about today, it’s this:
• #32: We have to remember that although all of us here at AWE this week are “true believers” in the power of AR, the potentially millions of people out there who would be the end beneficiaries of the tech don’t see things the same way. And as much as we want to believe that AR as a technology has the potential to change the way people work, the reality, no pun intended!, is that people – people like those maintainers in the field – are really the ones that give AR its true value. We owe it, as much to them as for us, that we aren’t just focused on building “what can be built,” but rather, what “needs to be built.” I’ll leave it at that - thank you!