Episode Eighty Nine: Yet More Smarts; The Continued Tyranny; A Different Kind Of Overload
0.0 Station Ident
I've been back in Portland for about 24 hours and the panic has only just set in a little bit. First is that there are a whole bunch of threads that I need to pull on and constantly manage in terms of what I'm going to do next, and all I know is that I should have them organised in some way. And instead it was difficult to pull myself out of bed this morning just because of the sheer *weight* of things to do.
And again, the big problem has always been being trapped in my head and having the fear of doing the things that need to be done. It doesn't quite matter that, once I start doing those things (like writing today's episode, for example, which I was terrified about writing until I started writing it, upon which all the words kind of just fell out of my fingertips onto the keyboard) they become exponentially easier to do until they just get done. For example: last Friday I had a good meeting that I promised to write up, and ultimately ended up being half-consumed by the terror that what I had been told was a good, interesting conversation would instead be exposed as a whole mound of stuff-I-made-up when I wrote it down. Which might still be the case. You never know.
But anyway. I got through the day. And tomorrow, my wife and son get back. And it's another day.
1.0 Yet More Smarts
I got a bunch of good notes from people in response to my posts about Solid and one of the more provocative ones was about notification overload. One way of thinking about this is that as we move to an attention economy (or, at least, as some people *think* we're moving to an attention economy), the default is to try to grab as much attention as possible, at any given time, modulo some semblance of reasonableness.
It's interesting that we've seen some sort of convention for dealing with notifications when we *don't* want to see them: the example of people putting their phones face down on a table when sitting down for a meeting or eating is one of them (and one that was quite nice to see Nokia highlighting, I think, in one of their product videos). Notification overload isn't a new problem: it was information overload a few years ago when people like Linda Stone coined the term continuous partial attention and its associated email apnea on the knowledge that people genuinely hold their breath - and get stressed - when they're doing keyboard tasks like checking and writing email. (Check it: I bet as you're reading this, or as you've got an Outlook window open, you're subconciously *not* breathing regularly).
Continuous partial attention didn't really catch on - or it did, it's just that the mobile internet came along at the same time and we all kind of collectively shrugged our shoulders and got on with playing Clash of Clans as well as checking our email and hey did you see that Twitter push notification that just came in.
I know it's a disingenuous argument, but we've even got a terminology thing going on here: *push* notifications talk about the back-end technology that rely on a server passing on information to an end-point, but at the same time isn't it such a wonderful coincidence that they also push themselves to the foreground of our consciousness...
One of the things that I've noticed in the last three years of my experimenting with advertising was the value of the relationship, and how the inexorable push of social media has caused the nature of that ethereal brand relationship to change. My ex-employer likes to say that it creates 'provocative relationships between good companies and their customers', and what company doesn't want a relationship with its customer? And in a world of mobile computing and communications, the new relationship is mediated not just by broadcast push messaging (ie: television adverts) but by in-stream messaging (Tweets and Facebook posts) and also some sort of holy grail of interruptive push messaging (an exhortation to win the hour, for example).
The end-game of this, then, is: what happens when everything has its own IP address, its own route onto the internet and can talk not only to us, but to its sibling devices? You think the Professor had it bad when he could hear all the voices, you should think yourself glad that you won't be able to hear or sense the cacophony of IP packets coursing around you in the next few years.
Perhaps one way of thinking about this is this: given we're all thinking about conversational interfaces now and everyone's busy trying to anthropomorphise their devices and services so that they can back-door hack empathy into our relationships with them, whatever happened to "if you don't have anything {good, interesting} to say, then don't say anything".
I don't have a Nest, so I don't know how chatty it is. But right now, the mediation that we have between ourselves and chatty devices is pretty thin. One of the things I did lately when I reset my iPhone in a valiant attempt to fix its battery life was to go into Background App Refresh and have a long hard think about the apps that I *genuinely* needed to be updating in the background and then take a look at Notification Center and have a long hard think about the things that I *genuinely* needed to know about, as they happened.
It turns out that there aren't that many. Sure, I don't find out, second by second, if someone has faved or retweeted one of my tweets or if someone has hearted one of my instagrams or even if someone has liked one of my Facebook posts. But all of those notifications are aggregated and pooled together when I decide to check. In fact, one nice change to my life was when I accidentally left on do-not-disturb mode and found out that it led to a considerable reduction in my anxiety that day.
When we talk about doing good by the user and satisfying user need, this is one of those difficult questions: does the user *need* to receive those messages instantaneously? Should the default be on, or off? Should we always ask for push messaging notifications, and what sort of notifications should they be? I'm reminded of one of the exhortations of Steve Jobs when he pushed to get the bootup time of an Apple computer just that little bit quicker, and he persuaded his engineers using the argument of the aggregate: of all the millions of hours that would be saved just by shaving a couple seconds off.
One of the dreams, of course, is to model some sort of executive assistant or agent who would know what was important to you and effortlessly perform some sort of do-what-I-mean-not-what-I-say prioritisation of incoming messages and notifications. But hey, that's kind of relying on hard AI or at least significantly better bayesian filtering not for spam, but for *useful* stuff. If anything, I hope Apple figures out what the "missed" part of Notification Centre means, because it's pretty much a joke.
Anway. The corollary of attention being the most valuable thing to a company or a brand is that it's also the most valuable thing to a person. A brand might do well to respect that knowledge in order to gain trust.
2.0 The Continued Tyranny
For various reasons, I've been looking at the Medium essay I published nearly a year ago now; The Tyranny of Digital Advertising[1].
One thing that didn't quite come across in that essay was exactly what was meant by native advertising, which pretty much every publisher has jumped on in the meantime. Nevermind the fact that there's already way more than enough display advertising inventory on the internet in the first place (the marginal cost of putting up another possible location for display advertising is zero, for example), one of the things that I got confused about upon rereading my essay was that there are clearly two things that native advertising could be. The first one is the boring one: advertorial content. The kind that we'd get at my ex-employer where we'd work with Buzzfeed and they'd pitch us a listicle that would kind-of go with our campaign and we'd kind-of be okay with it, but it was kind of obvious that everyone would rather be doing something else rather than a space-filling listicle.
The other kind is the more interesting kind, and where I wasn't quite specific enough in my terminology. That is, native-meaning-native-to-the-web, the kind of non-scaleable, one-off, make-it-like-you-mean-it advertising that I don't think exists quite yet *because* there isn't a toolkit.
Again all this comes down to a simple question: what does it mean to be native to the web? To which I'd say the answer is the unassailable link. The uniform resource locator. The anchor from one place to another. That's the web. Sure, you can say it's lots of other things too, but I think they all fall down without the humble URL.
So the real question is this: if the link is what makes the web the web, then how can you use the link to make advertising that is built for the web, and could only be built for the web? That, I think, is native online advertising.
[1] https://medium.com/i-m-h-o/2bfa73373a9a
3.0 A different kind of overload
A friend pointed me on Twitter to Perception's latest motion graphics work, this time for Captain America: The Winter Soldier[1], this time with the observation that it was high time Hollywood was able to move past 45 degree angle vir2l style designs and that they should embrace flat design already (never mind the teal and orange palette).
A separate problem, though, is that of dashboard design. Sure, we all take cues from Hollywood, mainly because Hollywood gets the excuse to make things all shiny and, bluntly, not quite well thought-through for everyday use, but the thing about Nick Fury's car and the amusing slide-to-unlock aim and fire interface that Maria Hill uses at SHIELD is the sheer complexity of it. I get it: we need to understand that these are Serious Systems and that Lots of Important Things Are Going On, but the end result of it is that people think, when they're designing *consumer* systems, that regular people need to understand that things are Very Complicated too, instead of doing the hard work and simplifying where and when possible.
There's a storytelling reason for all this information overload. We, unlike Nick Fury, are not being ambushed in our Awesome Car, and we're quite happy to sit back in a plush seat with an unreasonably sized Diet Coke in our hand and sip whilst the scary fake police dudes try to break the window with their hydraulic battering ram. Nick Fury has serious shit to be getting on with, so from his point of view, he doesn't really need to understand that his car is trying to read the licence plate of the car that just obstructed him.
We, the audience, need to see these things for storytelling reasons. We need to understand that the Awesome Car is doing lots of things, is investigating a lot of things and is Working Hard. And I wonder if this is just a side-effect of that peculiarly American sensibility to not just work hard, but be seen to be working hard. Computer car is, of course, computing.
Sometimes it's interesting to see internals for us to understand fundamentals. Many of us speak of a sort of golden age when you could take things apart and they weren't quite so appliances. I certainly remember a time when I was in my teens and I spent an, in retrospect, entirely unrealistic amount of time reading about Intel reference motherboard designs and ATX cases and building my first few PCs. It was a powerful feeling: to know that I knew what components went where, how they worked together, despite the fact that a lot of the esoteric knowledge I was gathering wasn't *entirely* useful in the long run. But it did mean that I understood concepts like busses and clock frequency multipliers and power supplies and cooling and disk interfaces. That said, I'm willing to bet that everyone who remembers what it was like to have to resolve IRQ conflicts would quite happily never go through that ever again.
Anyway, that's a long roundabout way of saying this: Nick Fury's car, through its user interface, is explaining what's going on. We frequently don't get that in the user interfaces we deal with every single day. They don't, as it were, do a director's commentary on what's going on. Some of us intuit what's going on and recreate that director's commentary ("Well, it looks like what's going on here is that the local Wifi network is fine because I can ping the router here, see, but DNS lookups aren't working and traceroutes are timing out, and if I try this other device over there...") to try to understand what's going on on the inside.
We see some of these things at the other end of the stack when we run things like stack traces when things go *really* wrong so we can step through and figure out what's going on. But there's a middle ground where we just get gobs of information thrown at us, smart-city-dashboard style, and we don't quite know what to do with it. It doesn't serve a storytelling goal, because the information isn't *going* anywhere - no one has written a beginning, middle or end - in fact, it's just an endless series of iterating beginnings as the dashboard silently refreshes.
I know there are certain of you out there that are going to eyeroll at this, because it's just me banging my drum and repeating ad infinitum the value of storytelling. But dashboards frequently don't tell stories in the way that Nick Fury's car does. They don't say: I'm doing this, to find out that, which means this. They just show you the number of faces they've seen at a busy intersection. That isn't to say that there always is, or must be, a narrative, just that one way of dealign with information overload is by putting it into a sort of temporal context that makes it relatable to us other than just a time-series slice of data.
[1] http://perceptionnyc.com/content/captain-america-winter-soldier
--
As ever, I appreciate your notes. And you'll appreciate that I have a backlog of about 227 to reply to, which shouldn't discourage you from sending more. Because really, I like getting them.
See you tomorrow,
Dan