s12e06: Dan Hon is Away, Still
0.0 Context Setting
It's Tuesday, May 17 2022 and I am still sick.
Because I am still sick, this is a re-run.
"The Starship Enterprise" and "How Do You Solve A Problem Like Digital?" was originally in Episode One Hundred and Eighty Two: The Starship Enterprise; How Do You Solve A Problem Like Digital? published on November 1, 2014.
1.0 Some Things That Caught My Attention
1.1 The Starship Enterprise
I was having dinner the other night with Tom Armitage and we figured out what my next talk was going to be. We've been friends for a very long time - at the same time as crossing paths at university, we were both bloggers in the late 90s early 2000s in the UK. Tom knows that I have, if not a soft spot, then a sort of obsession with the world of Star Trek: The Next Generation - not necessarily in terms of the characters and stories, but more in terms of what the world of the Federation actually means. How hard you can push it and how much of it just ends up being silly, or falling apart at the seams.
I've written about this kind of stuff before - way back in episode 12[1] I was writing about the computing infrastructure of the world of the Next Generation, I mean, does *everything* in the Federation run LCARS? And, you know, somewhat tongue-in-cheek, who runs devops at Starfleet? Is LCARS open source? Can any civilian submit a pull request?
But no, this is the idea, and like most things, I can't remember how we got here. I think it was because Tom saw that I'd somewhat facetiously tweeted if LCARS supported animated gifs. Or maybe it was because I'd been thinking out loud and told him that I wondered how the Enterprise-D's Main Computer disambiguated names when routing communications requests. There aren't that many times (I haven't looked) where someone does the equivalent of *tap combadge* "Picard to Smith" and Computer says "there are seventeen currently Smiths on board, which Smith did you mean?" and Picard has to say "John" and Computer says "there are three currently John Smiths on board" and Picard has to say "John Smith, no bloody A, B, C or D".
(Of course, I realised as I wrote that last sentence that that's a pretty good example of Computer requiring disambiguation, and a pretty good episode to boot[2]).
The reason why the Enterprise-D is so big, the reason why it has hundreds of crew, the reason why it needs to haul around all of that *stuff* is that the Enterprise-D is chock-full of knowledge workers. "Enterprise" isn't just the name of the ship, it's a description. It's a tiny little city state - with everything that that city state needs - and it doesn't do it through magic hand-wavey unobtanium technology, it does it, horrifyingly, in *exactly the same way* we do things right now. With computer-augmented humans.
(Yes, I have a copy of the TNG Technical Manual. I know that's not how the Enterprise-D *really* works. Computer has over 47 kiloquads of processing power! No, this is taking what we have now, sticking it in the Next Generation universe as a way of thinking about what we have right now. Thank you for holding fire, Star Trek pedants.)
There's what, 10-15 bridge crew on the 'D. They haul an immense support infrastructure around with them. Here's what happens when Riker taps his combadge to be put through to Dr. Crusher:
Riker taps his badge and as it chirps, someone down there on Deck 47 sitting at an LCARS terminal, *just like a call center worker*, sees Riker's profile pop up on the screen in front of them. Computer is listening everywhere, always, just like Echelon and the NSA are - there are mics and cameras all over the place. Knowledge Worker (Ensign Grade) Barclay sees a transcript of pretty much everything that Riker's been talking about over the past few seconds, minutes and hours. There's a graph layout of everyone Riker talks to and how often. And Knowledge Worker (Ensign Grade) Barclay taps on the pre-selected, auto-completed icon of Dr. Crusher and routes the call. Because there's a *switchboard*.
Every single one of those LCARS terminals has a support ratio. At least 5% of the D's crew are desktop support staff, either replacing burnt out LCARs terminals or refreshing them. Some of the items are self-service, like the vending machines on the Facebook campus. But it turns out that a lot of the time, you just have to see someone down in IT.
Every time Data looks something up there's at least ten people behind him, several layers deep, preparing reports and checking on computer-augmented intelligence. It's kind of depressing, really: the entire infrastruture is set up so around 10-15 bridge crew can swan around and do away missions. The only genuinely piece of *absolutely amazing tech* that can work itself is the transporter.
The worst part, the absolute worst part, is the holodeck. Whilst 5% of the D's crew are desktop support staff, there are at least twice as many texture artists slaving away making sure that when Riker wants to program in Riker Risa Scenario Forty Seven Alpha, he's got the right Blender models to choose from in the library and the right textures to decorate everything. Assassin's Creed: Unity, the latest in the Ubisoft I'm-a-pretend-assassin-travelogue set in Paris, probably has up to 10 teams working around the world, say up to a thousand people recreating Paris for us to play in 1080p on next-generation consoles like the PS4 and the Xbox One. The future of the Enterprise isn't magically waving a wand and creating a new Holonovel like a sort of interstellar-holonovel-writing-month all on our own, it's more of a production line. Have you ever sat through the end-credit sequence of a Lord of the Rings Extended Edition Movie? Yeah, that. Now look at the Enterprise and every single one of those glinting portholes. That's a 3d modeller, rigger or animator, texture artist or whoever, slaving away so that Picard can pretend to be Dixon Hill again. That's the mission: to explore new worlds, to boldly go where no one has gone before and to create disruptive innovations in the VFX workflow/pipeline.
Oh, oh, and you want to know why the Enterprise is so big, *on top* of all of the staff? At least sixty of the decks (oh, I don't know how many decks there are) are just made of CLOUD. Do you have any idea how much iron it takes to bring a small bit of cloud with you? There's a scene that both Tom Armitage and I love from Danny Boyle's movie Sunshine where you see that the ship's computer is overclocked - it's submerged in liquid nitrogen or something so that its processors can run at a gajillion gigahertz because otherwise it turns out you'd need a billion of those processors to provide the voice interface. We see this when one of the characters has to plunge their hands into the liquid bath keeping the compute infrastructure cool.
The TNG Technical Manual has a little bit about how the Enterprise has a main computer core, it runs vertically, I think, through a few decks but: come on. We've seen datacenters. Look at all that back-end infrastructure you need! Most of it, again, is just racks and racks of nVidia/AMD GPU nodes so that Geordi can have a semi-plausible visual simulation of Dr. Leah Brahms (which is creepy enough in itself). At least you're in space so you don't have many problems cooling all of that stuff. Just open an airlock every now and then.
Think of the Enterprise as a regional office: 1,200 people, with ten or so Director/VP-level appointments at the top. Everyone else is just hanging around, clocking in and clocking out. This is what an Enterprise means: taking a little bit of culture, taking a little bit of infrastructure, taking a neighbourhood and making it self-similar so it can fly around in a metal bubble filled with air.
It's not a post-scarcity society. There are still jobs in Starfleet. They just all happen to be knowledge-worker jobs. Tobias Revell's closing keynote had a reference to a particularly dystopic recording of a call centre marketing call where the human recipient accuses the caller of being a robot and the robot - or, at least, you think she's a robot, you can see the rules and the branching trees performing the VoiceXML analysis and routing, picking up on keywords and protesting that she is not a robot, she's human. But the twist is this: not only was the call *coming from inside the house*, but it was operated by a human using the equivalent of a call centre sound board - a human listening to another human, but having no mouth and not being able to scream, communicating only through pre-recorded snippets that *cater for the fact that she will be accused of not being a human*. Mechanical Turks, all the way down.
The Enterprise's computer is the kind of computing system we all use right now: one where computers do things that computers are good at, and where humans kick in doing the things humans are good at, or even better, are *cheap* at doing, even cheaper than computers. Red Dwarf took this to the British satirical extreme, where a miles-long mining ship employs hundreds of *vending machine repairers and cleaners* because it's easier to get humans to do it than it is to get robots to do it. We use Protein Folding games right now to get humans to pattern match - to work in symbiosis (of a sort) with computers to examine things and click things just so. That's what's happening behind the scenes on the Enterprise. You wake up. You get breakfast from the Replicator, which menu today has been laboriously manually programmed by Chavez, two bunks down, you sit in front of your LCARS terminal and then you look at things and tap on them. For the rest of your life. But hey, space travel! I mean sure, we get episodes like Below Decks and get introduced to Ensign Laren and her friends every now and then, but that's really not what Star Trek: Call Center is like. Or Star Trek: Amazon Fulfilment Warehouse. Or Star Trek: Knowledge Worker.
(An aside: Tom and I were also talking about how stupidly surprisingly awesome Tom Cruise's Oblivion is because of the Big Dumb Object that hearkens back to classic Science Fiction, or at least a certain genre of it, and the idea that somewhere in Starfleet there's a Warehouse 13, or a Special Circumstances, or just the team that goes out and discovers the Genuinely Weird, Unexplainable, Really Fucking Terrifying Outside Context Problem stuff. Like the Whale Probe. I mean, that's super alien! But no, we have to have little episodes about relocating indigenous people so we can steal their youth-providing radiogenic field or whatever.)
There's of course the joke that Star Trek: The Next Generation is the way it is because it's the quintessential 1990s show - all curved edges, fake wood trim and Marriot In Space, but the other idea is that pretty much everyone else on the ship apart from our bridge crew are Uber-style "independent contractors" - and maybe even the bridge crew, maybe even Captain Kirk is an Uber-style independent-contractor-not-an-employee so the next time his sexually assaults a first contact species or accidentally destroys a civilisation, Starfleet Corporate can go "well, he's not *technically* an employee" and disavow his actions.
Yeah, the horrifying future of space travel is a super exciting thing for around 10-15 people and cube-farm hell for the other thousand-or-so travelling. Mundane, manufactured normalcy. Your job in space is *just the same* as your job on Earth.
[1] Episode 12 - Attention, Star Trek and Cars
[2] Relics at Memory Alpha
2.0 How Do You Solve A Problem Like Digital
I'm still mildly simmering after having hammered out the previous episode about the Samaritans Radar mess. And let's be clear: there's a lot of mess to pass around, and it's fair to say that there was at least *an* interesting insight in that people are saying things and how, well, can we be more empathic to those who're in our feed and sensitive to what they're saying. Which at some point after pulling on the thread for a while you come to: ah, well that feels a bit like a societal problem, how do we, as you say, "move the needle" and create a more kind environment for everyone so *everyone* can deal with mental illness better.
But this isn't about that. This is about the piss-poor situation regarding "digital" and making things that work, that make things better, rather than making half-assed stuff that just opens a can of worms of problems and now you have Two Problems but I suppose you have an advertising award, so you can just fuck right off.
I can post-rationalise and tell a reconstructed story of how something like Radar might have happened. For those of you who are (relatively) new to this newsletter, my "creds", as such as they are are that I spent the last four-and-a-quarter years doing "digital" stuff in agency land, first as as senior creative at W+K London, and then as a Creative Director at W+K Portland. I've seen briefs glittering in the dark off the gate to the planning department, I have. I've seen creative teams fly out to New Zealand to build a website.
And I've seen what happens when agencies tell themselves that they're there to Solve Business Problems.
There are so, so many things at play here. You have your main two players: the Client, Samaritans, who, given the lack of any evidence to the contrary, appear to be somewhat naive in the "digital" space. You have their agency, Jam, part of the Engine holding company, self-billed as "a social media marketing agency, and part of Engine. We create award-winning social media communication strategies for brands."
And behind all of that, I am assuming that there's pressure both at Samaritans and at Jam to "get" digital. Because "digital" is a thing that is happening, and it's changing everything or some things or maybe even no things and you'd better damn well have a strategy for it. It's doing that everywhere, whether you're a car service or laundering clothes or selling clothes or treating people for diseases or "reading the news" or watching a tv show or any of those things.
It is doing those things because "digital" is a transformative technology just like the way electricity was.
There will have been someone at Samaritans who wanted to Do Something. Probably, I'm guessing, because Twitter is a thing and young people are vulnerable and at-risk, and it's generally good advice that's been passed around for the last ten years or so in the digital space that you "go where the people are" and that you shouldn't be Kevin Costner and build things hoping people will just turn up. And it turns out that the young people that you, as an organisation wanting to help people with mental health issues, want to make sure that you're doing the best at serving younger people.
So you go find a group of people, an organisation, who can "help you reach younger people" and you find an agency like Jam, because hey, younger people "use the internet" and they definitely "do social" and what you definitely want to do is "communicate" with young people so they know about Samaritans and how Samaritans can be there for them when they need someone.
All good so far.
Problem: you're worried about young people knowing about Samaritans.
Solution: young people use the internet and social media, so figure out a way to "create awareness" so that young people know about Samaritans and how they can help.
This is, especially if you hook on to the "awareness" word, a communications brief. It's a "hey, we do a thing, and we want people to know that we do this thing". It leads, inexorably, to a campaign. With advertising, say.
And then everything goes to shit.
Because the way things are in agencies at the moment, most of them - if not all of them, really - are *shitting* themselves trying to figure out what "digital" means to them and what it means to what they do.
A very quick answer to the problem of "awareness" for the Samaritans would be to do a bog standard social media campaign and get people to talk about them. You know the kind, throw in a few hashtags. Because that's the problem, right? Get people to know that the Samaritans exist?
But agencies look at the new economy and the way the internet works and they get a bit unsure of themselves (sometimes quite rightly) because it turns out that actually, in a world where it's easier to create connections between demand and supply, there are different ways of satisfying or creating user needs. John V Willshire puts this succinctly as "make things people want" versus "make people want things". Advertising and the whole deal about "brands" has all been about making people want things. Digital doesn't *mean* making things people want, it means that there are new *ways* to make things people want, that people are able to discover and use on their own. The two approaches aren't contradictory - they work together a lot of the time.
It's terrifying for agencies to look at fast growing businesses that are able to, well, grow fastly *without* "traditional" advertising. The reason why those businesses grow quickly in the first place? They rely on new communications infrastructure and they serve latent need that hasn't been served. Sometimes those latent needs are *so great* - in the case of products like social networks through the ages - and they come with new behaviours that you don't really need advertising, as we've known it.
So then there comes this talk of expanding what it is an advertising agency does - they don't just, well, advertise, but they *solve business problems*. Most of the time that's OK: the business problem they're being asked to solve is something like "make sure people buy our stuff" or "hey, it turns out people think American cars aren't very good, but they actually are now, how can we get people to pay attention to us".
The problem is when you start expanding that definition. The reason why that's a problem is that - now, bear with me here - the kind of people who've been working in advertising agencies are the kind of people who a) have been making advertising and b) want to make advertising. There's a smaller subset of people who, over the last few years I think have actually been lied to a bit, and want to actually "solve problems creatively" and sometimes advertising is a way to do that, and sometimes not.
Let me put it this way: there's different types of creativity going on in the advertising process. When a client comes along to an agency and says "we appear to have sponsored the Olympics" and the agency says "well that's an interesting thing to do, you don't appear to have anything to do with athletic performance at all", it's an issue of creative problems solving to figure out exactly how and what a fast-moving-consumer-goods company *does* have to do with the Olympics. If you've been following at home, the resulting (genius, in my opinion) brief that goes to the creative teams is "Every athlete has a mother." And *then* you have the creative idea of telling that story and getting that story across, so that suddenly the act of being a mother to an Olympian is something that hopefully will connect with you on an emotional level. Two different kinds of creative problem solving, there.
Where it gets a bit more grey is where the brief can be interpreted as a business problem that can be solved not by communications, but by something else. Traditionally, this hasn't been in the remit of advertising or communications agencies because, well, clients come to them for advertising and communications and not, for example, brand new product ideas. And even if agencies do respond with product ideas, it's frequently an uphill struggle because of the client/agency environment: clients come to agencies for advertising (why else, right?) and the relationship is mediated, in the end, by the CMO and CEO, and *normally* it's OK for the CMO to buy an advertising or communications solution to a business problem because hey, that's in their remit, but if they want to buy a product or service solution to a *marketing* problem, that requires the sort of inter-silo collaboration that is apparently nigh-on impossible in this day and age, and a symptom in my opinion of rather disappointing executive leadership.
But, digital! Digital is a way of doing new things! Digital, because it means *everything* and because it commonly means *innovation* is a chance to do New Things, and advertisers and creative people love New Things because New Things get attention, and New Things plus Attention equals awards, and awards equal a promotion or a new job. Because that's the way the advertising job industry works. So you always want to do a new thing. And if you can do a new digital thing, well, all the better, because those Cannes Cyber Lions aren't going to award themselves.
So, let me draw a line. Digital advertising is not products and services. It isn't. Stop trying to behave like it is. You can't do both. You certainly shouldn't be trying to do both if you've never done the latter before. And if you're going to try to do the latter, try to build a product or a service having never done it before, but you're stupidly cocksure of yourself because you're "creative" you'll learn that no, the idea isn't everything, it's the execution, stupid, just like the way you spend 3 weeks in an edit suite, only this time it isn't 3 weeks, it's potentially *the rest of your life*.
I tried doing this. I tried doing this at a stupendously successful and well-regarded agency, one that has a well-deserved reputation for a) great insight and b) wonderful execution in terms of creative and emotional storytelling. I wanted to be able to use that attention to detail and craft and creative and emotional storytelling to make products and services that hadn't ever been made before, could never be made before, precisely *because* they had that advertising-level of craft and emotional resonance. And at the end, it came to asking them: "Look, I buy that you want to create provocative relationships between good companies and their customers. I also want to solve business problems. But is there also a giant unspoken asterisk there that says: *Only using advertising?" And, fair play to them, they've decided they're an advertising agency. Which means making advertising. Which is *one specific way* of creating provocative relationships and solving business problems.
But, it turns out that those things that *I* wanted to make wouldn't be ads. They wouldn't be communications. They might solve *exactly the same problem* as an ad brief, but they wouldn't be advertising.
And I felt good about doing that, *because I'd done that before*, and because I know how to do that. And because I know the people and team I'd need to put together to do something like that, and it looks hardly anything like the team that creates a 60 second tv spot.
The pressure in a lot of agencies is still do a New Digital Thing. It doesn't help that products like the Nike FuelBand win big at Cannes: it makes other agencies think that they need to do things too, and it's certainly true that the FuelBand "solves" a problem and "does branding" for Nike in a way that "traditional" comms can't or won't. But it's a fundamentally different thing.
Having the idea for something and being able to deliver it are very, very different skills and to quote the apocryphal, fictitious Sorkin/Fincher Mark Zuckerberg, "if you'd invented Facebook, you'd have invented Facebook" - which includes not just coming up with the idea but actually turning up every day and delivering it until it has over a billion monthly active users and still not stopping. There are hardly any people in agencies anywhere - agencies like R/GA and AKQA somewhat excluded - where there are a mass of people with that kind of non-campaign experience.
So, we're at Jam. A brief comes in: get young people to understand that Samaritans exists and use "social" in some way, because that's how young people express themselves these days.
The sexy idea, the alluring idea, the one that you're supposed to have if you're a creative team these days is a digital one. One that solves the problem. Not a service - you can see how Radar is talked about publicly - Radar is an App. An App that does X! That would certainly solve the problem.
Those of you who have actually built apps know what it's like. You know how much is involved. Guess what: making tv ads is really hard, too! It turns out though, that the skills involved in making a standout award winning tv ad are *not necessarily transferrable* to making a stupendously *good* and successful application. Because why would they be? Sure, there are skills that are applicable anywhere: knowing how to make decisions, having the right taste, editing, and so on. But they're fundamentally different things!
It's easier in the making-a-tv-ad world, because all of that expertise is abstracted away behind a production system roughly a hundred years in the making. There are *hundreds* of people involved in tv shoots. Right from the gaffer all the way to the edit assistant to the editor to sound design through wardrobe and location and production accounting and client management and rights clearance and so on. All those things! And, at the same time, a whole bunch of experience in the form of the creative team having *practised how to make ads a lot*.
That practice just isn't there most of the time when it comes to creative teams at advertising agencies. It just isn't.
But, this is the environment. A swaggering one that solves problems. That wants to win awards and make new things and get attention. One where four creatives, a creative technologist and a project manager can make an App. No matter what kind of app it is!
Where the idea is alluring, where the idea gets presented to executive creative directors who decide: Yes! This is a thing! This is a good idea! Go and make it! And literally hardly *anyone* in the building has done anything like it before. And why does this thing have to be done, why does it have to be made? Out of some sort of fear of irrelevancy?
Let me turn it around: it's easy to make fun of startups that pick up a Canon 5D MK III and shoot an ad. It kind of works. It's not great, but it kind of does the job. But, advertising creatives say, it's not done *properly*. It could be done so much *better*. They don't know what they're doing.
Yeah, well shit goes both ways.
I'd say this: if you're at an agency and you want to make "apps" and *most of your experience* in the industry has so far been things like tv spots or microsites, and if your app idea is something marginally more complicated than a soundboard or an Instagram clone (or, *even* an Instragam clone) or you're having delusions of grandeur and think you can sell through an Uber for X, then not only do you have an uphill struggle, but your job makes you practically unqualified for the task without a whole bunch of teamwork and, honestly, stepping back and being less involved. In other words, having the idea and then handing over to someone else who's had practice in doing that thing. Sure, this might sound like the business as usual of handing over to a production agency, but here's the thing: that situation might not be any better either, because a lot (certainly not all) of those production agencies will just make what you tell them to make, especially in an agency/production shop relationship. I don't envy you. You're being told by your colleagues and by the industry, and all the awards around you, that you'd better be shit hot at Making Stuff now, and Making Stuff includes a whole bunch of digital stuff more than culture-on-digital-platforms.
And the thing is, agencies are - more or less - *good* at making culture-on-digital-platforms. Because it turns out: that's advertising! But making apps and services is pretty much the equivalent of thinking that if you've done a few 60 second spots, you can knock out a - to a certain extent - a 4 hour blockbuster trilogy of Peter Jackson proportions. You can't. And frankly, *no one should expect you to*, because it's not like you've done that before. It's an unrealistic expectation that will lead to terrible work all around, *apart* from the fact that the ad industry will see on-the-surface successful work - just like the hardly-alpha of Samaritans Radar and start giving it awards. Give me a fucking break. Why does this thing smell like a marketing campaign? Because it was made by people who make marketing campaigns.
That's just the agency side. The other side is the from-the-outsides somewhat-shambles of Samaritans rolling this thing out in the first place and not understand what a big deal it is and what it means. This is, of course, Samaritans behaving like they need an Electricity Strategy and going out and finding Electricity Experts (who have quickly rebranded themselves such after having spent a lot of time being instead Horsepower Experts because what they're really about is Power when really a Very Different Thing is happening). Because this is the hard work: this is the idea that Samaritans, *like every organisation and company out there in the world* needs to figure, in some way, how it's going to react to electrification. But all it has to do is to not fuck up.
We've had the internet for a while now. There are certain things that we know about. The work (I'm sorry, but it's pretty much inevitable I'm going to make this reference at this point) that GDS are doing isn't New and it isn't Shiny. Not in the way that the context around Radar stinks. No, Radar stinks because it's a Samaritans problem and it isn't an outside-Samaritans problem. *Samaritans* should've done the work behind Radar, and it should've been Jam's job to *tell* people about Radar, if people even needed telling. I buy that there are academics involved to make sure the right questions are asked. But the whole thing stinks of someone at Samaritans not knowing what electricity is and throwing something out there without knowing the context in which it will be used.
In other words, no equivalent or visible "chief technology officer" or "chief product officer" - not that those are necessarily good roles to have because more often than not they result in siloisation, but bluntly, this: where's the service manager who runs this? How does this not feel like a campaign? Is this something Samaritans are going to run for months?
I got taken to task - or at least, it felt that way - by James Aylett for putting the Samaritans in a difficult position. If they don't know about digital, then how are they supposed to know about digital? Well, guess what: if you're on the exec or leadership team of a company or on their board then it's your *job* to figure it out. It's your *job* to hire someone who knows what they're doing. It's your job to be sure that they *know* what they're doing and that they're not some fucking huckster who're going to sell you on a hashtag strategy. And believe it or not, *good* digital people exist, because like I said, we've had the internet for quite a while now. In fact - just like GDS - sometimes those people even already exist *inside* an organisation, they just haven't been given the room to do the smart thing.
For crying out loud, this is the way that the world *is* right now. If you're in charge of a company - say, a bank - and for whatever reason *no-one* is able to figure out whose job it is to publish a page with Routing Number information *despite* it being the number one organic search query, then you're not doing your job. You're a *bank*. You provide *banking services*. You don't care *how* those services are provided, you just make sure you do it in the best way.
If you're the Samaritans and you provide assistance to at-risk populations with mental health issues, then you damn well figure out how to do that across *everything*. That's your job.
You don't let "digital" be a black box that someone else gets to deal with. That's not trying. Not trying is a signal. That means you're not bothered about *what the point of your organisation is* and you're more worried about the existing ways you deliver whatever it is that you do. Digital is the *one* thing that changes how you can being doing business and how you can be serving people. Not having understanding of it internally and relying on someone else for it is one of the stupidest things you can be doing and you should just go and fire yourself.
I'm still sick, and this is now my mumbleth negative rapid covid test in a row. How are you?
Best,
Dan