Dark Matter is growing at a pretty fast clip these days — and that growth has had me thinking about how to transfer the value of scale to you, the readers (see also: Metcalfe’s Law). Stealing a turn from Anjali Ramachandran’s wonderful ‘Other Valleys’ newsletter, I’ve decided to try something new:

If you’re trying to fill a job in your organization, let me know and I’ll post the role to Dark Matter. Several thousand misfit toys around the world subscribe.

There is no charge. Reply to this email with details, ping iandfitzpatrick at gmail, or DM me on the Twitters (@ianfitzpatrick). I reserve the right to edit for reason and decency. We’ll see how this goes.

This week in talking leisure communism with Ayn Rand:

Distinguished fella (and Dark Matter subscriber) Chris Butler has a new podcast, The Liminal, focused on the in-between states of our lives. It covers extraordinary ground over the first two episodes, from Charles Fort to binaural beats.

* * *

Three standouts from ’26 Things I learnt in India’, by Martin Weigel (4min):

• Your ideas about the primacy of the individual and the nobility and urgency of seeking ‘self-actualisation’ do not wash here, buddy.
• You’re an employee. Pretty much everybody here is a business owner.
• People are aware of the price the West pays for its untrammelled, selfish conception of individuality.  Having seen the consequences, they’re not convinced they want to pay it.

The whole list is fantastic, because of course it is, because Martin. He’s got a fair amount to say about the value of distancing oneself from the routine and familiar, from contexts, and from ideas of ‘old’ and ‘new’. More immediately, a simple reminder: our generalizations about the world reflect a need to, paraphrasing Anais Nin, “see the world not as it is, but as we are”.

* * *

Unrelated/completely related: work is underway on the development of Amaravati (14min) — an Indian city being designed and constructed from the ground up, projected to house a population of 11 million by 2035.

Consider the implications of this idea on what you do (they almost certainly exist):

If anything, the focus of inspiration for what makes a good city has shifted from West to East. For centuries the great cities of Western Europe, and then the big cities of the United States, were the aspirational reference points for any up-and-coming metropolis. Now Asian city planners are mostly seeking models elsewhere in Asia, for the simple reason of scale. “There is no city in Europe that has the density of Bombay or Beijing or Shanghai,” Chua told me. “You use the Amsterdam model, you would be dead. You can no longer look to Europe and America for any lessons.”

* * *

Now, pair that shift with median age by continent, via the wonderful @Amazing_Maps account

This week in fake news of the near future:

You’ve likely seen the work done by a research team at the University of Washington in fabricating new video clips of President Barack Obama (6min) using small snippets of recorded text and a neural net to dynamically generate realistic mouth movements. It’s fascinating, terrifying, brilliant work. An especially-potent excerpt from the university’s own website:

Previously, audio-to-video conversion processes have involved filming multiple people in a studio saying the same sentences over and over to try to capture how a particular sound correlates to different mouth shapes, which is expensive, tedious and time-consuming. By contrast, Suwajanakorn developed algorithms that can learn from videos that exist “in the wild” on the internet or elsewhere.

“There are millions of hours of video that already exist from interviews, video chats, movies, television programs and other sources. And these deep learning algorithms are very data hungry, so it’s a good match to do it this way,” Suwajanakorn said.

Put another way: the full history of recorded video is potential training data for new footage of each of us expressing ideas we’ve never had. The implications are staggering.

* * *

When you’re done there, read Kenneth Stanley on Neuroevolution (7min). Neuroevolution?

Put simply, neuroevolution is a subfield within artificial intelligence (AI) and machine learning (ML) that consists of trying to trigger an evolutionary process similar to the one that produced our brains, except inside a computer. In other words, neuroevolution seeks to develop the means of evolving neural networks through evolutionary algorithms.

When you’re done there, Kernel.

* * *

Surveyor, by the New York Public Library, is spectacular: a higher-tech Mechanical Turk for mapping the locations of archival city photographs. You’ve been warned: it’s an absolute rabbit hole.

* * *

Do read Jeffrey Inscho’s piece for The Studio on mobile usage benchmarking at the Carnegie museums (7min). There’s nothing especially novel in the findings, but his team’s approach is absolutely worth replicating. Fundamentally, organizations and institutions of all stripes are still really awful about benchmarking customer and user behavior.

This week in building your own Accenture with Raspberry Pi:

I was exchanging Tweets recently with Farrah on the possibilities presented by the phrase ‘how might we?’, increasingly pervasive in organizations of a certain disposition. I suggested a simple modification, replacing the word ‘how’ with ‘in what ways’, the latter being divergent, the former driving convergent thinking.

I was reminded of that distinction this week in Scott Smith’s wonderful piece on the power of Future Design to align perspectives (6min). He writes:

As is often the case in structured explorations of the future, the teams’ sharing of their own future maps—and sense opportunities and concerns in their own words—allowed new understanding to surface. By making their own forecasts and insights visible to the other team, and by taking the time to debate differing data and understandings, each group deepened their own pool of possible strategic pathways.

Which, in turn, reminded me of this gem from Nick Harkaway’s The Gone-Away World:

“You dont make strategy so that there’s one path to victory; you make it so that as many paths as possible lead to something which isn’t loss.”

Quote originally plucked by Mr. Davies.

* * *

Definitely read Eric Karjaluoto on starting a design studio (4min), even if you won’t be starting a design studio. The advice is evergreen, especially this:

Be easy to hire. I have a friend who’s smart and well qualified, but he can’t find steady work. My hunch is that no one hires him because he makes the process unnecessarily difficult. He complains about their HR software. He questions whether their interest is legitimate. He gets frustrated when the process carries on. Don’t fall into this trap. Make it easy for clients to try you out, see how you work, and get comfortable.

* * *

* * *

Platforms are not technologies but scalable collaboration agreements.

an interesting shift in understanding this idea of platform happens when you start thinking of the organization as a platform for collaboration and knowledge creation between team members and contributors. In other words, applying this model to any company, as a way of framing and structuring it, using platform design thinking ideas instead of hierarchies. Extending the model beyond its original use in creating and understanding Marketplace Platforms

* * *

This week’s a bit lighter on links than usual, as I’ve been completely consumed by both work and Mike Bise’s incredible archive of GAP in-store playlists — which reveals considerably more Saint Etienne than I recall overhead in the halcyon days of early-90’s New Hampshire.

Thanks, as always, to those of you who share Dark Matter with others. If you’re the thoughtful sort, a request: reply to this email with a succinct one-to-two sentence description of the newsletter.

I’ve been working, slowly, diligently through Jan Chipchase’s extraordinary Field Study Handbook, which arrived along with a smaller, paper-bound book called Sustainable Data. Over the last week or so, I’ve returned several times to this passage from the latter:

“As more data comes on stream, revealing what people are doing and how, there is a growing danger of people being treated as little more than lines in a database, stripped of personality and context, there solely to be mined and monetized. The massive scale of wholesale data collection has allure, particularly at the organizational level, coddled by numbers. Unaware that there’s little more to life than what they are able to measure, over time, the organization’s palette becomes calibrated to bland food.”

This week in you’re over-using the word ‘liminal’:

From a Creative Review interview with photographer Gregory Crewdson (6min — you absolutely should read it in full), this gem:

“It’s not entirely fictional and it’s not entirely truthful. It’s consciously taking places that mean something to me and have connections to my life but also working with a theatrical set of conventions and cinematic images and use of heightened colour and light … all that’s put into a mix to create some kind of ‘subject of truth’, I would call it. It’s not evidential truth – it would do you no good in the court room, but does you well in the gallery.”

which, in turn, reminded me of this from Russell Davies in 2011:

“Because that’s what we need to add to so many things, to give them that extra neccesary magic. A pretending layer. So it’s not just a useful or beautiful or functional object – it’s got some little nod to who we’re pretending to be when we’re using it.”

* * *

Mark Pollard, Aussie, gentleman, on the Twitter:

“Young strategists start out like young rappers — they put words together because they can. After a while, they start with a point and then find the words.”

* * *

There’s a teaching moment for young designers and strategists in BlueCross’ ill-conceived notion of sending marketing materials in a USB-powered direct mail piece (3min). Really.

* * *

All the service framework Reddit channels are belong to Made by Many, who have launched a professional service and product design development program (7min) that resembles a hybrid of A Book Apart and IDEO’s Human Centered Design framework. Sadly, it’s not yet available for public purchase.

* * *

Building on a post I shared a few weeks ago: do read Robin Wong on slow innovation and the three parts of the newsroom (5min): factories, workshops, and labs. It’s a piece that Richard Sennett could be proud of.

This week in wait until Kaytranada gets his hands on scikit-learn:

A bot called ‘My Handy Design’ is creating and marketing cell-phone cases on Amazon using random stock and rights-managed imagery. Items appear to be removed for copyright on a near-constant basis. As a bot, My Handy Design is tireless in its Negativland-like pursuit of trademark infringement.

* * *

A lovely piece from Robert Ito for Google on the important role that gender balance plays in machine learning (12min). To excerpt liberally:

“Programmers give virtual assistants a female voice because, well, men and women alike tend to prefer one. “But it perpetuates this idea that assistants are female, so when we engage with these systems, it reinforces that social bias,” says Chou. Many of the field’s best minds worry about what’s going into real-life AI systems—and thus what’s going to emerge. That’s where the push for greater diversity in AI comes in. Little of this will be easy. But its proponents are smart, resourceful, and committed to the cause.​”

Google also just launched PAIR: People + AI Research Initiative, which is worth a look if you’ve not seen it.

* * *

John Robb’s written a short piece on ways in which AI creates new jobs (4min). Some are obvious, others not so readily apparent.

* * *

Artificial Intelligence Now is especially interesting: an initiative from Kate Crawford and Meredith Whittaker with a focus on four areas of learning:

1. the role of AI in rights and liberties
2. the relationship between labor and automation
3. how AI impacts bias and inclusion
4. the impact to safety of AI deployments in institutional technology infrastructure

This week in a future gleaned from a childhood spent playing Fatal Frame 2:

From Lewis Gordon’s meandering, spectacular review of the new game The Signal From Tölva (9min), in Heterotopiaszine:

Tölva’s derelict industrial spaces feel similarly empty. But there’s another symptom of neoliberal economic policy that the game touches upon. There might not be a single human in the game but there are robots. Many of them. The game’s workforce is entirely automated aside from you, the player, who has managed to hack into the Surveyor network, able to assume control of a drone and exercise some semblance of agency roaming the wilds of the planet.​

Read it. Really. Please.

* * *

Adapted from a talk at SXSW, a New Inquiry piece by Joerg Blumtritt, Simone Browne, and Heather Dewey-Hagborg on ‘subverting biopolitics’ (8min) makes for excellent reading. This, in particular, stuck with me:

This biopunk future will come in three ages:

1. Quantified bio: 23andme-like diagnostics become available at grocery rates for everybody. Just as the iPhone tracks every step I take, people will have their genomes continuously monitored, including their microbiome and all microorganisms around their house, gardens, pets and cattle.
2. Evolution as will and representation: Genetic modifications are commoditized. Sanitation, hygiene, and most of medicine has moved from chemistry to biology, substituting hydrochloric acid or isopropyl alcohol with enzymes and phages.
3. Siphonophora humanity: Our modified bodies are enhanced with all kinds of additional bio-receptors. These work similarly to taste buds in their method of gathering information about our environment and people within smelling distance while we pass by, without any need of cognitive processing, will feed directly into our metabolism.

* * *

How many unintended micro-aggressions will I perpetrate against the machines? wonders Matt Jones. This is brilliant.

* * *

Finally, from a CBC article on a small Canadian town that depends almost exclusively on Amazon Prime for supplies of commercial goods (4min):

Canada Post says the Iqaluit post office is one of the busiest in the country and parcel shipping to the remote office is increasing at two to three times the national average. In the first five months of 2017, the post office delivered 88,500 parcels. That’s an increase of 27 per cent over last year and averages out to around 12 packages per person in Iqaluit. While those numbers aren’t specific to Amazon orders, Marineau-Plante, peeking behind the post office counter, estimates 80 to 90 per cent of the parcels bear the Amazon logo.

* * *

Until next week. Be well to one another.

* * *

The basic premise of the argument is that network effects make it hard for consumers to switch to new entrants. Switching costs was the basis of mobile number portability but there were not the network effects issues as you could call anyone on any network. Facebook is a different story. If you were to switch to another social network, let’s call it Newbook, you could not read your friends posts on that network and your friends could not read your posts on Facebook. Not surprisingly, that is a big problem for Newbook’s ability to compete and Newbook — even if it were far superior to Facebook for some (or all) consumers — would not get many (or any) customers.

* * *

* * *

* * *

This week in everybody in Cannes is totally talking about how scent is the new UI:

The London Squared Map (4min) is a fascinating idea with a challenging remit:

How can a city be reshaped to allow for a more even presentation of data without obliterating the forms that make it a recognizable space?

Put more simply: how does one design to avoid the distortion normally built into map-based data visualizations? After the Flood have done some really smart work here. H/T to Matt Webb for the link.

* * *

Joanne Schofield has written a smart primer on content design (3min) for the Co-op blog. Circulate it generously within your organization.

* * *

Racked has a fabulous profile of ‘indie artisanal perfume pioneer’ Frederic Malle (5min) that you should absolutely read. Really. Still not sold? Consider the opening:

“I was at Chateau Marmont yesterday in the elevator and there was this girl preparing for a party, and I was really sad for her because she smelled like a Duty Free.”

You’re welcome.

* * *

Shipping containers are the new Sarah Records singles: (4min):

He has been keeping his eyes out for a refrigerated Maersk box, which he has never seen. “Maersk might not be the most boutique one to spot,” he said, “but it’s my favorite as a layperson in the world of container spotting.”

* * *

This week in ‘I hear that you and your band have sold your guitars and bought a DMP’:

Preach, Paul Graham:

“Markets don’t work for everything. Truth is one place where they fail.”

* * *

Plenty of people have been circulating Andrew Chen’s excellent essay on the challenges of growth at the end of a tech cycle (9min), and with great reason. He outlines six key trends that serve as meaningful obstacles to product growth, namely:

1. mobile platform consolidation
2. competition on paid channels
3. banner blindness
4. superior tooling
5. smarter, faster competitors
6. “Competing with boredom is easier than competing with Google/Facebook”

The most interesting to me is the parity produced by the fourth factor, superior tooling, through products like Mixpanel that commoditize cohort analysis. We’re working at a really interesting time when complex analytics tools are becoming commonplace. Read the whole piece, please.

* * *

Zillow is threatening to sue the McMansion Hell blog (3min) — always a sign that things are going well.

* * *

I finally got around to Ramzi Yakob’s piece on the enterprise value of first-party data (7min). You should, too. This is especially good:

To me, ‘meaningful use of data’ is any use that has a positive contribution to one of the three ways you can grow a business:

1. Increase usage of the existing product or service (more people and / or more often)
2. Increase the value derived from each instance of use (higher price)
3. Increase the utility of your business by serving a larger number of needs (new products & services / bigger share of wallet)

* * *

Sam Ford on ‘slow innovation’ (6min) is wonderful:

Slow innovation is the realm of pattern recognition: searching for emerging developments outside the organization’s immediate line-of-sight or that may be happening steadily, but not rapidly.​

* * *

This week in training Watson on a steady diet of vintage Geoff McFetridge prints:

* * *

* * *

“(Organizations) have to invest to create what our colleague Todd Supplee calls data factories: systems that can combine data from proprietary, third-party, and public- and partner-generated sources and extract value. While doing so, they must build the capacity for data governance and be sensitive to norms, regulations, and expectations surrounding transparency and privacy.”

* * *

Using neural networks to explain neural networks (better on desktop than on mobile).

* * *

Until next week.

If you read nothing else this week, make it Longreads‘ excerpt from Adam Greenfield’s new book Radical Technologies— which is next up in my reading list — on the sociology of the smartphone (27min). It’s a long piece, yes, but with some absolutely bang-on points. For example:

“The individual networked in this way is no longer the autonomous subject enshrined in liberal theory, not precisely. Our very selfhood is smeared out across a global mesh of nodes and links; all the aspects of our personality we think of as constituting who we are—our tastes, preferences, capabilities, desires—we owe to the fact of our connection with that mesh, and the selves and distant resources to which it binds us.”

Hat-tip to reader (and Almighty co-founder) Chris Smith for the link.

* * *

This week in be the algorithm you want to see in the world:

Now you can own that co-worker who started dropping oblique references to neural nets and machine learning in morning stand-ups: Machine Design has published a quick guide to the distinctions between machine learning and AI (3min). Commit it to memory.

* * *

It’s Tim Hwang’s world, and the rest of us occasionally get to occupy it: Y-Combinator has a fascinating, wide-ranging interview posted with Google’s new Global Public Policy Lead on AI (12min) that gets to some really interesting points on the relationships between government and AI policy. To wit:

one of the most interesting aspects of the GDPR, which is a new privacy regulation in Europe, is the potential for this, what they call kind of a rights explanation. So the idea is for certain kinds of automated decision-making, it might be so significant as to require or give citizens the right for that system to be able to produce some kind of human understandable explanation for what it’s doing.

* * *

All of the cool kids in Cannes this week are going to be talking about  “quantum supremacy” (5min) is going to be a really important idea over the next few years.

* * *

What happens when you put portable DNA sequencing in a $1,000 package you can fit in your pocket (7min)? Things we never imagined. The possibilities presented by MinION are mind-boggling.

* * *

This week in the lost productivity attributable to clever GIF-related Slackbots:

It’s been true for a while, but it’s still wonderfully refreshing to hear someone like BBH CCO Pelle Sjoenell say it into a microphone (5min) (in a terribly candid interview with Shots):

Historically, the only route to mass communication was to hire an agency, that’s why we still bill by the hour like lawyers. But now someone on YouTube could make better ads than I do. They have the tools, the tech, the access and we have to compete with that. ​

* * *

Mike Davidson’s written a thoughtful piece on a remarkably under-addressed topic: how to give product feedback that’s useful to product teams (6min). Bookmark and share liberally.

* * *

It sounds heretical, but Amir Salihefendic’s piece on why Doist quit Slack cold turkey (10min) is a story I’m hearing with increasing frequency. To summarize:

• it’s addictive
• it’s built for shallow conversations
• it’s disorganized
• it only simulates transparency

I’ve not tried the product their team developed in response, Twist — but there are some nice notes within the piece on design decisions that operated in direct response to the perceived tyranny of Slack’s UI (the online status indicator chief among them for me).

* * *

Lara Hogan’s Github repository of HR docs and resources from her time at Etsy is an absolute goldmine. Download immediately or bookmark forever, kids.

* * *

I wasn’t going to do much with Amazon + Whole Foods, but now I am because Ben Thompson. His point: Amazon acquired a customer, not a retailer (12min) — a point laid out in extraordinary detail and with a strong understanding of the dynamics of the grocery business. It’s wonderfull stuff.

* * *

This week in ‘I’m a voice designer’ is the new ‘I’m an organizational designer’:

Anab Jain is a superhero, and her work on alternative futures at Superflux is the kind of vocation many of us dream about building for ourselves. Her April TED keynote is finally available, and it’s extraordinary. Seriously, the best fourteen minutes you’ll spend today.

* * *

Ana Andjelic has, as usual, written a really on-point piece on the choices facing fashion houses (4min) as their industry, too, moves from a focus on the corporation to a focus on the customer.

* * *

Fjord has shared a set of principles for voice UI design, wrapped up in their own nifty web UI. Some of this is well-trod, but other parts break some new ground. I’m particularly drawn to reminders that in conversation, ‘everything happens in sequence’. Well-worth a bookmark, though it’s a clubhouse leader for most-cited report in 2017 pitchware writeups.

* * *

Amal Graafstra and his friends are norm-coring human chip implants (4min), folks. From a piece this week by The Institute for the Future:

“A good implant,” Amal explains, “becomes so integrated with your daily life that it disappears. It’s unmanaged. You think about it about as much as you think about your kidneys doing their job. Every time I come home, I go to my door, I swipe, I grab the doorknob. I don’t think about it at all.”

* * *

Grant McCracken has a smart, short piece this week on the validity of using Google Trends data to make life decisions (4 min). I’d love to see him build out an entire series on this topic, as I think it lends itself well to exploration. In the meantime, are you caught up on his Artisanal Economies Project?

* * *

As always, my thanks to each of you who makes room in your inbox for Dark Matter. Thanks for all of the kind words on the Twitter. If you’ve enjoyed it, why not pass it along to a colleague or co-worker.

Until next week, I leave you with a complete history of mutually assured destruction (9min).

Six months after the sale of Almighty, I’m evolving the Dark Matter newsletter from an agency project to a personal one. The original archives are still available on the Almighty website. I’ve moved future editions to a new home at TinyLetter.

Issue 02.01 featured a missive from the wonderful Artisinal Economies Project, where old friend Sam Ford and the legendary Grant McCracken are doing amazing work. It also touches on the ‘globally inferior Nash equilibrium’, as told through the lens of Kenyan motor vehicle crashes, and the BBC’s work on binaural audio.

You can sign up here, if you like.