San Francisco hardware-ish coffee morning

Last week's coffee morning was awesome -- 20 people, a bunch of demos: music-box software, razor handles (not all hardware is internet-connected), battery monitoring tech for sub-Saharan Africa. Tons of chitter chatter. I promised to send an email with a list of who was there, I'll do that soon.

But first! Next week I'm in San Francisco, so I figured, well, we could have one there. Let's try this...

Thursday 6th August, 9am-ish for a couple of hours. Sightglass Coffee, Soma district (270 7th Street, San Francisco.)

Here's what happens. It's so informal, no introductions. We just find a table and talk about the weather and the cricket. It's nice-not-compulsory to be interested in the hardware world... do bring a demo if you are, I'm always curious to see what's going on.

I've never been to Sightglass before. It might be terrible. I'll try to make the world's tiniest sign on a post-it.

Anyway ALSO I'll be at Foo Camp which is this coming weekend, so do say hi if you'll be there too.

It'd be lovely to hang out. Londoners, normal service will be resumed soon. Keep your eyes on the announce list.

What's new on Machine Supply

Okay! So since I first tweeted about Machine Supply 48 hours ago, there have been 46 books recommended by people-who-aren't-me! In case you missed it, here's where I explain what Machine Supply is.

And, for example, here's my recommendation for Wild Life.

I have also earned Amazon affiliate fees totalling - drumroll - $3.30.


To celebrate I've added a simple way to see new recommendations -- once you're signed in, there's a "What's new" page which lists the 15 most recent.

tbh I'm not totally happy with the functionality, but it'll do as a start.

Machine Supply

I read a bunch of books -- here are the books I read in 2008 which was a particularly good year. Some books are comfort blankets (Red Mars, Kim Stanley Robinson), some are like the best hikes: a steady workout on the muscles accompanied by epiphany after epiphany after epiphany (Philosophy & Simulation, Manuel DeLanda). Ursula le Guin makes me forget where I am. Three Men in a Boat (Jerome K. Jerome) makes me laugh out loud, and was the first book recommended to me by Angela. We're now married. So.

Last week I was having a beer with Ben and Tom (literally everyone in this industry is called Ben or Tom or Matt), swapping sci-fi recommendations. It wasn't for finding new books, or at least not exclusively -- knowing what books someone loves is to know a person. I read 104 books in 2008, that was tough going. In the maybe 70 reading years I have available - mod a life-extending singularity cascading its way into reality - I could read a maximum 7,280 books. At all, ever. There are 6,000 books published every day. Knowing what books someone loves is to know their perspective and their journey, to have something special in common, to share a language.

I heard once that geeks come in two flavours: those who read A Thousand Plateaus; those who read Godel, Escher, Bach.

I'm ATP through and through. It changed my life. Here's chapter 1 as a PDF, I used to keep it printed by the door to give out to Jehovah's Witnesses. It's a philosophy roller coaster, a call to arms. Didn't get on with GEB.

I'm Starship Troopers not Dune, The Beatles not the Stones.


Anyway, I like to collect book recommendations. Sometimes I even read the books. At conferences, for years, I've asked people for their 3 recommendations.

Not favourites. Not the books they think I ought to read. Just 3 recommendations, whatever's on their mind. I try to find a board and some post-its and get people to share. Here are some recommendations from Design Engaged in 2004 where I met so many friends for the first time. Here's Matt Jones' version of the same question from Foo in 2014 -- I wasn't there, but touchingly the board is titled "The Matt Webb question: What 3 books should I read this year?" Thank you! I'll be at Foo in a couple of weeks, let's do the same session.

I love to share my recommendations with other people. Here are the books I read in April and May 2015.

So I made a website.

Machine Supply

At Machine Supply I can make a book recommendation by pasting in an Amazon link and writing a short paragraph. Then when I share a link to that (on my blog or on Twitter), my reason comes joined together with two Amazon links... one to the US site and one to the UK site. That's always been a niggle for me, to bundle those things together, to make a recommendation which is easy to share.

I'm classing this as a hobby, which means I'm trying to make the kind of website that I'd use. I'm not a hugely early adopter generally. I don't spend much time kicking the tyres of online services, I need encouragement to keep using things because I'm enormously forgetful, and I'm hugely sceptical about putting words I write into other people's databases rather than plain text on my own laptop.

All of which means -- that's what I'm making. A website to make it easy for me to share book recommendations. Here's my recommendation for The Peripheral (William Gibson), and here it is again as it appears on Twitter.

What was amazing -- and honestly what I hoped would happen, and what I'll make sure the site encourages to happen, but didn't know whether it would happen or not - what was amazing is that a few friends tried out Machine Supply when I tweeted about it yesterday.

And already I've seen @blech recommended Spacesuit: Fashioning Apollo. (Now bought on Amazon.) And @chrbutler recommended The Book of Strange New Things - which I also love - and by the way mentioned four other books, one of which is a deeply loved favourite of mine, and the other three I hadn't heard of. So those are now on my books-to-check-out list.

What next?

As it says on the front page, Current status: Pre-pre-alpha, hobby. Links will break. Cities will fall.

I've got a hobby! Haven't had one of those in a while.

Have a play. Let me know if anything breaks. My aim is to make a handy, finely-tuned little crystal. Any and all ideas welcome.

Machine Supply is over here.

Filtered for unknown lands


Two utterly gorgeous Twitter bots:

  1. @infinitedeserts, an infinity of deserts, each more infinite than the last. ASCII-art endless horizons and open sky. I can see the mesas, I can see the desert sun.
  2. @mothgenerator, every few hours a new, random, computer-generated moth, with picture and name. lepidoptera automata.

The last couple hundred years have been anomalous, historically: we've run out of frontiers. Now humanity is pushing on two, outer space and phase space -- the space of all possibilities, explored with algorithm probes. Who can say what we'll find.


List of our dwarf planets, closest to the Sun first:

  • Ceres
  • Pluto
  • Haumea
  • Makemake
  • Eris

But Pluto shouldn't be categorised as a dwarf planet -- we've found out that it's a binary planet with four chaotically orbiting moons. What do we even call that?


Aquaterra, the various lands now under the ocean previously populated by humans, roughly the size of North America.

"When scientists do mention aquaterra, they often call it a 'land bridge' as if ancient people only used it to get from one place we know today to another place we know today. This was not just a bridge. When sea level was low, aquaterra was a vast coastal plain with population densities at least as great as those in the lands above. There were houses, roads, villages and possibly cities. It was all coastal, all flat, and mostly tropical - clearly the best place to live during the ice ages."


an area of land, now lying beneath the southern North Sea, that connected Great Britain to mainland Europe during and after the last Ice Age. It was then gradually flooded by rising sea levels around 6,500 or 6,200 BC. ... It was probably a rich habitat with human habitation in the Mesolithic period, although rising sea levels gradually reduced it to low-lying islands before its final destruction, perhaps following a tsunami caused by the Storegga Slide.


Radio Aporee, found via Warren Ellis' newsletter Orbital Operations where he describes it as a constant stream of field recordings from all over the world. More radio. Continuous synthetic rain.

At the core of, a stochastic audio engine generates a realistic rain shower by randomly drawing sounds from different categories such as light rain, heavier rain, thunder, and water sounds.

Short story. The Library of Babel, by Jorge Luis Borges.

The universe (which others call the Library) is composed of an indefinite and perhaps infinite number of hexagonal galleries, with vast air shafts between, surrounded by very low railings. From any of the hexagons one can see, interminably, the upper and lower floors.

The books in the library are infinite, and the text - of 25 letters - appears to be random - the formless and chaotic nature of almost all the books - no two identical.

One which my father saw in a hexagon on circuit fifteen ninety-four was made up of the letters MCV, perversely repeated from the first line to the last. Another (very much consulted in this area) is a mere labyrinth of letters, but the next-to-last page says Oh time thy pyramids.

Coffee morning next week

It's been super ages since the last hardware-ish coffee morning. I'll be hanging out in Old St next week if anyone wants to join...

Thursday 23rd July, 9.30am for a couple of hours, at the Book Club (100 Leonard St).

Here's what happens. i.e. v informal; just chatter; it's nice if you're interested in the hardware world but not mandatory; it might just be me doing my email all morning or it might be a few of us.

Anyway it would be lovely to hang out so do come along. See you next Thursday!

Two other things

  1. Email. You can subscribe to get infrequent coffee morning reminders to your inbox, join the coffee morning announce list here.
  2. San Francisco. I'm going to be in the Bay Area in a couple of weeks, so I'd be up for doing a hardware-ish coffee morning in San Francisco if there's any interest... say Thursday 6 August. What do you reckon? Would you come? Where would we find coffee and a table or two?

Filtered for coherent narratives


The Phantom Time Hypothesis suggests that the early Middle Ages (614-911 A.D.) never happened the implication of which is that Charlemagne was a fictional character and that the year is not 2015, but actually 1718. Somebody jumped the calendar forward; documents were forged.

Mixtape of the Lost Decade: evidence is mounting that points to a 'lost decade' between what we now remember as the 1970s and 1980s. Art, toys and music are all rediscovered -- a distinct era, the 19A0s.

The Internet was better during the 19A0s.


The city of Guntrum at OpenGeofiction, a Google Maps-style collaborative fictional world...

This world is set in modern times, so it doesn't have orcs or elves, but rather power plants, motorways and housing projects. But also picturesque old towns, beautiful national parks and lonely beaches.

(About.), founded 1999, was a web search engine that mapped results to a virtual reality representation of the continent of the same name. More:

The display of search results was a 3D landscape, complete with clusters of structures (related topics) and multilevel buildings (important sites).


The impetus for the Visual Net is Mr. Bray's long-held belief that users find a shared landscape a comfortable, intuitive way to explore various types of information ... a "shared landscape" makes complex arrangements of data usable by the human mind.


A timeline of events in the history of the Pokemon world.

Transformers: A History.


Surkov is one of President Putin's advisers, and has helped him maintain his power for 15 years.

Adam Curtis on Vladislav Surkov and non-linear warfare [video]:

[Surkov] came originally from the avant-garde art world ... what Surkov has done is to import ideas from conceptual art into the very heart of politics. His aim is to undermine peoples' perceptions of the world, so they never know what is really happening.

[creating a politics where] no-one was sure what was real or fake ... A ceaseless shape-shifting that is unstoppable because it is undefinable

A war where you never know what the enemy are really up to, or even who they are.

[using] the conflict to create a constant state of destabilized perception, in order to manage and control.

We live with a constant vaudeville of contradictory stories that makes it impossible for any real opposition to emerge, because they can't counter it with any coherent narrative of their own.


Surkov published a short story in 2014, just before the Russian invasion of Crimea, Without Sky, set in the future, after the 'fifth world war.' Review in the LRB:

It was the first non-linear war. In the primitive wars of the 19th and 20th centuries it was common for just two sides to fight. Two countries. Two groups of allies. Now four coalitions collided. Not two against two, or three against one. No. All against all.

Surkov: The only things that interest me in the US are Tupac Shakur, Allen Ginsberg and Jackson Pollock. I don't need a visa to access their work. I lose nothing.

Russia. Pollock. Modern art was a CIA weapon -- The Central Intelligence Agency used American modern art - including the works of such artists as Jackson Pollock, Robert Motherwell, Willem de Kooning and Mark Rothko - as a weapon in the Cold War. Why?

Because in the propaganda war with the Soviet Union, this new artistic movement could be held up as proof of the creativity, the intellectual freedom, and the cultural power of the US. Russian art, strapped into the communist ideological straitjacket, could not compete.


The centrepiece of the CIA campaign became the Congress for Cultural Freedom, a vast jamboree of intellectuals, writers, historians, poets, and artists which was set up with CIA funds in 1950 and run by a CIA agent.

Several inches were cut from Jackson Pollock's Mural by Marcel Duchamp in 1943, so it would fit in Peggy Guggenheim's apartment.

Those inches of canvas have never been found.

Filtered for bad things


Mountains that talk about bad things.

It's weird:

two friendly mountains loudly reciting tweets from around the world. ... [a tweet using the word 'bad'] is picked up and converted by a text-to-speech engine, then loudly recited by the mountains in realtime.

You can also visit this world by using a smartphone to have a 360 degree VR experience as one of the villagers living below the mountains.

Bit shonky in that sometimes the voice of the mountains disappears for me, or the tweets doesn't come through. Reload and retry.


Osper is mobile banking - and a credit card - for kids. Jeez, I'd happily use this.

There's a new mobile-only bank coming to the UK, called Atom.

I'm into this. Unbundle the banks. Experiment with different interfaces for consumer banking.


Hey, so what if the dinosaurs were raptured? Like they were all good Christians and they all ascended and the mammals and birds are the Left Behind and it wasn't a meteor after all.

That's what I thought this book was about, but it turns out Rapture of the Raptor is dinosaur erotica instead.

See also: Taken by the T-Rex.


Crystal shows you the best way to communicate with any coworker, prospect, or customer based on their unique personality.

It's an email plug-in that tells you what phrases to change.

Crystal analyzes public data to tell you how you can expect any given person to behave, how he or she wants to be spoken to, and perhaps more importantly, what you can expect your relationship to be like.

More on conversational UIs

ICYMI, last week I dropped a ton of links + speculation on text messaging as user interface... Read it here. Alternatively catch up with:

  • Futures of text by Jonathan Libov of Union Square Ventures is a far, far better article than the one I wrote: A survey of all the current innovation in text as a medium. Plus: animated GIFs
  • Wired: The future of UI design? Old-school text messages is a quicker, more readable overview, and some neat extra points... It may always feel silly to talk out loud to Apple's virtual assistant; maybe Apple should let us text Siri instead.

I wanted to add a few more links.

Lark is a weight-loss coach that communicates with you exclusively through messaging.

Hello Lamp Post (detailed project page) is a playful SMS platform, inviting people to strike up conversations with familiar street furniture using the text message function of their mobile phones. Including escalating intimacy:

To help players feel as though their relationship with objects could develop, we built in a friendship mechanic - initial conversations would be a bit small-talky, about the weather and observations on the local environment, but on repeat visits the questioning of the objects would change, to focus on opinions, memories and beliefs.

(Unique qualities of text-based conversational UI... user-initiated conversations and app-initiation conversations feel the same, unlike regular apps; the element of time allows pauses and rhythms, like free-to-play games; it's how we already talk with our friends.)

Designing for text-based interfaces is going to take some experimentation.

What is conversation? is some decent theory... might be useful as a framework to talk about how conversations are structured and what's they're for. (Thanks @matt_thinkux.)

The word "just" creates a parent/child relationship. The article is in the context of women in the workplace, but this is an important point about language: Should a bot display deference? What should its stance be?

I'm definitely more into how all of this feels -- Alexis Lloyd (at the New York Times Research & Development group) wrote up her experiments: Our friends, the bots? I was curious to see what it would feel like to have a bot that was trying to engage as part of a social group

I haven't yet found the right words to characterize what this bot relationship feels like. It's non-threatening, but doesn't quite feel like a child or a pet. Yet it's clearly not a peer either. A charming alien, perhaps? The notable aspect is that it doesn't seem anthropomorphic or zoomorphic. It is very much a different kind of otherness, but one that has subjectivity and with which we can establish a relationship.


The conversation about how to define the bot's relationship to us really elucidated the idea that we are moving toward one member called "non-human mental models". We are beginning to understand machine subjectivity in a way that is in keeping with its nature rather than forcing it into other constructs, like a person or an animal.

This I love.

It's not just bots. How do we speak with non-humans, on their own terms? What does a bot want? Or a penguin, or a rock, or the military-industrial complex. Do we need human translators who can hold empathy for them on our behalf? Do we need a speaker for the thermocline? See also: The Author of the Acacia Seeds, Ursula K. Le Guin.

There's a hashtag used by speakers for the bots: #botALLY.


we are kind and gentle botmakers, allies to bots of all kinds and creeds

Found via that tag, a tool to help make Twitterbots: Cheap Bots, Done Quick!

e.g. @infinitedeserts, an infinity of deserts, each more infinite than the last.

(I'm no stranger to twitter bots, I made a presence machine and retold 99 Secrets -- both now silent.)

More on writing twitter bots, without code. More on writing twitter bots, with code.


Telegram Bot Platform. (Telegram is a messaging app with 60+ million monthly active users; it's growing fast.)

Bots are simply Telegram accounts operated by software - not people - and they'll often have AI features. They can do anything - teach, play, search, broadcast, remind, connect, integrate with other services, or even pass commands to the Internet of Things.

Neat about Telegram's approach, #1: Bots can now provide you with custom keyboards for specialized tasks (examples are shown). Any good bot platform is going to have to do this, typing is too cumbersome otherwise.

Neat about Telegram's approach, #2: any message from your bot forwarded to a person or group is a messaging equivalent of a retweet - bots are viral.

The really unique feature about conversational UIs is that messaging is social. Introductions can be made. Bots can take part in group conversations; facts can be remembered and shared. There's a figure and a ground.


Filtered for computers and birds


I... look... just... this super freaky image of squirrels is a picture drawn by a computer on its own.

This squirrel has a weird amount of eyes, yeah? And seems to be made at least partially of dogs? Check out its weird rear appendage, which is composed of slug tentacles that are themselves composed of birds. A two-headed fish lurks in the foreground, and upon reexamination the background is not mere swirls, but a warped, repetitive city

More of the backstory on the Google Research Blog.

It's a crazy process that works by turning up the gain on an artificial computer brain. Giving it acid...

We ask the network: "Whatever you see there, I want more of it!"" This creates a feedback loop: if a cloud looks a little bit like a bird, the network will make it look more like a bird. This in turn will make the network recognize the bird even more strongly on the next pass and so forth, until a highly detailed bird appears, seemingly out of nowhere.

Here's a gallery: Going deeper into neural networks.

Cloudscapes with hidden cities. Leaves that turn into birds and insects. Angels in the architecture.


There are so many things to love about Paul Ford's epic Businessweek-takeover essay What is Code? but I'm taken with his insistence on treating the computer as a physical thing -- which is of course it is, all teeny-weeny electrons and stuff...

What are the steps and layers between what you're doing and the Lilliputian mechanisms within?

Just as the keyboard is waiting for a key to be pressed, the computer is waiting for a signal from the keyboard. When one comes down the pike, the computer interprets it and passes it farther into its own interior.


What do they call turkeys in Turkey?

Answer: Hindi.

In Hindi, a turkey is called Peru, which is itself a borrowing from the Portuguese.

The Language of the Birds is an angelic language, a divine language which predates and supersedes human speech, used by birds, understood by the initiated.

There's a funny-peculiar reference in Understanding Media (Marshall McLuhan) to Jason's quest for the Golden Fleece, and his sowing of dragon's teeth which grew to become the spear-tips of a new army. McLuhan enlists the legend of the dragon's teeth as an allegory for the development of the alphabet and, at the same time, bodies of armed men.

Saying thank you in Hindi destroys intimacy.


Data furnaces:

Microsoft has a lot of servers, mostly sitting in large data centres, producing huge amounts of heat -- heat that is a massive nuisance to deal with. ... Instead of finding a novel way of transferring waste heat away from the data centre, the research paper proposed that the servers themselves should be placed in homes and offices, where the waste heat could be used directly.

Data furnaces arrive in Europe: Free heating, if you have fibre Internet. Nerdalize is rolling out eRadiators in the Netherlands, providing 1000W of heat.

On conversational UIs


When trumpets were mellow

And every gal only had one fellow

No need to remember when

'Cause everything old is new again

-- Peter Allen.

Stand back folks. I've not spent any time editing and now I'm going out. This is stream of consciousness, and it's long.


There's that bit in the great article on Chinese mobile UI trends about how there are no websites, there's just messaging. And not only that, some weird mish-mash of talking robots and customer service people:

Many institutions that otherwise would have native apps or mobile sites have opted instead for official accounts. You can send any kind of message (text, image, voice, etc), and they’ll reply, either in an automated fashion or by routing it to a human somewhere. The interface is exactly the same as for chatting with your friends

You know, and why the hell not. I have one language to use with apps (pointing, tapping, swiping) and another with my friends (chatting). Why not chat with my apps too?

So as Benedict Evans - mobile and technology analyst extraordinaire - points out, messaging is the new app platform:

[In WeChat, in China] You can send money, order a cab, book a restaurant or track and manage an ecommerce order, all within one social app. So, like the web, you don't need to install new apps to access these services, but, unlike the web, they can also use push and messaging and social to spread.

The other piece of the puzzle here, Evans continues, is the smartphone notifications panel:

That pull-down panel aggregates activity from everything on your phone, and Google and Apple have made notifications actionable and given them payloads. ... More and more, one's primary interaction with any app, social messaging or otherwise, is a little pop-up with a button or two.

So I've long been interested in the idea that "next actions" should float away from their apps and come together in a single place... SNAP was my 2008 take on this.

But I guess the 2015 twist is that everything old is new again, and we're dealing not just with actionable notifications, but robot-generated text that we can have an actual conversation with.

Which is Twitter's fault.

Nowhere is it more evident that The Web is a Customer Service Medium. Just look at @AskNationwide (with replies) or @British_Airways (with replies). It's all conversations with customers.

The canonical automated version of this is @andy_house (2009), aka the house that twitters, or see House of Coates for a more up-to-date take on the tweeting smart home.

Now imagine it wasn't just an activity feed, but you could talk back.

A big bit of the current excitement is the rise of Slack for workplace comms and its embrace of bots. Which takes us to Ben Brown's insanely incredible insight: What happens when you start automating workplace processes?

What if there was a meeting runner bot that automatically sent out an agenda to all attendees before the meeting, then collected, collated and delivered updates to team members? It could make meetings shorter and more productive by reducing the time needed to bring everyone up to speed.

We've just been through an era where management has been regarded as the essential scarce resource of a business, and operations and technology are functions to be outsourced to fungible workers like so many cogs. But what if the core business resource is human ingenuity, and it's management that can be turned into software... automated and optimised?


Digit is an automated savings tool: Every few days, Digit checks your spending habits and removes a few dollars from your checking account if you can afford it.

The kicker: You communicate with it via text message ("Great, I've moved $10.00 to digit"), they have no plans for an app. And what's interesting to me is that it has adaptive behaviour... and maybe because of the text message interface, this Digit review semi-anthropomorphises the software:

At first, Digit was really cautious with my money ... But over the next couple weeks, as my balance recovered from holiday spending, it got a bit more ambitious

Software isn't "cautious" or "ambitious", those are qualities of alive beings. But maybe it serves us to think so.

Walkadoo is a walking game that encourages activity; you communicate with it by text message. Related: Autom is a robot weight-loss coach with big blue eyes. You lose more weight because you regard it as having a mind.

Rhombus is an e-commerce platform for shop to chat with customers by text message... and also accept payments, within the conversation. Related: Twitter's in-feed Buy Now button which is a game changer if widely rolled out.


One of the problems with text interfaces is text entry. Keyboards suck, especially on mobile devices. Typing also introduces a discoverability question: How do you know what words are valid right now, or the right grammar to use? How do you make complex statements?

In the game Lifeline (iPhone/Apple Watch) you're texting an astronaut called Taylor who is marooned on a moon. It works in real-time... when Taylor hikes to a location an hour away, you won't hear from her till she gets there. You text back, but it's not free text entry, you only get two options at a time. Works well on the smartwatch too.

Despite the constrained responses, it still feels conversational. Enough that the first time I killed Taylor - by freezing to death based on advice I'd given - well, ooof.

Lifeline was prototyped in Twine, the visual programming language for writing interactive fiction. See also Inform 7 where the code-behind-the-fiction reads like a book but every word has its code-meaning too, like casting a spell in a stupid universe, like talking to a golem. e.g.:

The dark doorway is an open door. "The doorway looks ordinary enough, but it's so difficult to be sure with the unaided eye." The dark doorway is scenery. The dark doorway is not openable. The dark doorway is west of Longwall Street and east of Turret Roundhouse. The dark doorway is trapped.

Another take on text input:

The web-based game A Dark Room (also available for iPhone) is astounding. Half text adventure, half point-and-click. There's something about clicking stoke fire and the words turning into a progress bar while the fire burns down... The communication of the element of time.

Hangkeys is a neat hack -- it's a custom iPhone keyboard that makes it super easy to play Hangman over SMS, Whatsapp, or other text message services.

Meet by Sunrise is a custom smartphone keyboard that integrates with your calendar: Instead of typing times and locations, you tap them instead.

Matt Galligan imagines tap-able buttons in text messages: What if instead of installing an app, we might instead allow a service to chat with us via iMessage?

Writing code... Swift allows emoji for variable names. There's something interesting about this. Variable names like "a" or "theCounter" or "dimensions" are meaningful... but what about underlying feelings they carry? The "ii" counter of a tight inner loop always has a zing for me, it's the twang of high-tension power lines.

So can emoji carry more meaning, or meaning along a different axis? What could we use this for -- instead of "houseAddress" just have a picture of a house; instead of saying errorDescription just use smiling pile of poo emoji.


I was noodling with conversational UIs in 2002, back when AIM was a thing. What preoccupied me then - and what interests me most now - is how to make an automated conversation with a bot not boring, and (more importantly) not shit.

So one of the things I found before: The more the bot acts like a human, the more it will be treated like a human. And the more that happens, the more likely the bot will have to say "I don't know what you mean"... which is lame.

Our guiding light is The Jack Principles from the 1995 quiz show video game You Don't Know Jack. In short, how to direct the conversation so a user will never be in a position to ask a question the machine can't answer.

(Tom Armitage collected a link to this and more when he ran a chatbot project at Berg back in 2010.)

Something else surprised me about authoring conversations, something almost contradictory...

  • Leading a user through a conversation by the hand is great... once. It's like a phone tree: a branching tree of questions and multiple choice answers where nothing can go wrong. But if the user wants to change their answer, compare different possibilities, or run through the conversation a second time: It's tedious and frustrating.
  • But the alternative is a wide open space: The user can say anything, and it's up to the bot to interpret and respond. However that introduces a discoverability problem. You can be chatting to your bot about what's on TV tonight, completely unaware that it also knows what movies are showing nearby. Or let's say you do ask about movies, and then you say "well how do I get to the theatre?" and suddenly it's dumb again. Using Siri is often like that.

I guess what I'm asking is how does a user have a theory of mind about a bot - a conception of its stance, intentions, domain of knowledge, etc - and how is that communicated.

My take back in the day was to organise knowledge into domains, within which a tree structure would be possible but avoided. To summarise how it worked:

  • I built an AIM bot called FingertipTV. You could traverse it like a tree: "now" (reply: "a list of channels and what's on now"), "bbc1" (reply: now and next for BBC1), "9pm" (reply: what's on BBC1 at 9pm and the following show)
  • Resolution varied. "now" would result in a list of every channel but just the show titles (and you could hit a number to get the full description). Getting more precise with the channel and "9pm" would show a full show description.
  • Once you'd learnt the vocabulary, everything allowed shortcuts: The user could gradually become virtuoso. So the previous exchange could be replaced with "bbc1 9pm"
  • There was a limited amount of history. So saying "bbc1" then "later" would show what was on BBC1 later... but saying "bbc1", "9pm", "channel4" would show Channel 4 right now. The 9pm itself was forgotten. Use any excuse to limit the memory of the bot, because that constrains what the user has to hold in mind
  • I added functionality to search movie listings... but encapsulated that in another AIM bot called Cinebot. If you asked about movies to FingertipTV, you'd get the reply "I can't help you but my friend can," and Cinebot would reach out and start a new conversation. Now you'd met Cinebot, and could add their name to your buddy list. Small, understandable, almost-stateless bots that talk to you and to each other

My point, I guess, is that a new medium needs a new grammar and conversational UIs are definitely a new medium.

For one -- they're intrinsically social. If I'm chatting with a bot in iMessage about what movies are on nearby, shouldn't I be able to turn that into a group chat with my partner? And does the bot conduct two separate conversations, one with each of us, or assume we're both searching for the same movie?

We'll need app frameworks to help author these bots, and the frameworks will make assumptions about how conversations work in small groups.

And you know what, we're still in the PC era, the era of personal computing. We don't really know how to use computers in small groups, how to use interfaces collaboratively.

Another big question for me is what happens when we have many of these bots...

How do traverse multiple knowledge domains, discovering features and adapting how we speak as we go? Is it going to be like FingertipTV and Cinebot, loosely coupled domains of differing knowledge and vocabulary? Or more like Siri where the same voice can tell you everything from "directions home" to "how long do dogs live" (to pick two examples that it's giving me right now).

Maybe it will be like Twitter, where everyone I follow is in a single stream -- but I miss what they're saying? Or like my apps on my phone, where many apps have their own activity stream... and they fight so hard to get heard that they constantly spam my notifications panel?

My head goes somewhere quite speculative,

and that's text adventures.

Dan's story, Text Only:

What now? >


You go North

The batteries on your Discman are almost depleted.

Or Julian Dibbell's memoire My Tiny Life, a tale of LambdaMoo, a collaboratively built text environment inhabited by objects, bots, and people.

There's a strength in spatialising information -- of arranging these bots - these domains of knowledge and differing patterns of interaction - into a web, or on a map. Some bots are closer to other bots.

So part of me wonders... what if I saw the activity feed from my smart home all together in one place, and then when I went north, say, that would be the activity feed from my social networks. Or instead, in my smart home I'd find my TV, and inside that we could have a chat about what's being tivo'd tonight.

I don't want to be too literal. But maybe we need an architecture to arrange all these bots and feeds and conversations, etc. And while our experience of the conversations will vary (we'll be friends with different bots) the architecture will be shared: Arbitrary, but shared. A cyberspace:

A consensual hallucination experienced daily by billions of legitimate operators ... [a] representation of data abstracted from banks of every computer in the human system.


And don't throw the past away

You might need it some other rainy day

Dreams can come true again

When everything old is new again

When everything old is new again

I might fall in love with you again


ps. More on conversational UIs.