Today’s topic is anticipation and context, two themes which drive a great deal of our human interactions. Even when two people don’t know each other very well, a shared context gives them an enormous head start in mutual understanding. As I started to write this, I carried out a little experiment. I typed into a Google search page “who won the cricket world cup?”, and Google correctly guessed that I was interested in the most recent one – last year – and gave me the right answer.

Then I typed in “where was it played?” – a question that any human conversationalist would recognise as directly referencing the first question. Of course the Google web search engine has no such context, and the first page of suggested links included some pages on table top board games, medieval music, the date when the musical Hair was released from censorship (1968), and a song by the Dave Matthews band. Can you imagine a set of answers like that turning up in a pub conversation about cricket?

Was this a fair test? Clearly not, one might say. Yet it shows how even a hugely successful search engine can be woefully inadequate at recognising context. If we are going to think of software as intelligent, it has to be able to offer suggestions which are contextually appropriate – answers to questions, for sure, but also behavioural changes that adjust to changing situations and an awareness of what I am actually seeking. If something cannot adjust like this, we are unlikely to think of it as intelligent.

Now, all of the major players in the online world are aware of this, especially when the results are being delivered to a mobile phone where my patience level, reliability of connection, and willingness to trawl through dozens of potential matches are all pretty low. We want the system to know enough about us that we don’t have to pedantically explain the same stuff over and over again. Mobile search is going this way, and mobile route planning is already there.

But… the flip side of that is security, or trust, if you prefer. Do you want the major online players to know so much about you that they can anticipate your every whim? Who else gets to know that much about your actions? In the debate about convenience versus privacy, many – perhaps most – people actually want convenience. Privacy is hard work: it is technically difficult and, usually, inconvenient. Lots of people can’t really be bothered, and simply trust that what they are doing is insufficiently interesting to attract the wrong kind of attention.

Which brings us to Far from the Spaceports. Mitnash and Slate are, in part, ethically motivated hackers. They have expertise in the art of cracking into someone else’s code, but they’re doing it in the interest of tackling crime rather than committing it. But it’s a fine line, and some of their actions probably cross a moral line somewhere:

“I just don’t know what the pair of them will get up to once they get in to the system. They’re going to take this a lot further than I would”…

A purist would say that what we planned to do was not exactly legal, any more than the code inspection at the relay buoy was. But then, there were not exactly any laws that applied to this situation…

But hacking aside, one of the most prominent features of the relationship between Mitnash and Slate is their ability to anticipate what each other wants next. For all the differences in their respective hardware systems, as a pair of coworkers, colleagues, and friends, they work on a rather unconscious level with each other.

As an aside, and I’ll be saying more about it later in the week, there’s a Facebook launch event for Far from the Spaceports next Monday, December 7th, 7pm-9pm UK time.

Facebook event banner

Carrying on the series about human-machine relationships, today’s topic is intimacy. I’m not proposing to talk about sex specifically – nor do I in Far from the Spaceports – but a much wider spectrum of close relationships.

In the book, Mitnash has a long term human partner back on Earth. She’s called Shayna, and we only actually meet her in one scene near the start, though she is a regular background presence throughout.

The main relationship that we see is with Slate, his working partner, who also happens to be female gendered. She has no physical form that would distinguish her from any other virtual persona, and with a bit of preparation can adapt herself to a wide range of available hardware.

So their relationship is not on the basis of bodily shape – I didn’t want to write an android book, and the difficulty of getting Slate close enough to the action to be useful is an important narrative ploy. But clearly they are a close-knit couple. As Slate comments to Mitnash about a particular data file she has intercepted,

“there’s actually more about me in the packet than Shayna.”

To which Mitnash replies,

“best not to tell her that, if you don’t mind.”

Their intimacy, the way I see it, rests on two things. Firstly, they share intense and difficult experiences together, supporting one another in them to the best of their ability. But secondly, they communicate with one another in a direct, constant and intense manner. Use of a cochlea implant and subvocal transmitter – originally simply to avoid having to speak out loud in situations where this would be awkward – means that Mitnash communicates not only what he is consciously framing in thoughts, but also a whole other level of half-framed thoughts and ideas.

“Slate, how much do I talk to you without knowing it?”

“All the time, Mit. You murmur to yourself while you’re thinking, and you subvocalise throughout the day. There’s very little about your thought life I don’t know. Or your fantasy life. You’re whispering to me almost all the time.”

“I suppose that means you know all sorts of things I have never told Shayna.”

We are clearly a very long way from this level of artificial intelligence just now. All of the major players in today’s online world have been working on this – Apple’s Siri is probably the best known, but there are many others. At the moment they are all quite gimmicky – after asking Siri what the meaning of life is, and showing your mates that you can send messages and be reminded about events, most people get bored with him (or her in some countries) and the level of interaction drops. Siri and that whole current generation of virtual assistants are just not interesting enough.

Sounding relational, as opposed to encyclopaedic, is a really hard problem in machine intelligence. I think most people remember with dislike Microsoft’s Office Assistant, with its cheerful chatter like, “it looks like you’re writing a letter… can I help?”. I actually thought it was a brave effort back then, but obviously I was in a minority and the whole idea was quietly dropped for another day.

The single best known benchmark for all this is the Turing test – basically you are allowed to chat without being able to see the other person, and have to decide if you are talking with a person or machine… without asking leading questions like “what are you made of?” Part of the test – certainly in the way it is conducted nowadays – is seeing how the entity at the other end deals with abrupt changes of direction in conversation, with ambiguous or poorly defined statements, and with questions where the speaker cannot possibly know the answer.

To date, nothing yet built does very well at the Turing test, despite massive improvements and changes in recent years. As I said, it is a really hard problem, and the numerous “digital assistants” already in use, succeed primarily because they are operating in a very limited domain, with a very constrained set of questions. Do I think we will get there one day? Yes indeed, but I don’t think it will be for a few years yet.

Next time… anticipation and context

I thought for the next few blogs I’d talk a little bit about artificial intelligence, seeing as how the relationship between the human investigator Mitnash and his virtual partner Slate is at the heart of Far from the Spaceports. Quite a few years ago now I used to work in AI, though at the pattern recognition end rather than personality creation.

AI has been a key strand in science fiction for many years, long before it came anywhere near possible in reality, and there is a long history of making entirely wrong guesses. Asimov’s earlier books certainly saw a key role for AI, and his Three Laws of Robotics rapidly became a basis not only for his mobile robots but also as a framework for the way other people thought about AI. But for what you might call serious work, like managing a company or a nation, Asimov was locked into the idea that the machines would be physically huge, filling whole buildings, and would need whole squads of highly specialised operatives to make them work. The concept of virtual environments which were geographically dispersed, like a company network, or indeed the Internet as a whole, escaped him.

Other writers or film makers had different blind spots. One often comes across fictional computers which are able to carry out vastly complicated calculations, analyse and direct the course of spaceships or the economies of worlds – yet output the end results of their deliberations on paper tape. It seems that the hardest things to get right are the interfaces that connect the human and machine worlds.

Often, authors have signalled the presence of artificial intelligence by means of stilted or artificial speech, failing in various ways to match human expectations. The android Data, in Star Trek, could never manage verbal contractions, so always said things like “I can not” rather than “I can’t”. This failure to attain informal speech lasted until the installation of an “emotion chip” which among other things upgraded his language faculties. Apparently verbal contractions are emotional rather than grammatical!

So I wanted to portray the relationship of Mitnash and Slate as one of normal intimacy between friends and coworkers. Each has the advantages and limitations of their particular “physiology”, and hopefully each emerges as a distinct personality. This led to a number of specific choices in the book, a couple of which I want to expand on today.

1. Slate, and the other personas, have a definite gender. Slate happens to be female, while some of the others are male. I’ve left it to readers to decide what this means, since she has no external biological indicators of gender. Some people will like the ascription of gender to machines, and no doubt some will not. There’s a sense in the book that machine gender is a relatively new advancement – Slate describes one particular persona they meet as “male, but only just”.

2. I didn’t want the baggage of clumsy language to get in the way of the relationship. So Slate is chatty, informal, but technically skilled and quirky in the way that a professional human coworker would be. Her communication is not only verbal in the strict sense, but includes a number of nonverbal noises that communicate things like satisfaction, frustration, encouragement etc – again, just like a human colleague does.

In terms of current technology we are a very long way from actually developing a persona like Slate. In recent years there have certainly been substantial breakthroughs in both hardware and software, but nothing I have yet seen persuades me that we are going to see virtual intelligences of this quality in the next decade or so. Within a century, perhaps – though this guess may be as far from the truth as guesses that others have made in the past.

Don’t forget – Far from the Spaceports is now on preorder: follow links to

• Amazon.co.uk
• Amazon.com
• Amazon.in
• and other global Amazon sites – search by name!

Next week… the role of man-machine intimacy…