Let’s Talk! Human-Computer Interaction: Dialogue, Conversation, Symbiosis (2)

[Part 1]

I chuckled when I read Rebecca Solnit describing her 1995 life: She read the newspaper in the morning, listened to the news in the evening and received other news via letter once a day. Her computer was unconnected to anything. Working on it was a solitary experience.

Fast forward 20+ years and her computer, like most other people’s, feels like a cocktail party, full of chatter and fragmented streams of news and data. We are living permanently in Alvin Toffler’s information overload. We are creating more data per second than we did in a whole year in the 1990s. And yet, data or information exchange is why we communicate in the first place, so I wanted to ponder here, how do we talk using computers?


Originally, you had to ask computer scientists like me. And, we had to learn the commands of the operating system we were using say, on a mainframe with VAX/VMS or DEC; on a networked workstation with UNIX, or a personal computer which used MS/DOS.

Then, we had to learn whatever language we needed. Some of the procedural languages I have known and loved are: Assembler, Pascal, COBOL, ADA, C/C++, Java, X/Motif, OpenGL (I know I will keep adding to these as I remember them). The declarative PROLOG, and (functional, brackety) LISP, and scripts like php, Perl, Python, Javascript. The main problem with scripts is that they don’t have strong types, so you can quite easily pass a string to an integer and cause all sorts of problems and the compiler won’t tell you otherwise. They are like a hybrid of the old and new. The old when computer time was expensive and humans cheap so we had to be precise in our instructions, and the new computers are cheap and humans cost more, so bang in some code. Don’t worry about memory or space. This is ok up to a point but if the human isn’t trained well, days may be lost.

As an undergraduate I had to learn about sparse matrices to not waste computer resources, and later particularly using C++ I would patiently wait and watch programs compile. And, it was in those moments, I realised why people had warned me that to choose computers was to choose a way of life which could drive you mad.

How things have changed. Or have they?


When I used to lecture human-computer interaction, I would include Ben Schneiderman’s eight golden rules of interface design. His book Designing the User Interface is now in its sixth edition.

When I read the first edition, there was a lot about dialog design as way back then there were a lot of dialog boxes (and American spelling) to get input/output going smoothly. Graphical-user interfaces had taken over from the command line with the aim of making computers easy to use for everyone. The 1990s were all about the efficiency and effectiveness of a system.

Just the other week I was browsing around the Psychology Now website, and came upon a blogpost about the psychological term locus of control. If it is internal, a person thinks that their success depends on them, if it is external their success is down to fate or luck. One of Scheidermann’s rules is: Support internal locus of control, so you make the user feel that they can successfully achieve the task they have set out to do on the computer because they trust it to behave consistently because they know what to expect next, things don’t move around like the ghost in the wall.

Schneiderman’s rules were an interpretation of a dialogue in the sense of a one-to-one conversation (dia means two, logos can mean speech) to clarify and make coherent. That is to say: One person having a dialogue with one computer by the exchange of information in order to achieve a goal.

This dialogue is rather like physicist David Bohm’s interpretation which involves a mutual quest for understanding and insight. So, the user was be guided to put in specific data via a dialog box and the computer would use that information to give new information to create understanding and insight.

This one-to-one seems more powerful nowadays with Siri, Alexa, Echo, but, it’s still a computer waiting on commands and either acting on them or searching for the results in certain areas online. Put this way, it’s not really much of a dialogue. The computer and user are not really coming to a new understanding.

Bohm said that a dialogue could involve up to 40 people and would have a facilitator, though other philosophers would call this conversation. Either way, it is reminiscent of computer supported cooperative work (CSCW) a term coined in 1984 that looked at behaviour and technology and how computers can facilitate, impair, or change collaborative activities (the medium is the message) whether people do this on the same or different time zone, in the same or different geographical locations, synchronously or asynchronously. CSCW has constantly changed and evolved especially with the World Wide Web and social media.

I remember being at an AI conference in 1996 and everyone thought that the answer to everything was just put it online and see what happened then. But just because the WWW can compress time and space it doesn’t follow that a specific problem can be solved more easily.

Monologue to Interaction

The first people online were really delivering a monologue. Web 1.0 was a read-only version of the WWW. News companies like the BBC published news like a newspaper. Some people had personal web pages on places like Geocities. Web pages were static and styled with HTML and then some CSS.

With the advent of Web 2.0, things got more interactive with backend scripting so that webpages could serve up data from databases and update pages to respond to users input data. Social media sites like Flickr, YouTube, Facebook, Twitter were all designed for users to share their own content. Newspapers and news companies opened up their sites to let users comment and feel part of a community.

But this chatter was not at all what Bohm had in mind, this is more like Solnit’s cocktail party with people sharing whatever pops in their head. I have heard people complain about the amount of rubbish on the WWW. However, I think it is a reflection of our society and the sorts of things we care about. Not everyone has the spare capacity or lofty ambition to advance humanity, some people just want to make it through the day.

Web 3.0 is less about people and more about things and semantics – the web of data. Already, the BBC uses the whole of the internet instead of a content management system to keep current. Though as a corporation, I wonder, has the BBC ever stopped to ask: How much news is too much? Why do we need this constant output?

Social media as a cocktail party

But, let’s just consider for a moment, social media as a cocktail party, what an odd place with some very strange behaviour going on:

  • The meme: At a cocktail party, imagine if someone came up to us talking like a meme: Tomorrow, is the first blank page of a 365 page book. Write a good one. We would think they had banged their head or had one shandy too many.
  • The hard sell: What if someone said: Buy my book, buy my book, buy my book in our faces non-stop?
  • The auto Twitter DM which says follow me on facebook/Instagram/etc. We’ve gone across said hi, and the person doesn’t speak but slips us a note which says: Thanks for coming over, please talk to me at the X party.
  • The rant: We are having a bit of a giggle and someone comes up and rants in our faces about politics, religion, we try to ignore them all the while feeling on a downer.
  • The retweet/share:That woman over there just said, this man said, she said, he said, look at this picture… And, if it’s us, we then say: Thanks for repeating me all over the party.

Because it is digital, it becomes very easy to forget that we are all humans connected together in a social space. The result being that there’s a lot of automated selling, news reporting, and shouting going on. Perhaps it’s less of a cocktail party more of a market place with voices ringing out on a loop.

Today, no one would say that using a computer is a solitary experience, it can be noisy and distracting, and it’s more than enough to drive us mad.

How do we get back to a meaningful dialogue? How do we know it’s time to go home when the party never ends, the market never closes and we still can’t find what we came for?

[Part 3]

2 Replies to “Let’s Talk! Human-Computer Interaction: Dialogue, Conversation, Symbiosis (2)”

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.