The Continuous Construction of the Computer User: Visions and User Models in the History of Human-Computer Interaction - Michael Friedewald

Reason for Blog Post
Part of any master's degree involves doing a lot of reading and researching, and studying video games is no different.  This blog has been commandeered as a forum through which I will be reviewing various readings for my Human-Computer Interaction class.

And so this journey begins with a quick review of The Continuous Construction of the Computer User: Visions and User Models in the History of Human-Computer Interaction by Michael Friedewald.

Disclaimer
Let me begin by say that I am somewhat prejudiced against the academic style of writing. I think I'll write a blog post about that later. I personally think that many academic writing conventions are completely counterproductive to, well, academia. But that's a conversation or another time.

Let me also say that I did not like the writing style of this paper, so while I found it plenty informative, I still had a negative opinion of it!

Reading Purpose
For now let's just say that Michael Friedewald's essay was assigned as little more than a quick survey of the evolution of Human-Computer Interaction. It was featured early on in the book Total Interaction. And it did a pretty good job as a survey. Important names and developments are all sprinkled in for review.

As I've been a student of Digital Media for almost five years now, this paper serves as a nice neat little tool for  matching a bunch of names to their corresponding breakthroughs and stoking the fires of my memory. I've studied all of these individuals/devices before, but as their names were never applicable for longer than a test or two (and I'm terrible with names to boot!) they only stuck with me for as long as I was going to need them.

I thought the information graphic prefixing the essay was rather interesting as well. It's completely un-intuitive to look at (in that being able to interpret it relies on an intuition I appear to lack. Actually, I usually find info graphics unnecessarily opaque) but I can tell that it's trying to show a map of the author's influences and/or interests so we can get an understanding of what kind of essay we're going to read.

Writing Style
Now I'm a bit of a writer- if I wasn't going to become a game designer I'd want to be a novelist- so his writing style is almost as important to me as what he was saying.

As a virgin reader of this writing, with little to no information on his background or why he originally wrote this article, I can say that the entire thing felt mildly passive-aggressive, non-committal, and slightly pouty- like an old man who's grumbling that he can't understand his mobile phone, and certainly like he can't exactly place his finger on what he's writing about. His paper appears to be an amalgamation of him reporting on historical trends he's observed, using strange wording to hint at his feeling towards certain theories, and then evidencing a sudden and melancholy complete lack of faith in technological gurus in the last five sentences

The writing style is almost coy, mildly hinting at displeasure without actually saying what displeases him. He primarily lists historical events, only to suddenly throw in a paragraph- quite out of nowhere- at the end that suggests he feels things are spiraling out of control and that technology is leaving humanity in the dirt. Which he doesn't support with anything else, throughout the entire course of the paper (he even kind of contradicts it).

This all is a manifestation of one of my biggest problems with academic papers: frequently they refuse to be read by anyone who does not already have a comprehensive understanding of the social-political discussion that they originally took place in. They are so vague, subtle, and scattered that they cannot easily be interpreted without the comprehensive context.

But I digress.

His Point
From what I could gather from my initial read-through, Mr. Friedewald was attempting to give a short history of human-centered interaction design first by defining and discussing the development of technical determinism and social determinism, and then discussing a transition into social constructiveness, which he appears to support.

Historically, he does a good job discussing major contributors to the field and their inventions/ideas: Allan Kay, Vannevar Bush, Douglass Engelbart, Marshall McLuhan, Memex, Dynabook, the mouse, teleconferencing, the medium being the message- all good stuff.

He was obsessed with software agents, as well. To be honest, I don't think the idea of a software agent is a bad idea; I think its just beyond our current technological means to implement one well at this point in time. For the uninformed, a software agent is something like Microsoft World's Clippy. In theory, it acts like a digital personal assistance. In practice, it has only been able to severely annoy anyone who comes into contact with it.

Dynabook (Allan Kay)
I got particular pleasure out of reading about the Dynabook, because reading Kay's description for the Dynabook was like reading a spec-sheet for an iPad- and I just purchased an iPad2 recently.  At the time, Kay was restricted to the vocabulary and ideas of his time. For example, he described the Dynabook as, through an interface to a data network, being able to access the accumulated knowledge of mankind.

Now he was writing this before the truth birth of the internet, and he could never possibly know that this 'accumulated knowledge of mankind' would be accessed by things called 'Google' and 'Wikipedia', but he got pretty darn close.

Jerome Bruner
I had also never read Jerome Bruner's thesis concerning how humans break down and store experiences as mental models. The three types of abilities for how we make these models are: Sensomotoric, Iconic, and Symbolic. Want to hear how they apply? Imagine a child trying to learn the skill of 'jumping' for the first time. Sensomotoric abilities permit the child to look at another person jumping beside them, and then mimic the same actions. Iconic abilities permit the child to examine a picture of a person jumping, interpret the picture, and then mimic what the person in the picture is doing. Symbolic abilities permit a child to hear a parent describe the action of jumping, form a mental picture of how it must be done, and then try to mimic that mental picture.

Why is this applicable to video games/human-centered interface design? Easy: The designer needs to be able to create an interface that people can learn to use. An interface that users either already have a mental model for, or which users can quickly and easily build a new and accurate mental model for.

GUI
The author didn't particularly care for the desktop metaphor; something on which he and Allan Kay agreed. He lamented the failure of Microsoft "Bob" (Look it up, you're laugh. It's like a giant room of Clippys). I think it's important that authors continue to draw attention to the fact that everything about the computer and its current metaphors are fairly arbitrary. The Desktop is just a metaphor after all, and yet it structures everything about how we interact with computers. Inventive young people should know this, and be encouraged to look for alternatives, should they desire.

The author pointed out, amusingly, that many younger computer users from the "Nintendo Generation" who have gathered experience with computers from their childhood contribute to the fact that an office-based desktop metaphor is no longer necessary and may even be restrictive.  I agree with this point: young technologically savvy people definitely have different needs than the older crowd.

I also thought it was unique to note that a great many new computerized devices eschew the model of the desktop. My in-car computer, GPS, and mobile devices all lack desktops.

Old vs. New
The author spent a great deal of time discussing the LOGO programming lesson and Kay's discovery that people did not necessarily need a large amount of training in order to use and even develop for computers. Previous to this, it had been assumed that people had to be good programmers before they could use the computer effectively.  At this point, before the birth of GUI, his assertion was that a good programming language could be usable by people even by people with very little training. This translates forward into the idea that all human-computer user interface should be conceived so that the humans can understand and work with them with very little difficulty.

Yet after all this discussion and hope for the future, the author ends with the strangest conclusion, including sentences like: "An abyss is yawning between the feverish promises of cybergurus and the experience of typical computer users, who again and again {...} wonder where their data disappears [to] when the computer crashes, who have no idea why they can't get into the Internet. It seems that up to now, the digital revolution has swept along only the computer - and left the human behind."

It was actually sort of odd that he concluded in this fashion. The rest of the paper suggested that developments in HCI were making it easier for laymen to use computers. Wasn't he just a few seconds again simultaneously suggesting that A) products were being made that took into account the metaphors of traditional users like office workers, B) that a new generation of computer-savvy young people was coming up that could handle tasks like coding, and that tools are also being provided for them? Of course we can always improve, but this conclusion seemed rather melodramatic in light of the buildup he made. But on that note he appeared to disapprove strongly of technical determinism, so this may be a cautionary word of advice he's giving out to the cyberguru community.

What New Things I Learned
I will definitely be highlighting this article and using it as reference.  It's useful, and it perked my interest in looking at Allan Kay's work a little. I like going back in time to study the problems of previous times, and since I'm interested in iPad development, taking a look at what Allan Kay thought a Dynabook could be used for is actually quite interesting. To extrapolate data into the future and predict where products will be in a few years, you need not only to understand the current ideas of the present, but also the ideas of the past. After all, you need more than one point to plot a line!