Christopher Dixon

June 5, 2008

Course Content

There is great potential within the realm of human-computer interaction that opens up innumerable technological possibilities. Practically anything is possible as technology advances. We now have digital music libraries that take up only inches of space when smaller collections took up many feet of space twenty years ago. The rise of mobile devices would have been unfathomable fifty years ago, but yet, progress is still made. An important point the text makes, however, is that designers of technology must decide what services they will provide both today, with a look towards the future, and open to developments as they occur.

This course discusses the design of “interactive products to support the way people communicate and interact in their everyday and working lives (Sharp, Rogers & Preece, 8). While the promise of one universal device and software design remains, today’s reality is that interfaces must provide distinct services and provide them well, based upon widely variable user needs. Let’s list a few basic principles for interactive design:

Redesigned or new interfaces must provide improved services;
Interface design must make the technology’s services clear and explicitly define ways for users to access those services;
As much as possible, designs must facilitate natural communication; and
Only through thoughtful and careful consideration of needs and technological constraints, in conjunction with user interests, can useful designs be created.

The book enumerates many processes in order to accomplish useful designs. It draws attention to two particular facets that should be mentioned: multiple opportunities for user feedback and the cycle of feedback and revision. Because people process information differently, their attention goes to different things, and their conceptual understandings differ, it is necessary to produce a design that is clear across a population instead of a user sub-group. There are very few things these days, with the democratization of information, to design an interface specifically for a user group (in exclusion of others). It is likely that the user populations for an interface are sufficiently large enough and public enough that the design must take into account people from all backgrounds. Even if an interface is designed for specific groups, will those groups remain the same for the interface’s lifespan?

By inclusively designing an interface with the users in mind, any interaction will tend to be more successful than when “experts” disregard user opinion. It is by asking for user feedback that designs ensure the interface fulfills their needs appropriately. Each time the design is updated, users should again be asked if it makes sense, the reasons why, and further suggestions until a dependable, revised interface is created.

This broad, user-centric approach is particularly invaluable when designing for social interactions. In particular, computer-supported collaborative learning (CSCL) requires smooth communication for effective knowledge building activities. In order to match up user requirements with the CSCL community, let’s look at the goals an interface should strive for:

Must be able to support synchronous activities including ways to talk to each other (text or via voice), share resources and point to particular data
Must be able to support asynchronous learning as a stable repository of knowledge and record of interaction
Must be able to describe the status of group members (online or offline, what they’re working on, etc.)
Must be able to assimilate emerging technologies and adapt to particular group dynamics

An interactive interface supporting CSCL should try to include features that mimic important features of both face-to-face interaction and computer-based tools. Collaborative group space is important for work utilizing the CSCL framework. For group members to learn create shared knowledge, it is necessary not only for thoughts to be shared, but also pieces of knowledge like images, text quotes or data sets. With Web 2.0 technologies, integrating social tools is rather natural and shouldn’t be a great leap forward for most users. By integrating industry-standard productivity software as a last step, an interface can truly serve as a mobile office.

I caution that while a collaborative environment should be able to handle any interaction the group requires, an everything-to-everyone solution surely cannot adhere to the goal of interface simplicity. One way to incorporate the possibilities of everything without actually incorporating all component parts, is to rely on open-source technologies and an adaptable interface (e.g., widgets). This flexibility will allow groups to tailor interfaces to their particular needs and dynamics.

In order to ensure the user perspective is being taken into account, an evaluation of some sort should occur. Initially skeptical of heuristic evaluation, I now advocate to use standard accessibility guidelines to evaluate any collaborative interface. Here are the component parts of a heuristic evaluation (Sharp, Rogers & Preece, 686-687):

Transparency of system: ensure a system response, even a busy signal, when users interact with the interface
Keep real world virtually: keep world conventions in the virtual space so users have established, ready-made guidelines for interactions
User control and freedom: allow users to take steps backwards and forwards as their needs warrant
Consistency: integrate commonly used references into the design for clarity
Error prevention: ensure a smooth working interface and safety features to ensure problems don’t occur initially
Recognition and not recall: stay away from requiring users to memorize site features and rather make the interface make sense so that they can figure things out for themselves (even if the task is new)
Efficiency of use: allow for shortcuts for unnecessary or repeated tasks
Proper design: make the interface easily understandable and navigable
Help users recover: users should be able to know when errors occur and the system should provide possible solutions to fix errors
Documentation: users should be able to find additional information on the interface to increase or find relevant functionality

As with development and evaluation in many sectors of life, developers should facilitate interface creation by users instead of functionality being dictated by service providers. The extra time and money in development will save from project overages and lateness and ensure product longevity (relative longevity).

Group Project

Our Group Project sought to capitalize on emerging social technologies like Facebook and blogs, and marry those with research tools in order to support CSCL research. It is quite natural to build upon established technologies that foster communication. Group F’s prototype sits within the Facebook site in order to easily demonstrate the many possibilities for collaboration including citulike, library resource support, discussion groups, personalized workspaces, and posting contact information. While a usable prototype would be removed from Facebook, it was thought to be the best option for class purposes.

While the prototype shows possibilities for asynchronous social networking, it lacks meaningful support of synchronous dynamics. While chats can occur, the interface will have trouble with groups collaborating and sharing ideas as the VMT software provides. I suspect the workspace feature included in the design is likely not powerful enough for effective group learning. While asynchronous space can easily be ready-made, the synchronous environment would be dramatically improved by incorporating specialized features.

A synchronous environment could provide features such as web browsing, web content creation, multimedia hosting and creation, graphics programs, training modules, training interfaces, and virtual desktops, among other possibilities, in line with group needs.

There is also no easy way in which to include productivity software with today's technology and is surely a problem in other applications as well. While Google’s online productivity suite is an improvement, it still lacks integration in a group-learning interface.

Facebook attempts at keeping tabs of account holders should they wish to list their current activities (ala twitter) and I think the group sought to incorporate this. The class seems in favor of allowing people to share their current activities to promote group awareness.

With more time, I foresee the inclusion improved synchronous operability and further refinements with the CSCL community in mind (with appropriate community feedback, of course).

Course Dynamic

My honest opinion is that the course is trying to do too much. My experiences during the first few weeks of class involved a large amount of time collaborating with group members, reading book chapters and additional articles. It easily took 20 hours those first few weeks, which is time I don’t have. I’m unsure of the norm for class hours, but another of my classes takes around 5 hours per week, which is probably low.

While the tasks are weighty, the content is good and valuable in the overall discussion. I’m actually advocating for more course streamlining than substantive alterations. Perhaps individualizing the group readings and only including the group project towards the end, when the textbook readings are lighter, is an option. I hesitate to be this honest, but I only voice my opinion and its only one voice.

Here are a few reflections on the group work:

I have little problem with group work, but it is very hard to maintain the dynamic over an entire quarter. I’m an individual learner and tend to do assignments spread over many days, pieces at a time really, and usually at the last minute. I understand that group work is valuable in any professional circles and in my professional life I’m all for it and actually good at it. However, scheduling work, personal responsibilities, and differing schedules between younger undergraduates and older graduate students (I’m one) is perhaps too difficult.
Early on, I had to put homework aside in order to concentrate on last minute preparations for a conference of which I was a co-chair. It was certainly my choice, but my priority went with the activity that had implications for my professional reputation (I committed to this 2 years ago). [I didn’t really expect the 12 hours days I was spending either.] It’s hard to feel like part of the group when you’ve failed your group members.
Since I wasn’t part of the initial prototype, I felt little credibility in offering opinions for change with that interface. I know I would take it personally if someone not included in the initial discussions offered even constructive criticisms later.
Even though I’d like to be able to schedule weekly meetings, my schedule is dictated by family circumstances. My weekends are my chance to catch up on school, household chores and sleep and these are often interrupted as my wife works weekends and we have a small child. Please understand this and the points above are not excuses, but are simply included to illustrate the implications of long groups projects within the reality of life. I opted for online because of its flexibility and that my study dynamic usually includes intense highs of activity with valleys of inactivity.
My personal opinion is that I would have rather prototyped something myself (even thought my midterm didn’t go over well) and started the collaboration afterward.

I think the aims of the course are good. Collaboration between undergraduate and graduate students is a novel goal. But that also leaves group members out of some assignments (like the presentation) as well as different levels of work (required readings). I assume it’s an attempt to share knowledge and you’ll have to see if was successful and whether my opinion is an outlier. It may be just that. I haven’t had time to relax and recoup in two or three semesters and I’m exhausted and frustrated at the end of my graduate school career (hopefully) and thus venting. I hope these candid observations will be taken constructively rather than a personal attack on anyone.