Post 3: Applying Theories

This week we are applying the theories we learned about in the previous weeks in our Generative Content Creation course to concrete examples of generative art.

We were lucky to have been presented with a big list of examples with loads of different generative works. Some of them are downloadable, some are web-based, some fall more into the category of tools and some are just things to play with.

One of my favorites among the links was The Generator Blog, which indexes a huge amount of small generative software. The picture above is from an Atari 2600 Label Maker application, which I found on the very same blog.

I scanned through the list of software on that site and I found some of the stuff on there pretty interesting in terms of the context of how these small generative tools are used for the purpose of social satire. There is a lot of generative software that is humorously critical of some sort of social phenomena.

The reason why it's interesting, particularly to me, is twofold: first, I always maintain that any artwork should be evaluated based on its relationship to society; and second, it's interesting because they reveal something about our attitude towards generativity in general.

I remember Philip Galanter mentioned Jackson Pollock in his article on Complexity Theory in the readings of Week 46. The question analyzed there was whether or not he could be considered a generative artist, but, regardless of that, Pollock, I think, serves as a very good example of what I mean by the importance of "relationship with society" in terms of good art.

Essentially, I believe the best pieces of art always indulge in some form of social upheaval. Art should always aspire to rebel against a sorry state of affairs, in a way. You can do that exceptionally well with art, because it seems to have a unique and efficient ability to call into question the very fabric of how a society comprehends itself. Hence, to me, the goal of art should always be to engage society in dialogue.

So what could possibly provoke dialogue in the seemingly random squirts of paint in Pollock's works? It's very easy to invoke the standard objection to abstract expressionism of the art philistine here - that I could probably whip something similar up myself, and dismiss it as a fad.

Visually speaking, perhaps one could indeed achieve a similar aesthetic. However, the point is that you could probably never achieve the same social significance in terms of what Pollock meant 60 years ago in a post-war moment in time, no matter how algorithmic your technique. After all, what was produced was not interestingly coloured wallpaper, but instead a statement on creation in general, as well as on the principles at the heart of society that govern it.

So, I'm always looking for the social commentary of art, and what's interesting with some of the little pieces of software on The Generator Blog is that generativity itself is used as a device to make that social commentary.

I've selected a few generative works that I would cheekily categorize as rebellious against "bullshit" - the meaningless marketing and corporate fluff that pollutes our media, in which you use the generative process to bypass any sort of effort or consideration towards certain mannerisms.

Prime examples include apps like Corpspeak: infinite corporate bullshit generator:

As you can see, it automatically generates long sections of text full of jargon-infected gibberish and it is obviously a satire on the general barren nature of press releases and the corporate attitudes and practices that result in them.

You have the basic characteristics of generative art present in this example. There is an underlying system of language rules and an algorithm to arrange the jargon dictionary entries into something that mimics sentences. The software creates the work with a significant amount of autonomy and the author gives up some of his creative control in the process. In fact, in this particular case, you might say too much control is given up, as the only input from the "author" in the standard version is the press of a button. However, there is a customization option where you are free to modify the source code and dictionary to make the generator behave differently.

It brings up an interesting question, though: how much autonomy can a generative piece of art have before the author loses his authorship entirely? Is that even a logical possibility?

Let's say, for the sake of the argument, that we are indeed dealing with a piece of art here, as it might be argued that a text generator is not art, but it doesn't really matter in the context of the question.

Assume the same bullshit generator is on display in an art gallery installation - what exactly is the piece of art here? Is it the generated result or the generator itself?

Let's look at a couple of more similar examples like this. For instance, the Bullshit Job Title Generator - similarly directed satire and similar execution as compared to the press release tool:

Here you just have three columns of dictionary entries which are grouped together with the method of randomization. All you need to do is click on the button to generate a job title.

An identical process can be found in the Web 2.0 Bullshit Generator, which I think is particularly excellent for generating all the stupid mash-up terms that you need to underline that the field you are working is so cutting edge that you are constantly struggling to find the vocabulary to describe it:

Or what about this Mission Statement Generator - particularly interesting to me today since I am actually supposed to come up with a mission statement for my work:

All of these examples share the common feature of effortless, one-click generation, which is almost essential to their message - that the contribution to progress of society of such mission statements, job titles and press releases is about as significant as a single mouse click.

So, on the one hand, you have a message in the effortlessness, but on the other, you have a very high degree of autonomy with the generative tool, and I think the level of autonomy brings up some questions about the nature of authorship with regard to such applications, and the relationship between the author and the work in terms of the generated result and the generator itself.

In the above cases, the authors of the the works can be no other than the people who programmed the generator. You might as well set it to load automatically with the page and you would eliminate my input the process altogether, so the only person who can claim authorship is the author of the source code.

But what if set up my own piece of generative art based on certain preexisting generative tools to which I add my own artistic configuration? Do I share authorship with whoever devised the generative algorithms, or is the work of art defined solely with the result of the generative process?

For instance, it seems counter-intuitive to say that a piece of electronic music is authored by the creator of the drum sample humanizer algorithm that I was talking about in a previous post. Or that you would owe royalties to the person who pushed through the idea of a parameter variable randomizer module on a software synthesizer, as you are benefiting from from a sound that is the result of nothing more than his design.

Yet you would probably agree that my push of the button does not constitute authorship of the resulting corporate bullshit of the tools outlined above in any conventional sense of the word. The authorship of the satire belongs to the programmer, not me.

Of course, I'm analytically all over the place here - mixing concepts that belong to the realm of intellectual property law and more philosophical ponderings about ownership in general, but the question could be worth while exploring nevertheless, as you might have more and more generative tools emerge in the near future, and the boundaries between author and audience, designer and user could blur even further.

But beside the questions of authorship that might arise from the relationship between the creator and the creator of his tools, my chosen examples also seem to illustrate a certain attitude towards generativity.

It's as if the generative process itself is derogatory. The message, after all, seems to be that something is so dull, stupid and predicable that it can be generated.

I was reminded of the last task in which we had to read up about effective complexity as a measure to evaluate generative art - that the ideal amount of generative complexity in an art work is determined by the balance between complete order and total disorder.

So, perhaps, if people seem to be prone to attributing a derogatory meaning to generativity, the general state of generative content creation is still perceived as being in a phase characterized by its inability to achieve effective complexity?

It might be a stretch to base a claim one way or the other based on just a rummage through some of these examples, but I was nevertheless left with these types of questions after having played around with them a little bit.

But I think there is definitely an attitude towards generativity that seems to be shot through with a certain amount of pessimism. It seems to be more associated with automation, drone-work, the lifelessness and soullessness of a robot algorithm, and that's a pretty strongly loaded evaluation. I just wonder if generative art as a whole suffers in some way in the likely presence of this predisposition?

I'm going to wind this post up now, as I fear that I've gone slightly over budget in terms of length, but I want to leave the questions about authorship and the perception of generativity here as a foundation to possibly build the final project upon. It would be nice to utilize some sort of ideas that would really push both considerations to their limits and see if any clarity about them could be found in their practical application.

References:

Galanter, Philip. 2003. What is generative art? Complexity theory as a context for art theory. In In GA2003–6th Generative Art Conference

The Generator Blog
http://generatorblog.blogspot.com/

Atari 2600 Label Maker
http://www.labelmaker2600.com/

Corpspeak: infinite corporate bullshit generator
http://lurkertech.com/corpspeak/

Bullshit Job Title Generator
http://www.bullshitjob.com/title/

Mission Statement Generator 2.0
http://www.evilgeniusmarketing.com/mission-statement-generator.cfm

The Web 2.0 Bullshit Generator
http://emptybottle.org/bullshit/index.php

Task 14: Final Reflections

Our New Interactive Environments course is coming to an end and as a final task we were asked to write about our experience and what we've learned in the past weeks.

I remember I enrolled to this course expecting it to be an overview of all the cutting-edge emerging phenomena and tools that our field of research is working on. Obviously, that was all down to the title of the course, because that's what I imagined would fit most with what that title describes.

I was excited after the first meeting was held because I got the impression that not only were we going to look at these tools, but we were also going to explicate some of the common underlying mechanisms that are going on in any given environment and its community.

That was something particularly interesting to me because I often felt that I lacked such a fundamental understanding of what is going on in these environments in my work, especially with regard to digital communities. I was looking specifically for insight into how online communities are set to flourish - how you discover what people in it are really after, what moves them, how you can create meaningful experiences in cyberspace, help shape digital lives and how you can create environments where people can live them out more easily.

I've mentioned before that we deal a lot with producing video blogs for festivals and showing them to people in online environments. The problem is that in the process of doing that we are constantly struggling to grasp the fundamental mechanisms present in the environments that we produce content for. How do you design a meaningful piece of video that, on the one hand, is advocating the goals of the festival and, on the other, really speaks to and enriches the experience of its community?

I'm deliberately bringing a very specific example because I want to illustrate that, to me, the most important part about understanding the build-up of any sort of digital environment is the qualitative dimension of the relationships within it. You need a framework to explain both how you create or describe environments, but, fundamentally, that same framework should explain how you work best with that environment.

We looked at Activity Theory as one foundation for understanding the structure of how people work together with a given motive to create this meaningful experience. But, unfortunately, I don't think it alone can come close to explaining what goes on where the rubber meets the road in terms of actually designing for specific experiences within that system.

To me, it describes the ontology of an activity system environment - it compartmentalizes it's core elements and the relationships between them, but says very little about the quality of relationships between those components.

So, if I were to point out something negative, its probably that I didn't find the framework for such specific community analysis that I was so looking forward to learning about.

But it was a mislead expectation in the first place. I've come to understand that the titles of the courses often have little to do with the actual content. I think this course was obviously geared towards Activity Theory from the very beginning - we started off with the study planning activity mapping exercise and we ended up focusing mostly on that framework in the end, so I would just recommend renaming the course to avoid confusion in the future. There is nothing wrong with a straightforward title like "Introduction to Activity Theory", which this course undoubtedly provided for me.

I was happy that I got to read some articles after a long period of focusing on work. I think the most important thing that I've taken on board from Activity Theory is that it crystallized for me the notion that the context of our activities is indeed dynamic, dependent on and shaped by constantly morphing artifacts and processes.

And I always maintain that there is a social significance to that idea - in the same way you can't grasp concepts adequately in our research field with the help of the linear, isolated models of old, there is a similar call for change in how we conceptualize things in our society.

I do believe that digital technology has the power of changing the way we go about our activities in life and if I accept that notion, then I see no way of avoiding the upshot of it having significant social as well as political influence. I've become slightly fascinated by the philosophical consequences of the idea that digital technology is shaping society itself and I might even be considering writing my master thesis on the subject.

From an organizational standpoint I have to say that I liked the no-excuses approach to producing the tasks. I also think it was clever that you didn't create a weekly routine for posting tasks, as this allowed you to keep the students on their toes and more lucid with regard to work.

There was, of course, the incident where we had to post an assignment within very short deadline, which ruffled a lot of feathers, and I think that was probably a mistake on your part. It didn't really affect me because I happened to have time for it, but I also create my own work schedule and do not have kids, so I understand the backlash from some of the others.

It would also be a good idea to publish at least a conceptual plan for all the weeks describing what the course will be looking into and where it is heading, because when we did the first task of mapping our study planning, nobody really understood why, and, consequently, didn't necessarily want to do it. I think it makes sense now, and I understand that you wanted to see maps that were not influenced by any theoretical frameworks, but providing an general sense of direction would go a long way with motivating the students.

I liked that there are no hard feelings if I am late on or missing some of the tasks - that its just a question of how many points I receive, as I had a pretty intense period at work during the course and that made me miss two posts. I sincerely hope that I'm above the threshold for a pass, even though, as a general rule, I'm opposed to the idea of receiving the maximum amount of points for a work that is more or less adequate simply by posting by the deadline - I think you've hinted that this would basically be the criteria for assessing the works.

Please don't fail me :), but consider, perhaps, focusing in the future more on the quality of the posts as well, as it does a lot more to help the students gain more actual knowledge if there is some discipline as to what people are saying.

I wish you all the best and thank you for the course!

Task 13: Redesigning and Re-Instrumentalising Activities

This week in our New Interactive Environments course we were asked to think of an activity that could be redesigned to be more efficient and reasonable.

For me, the first thing I can think of that needs a major overhaul is e-learning in general. And by e-learning I mean absolutely all of the operations that are carried out electronically at the university - be it administrative operations, like enrolling for courses and applying for stipends, taking e-courses in their respective e-environments, or even online student socialization in general. I would propose to put them all together with the help of a central electronic environment.

I sometimes wonder why these things haven't been integrated a long time ago. After all, these operations do not seem to serve any other purpose than to facilitate the activity of our studies - the motivation behind them is the same. So it's often frustrating to deal with a heavily scattered body of functionalities that, ultimately, serve the same purpose.

We recently got a new information system at TLU, which seems like a real blessing after its terrible predecessor. However, at the heart of it, it's still looks and handles like a decade-old environment. I realize that the visual side of things is often governed by stricter usability guideline requirements for public institutions, nor do I care that much about that, but, to me, an environment where you can't, for example, generate a Google Calendar automatically based on the courses you registered to, is a bit behind the times. The same goes for any other platform that you are already using for managing your e-mails and documents, which could really be integrated into the process to make it more seamless. Not having the university information system integrated into your real digital life forces you to spend a lot of resources on doing redundant tasks which can really be obliterated given the right technology.

What's more, I have used at least 5 or 6 different stand-alone Moodle-type learning environments in my time as a student - all essentially offering the same basic functionality in their own uniquely flawed way. Add to that the countless blogs, wikis, intranets and web directories that courses utilize and you have a right mess on your hands, because you can't keep track of them all. If you want any sort of archive of your studies, you have to build your own database.

And that is only one example of a huge waste of resources which could be avoided simply by having a central environment where all of the course materials, assignments, projects and feedback can be hosted and aggregated in the first place. It's not like different courses use some sort of special instruments for their studies - the things that you usually need are blogging tools, wikis, discussion boards, live conferencing tools, file sharing tools and a system for messaging and socializing. Add to that adequate features for integrating content from external sources and you should pretty much be golden.

There are more applications, of course, that should be added to this central environment, but the point is that most probably only a finite number of tools is needed and it is absolutely viable to have the mother of all e-learning environments where you can log-in and perform absolutely all of your school-related operations.

From an Activity Theory standpoint it almost seems that we have one activity of "studying", and you can identify the subject, object, the tools, the rules, the community and the division of labour, but the interaction between all those elements is suffering on the count of not having the right tools. There seems to be a lot of redundancy in the instruments and in the communication between the subject and the community.

So I think it is definitely possible to re-instrumentalize and re-organize the activity of studying in TLU with the help of emerging technology. I would first propose making an all-encompassing study of organizational efficiency both from the perspective of the administration of the university and from the perspective of students, of what sort of electronic platforms are needed for schoolwork and how students and lecturers are building their own systems for communicating vital information related to academic work. After all this has been mapped out, I would suggest investing in creating a single environment through which all curriculum planning and management can be carried out on the university side and every conceivable electronic operation can be carried out by students.

If you get the study right, if you engage in intelligent dialogue with all the parties involved and you find the right people to design the environment, I think you would save money in the long run, it would make everything much more efficient and it would also add to the quality of education.

Task 11: Our Components Versus Components of Activity Theory

This week we are comparing Activity Theory elements to a list of elements we collectively jotted down in a Pirate Pad session after we had drawn up some sort of diagrams for how we plan our school studies.

Looking at the list it seems we unconsciously described most of the elements in the structure of Activity Theory or presented some keywords that are essential to it. For example - we mentioned the time frame, learnability, aims and goals, motivation and that the process is influenced by its surroundings. It sounds very familiar after reading up about Activity Theory because of its focus on dynamics over time, its distinction of goals on the actions level and motives on the activities level and the importance of context.

We also included the various participants and actors which can be summed up as the community, with Sebastian's help we arrived at the different roles or that play a part in the process or, in other words, the division of labour, we mentioned a whole heap of tools and we even managed to explicate that there are restrictions and rules to consider.

What we missed, however, was the subject and object part of Activity Theory's structure. We didn't seem to focus on any perspectives in the system other than our own. And also, while we mentioned time frames, context and interrelatedness, we were probably too focused on the linear, flowchart-like processes we had drawn up to really adequately describe how dynamic an activity can be. We had rigid paths with start and ends points, with decisions in between, mostly single interactions with tools with a focus on the sequence of elements, but it doesn't quite have enough dimensions to describe all the simultaneous relationships that are involved in the transformation process of the activity.

However, if I were to point out any irrelevant components in our list, I would be in trouble because it seems to me they are all more-or-less acceptable terms, provided that they are recategorized or regrouped under their respective umbrella terms. Perhaps some of the evaluative terms are a bit more tricky to place, but I guess you can have an activity that incorporates actions and operations that are directed towards improving quality, for example.

I think its fair to say that our list is quite concurrent with the Activity Theory framework after analyzing what it contains, but then again, it seems to me that Activity Theory is particularly good at making itself compatible with almost everything. That's why it's so hard to say whether there could be anything additional that the Activity Theory framework would benefit from.

Perhaps it should somehow incorporate more of the rigid influence of the flow chart processes to help define the transformation process's scope and levels more clearly. It seems to be too easy to resemantize your analysis to fit Activity Theory by just compartmentalizing all the different elements to the levels of operations, actions and activities and choosing some arbitrary boundaries to the system.

The trouble with this universal suitability is that when you analyze any given system and find that it indeed can be grasped within the framework of Activity Theory, what is the point? You might find a load of elements that you can attribute to the different levels of an activity system, but does that really tell you much about nature of the interactions between them? I was left with the question of whether there are any clear normatives provided by Activity Theory with regard to how a system should be like?

So with that in mind, the evaluation criteria might actually be one of the more interesting things to incorporate into Activity Theory.

Generative Art

This week in our Generative Content Creation course we are looking into what the term "generative" actually means and we are reading up on how generative processes are used in the creation of art and literature.

To me personally, generative art forms have always been associated with the use of some sort of randomizing algorithms. Perhaps that is down to the fact that I mostly only use generative audio software that is based on numerical randomization of some sort of parameters. For instance, a lot of randomizing is used with drum samplers in an attempt to make the drum hits sound more human. By adding random imperfections to the synchronization of the hits you simulate a human being's natural imperfection in timing, which is one of the more effective methods to trick your ear into passing a sampled track as real.

There are other examples, mostly found in the modules of software synthesizers, however, equating randomization alone with generative art forms would be a mistake, as there is definitely more to it than that.

Coming back to the drum sampler analogy, to achieve the best results in ear-trickery by timing alone, you have to take into account the not-so-random aspects of how any given song invokes emotion. Whether you are after explosive power or an effortless, lazy vibe, you need to express timing that is characteristic to drumming communicating those emotions, which in this case might mean you would jump the gun on the snare or lag behind on the kick.

So there are different levels to such elements of generative creation - randomization might address some basic psychological aspects of our hearing, but generating intonation and expressiveness requires something more.

The first thing to remember, then, is that randomization is only one of the possible ways to generate art. But how would one define the notion? What else can be considered a generative method for producing art?

There are a couple of definitions I have taken a look at, the first of which deals with art forms in general:

Generative art refers to any art practice where the artist uses a system, such as a set of natural language rules, a computer program, a machine, or other procedural invention, which is set into motion with some degree of autonomy contributing to or resulting in a completed work of art. (Galanter 2003: 4)

Another one definition deals with generative literature:

I call “generative literature” a literature where the texts are produced through a computer by means of a set of formal rules, the use of any kind of algorithm, specific dictionaries and eventually knowledge representations. That means a literature of which the author does not write the final texts but which only works at the level of the high rank components such as: conceptual models, knowledge rules, dictionary entries and rhetoric definitions. (Balpe 2005)

Immediately you can see similarities in the two definitions. The most obvious is the use of a set of rules or algorithms to produce the work. Perhaps less obvious is the requirement that the author surrenders some or all of his control to the generative procedure. A certain level of autonomy is expected with generative art, which means that the system has to exhibit the capability to produce all or parts of the work independent of the author.

The literature definition includes the computer as that system, however, generative creation does not seem to be necessarily dependent on computer technology. Generativity does not necessarily refer to the technology involved, but rather describes methods of generation that are in fact applicable to a wide variety of approaches.

This is further backed up by claims that generative art predates computer science by a long way. To be clear, other forms of generative creation besides randomization, which are said to conform to Galanter's definition, include symmetry and tiling, fractals, genetic systems and Lindenmayer systems (Galanter 2003: 12).

Examples of symmetry and tiling can be found in art artifacts dating back thousands of years (Galanter 2003: 15). Which gives all the more credence to the notion that you don't need a computer to produce generative art of literature. The fact that most operations are performed on computers nowadays should not take away from the theoretical possibility that the generative process can be carried out by any sort of system that meets the requirements of the definition - even a human being.

From a scientific point of view the subject of generative creation seems to have been conceptualized within the frameworks of some mainstream theories. One of which is complexity theory, which is justified with the notion of "the system" in the definition of generative art (Galanter 2003: 12). Since complexity theory is held to be the paradigm of understanding any sort of systems, it only makes sense to turn to this particular field.

Complexity theory is capable of providing a framework which can explain the requirement for moderation between complete order and total disorder in art. It is postulated that the art work is evaluated based on how well it achieves a complexity in its structure - generally speaking, uniform repetition is not likely to please and nor is a complete lack of any pattern. Complexity theory offers the measure of "effective complexity" to tackle this balance (Galanter 2003: 10).

The systems themselves are described as chaotic by complexity theorists, rather than random. A system displaying chaotic behavior does not have linear dynamics and it is difficult to predict what happens to it over time. However, the crucial difference is that some set of rules and laws still exist which enable one to determine cause and effect within its confines (Galanter 2003: 6). A random system, conversely, exhibits no such features.

Speaking of different scientific fields, however, one of the topics I find interesting is artificial intelligence and while reading up about generative art I started to wonder what sort of relationship AI might have with all of this? Could generative art somehow be integrated into the research field of computer science - taken as an aspect of machine intelligence?

Ignoring, for the moment, the previous discussion about how computers were not essential to generative creation, the reason why I'm interested in intelligence with regard to art is that I'm interested in the possibility that these generative processes based on rules and algorithms might actually be central in the production of simulacrum intelligence. I would have wanted to read more about this possible relationship.

References

Galanter, Philip. 2003. What is generative art? Complexity theory as a context for art theory. In In GA2003–6th Generative Art Conference

Balpe, Jean-Pierre. 2005. Principles and Processes of Generative Literature. Dichtung-Digital

Task 10: Applying Activity Theory to Practice

The activity system of NIE

Tools
The tools that we use are weblogs, the course wiki, Skype, articles and other resources, the Flashmeeting tool, e-mail, and I think we used something called Pirate Pad once for our group sessions. I also use PDF reader software, a web browser, graphical design software, online dictionaries, the IMKE online schedule and the TLU website. Some students have probably used concept mappers and other visualization tools. Perhaps I should add my mental artifact for scheduling studies to this list, as it was present in my mapping in an earlier task.

Subject
The subject of the activity system is the student.

Object
The objective of the activity system of a course is to gain knowledge, so I would say that the object is knowledge.

Rules
There are weekly assignments posted and submitted on weblogs, which can net you a 10 point maximum score for each assignment, minus 2 points for every day that you are late. The number of tasks is unknown and so is the maximum amount of points attainable, but in the end the course is passed at a 70% threshold score. Online meetings are scheduled by voting on a selection of dates and times set by the lecturers. The date and time is decided by popular vote, however, online meetings do not count towards your total score because not all students can be present at the most popular time. If you submit assignments early, you will get feedback on it and you can modify your submission without being penalized. There is a "no excuses" policy on missed deadlines, however, if you notify the lecturers promptly about an assignment that you know you will be late handing in and provide a valid reason, you might get an extension on the deadline.

Community
The activity system community is comprised of students and lecturers primarily, but secondarily you might include the university administration and whoever is responsible for keeping all the blog sites and wikis running on the server side.

Division of Labour
As I understand it there is a basic division of labour into students and lecturers, but the former has some additional levels of division - Terje seems to post all the tasks and feedback, but I have no idea what goes on behind the scenes - who deals with the department at the university, how the curriculum is produced, who sources the materials and decides on assessment criteria etc. So from that perspective I imagine there could be the role of the administrator, the researcher, the reviewer - each not necessarily taken on by one particular person.

The Activity System of PLENK 2010

Tools
As far as I understood, the PLENK 2010 course uses a course wiki, a daily newsletter for information, articles and resources, a course forum in Moodle and the Elluminate tool for live sessions, which are also recorded and stored on the course website. They also seem to use Delicious for bookmarking and they keep track of Twitter posts of the students. Students Twitter accounts and blogs are bundled into RSS feeds. Other environments used are Slideshare, Wikispaces for diagrams, Bit.ly for short URLs and TED talks. I noticed one of the first assignments was to produce a concept map with CMAP and the final project of the course was described with the terms "podcast, slidecast, Articulate presentation, video recording, Second Life presentation", so you would have to include all the tools necessary to produce either one of these. Students also need a web browser and a PDF reader to perform basic tasks.

Subject
The subject of the activity system is the student.

Object
The object is knowledge.

Rules
There are reading assignments for each week and also a live sessions for each block. As far as I understood, the tasks related to the weekly reading assignments are optional. Assessment is based on a continually developed concept map and a final project, which can be submitted for peer-review mid-way through the course. There seem to be two Ellumination sessions per week - on Wednesdays there is one with guest speakers and on Fridays there is one dedicated to weekly reviews with course facilitators. If you have a question, you are free to post it in the discussion forums in the corresponding weekly threads and other students are encouraged to suggest answers. Collaboration is a keyword - for example, the course wikis can be edited by participants of the course.

Community
The community is comprised of students, facilitators, guest speakers, administrators.

Division of Labour
It seems that you can identify the roles of students, peer-reviewers, facilitators, guest speakers, administrators.

Drawbacks of Activity Theory in Analysis

I have to say one of the key things I had trouble with was determining the scope of how far I should go with the inclusion of different tools, community members or labour division. It seems hard to define a clear boundary of where one activity system ends and another begins. Do I have to include such things as internet service providers in the community analysis? Perhaps the designers of more or less ubiquitous learning environments? They certainly seem to be integral to the learning process. What about computer manufacturers?

The same goes for tools - it's stupid to list my desk and chair as tools in the activity systems, even though I certainly wouldn't be posting this task standing up.

Also, I'm not that sure about the distinction between community members and divisions of labour. How do you classify or define community members and their divisions of labour? Can both of these terms apply to both individuals and groups of people?

My first instinct is to classify the members according to their assignments of labour, but it seems in practice labour division could almost be identified by the different roles that can be taken on by any number of community members. For example, in PLENK you have facilitators who are responsible for the content of the course, but also perform the role of moderators in the forum. Students perform the role of students as well as instructors in the process of collaborative exploration and peer-review.

Differences and Similarities

The obvious difference between NIE and PLENK is the number of course attendees and facilitators. Consequently the rules of the game are very different. The structure of NIE's activity system is much more rigid and focused on interaction between the subject, tools and object, while PLENK seems to have heightened emphasis on interaction with the community as well as a broader division of labour. There is a much wider selection of tools in use within PLENK.

The obvious similarity lies in the subject and object of the activity systems.

Task 9: Exploring Activity Theory as A Framework for Describing Activity Systems

The motivating force behind the emergence of Activity Theory seems to be a theoretical crisis in the study of Human-Computer Interaction. Traditionally, the underlying theoretical framework or methodological approach of HCI has been consumed with what is called the classical Cartesian ideal of cognitive science. That is said to usually entail the use of the experimental apparatus of laboratory-oriented classical psychology, which is borrowed from natural sciences.

However, a number of problems arise from this apporach. For instance, this classical, laboratory-centered approach is prone to isolate the human being not only from their social context, but also from the context of their behaviour in general. By concentrating only on certain individual acts within human-computer interaction, it fails to recognize that the actions performed are really interrelated and that humans regulate and coordinate their behaviour depending on a larger context and are not passive elements in the system.

Also, traditional frameworks lack the tools to understand the dynamics of HCI over time. HCI research is usually carried out in a very limited period with relatively inexperienced users, however the methodology does not focus on how humans develop their interface skills long-term. Excluding this focus means that the design process tends to overlook this gradual development, which is a crucial characteristic HCI.

The design process in practice usually incorporates user perspectives in iterative design, because that has been found to produce the best end products. There is a whole discipline called "user centered design" that deals with this. However, this practice, once again, lacks a unified theoretical background.

This crisis has prompted a search for a new framework - either in the form of a modification of the existing psychological frameworks or in the form of a completely new theory. The underlying assumption, however, seems to be that whichever framework can explain how the subject matter has changed in a way that the previous theoretical foundation has become inadequate, could itself be the new basis for knowledge in the field of HCI.

So there have been attempts at re-conceptualizing things with the help of other research disciplines - for instance, with the aid of Information Systems Resource. But while it perhaps explains the emergence of the crisis in terms of the general development of computer technology and computing, it is still reported to lack the theoretical cohesion to offer any relief in the crisis.

This is where Acivity Theory comes in. It's theoretical strength compared to previous approaches lies within its focus on the unit of activity, rather than action, as was prevalent before. Activities are of a higher order as compared to actions - actions might have their own concrete goals but they lack the means for understanding how different actions and goals are related to each other and how they change over time. Activities enable wider understanding of the general motivation behind all the goals of the individual actions.

Activity is held to be the minimal unit that permits the study of the meaningful context of actions. It is characterized by having a history of its own - it has a tendency to produce traces of itself that form the basis for its new phases. These traces are called artifacts.

The benefits of using Activity Theory as a theoretical base for understanding HCI seem to be that it can be applied to multiple levels of any given problem, integrating different steps of the design process into a unified framework. The embedded context also means that it can be suitable for studying social interactions and, as already mentioned, it makes it possible for studying how interaction changes over time.

References

Kuuti, K. (1995). Activity theory as a potential framework for human-computer interaction research. In. B. Nardi (Ed.), Context and Consciousness: Activity Theory and Human Computer Interaction. Cambridge: MIT Press.