This week is going to focus on interaction as a medium. After looking at different kinds of media, and how some content works better than some other content through these media, we'll focus more closely on the specificities of interaction.

This will start from a high-level, conceptual perspective, and focusing all the way to the specifics of user interface design and user experience design (along with expliciting the difference between the two!).

In The Art of Interactive Design , Chris Crawford defines an interaction as a conversation, between multiple agents who, one after the other, "listen, think, and speak", which in computer terms, would consist of "input, process, output".

The richness of an interaction comes from the care in the listening experience ("is everything that is being said, being heard?" -think of non-verbal cues), to the depth in processing ("what is it that is really being meant, and how can I extract that meaning to serve as the basis for my response?") to the quality and clarity of the output ("is what I'm saying clearly reflecting the results of my thought process? does it leave room for further exchange?")

Whatever the response/output is, it's always important to leave room for further action. Even when an error happens, the handling of the error should be done "gracefully" (e.g. "This operation did not succeed. Would you like to try again? Or try a different operation?)

some of the requirements for interaction include: - being able to communicate exactly what you intend, - having an equal interest in both parties, - having a common factor, common references, - being able to receive information,

links -> recommendations or suggestions

Most interactive systems include: the developer, the system and the user.

In terms of the interactor, it's a significant switch from user: user means passive, while interactor means active. Just because one is on the final, receiving end of an event (even if it's an interactive event), doesn't mean that they do not re-appropriate the event for themselves, and engage with it in non-expected ways. There is a thin line between making a system error-proof (making sure the interactor doesn't break the system by inputting the wrong data), and making the system fool-proof (making sure the interactor doesn't do anything too complicated because the assumption is that the interactor can't know as well as the designer).

The concept of agency is tied to the second meaning: an interactive system should enable its interactor, allowing them to do as much as they would like to do, as much as they can imagine, and perhaps even more. A good example of such a thing is the LEGO brick. A system which provides agency is a system which makes you responsible for what you want to be responsible for. On the opposite, a system like an automatic slide door doesn't provide much agency: it's either fully open or fully closed, and doesn't exactly tell you what is need for either of these states to change.

The need for transparency in interactive systems comes from this similar idea. Thinking that a system is intuitive relies on a particular number of assumptions: you already know shortcuts, design heuristics, patterns and processes. A transparent system, on the other side, is a system which exposes itself to the interactor, and only assumes that she would be able to make sense of it, given a good interface. Robert Henke's CBM8032 is a good example of the things someone can do if they're given access to all the relevant parts of a system.

Additionally, interfaces on desktop between Mac, Linux and Windows assume very different behaviours of their users by showing them different amounts of information

The computer can "think". That is, it can act and respond somewhat dynamically and contextually. If the programmers can think through the possible future situations that the computer will be confronted with, it will give the impression that the interactor is having a "smart" conversation with the system.

In short, digital media is different from other media in that it can reconfigure itself as it goes. The problem, as we've seen with Licklider the previous week, is that thinking through possible problems in advance turns out to be more complicated than expected. So the answer at this point is either to establish the digital product you have as either close-ended or open-ended. Close-ended means that you are not providing anything beyond the basics of the product can do: the scope is tight, the interactions safe, and the "world" narrow. On the other side, open-ended means that you leave some room for interpretation to the interactor. It might not be as rigorous—in the scientific sense of the term—but it will provide a different kind of engagement, by involving the interactor deeper into the meaning of what she is presented with. Open-ended systems involve some kind of gestalt (the whole is greater than the sum of its parts).

The image above exposes a kind of open-ended system: Joseph Weizenbaum developed a program called ELIZA , a chatbot which would essentially "bounce back" questions that the interactor would input. For instance, if you would type "I'm unhappy", ELIZA would respond "Why are you unhappy?". While the computer program has no idea of what "unhappiness" is, the interaction still elicits interesting thoughts in the interactor's mind.

This description was first developed in hamlet on the holodeck by Janet Murray

The spatial component comes from this possibility of navigation, of going from one place to another. There are locations (websites), pathways (hyperlinks), landmarks (icons) which allow us to both explore and keep track of where we are. How do we make sure that the interactor never feels lost?

The procedural component relies on the fact that we are dealing specifically with digital media. Through a computer, it becomes possible to set up rules, procedures, in order to allow for everything else (space, encyclopedia, participation).

The encyclopedic component is the consequence of working with networked media. If one is connected to the Internet, there is no pre-existing reason as to why one shouldn't be able to access any information that is on the Internet. A good example of that is Assassin's Creed's Discovery Mode through which you walk around the game world and learn of the "real world" equivalent of what you're seeing.

So, as we start developing interactive experiences within the context of museums, it's important to keep in mind those things:

• How are you enabling an interesting conversation? How responsive and predictive is your system. For instance, if the visitor takes one action (view an object up close), then would be the next action (e.g. viewing other objects? viewing advanced information related to the same object?)
• What kind of world are you creating? What are its multiple components, and where do you put the emphasis? Is it a social world, an economic world, a world connected to other worlds, etc.?
• How is meaning co-constructed? How do you let the visitor explore them in more details, at their own pace, in their own way? And to what extent do you ensure that all visitors get the same amount of knowledge, and each visitor gets their own version?

-> interface: appearance, information, functionality (number of buttons, the number of clicks, shortcuts, etc.)

-> experience: atmosphere and feeling (color, fonts, animations, sounds)

User Experience is comprised of touch points, of particular moments in which the public comes into contact with the product. For instance, the moment that someone books a ticket online for a museum, the moment someone enters the museum and goes through security check, the moment someone asks for directions at the information kiosk, etc. And how all of these come together in a coherent moment.

On the other side, User Interface is the collection of specific graphical elements which help create a metaphor for the interactor to understand how to manipulate and communicate with such a system.

In other words, UI is what we see, and UX is how we feel when we use something (the overarching ceremonial), and how we feel when we're done using it.

User interfaces are indeed a metaphor, either building on previous knowledge by the interactors (e.g the Trash on a desktop isn't really a trash, but it does the same thing as a trash does), or by developing new habits (the hamburger icon ).

The point of a good UI is to have itself be forgotten. It should be as invisible as possible, while at the same time being "obvious". This pinnacle of obviousness was reached in Apple products, particularly with the iPod and the iPhone, which didn't come with a user manual!

A newer medium also develops on top of previous media, in a phenomenon called __remediation__ (writing integrates parts of speech, photography integrate parts of drawing, cinema integrates parts of photography, etc.)

Google Docs started with a page layout in order to get the users to understand that this is a familiar product. Once they are familiar with it, then the pages can be abandoned, since there are no fundamental reasons for having pages on the web (i.e. there's no paper!)

UI elements can be separated in three categories:

• graphical elements (buttons, icons, layout, typefaces, etc.)
• input methods (tap, swipe, hold, left-click, right-click, shake, etc.)
• system messages (confirmation, errors, dialog boxes, alert boxes, etc.)

In their design guidelines for iOS, Apple separates them in three parts: - navigation bars - action sheets - buttons

Interaction Design is halfway between UI and UX, it is the art and science of providing the user with an environment in which her actions are meaningful—they are acknowledged and acted upon in a way that is coherent with what she expected.

All of these elements are explained thoroughly in Jenny Preece's Interaction Design reading for this week.

An element is diegetic if it makes sense with the content of the work.

For instance, take a roadtrip movie playing a soundtrack of a rock song distorted as if it was coming out of a car radio. The rock song is diegetic.

Compared to a science-fiction movie taking place on a space station with a massive philharmonic soundtrack (think Hans Zimmer) is not diegetic (the orchestra is nowhere to be seen on the space station!)