Cornell Language and Technology

exploring how technologies affect the way we talk, think and understand each other

Tuesday, January 31, 2006

Non-assigned But Relevant Post About Dinosaurs

Hey everyone, I know this isn't really related to the course, but I thought it was amusing and it is topically relevant.

If you are not familiar with Dinosaur Comics, Dinosaur Comics is a webcomic whose point is basically to see how many scripts this guy can write that will work in the exact same panel layout made with clipart. It's extremely funny, and yesterday's comic was about lossy networks (which you could argue conversation is a type of) and paradoxes of infinite recursion. So it's basically dinosaurs talking about CG-iterative. Comic is here.

Jeff, Amy: if I shouldn't have posted this because it wasn't assigned, please yell at me and I'll delete this.

Assignment #1 Option B

According to Clark, “entries and exits have to be engineered for each joint action separately” (37). With instant messaging, establishing clear boundaries for entry and exit often pose problems unique to its medium.
Agreeing on a point of entry is difficult; although one person may initialize by sending a message to a friend who is online, the friend could either be away from the computer, be occupied with another project online, wish to wait awhile before responding, or simply choose not to respond. One way in which AIM deals with this situation is that users have the option of having a noise sound when a message is received. In this way, a friend’s invitation for entry into conversation can always be clearly known.

In instant messaging, one person could also quickly terminate conversation without warning. With regard to this issue, AIM maintains the automatic away message. Although the user has not explicitly posted a message to establish his status as “Offline”, when a message is sent to him another one appears indicating his status as well as the time in which he signed off. However, if the user has simply chosen to not respond while still being online, the exit is ambiguous; his friend could sit at the computer for several minutes waiting for a reply. A similar situation occurs when a conversation lags without both parties having signed off. At this point, no clear exit has been agreed upon. When one person eventually decides to go offline, he now does not know whether to simply indicate his exit by signing off or to explicitly send a message saying goodbye.

In addition, each phrase of the instant messaging conversation follows a similar pattern to face-to-face conversation as described by Clark, being “aperiodic, unbalanced, and alternating” (87). However, while in speech exits are helped in their indication by intonation, in instant messaging the exit for each phrase has to be clearly designated with either syntax or the passage of time. This is due to the fact that conversational turns do not end when one message is sent. For example, if someone receives the message “I am going to Alaska this summer”, he would probably wait for a few seconds to see if it is followed by another phrase. If enough time has passed, he would respond with a message such as “That’s great!” However, if the first message was followed by “What are you doing?” to clearly indicate its exit, the user would not have to wait before responding.

#1 – Option A: Language Conventions

According to Clark, a convention arises out of the need for a community to solve a “recurrent coordination problem”, such as businessmen coming up with a time for weekly meetings. In order for something to be a convention, it has to be 1. a regularity, 2. somewhat arbitrary, 3. be a common ground, 4. a coordination device, and 5. used to solve a recurrent coordination problem. Language, too, is a convention, argues Clark, in that it has the five characteristics listed. Specifically, he introduces the notion of signaling doublets, which are pairs of conditional statements that are conventional. Furthermore, Clark breaks down the different forms that language conventions might take on into four categories—lexical entries, grammatical rules, convention of use, and conventions of perspective.

The example of a language convention that I will use comes from a subcategory of lexical entries—idioms. For example, everyone knows that if one is “caught red-handed”, it means that they’ve been caught in the act of a crime. The fact that everyone knows this makes it a regularity. It is somewhat arbitrary because of historical significance. Had history happened differently, this idiom might not even exist, or, for all we know, “saw green-footed” could have meant the same thing. Thirdly, it’s a common good because it’s common knowledge—everyone understands it. Fourthly, it’s a coordination device in that it is a “mutually recognized key” in people’s understanding of it. And lastly, it can solve the recurrent problem of having to say “I caught him in the act of committing that crime” by shortening it to “I caught him red-handed.”

In the recent decades, the rise of technology has introduced interesting conventions related to technological settings. For example, “brb” is an acronym for “be right back”, often used in instant messaging. This convention supports Clark’s model under the condition that both partners are aware of how long one member will be away for and how long the other will have to wait. Once again, it’s regularity because everyone uses it and understands it. It’s arbitrary because it could have been “rb”, or something else completely. It’s common ground within the teenage to the working adult population because it’s shared information that “brb” stands for “be right back”. It's evolvement into a coordination device occurred because it’s the “key” to solving the recurrent coordination problem (criterion five) of not having enough time to type out “I’ll be right back” or something similar. Especially crucial is that, since there’s no immediacy in online chatting, the speaker has to find a quick way to let the addressee know that he/she won’t be able to respond to whoever he/she is talking to for a short amount of time for whatever reason. This way, there’s no miscommunication. When “brb” is sent, the addressee immediately knows not to expect a reply for a short time.

However, the difference in the interpretation of “be right back” can pose a potential problem. How long is the speaker going to be away when they say “be right back”? Thirty seconds? Five minutes? Or an hour? Most people probably take “be right back” to mean a couple of minutes at most. However, some people might interpret it to be longer. Or, sometimes, the situation at hand dictates that the speaker be away for a longer-than-expected time. If an explicit agreement or another coordination device between the two chatting partners is not established, then miscommunication and discoordination might result, making the addressee confused, and even frustrated, or worse, angry. In this case, then, the acronym would be an ineffective coordination device, and thus fail as a language convention in that it didn’t completely solve the recurrent coordination problem of not having enough time write out and explain to the addressee that some situation has occurred at his end and that he’ll have to “be right back”.

#1 31Jan06 — A — Emoticons as a Lingual Convention

Emoticons in online communication, whether it's AIM, IRC, forums, or even email, are an excellent example of a language convention in a technological setting. In fact, emoticons make up a sort of conventional signaling system for the community of internet users.

Emoticons developed as a convention in online communication because of the associated problems with conveying emotions through hastily jotted text. Particularly in chat, where the user does not have time to sufficiently consider or edit their words in order to convey their intended emotion, it is very important that participants be able to quickly and intuitively indicate emotions to each other. Emoticons fill this need nicely: most of them require fewer than 4 keystrokes, and the most commonly used of them closely resemble the types of expressions they are meant to indicate. They certainly allow people who only have text interfaces to coordinate their understandings of the emotional content of each other's statements, and this makes them a very valuable convention since emotional intent is extremely important in building common ground (e.g. is "That's some 'stache!" sarcastic or sincere?).

Although they are a relatively young development in language emoticons have become fairly stable and regular. There are slight variations, such as noseless emoticons (compare ":-)" and ":)"). There are many exceptions to this, such as the drunk smiley ":*)", where the nose actually is the most vital part of the construction for communicating the desired point. Nonetheless, I would argue that in most cases where different constructions are used to indicate the same emotion the differences are superficial, such as with the dropping of noses, and so emoticons do display a great deal of consistency across their usage.

Another requirement Clark gives for conventions is that they be at least somewhat arbitrary. In keeping with Clark's examples of different greeting conventions in different cultures, I think it's useful to compare American and Japanese emoticons, which have developed in very different ways. Where an American might type ":-)", a Japanese person might type "( ^ ^ )": the only character these two constructions share is the ")", and they use it to indicate completely different parts of the face. It's also interesting to note that where American emoticons are meant to be read left to right, just like normal English text, Japanese emoticons are meant to be read vertically, just like a person's face in face to face conversation.

From all of the evidence of how emoticons have developed and are used, it seems they fit and support Clark's formulation of lingual conventions quite well. They are semi-arbitrary, consistent and well understood within communities, and efficient at solving the problem of how to convey emotion through text that lacks the subtlety of voice communication. If you're interested, here are lists of common and not-so-common (sometime completely in jest) Japanese and American emoticons.

~will out.

Monday, January 30, 2006

Assignment #1 option C - 1/31/06

For any number of parties to engage in conversation, their knowledge at that point must include some amount of common ground. Common ground is comprised of the knowledge and beliefs shared by a group of people. For two people to converse effectively, they must each i) have some piece of information and ii) be certain that the other person fulfills constraint i.

Common ground may be divided into two large subcategories – communal and personal. Communal common ground defines the knowledge shared among people with the same hobbies, profession, nationality, or any other classification that may be attributed to a community of people. Generally, these classifications relate to specific cultural groups. Once two parties identify one another as being a member of a certain cultural community, they are able to assume that they both share knowledge about that community and thus converse freely about that joint knowledge without worrying about a possible lack of understanding. If two people recognize one another as American citizens, for example, they can both safely assume that they each know what the American flag looks like.

Personal common ground links specific people together through shared personal experiences and events. If person A and person B are aware that a book is sitting on a nearby table, A can ask B to retrieve it for him or her. When two friends go to a concert together, the experience gets added to their common ground, allowing them to have a conversation about it in the future. The more common ground shared by two people, the more intimate their relationship.

Communication technologies such as the Facebook rely on common ground to connect groups of people. Facebook and similar services allow users to browse through members who share the same interests they do. Users can essentially become acquainted with a person without ever talking with him. The person being researched is not even aware that some other individual is learning such a great deal about him. Normally, this might be considered an invasion of privacy, but the Facebook invites and promotes such activity. In accordance with Clark’s model of common ground, once one Facebook user finds another with some shared interest, background, or personal experience such as a mutual friend or event they both attended, he can feel free to initiate a conversation about this joint knowledge. Through the use of the Facebook, two people no longer have to meet one another to discover whether or not they are compatible. These technologies establish a familiarity between two people that would normally be gained through typical introductory conversation. Thus, the first time they talk, it can appear as through they have known each other for years.

Saturday, January 28, 2006

Assignment #1 - Tuesday Jan 31

Assignment #1 is concerned with chapters 3 and 4 in Clark (1996). Your 350-450 word post must be online before class (11.40) Tuesday Jan 31. Your comments on the posts must be online by Thursday, Feb 2.

Remember that posts will be evaluated on three dimensions: how well you describe the relevant theoretical concepts, how well you integrate your personal observations with theory, and your writing. Comments will be evaluated on the degree to which they add new or critical insights to the post to which they refer. Amy will evaluate everyone on assignment 1 to give you some feedback on how you’re doing.

Good luck!

Choose one of the three options:

A. Describe a language convention, as discussed by Clark. Identify a convention that you have observed in a technologically based setting (e.g., IM, videoconferencing, mobile telephone, text messaging, etc.) and discuss how or why the convention evolved. Be sure to discuss whether your observation supports Clark's model or raise questions about it.

B. Clark discusses how joint actions involve an entry, a body, and an exit. This kind of coordination is critical for any joint action, but mediated language use may need involve extra work to coordinate entries and exits. Discuss the notion of entry, body, exit and how a specific mediated communication space deals with this issue (e.g., online status and away messages in IM, ringing vs. vibrating phones, etc.).

C. Describe the concept of common ground, as outlined by Clark (don't worry about differentiating the different representations of common ground). Discuss how communication technologies, such as Facebook, play a role in how common ground is shared between people. Will it affect language use between people according to Clark's model?

Thursday, January 26, 2006

#1 Intro

Hi everyone, I'm sorry that my post is late - like Will, I also could not log into Blogger this morning. I guess I'll just jump into the post now:

I'm a junior Information Science and Fine Arts major, which means I'm in both AS and AAP. Doing two degrees is crazily hectic but the classes are really fun. What's also great is that I get to meet engineers, artists, and now Comm students! In Art I'm concentrating in Combined Media, which is mostly electronic imaging stuff. I was born in Texas, lived in Taiwan for awhile, then Texas again, and recently my family moved to Shanghai. Here's a random fact about Shanghai: sending a text message over the phone is cheaper than making a phone call, so you see people text messaging each other all the time. Hmm... my favorite movie is AI, but I know most people hated it. One of my favorite sites is explodingdog.com, where the artist takes prompts sent to him through email and makes a cute and sometimes ironic picture from it.

I'll end with a corny joke: Two atoms were talking. One says, "Oops, I think I lost an electron." The other says, "Are you sure?" The first one replies, "Yes, I'm positive!"

See you in class!

-Jennifer

#1 Hello, World.

As a preface to my post, I would like to say that it appears the Universe was conspiring against me on this assignment. First, as I said in class today, I was unable to complete my post yesterday afternoon because I spent my evening at the Daily Sun, and when I got home near 1 I couldn't post because Blogger's scheduled 30 minute outage turned into a 12 hour outage. I don't think the site actually came back until probably less than an hour before class, because it still wasn't up when I woke up. I was going to post immediately after class, but RedRover was misbehaving in my lecture hall. Long story short: I'm sorry for the lateness of the post. All that said, on to the post.

Greetings, all. I'm Will Fleming: I'm a junior Information Science major in the Engineering school. My professional interests are largely in intelligent systems, game design, and web development. I also haven't the slightest idea what I'm going to do after I graduate.

Outside of class, I work at CIT as an Applications Developer and I'm the Assitant Design Editor at the Daily Sun. Well technically, I'm not the Asst. Design Ed. at the Sun yet: I'm still in training. But I already held the position a year ago, I'm the only person running for the position right now, and the incumbent editors are out of the country, so for all intents and purposes I think I've got the job.

My free time is eaten up mostly by music, movies, books, and my personal web development projects. I'm a bona fide music junkie, and listen to mostly pop, rock, indie, folk, electronica, dance, and classical with occasional excursions into the realms of hip hop and jazz. My taste in books and graphic novels is somewhat less eclectic, but some of my favorites would be anything by Neil Gaiman or Alan Moore, the Sherlock Holmes stories, and Lord of the Rings. Although most of my development projects never reach a level of maturity where I take them public, I do keep a blog at http://jwock.org.

I think that's really all I have to say about myself. I guess I also drink too much coffee, and miss my pets when I'm in Ithaca. Yeah, that pretty much sums me up.

Bye now!

~will out.