Panda in a petri dish

"A Pentagon-funded group, the Highlands Forum, is investigating the possibility of fighting terrorism by postulating that terrorist networks are similar to the online communities found in MMORPGs. By studying interaction within a community like Everquest, the military hopes to find hints on how to crack Al Qaeda."


DARPA Awards Contracts for Pioneering R&D In Cognitive Systems
Date: Friday, Jul 16 @ 13:18 PDT
Topic: Archive


The Defense Advanced Research Projects Agency (DARPA) has awarded contracts to Carnegie Mellon University's School of Computer Science (Pittsburgh, Penn.) and SRI International (Menlo Park, Calif.) to develop software that could significantly advance the science of cognitive systems and revolutionize how computers interact with humans.


The program, called Perceptive Assistant that Learns (PAL), is expected to yield new technology of significant value to the military, business, and academic sectors. It will spur pioneering research in cognitive information processing, including areas of artificial intelligence, machine learning, knowledge representation and reasoning, machine perception, natural language processing, and human-computer interaction. Through the PAL program, researchers will develop software that will function as an "enduring personalized cognitive assistant" to help decision-makers manage their complex worlds of multiple simultaneous tasks and unexpected events.

DARPA’s Information Processing Technology Office (IPTO) manages the PAL program. IPTO is the DARPA office responsible for the ARPANet, the Internet, large-scale parallel processing, and many of today’s critical computing technologies. Part of IPTO’s current mission is to develop cognitive computing systems – systems that can reason, learn from experience, take advice, explain themselves, and respond intelligently to situations never encountered before. Technologies developed in the PAL program will make important contributions far beyond the realm of smart assistants. If cognitive systems become a reality, we will have computer systems far better at adapting to their users' needs, and far better at coping with unexpected inputs than today's systems. In addition, cognitive systems will become more responsible for their own maintenance, do a far better job of warding off security attacks, manage internal resources more effectively, and have drastically reduced development and deployment costs.

Carnegie Mellon University’s effort under PAL is called RADAR, for Reflective Agents with Distributed Adaptive Reasoning. The system will help busy managers to cope with timeconsuming tasks such as organizing their E-mail, planning meetings, allocating scarce resources such as office space, maintaining a web site, and writing quarterly reports. Like any good assistant, RADAR must learn by interacting with its human master and by accepting explicit advice and instruction. The RADAR project draws on Carnegie Mellon's expertise in artificial intelligence, machine learning, natural-language understanding, and human-computer interaction. DARPA has awarded Carnegie Mellon $7 million for the first phase of a five-year contract.

SRI has dubbed its project CALO, for Cognitive Agent that Learns and Observes. The name was inspired by the Latin word “calonis,” which means “soldier’s assistant.” The CALO software, which will learn by working with and being advised by its users, will handle a broad range of interrelated decision-making tasks that have in the past been resistant to automation. It will have the capability to engage in and carry out routine tasks, and to assist when the unexpected happens. DARPA has awarded SRI $22 million for the first phase of their five-year contract.

The SRI contract includes subcontracts with 20 additional research institutions: Boeing Phantom Works, Carnegie Mellon University, Dejima Inc., Fetch Technologies Inc., Georgia Institute of Technology, Massachusetts Institute of Technology, Oregon Health and Science University, Stanford University, State University of New York - Stony Brook, University of California at Berkeley, University of Massachusetts at Amherst, University of Michigan, University of Pennsylvania, University of Rochester, University of Southern California and its Information Sciences Institute, The University of Texas at Austin, University of Washington, University of West Florida’s Institute for Human and Machine Cognition, and Yale University.

Researchers for both project teams will themselves use the PAL software during its development to ensure that it satisfies all of the important needs and requirements, including privacy, security, and trust. Technical progress will be assessed each year through a series of experiments and structured evaluations.

**** lining walls of SL house with foil and encrypting IM's *****

Yes Panda. I would love to attend.

-aimee

These are for live-fire excersizes. It does not involve SL, but team-play games, especially PVP games. They're discussing an agent that learns tactical and situational awareness.

Quake, Counterstrike, etc. all will be likely candidates, but most likely are government-sponsored games, such as American Army. Repeat: this is NOT a SOCIALLY AWARE SYSTEM. It is a TACTICALLY AWARE SYSTEM.

The ones which are not tactically aware are bunk - vaporware porkbellies. The same proposals have come out for decades, and they never get anywhere.

Edit: Although the first paragraph says something entirely different. Is that even a related paragraph?

-Ansi

This experiment has gotten a little attention ...check out my thread Research=advertising...it seems blogs and link dumps are spreading the word of our lil petri dish

Actually Ansi I was sorta joking... but I'm still lining my SL wall with foil.

I got an A- ... got some points off for bad spelling.



Well, I agreed with many of the points in your original post, actually. Just your ranting about an obvious misspelling and how all of your dictionaries were choking on a single letter out of place called out to me for a smarmy retort.

As far as the subject at hand, SL is a perfect place to do social research, so long as the researcher does not pull back the curtain and reveal themselves. Anonymity is key to success to uncontaminated observation.

Someone else mentioned Heisenburg's theory of uncertainty. I would say that only applies if the party being observed is aware of the observations. I believe this could be easily observed by the Pitts and others with a simple experiment... shell out the $10 for a basic account minus the Pitt last name, and throw yourself into the populous to observe anonymously. Afterward, return to the same group of people but be open about who you are and what you are doing there. The results will be vastly different.

As far as the ethics of observing us or doing social science in general in SL or other online world, I have to side with those that are saying it is ethical so long as the anonymity is maintained.

-Ghoti

Panda -- with the amount of forum attention this issue has received, and now that it has spread to other sites -- it's time to recognize that this is a larger issue that Mr. Parent and the Pitts have accidently stepped in.

Good. Isn't that what we're here for? To define the boundaries of the "new world"?

So Mr. Parent should realize that it is no longer a matter of negotiating with you alone, but with the larger SL community -- and he should also realize that he has, for better or worse, placed himself at the cusp of a bigger issue than he may have bargained for.

Which ought to thrill him -- if he plays this right, he could get an innovative, attention-getting academic article that will make him look like something of a pioneer....and at the beginning of his career.

If he plays it right and straight -- and listens, and takes the long view.

I couldn't agree with you more, Donovan (as you can see from my last post). If the professor choses to open this discussion up to the public in-world, I would love to have you participate.

Obvious to YOU perhaps, but I've seen enough oddly spelled, though entirely correct words to know that if I don't recognize something, chances are it's just obscure, not misspelled.



Such styles of research have often resulted in totally crap results. The Professor mentioned such in a blog or something, and he's not the first person I've heard it from.



Nope. It certainly INCREASES the 'uncertanty', but just the act of observing something, even 'anonymously' or 'silently', alters what occurs. That's the actual theory.

Ok since my last joke post went over like a lead balloon, why don't I just make a quick serious comment about research ethics.

While nobody is going to go to jail for simply breaking the generally accepted rules of ethics in psychological research, your work and YOU can be dismissed as unethical if you are in violation. A researcher's obligation to the subject(s) are as follows:

Do no harm:
I don't see a problem here. No matter how peeved we get, I don't think these students are going to be able to harm us.

Consent:
Yes. The subject of research needs to be able to give consent for the work to be considered ethical. Unless it is stated in the TOS, I do not believe a user's presence in Second Life automatically implies consent. We are here for a lot of reasons, but "to be the subject of sociological analysis" may not be one of them.

Deception:
It is generally accepted that SOME form of deception may be required to ensure that the subject's knowledge of the study will not skew the results. The level of deception should extend no further than what is necessary for the research to succeed, and should not violate the other rules.

Freedom for the subject to withdraw:
Easy to do in SL. We can just walk away. But it presumes consent in the first place.

Confidentiality:
The big no-no I see here is when the researchers are posting chat logs without concealing usernames. This is not only a violation of TOS, it is unethical.

Debriefing:
This is particularly important if deception is used. The researchers must complete the subject's understanding of the research including the techniques for data gathering, purpose, and results. This is also an opportunity for a responsible researcher to accept feedback from the subject, particularly when the subjects express a negative reaction to their treatment.



Ok so say Aimee had to design the research experiment. How could she do it?

1. As a representative of UPITT, I would approach prospective subjects, explain to them I am conducting sociological research that may or may not involve anonymous observation by researchers and/or other means. I would also impress upon them that not all of the subjects being signed up will actually be observed.

2. If and only if they agree, I would then observe them via alternate anonymous accounts at random intervals during the passing months of the semester. They DO know that a possibility exists that they are being observed...or maybe not! I think it's safe to assume the average user will let down their guard, especially after a few weeks of not hearing from UPITT.

3. If at any time the user feels uncomfortable with their participation, they can IM the UPITT representative and opt out immediately.

4. When the analysis is done, the user is debriefed, and all names and identity-revealing information is removed from the final analysis. The research is published for peer review, and the institution / researchers won’t look like a bunch of scumbags.



-aimee

I loooooved your last joke Aimee

YAY! YOU LUB ME!

-aimee

Right on tack Aimee...but I am still waiting for any of the proffesors involved to say that they are actually using this for personal or proffesional research. So far it is being used as personal experience writing exercises for freshman. But your outline could and should be if not used at least considered by any potential academic researchers who come to SL for the purpose of publishing research. When and if this happens, the people doing the research will likely be working solo and will be able to assimilate easily into the world with little notice. I doubt it will be a group of youngsters in an english 101 class.

As far as the class let the freshman write their papers and enjoy their experience in SL. If the residents feel that they need to outline codes of conduct for users in the experiment, can it not be done without accusing them of being disingenuos. They are young, fresh minds looking at sl from a different perspective. But everyone looks at sl from their own perspective. They could be future residents. They could be future builders, scripters, patrons and friends. Many have made the mistake in the past of posting chat logs. It is not right, but mistakes will be made. This a new realm. Many will try to find new uses for this media, we hope, many will make mistakes, find themselves uniformed, feel confused about the purpose of SL and possibly step on toes. It is our job to question and find answers but not to lynch every new idea that comes into the world.

Is it up to the real world to conform to the confines of SL or is it up to us to expand our view of what SL is and make ourselves maleable enough to grow with this environment thereby impacting the growth without stifling progress?

Yes Isis, actually this whole thing is starting to sound more like a creative writing project. I would put this on par with "mingle with people in the park then write about it". Mostly harmless and I think journalistic ethics (far less rigorous) are more applicable than experimental ethics in this case.

-aimee

p.s. I still think my joke was funny.

As I work on building, I notice a little green dot on the map. I peer over to the Digi building and see one of the elusive Pitts. I watch her (I will call her Mary to protect the innocent) for a bit. She unsucessfully attempts to sit in a hot tub in the building (guess they like to have fun too, as any regular noobie would. She explored the building a little bit and eventually ended up on the roof, I decided to make my move. I fly up to the building and say "hiya" Here is the conversation that followed (many smileys had to be deleted from this convo due to "too many images":

You: hiya
Mary Pitt: hi!
You: hehe i see you are relatively new
Mary Pitt: haha yes, for sure!
You: hehe
You: hehe how are you liking SL so far?
Mary Pitt: i haven't figured out a lot of stuff yet, but it's really neat!
You: hehe
You: ya it takes a little bit to get use to everything
Mary Pitt: definitely. im not even used to computer games either!
You: hehe
You: well this can get sooo addicting lol
You: especially if you are the creative type
Mary Pitt: haha yeah i believe it
Mary Pitt: yeah!
You: have you learned how to build yet?
Mary Pitt: um.. not really. i can make a nice wooden box!
You: yay lol
MaryPitt: heheh
You: hehe thats a good first step
Mary Pitt: hah yeah its something
You: hehe have you gotten to go shopping yet?
Mary Pitt: no! that always sounds like a good idea
You: hehe, the best place imho is midnight city
Mary Pitt: neat! where's that?
You: gets kinda laggy there at times, but it has the best fashion
Mary Pitt: haha perfect
You: hang on, lemme find the landmark
Mary Pitt: ok
Mary Pitt accepted your inventory offer.
Mary Pitt: cool, thanks
You: hehe
MaryPitt: Well, thanks a lot for being so friendly, I've got to go

She seemed kind of nervous, but nice. I rated her and gaver her a jewelry set (cause no lady should be without!) as she teleported out, perhaps to Midnight City.

My analysis: Little more nervous then regular noobies. Seemed like the kind of person that just wants to explore the world on their own. There were long pauses between pieces of conversation and I felt that I started most of the conversation. I will see in days to come if this is typical behavior of the Pitts.

I will watch their blog and see if this specific Pitt has anything to say about the interaction.

It's a "How to write an essay" class. In the college I went to, it was a -required- course. Every freshman HAD to take a "How to write an essay" course. Mine involved something like thirteen essays on coin collecting. Theirs will involve digital communications mediums like blogs and SL.

Quick update....


The professor and i have spoken, and we would both like to see positive come out of this situation. We will be co-hosing an event next Wednesday at 4:00 pm game time. His entire class will be there, and I would love to see a turnout from interested SL residents who would like to bring this discourse in-world.

Feel free to post here or IM me with any questions, comments, or concerns

Awww, I'm going to be at work for that. Bleh. Well, thanks for the offer anyways.

Filthy numismatist!

*snerk* Haven't heard that word in a while.

I learned it as an eight year old, clutching my red book and bargaining feverishly for an 1853-O half dime.

Emmy's Blog:

Conversation with Pitt who shall remain nameless (to protect the innocent..although I'm not sure who they are..). I was in IM as pandastrong was in mouselook blowing them off his property with..a Seburo, I think?

nameless pitt: and he knows that i was trying to help him
Emmy Guillaume: Apparantly he does not.
Emmy Guillaume: or he wouldn't be shooting at you.

The nameless Pitt began the conversation by shouting "EMMY, COME HERE"...since s/he couldn't get over the property line without being blasted. I'm like..um.."Say please?" but instead opened an IM to see if I could broker a Detente.

The nameless Pitt in question did tell panda they were sorry...right before they indicated 'but, it was your fault'. So Detente lasted about ...4 seconds (panda? 'bout right?)

We tried.

I will be at the Pitt meeting with Bells on.

What an interesting turnaround!

I think I've gotta go hunt for my own Pitt pet when I'm online, and study them. Groom them into a quality SL player. Convert them to the daaaaaaark side, mwuhuhuhuhahahaahah!!!!! *cough, cough, hack wheeze*

Anyhooo.

haha...I was just speaking with a friend of mine who works with the local paper and we said the same exact thing. (not about your joke...though I did end up with beer spewing when I read it the first time) Protecting sources is key to journalistic endeavors though they are likely to quote out of context and tell you that it is off the record only to post it in the next issue. That aside, I find this rather thrilling. Creating new paths and boundaries...We are really all creating a new world...hopefully we do a better job than we have done with the real one.

Emmy, you should sooooooo join the UN

And yes, I made a decision to shoot any and all Pitts off of my land until the quotes were removed from the professor's blog. And my trigger finger got especially itched during the Detente, when the Pitt tried to explain to me that he was sorry, but I REALLY SHOULD HAVE TALKED TO HIM OR THE PROFESSOR BEFORE STARTING THIS THREAD.

Anyone know of a "shoot upon the presence of irony" script that can be attach to a Seburo?

At any rate, after my talk with the professor he kindly deleted the quotes off of his blog.

For now, the Pitts are welcome back onto my land, and I hope to keep this permanent after hearing what the class and SL residents have to say at next Wednesday's event.

OOOOH.. and everyone is welcome to check out the INSANELY AWESOME build that Ariel Roentgen did on my land of an observatory.. complete with one way viewing glass!!!!! I'll post a pic when I get a chance.