Slashdot Mirror


Bot-avatar Pesters Second Life Users (For Science!)

holy_calamity writes "A bot-controlled avatar that tracks down lone avatars in Second Life and purposely invades their personal space has been created by UK researchers. The idea was to see if users value their virtual personal space. Bots avatars are not encouraged by Linden Labs — although this one is being deployed by academics, presumably spam-avatars (spavatars?) won't be far behind."

31 of 124 comments (clear)

  1. Personal Space by dintech · · Score: 5, Funny

    Sounds more like 'Second Wife' than Second Life...

    1. Re:Personal Space by Firethorn · · Score: 3, Insightful

      The first thing I thought of was a response:

      "Virtual Shotgun". For those who really want their privacy.

      While I don't normally understand why you'd play an online game to just be alone - from what I understand of second life you could have the equivalent of 'prepping', IE you're creating something to be shown later. Whether this is a house or an adult accessory, it doesn't really matter.

      --
      I don't read AC A human right
    2. Re:Personal Space by Opportunist · · Score: 2, Insightful

      Not to be alone. But I tend to choose my company and spend my (spare) time with whoever I want to spend it with. Bots and stupidheads are usually not in that group.

      I mean, imagine you have a party and someone keeps trying to sell his Amway crap, going on everyone's nerves. Wouldn't you throw him out?

      --
      We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
    3. Re:Personal Space by dintech · · Score: 3, Interesting

      Hold on, I post a comment about wifes and the first thing you think of is shotguns? :)
      You should be posting in this thread.

    4. Re:Personal Space by garett_spencley · · Score: 2, Interesting

      I think the issue really applies mostly to random situations in public. For example, the other day I was at the bank. It was not busy at all. There were two tellers and each was serving a customer. I was next to be served and there was no one else. Then a man comes in and gets behind me in line. He came up RIGHT behind me to the point where I could hear him breathe, almost feel his breath on the back of my neck and he reeked of bad after shave.

      I couldn't help but think to myself what the fuck is this guy's problem?! There is absolutely no one else in the bank. Mind your own personal fucking space asshole. It's not like it's busy as hell and the line was cramped and he had no choice. He chose to get so close to me as to make me feel very uncomfortable when there was more than plenty of space for us to both keep our distance.

      Had I been in New York I imagine I might have had the nerve to turn around and punch the guy in the nose. But I'm a regular customer at that bank and I knew I was going to be served any second so it wasn't worth making a scene.

      My guess is that's the type of situation that this "study" is examining. Whether people in virtual worlds would be as uncomfortable with random assholes getting too close as people in real life.

  2. Another headline... by Anonymous Coward · · Score: 4, Insightful

    ...researchers stalk people online to see if they mind.

  3. Statistics! by Aladrin · · Score: 4, Insightful

    "Out of 28 avatars approached this way, 12 simply moved away and 20 also responded via text chat."

    So they 'simply moved away' and some also responded by text? Then they didn't 'simply move away'.

    And 28 is a pretty small sample. Why bother having a bot for so small a sample? Wouldn't it have been a lot easier to just do it by hand? Or just let it run a few days before publishing the results? And those who stayed put... How many were idling (not even at their computer) and how many simply ignored the childish idiot that was harrassing them? (You don't have to play online games for long until you've met enough idiots and learn that ignoring them is the best possible course of action, especially the ones that want to get right up on you and do stupid things.)

    --
    "If you make people think they're thinking, they'll love you; But if you really make them think, they'll hate you." - DM
    1. Re:Statistics! by kebes · · Score: 3, Insightful

      Why bother having a bot for so small a sample? Well, it may be that for research purposes they want the behavior of the approaching avatar to be as consistent as possible. Therefore they use a bot to automate the approach so as to avoid the experimenter biasing the results somehow.

      That having been said, I totally agree with your post: 28 events is a ridiculously small sample size to try and measure the behavior of people in virtual worlds. Considering they went to the bother of writing a bot, one would hope they will leave it running for awhile longer to accumulate more data.
    2. Re:Statistics! by foobsr · · Score: 2, Funny

      28 events is a ridiculously small sample size

      But imagine how many nearly identical conference/research papers they can conjure up by slowly increasing the sample size they report on!

      CC.

      --
      TaijiQuan (Huang, 5 loosenings)
    3. Re:Statistics! by mattOzan · · Score: 2, Funny

      "Out of 28 avatars approached this way, 12 simply moved away and 20 also responded via text chat."
      So it looks like there's only 28 people left in Second Life? More than I would have imagined. They were probably just startled to see someone new.
  4. One flaw by SmallFurryCreature · · Score: 5, Insightful

    They observed that female avatars were less guarding about their personal space then males, a behaviour apparently the same as in real life.

    The flaw? Female avatars do NOT have to be controlled by a female user.

    Would a male playing a female mimick this behaviour? IF that is the case, that would make a far more intresting study. If it isn't then their measurements are flawed since they cannot tell what sex a user really is.

    --

    MMO Quests are like orgasms:

    You may solo them, I prefer them in a group.

    1. Re:One flaw by sw155kn1f3 · · Score: 5, Funny

      >> Female avatars do NOT have to be controlled by a female user.

      Huh? You mean that cute elf girl I've been dating... Oh shit..

      --
      - Arwen, I'm your father, Agent Smith.
      - Well, you're just Smith, but my father is Aerosmith!
  5. interesting test for science' sake by freg · · Score: 2, Interesting

    This begs the question... do Westerners stand further apart than Asians when chatting in a virtual world? I would guess so. For some reason I occasionally find myself backing up my avatar a step or two and facing it towards the avatar I'm talking to, without even thinking about it really.

    1. Re:interesting test for science' sake by Rogerborg · · Score: 3, Funny

      do Westerners stand further apart than Asians when chatting in a virtual world? I

      No, they stand the same distance apart; it would violate the laws of physics if one of them was "further apart" than the other. What a peculiar question.

      --
      If you were blocking sigs, you wouldn't have to read this.
    2. Re:interesting test for science' sake by Fulcrum+of+Evil · · Score: 2, Informative

      10 feet is a lot closer with a knife if you're in a cramped hallway.

      --
      "We returned the General to El Salvador, or maybe Guatemala, it's difficult to tell from 10,000 feet"
  6. IRB issue by ckolar · · Score: 5, Insightful

    The article mentions fleetingly at the end that the ethical issue is still up for grabs. I wonder if they actually got IRB approval for use of human subjects. Even though it is a bot that interacts with the other avatars, it is still an investigator-designed intervention into this space, they are collecting data in a deliberate and systematic way, and looking to generalize the results. The fact that they are collecting data without consent and using it in this manner strikes me as a violation of user privacy. Yes, I serve on an institutional IRB, and no, this would never pass in my institution. It is frightening that these researches imply that there is somehow a lower standard for virtual environments (it is not the avatar that is being studied, but the human on the other end) for the conduct of psychological experimentation.

    1. Re:IRB issue by kebes · · Score: 2, Interesting

      Yes, I serve on an institutional IRB, and no, this would never pass in my institution. Out of curiosity, why wouldn't it pass?

      It doesn't seem too different from psych experiments where a researcher stands in a public place and does something or asks a certain question. (E.g. a famous one was to have an attractive female and male person ask passerbys on a college campus if they wanted to go have casual sex right now.) I was under the impression that experiments in public places didn't require the explicit consent of each unwitting participant (provided, of course, that the experiment involves only normal everyday interaction, like asking questions, and does not induce undue stress on participants, etc.).

      I fully agree that any experiment involving humans (even over the Internet) should require internal review and approval. But why would this experiment not pass? It doesn't seem like the burden imposed on the unwitting participants is very large, considering that they are voluntarily participating in an interactive "public space" online and have no expectation of privacy or even peace (the bot wasn't doing anything that a human player couldn't or wouldn't).

      Of course another aspect of passing review is that the data recovered should be sufficiently robust, meaningful, and significant that it warrants intruding on the lives of subjects. Perhaps it is this criteria that you feel it fails?
    2. Re:IRB issue by ckolar · · Score: 2, Informative
      Well first off, my gut reaction is to reject everything. :) Just kidding. The problems (and I have not seen more than the article, so I am just making a lot of suppositions here) are like this.

      • 1. As other people posted in this thread, the terms of use for SL indicate that you may not use a bot to collect personal information, and so the users DO have an expectation that they are not being approached by bots for the expressed purpose of collecting personal information.
      • 2. As you mention at the end, the design is significantly flawed WRT yielding useful results.
      • 3. Without IRB review, there is no guarantee of risk abatement for subjects. Back to your street corner example, lets say that one of the people approached was one of the 25% of college women who have been sexually assaulted and the proposition causes her to have an unexplained (to the researcher) anxiety attack, or depressive episode. The researchers here (presumably knowing that they were being annoying) would need to have demonstrated that they could intervene if they created a potentially harmful situation for the subject.
      • 4. The very fact that you are in a public place implies that you are not being part of a research project unless otherwise informed to the contrary -- that is the informed consent part. The burden falls on the researcher. An IRB may approve situations where there is considered minimal risk (do people stand to the left or right on an escalator), but again, the researcher must first demonstrate that there is no risk involved the participants.

      I could come up with more, but this is starting to seem like work. :)

    3. Re:IRB issue by Atreide · · Score: 2, Interesting

      I was under the impression that experiments in public places didn't require the explicit consent of each unwitting participant

      this raises the question : "when connected online with other people, are you in public space or private space ?"

      * when online, i (presumably) am at home, in private space
      * only a part of my self & conscious is in relation with other people (the same if i am telephoning)
      therefore it can be seen as a huge private phone conversation ?

      after all there are regulations & debates worldwide regarding online privacy and what can be and cannot be done

      also the answer has repercusions on marketing (can you advertise), on law enforcement (can you be "watched" upon), on free speech (can a journalist report your statements without your consent because said in "public")...

      finally, it can be a private or public space but it can also be defined as a new type of "environment". This can lead to give such virtual world its autonomy and own regulations.
      "no sir, only my avatar commited the robbery of linden dollars, my 'self' is in a totally different world and cannot be prosecuted for robbery. it can only be prosecuted for benefiting the stolen money".
      --
      The world belongs to those who get up early. - I'm far from being the king of Earth then :-(
  7. Re:I for one welcome ... by cromar · · Score: 2, Funny
    Now, that would be a cool idea for a bot net experiment =oP

    I'm down.
    1. Steal top secret gov't research.
    2. ???
    3. Build mesh networked pan-handling robot emissaries.
    4. Profit!
  8. Bots avatars are not encouraged? by jhRisk · · Score: 4, Interesting
    A few choice selections from Section 4.1 of their TOS http://secondlife.com/corporate/tos.php

    (v) take any actions or upload, post, e-mail or otherwise transmit Content that contains any viruses, Trojan horses, worms, spyware, time bombs, cancelbots or other computer programming routines that are intended to damage, detrimentally interfere with, surreptitiously intercept or expropriate any system, data or personal information; I'd consider that detrimental interference. Also, there's this one

    (vii) upload, post, email or otherwise transmit any unsolicited or unauthorized advertising, or promotional materials, that are in the nature of "junk mail," "spam," "chain letters," "pyramid schemes," or any other form of solicitation that Linden Lab considers in its sole discretion to be of such nature;

    There are others that I believe apply to the utilization of a bot, potential exploits through bots (ex. spamming) or both. Also, what they're extrapulating from the empirical evidence is off IMHO as well.

    SL-bot observed pairs of normal avatars as they interacted. It found that users are, on average, six times more likely to shift position when someone comes to within 1.2 m. That backs up the idea that people also value their virtual personal space, say the researchers.

    I'm sure it had nothing to do with being courteous, putting the new character into view to inspect or anything else. Yeah, they wanted their "personal virtual space"... sure. Sounds like another misread on cause and effect at the expense of opening a pandora's box.
    --
    That's just my POV... no more, no less.
  9. One Ring to rule them all... by digitaldc · · Score: 2, Interesting

    Ring of power
    The ring connects the avatar to software that not only controls its actions, but can record everything going on around it. This is an extreme example of the way objects can control characters in Second Life


    Once again proving that Second Life is becoming more and more like Tolkien's world of Lord of the Rings

    --
    He who knows best knows how little he knows. - Thomas Jefferson
  10. Bots by Ash-Fox · · Score: 2, Informative

    Daily I end up banning a bunch of "naked ruth" bots (as others do) that don't seem todo much other than idle in various places.

    I don't know what they're doing, but if it's for research purposes, it's really getting to be really annoying. The banlist I have has exceeded over a hundred of these being banned and they keep coming back (under different names). This isn't the main grid which is considered public, it's a grid of private simulators (known as the valley sims) and there has not been any permission granted at all for research purposes in the simulators I help out in.

    It is at the end of the day wasting a lot of my time and I consider these bots without prior consent, harassment.

    --
    Change is certain; progress is not obligatory.
  11. Naked Ruth Bots? by MLease · · Score: 2, Funny

    Are they hot?

    -Mike

    --
    I'm sorry; I don't know what I was thinking!
  12. Hiro by hansamurai · · Score: 4, Funny

    So how many users chopped off the intruder's appendages with a katana and then had their homemade daemons clean up their handiwork?

  13. Ring of Power: Cute.... by nweaver · · Score: 2, Insightful

    An interesting part of the article was the discussion of the technique involved.

    Apparently, Lindon doesn't want bots, so you can't script avatars. But items can be scripted, and items can instruct avatars to do things. So you just script an item to instruct the avatar what to do...

    Trez cool. (Now if only you could make the item replicating and infectious.... :) )

    --
    Test your net with Netalyzr
  14. Oh, when THEY do it in SL, it's called research by Rogerborg · · Score: 3, Funny

    When I do it to teenage girls on the subway, it's called 6 to 12 months.

    --
    If you were blocking sigs, you wouldn't have to read this.
  15. Re:They did not get approval... by ckolar · · Score: 3, Informative
    Nice catch on that. By US federal regulations, a research project can only be "exempt" after an IRB reviews the proposal and declares it exempt. Sounds like a contradiction I know, but you are NEVER exempt from being reviewed, just judged to be exempt from additional monitoring/oversight (for low risk situations). A researcher may NEVER decide on their own that the IRB would declare a project exempt.

    Here are links to relevant sites:

    Appendix A: Federal Policy for the Protection of Human Subjects

    http://www.hhs.gov/ohrp/humansubjects/guidance/45cfr46.htm

    Appendix B: The Belmont Report

    http://www.hhs.gov/ohrp/humansubjects/guidance/belmont.htm

    A copy of our screening form and a link to the AERA grid for risk and ameliorative measures can be found at: http://www.imsa.edu/learning/research/hasrc/
  16. In related news... by CarpetShark · · Score: 4, Funny

    ...researchers stalk people online to see if they mind.


    "Online research" patent claimed by FBI and RIAA.
  17. avatar space invaders by xPsi · · Score: 2, Insightful

    Back several years ago when Star Wars Galaxies just came out, my brother did a similar avatar experiment on me. SWG only allowed one character per server per game copy so my brother went out and bought a second copy of the game and another computer to run a full time mining character on the same server as our adventuring characters (ah, the glory days when we still thought that kind of thing was important). He failed to mention this detail to me. One day we were out questing (using headsets) and this very random character came up to me in Mos Eisley and started following me around typing a stream of non sequiturs like "what's the frequency Kenneth? Reveal the blue bug rathouse conspiracy!" over and over. My brother and I had this extended conversation in the headsets about how random and rude this was until I finally caught on when the freaky character mentioned some inside joke (which took me aback initially). Invasion of "online space" is rarely Evil, but it can be really, really annoying and distracting. I doubt this is a surprise to anyone.

    --
    i\hbar\dot{\psi}=\hat{H}\psi
  18. At least it's a new excuse ;) by Moraelin · · Score: 3, Informative

    Well, if there's one thing I've learned in my days on MUDs is that there's always a minority who gets their kicks out of being assholes and annoying to everyone else. And they're always ready and willing to twist logic in the most incredible ways to argue why it's a good thing, and you should allow... nay, be thankful that they're doing it in your game. Among other things:

    - that it's great fun for everyone, and their victims who complain about it somehow don't know what they really want in a game. Why, they'd probably leave in droves if someone didn't harrass them.

    - that it's a pre-requisite for role-playing. (Apparently being killed again and again by someone 30 levels higher than you, and with battlecries of, "LOL! N00B! U SUCK! I FUCKED UR MOM!" is proper role-playing. In fact, the only kind of role playing.)

    - that it was testing, if they were using a bug against everyone else, and they were surely going to report it. They "tested" it 100 times a day for a whole month just to be really sure how it works, and submit a really really good bug-report, you know.

    - that the first amendment gives them a sacred right to say and do whatever they want, anywhere they want, and to anyone they want. And if you try to stop them, that's the road to tyranny and slavery. (Never mind that the actual text refers to the Congress, not to a privately owned server.)

    Etc, etc, etc.

    That it's for scientific research... well, now that's a new excuse. Just when I thought I had heard heard everything.

    But I hope that everyone will excuse me if I still see it through the eyes of a jaded old MUD coder. The primary aspect is that it's (mild) harassment, no matter in the name of what mis-guided idea or excuse it's done. It's inconveniencing someone else, so don't do it.

    Even if it seems like a mild annoyance at best, already there is no shortage of people annoying everyone else. And then there are people who come from a very stressful RL situation to unwind online. Even a mild annoyance just adds to the existing stress, when one is stressed enough. If someone came home after the boss riding his butt for 2 hours, dealing with clueless people for the other 6, and maybe add something like a visit to the dentist and/or an argument with his wife, the last thing he needs is an annoying newbie getting in his face all the time.

    And I might even shrug and move on if it were a genuine newbie who barely has enough WASD motor skills to get in that room at all, but not enough to maneuver himself in a socially acceptable position. But it being a (mild) harassment bot and justified as "research"... dunno... just feels... wrong.

    --
    A polar bear is a cartesian bear after a coordinate transform.