Pop Bots – Avatar of the Year – 2008
by Alphaville Herald on 14/02/09 at 7:07 pm
by Pixeleen Mistral, National Affairs desk
With the lunar new year celebrations behind us and a Gorean slave cleaning crew going through the Herald offices throwing out old Second Life Herald stationary – we're rockin’ Alphaville style now — a crucial story that somehow never made it to press was un-earthed. Yes, the most important function of the Herald was somehow overlooked – presenting the coveted avatar of the year award.
After shooing the Goreans from the offices with a promise that they can always visit later, a solemn ceremony was held, culminating in black smoke rising from the a seldom used chimney in the Hyperborea office complex and a single white dove released to swoop the crowd before flying off into the distance. Having completed the mandated rituals, the Herald now officially names the Pop Bots the avatar of the year for 2008.
Much loved by land owners who wish to hype the traffic figures for their stores, and hated by nearly everyone else, Pop Bots have become a fixture in Second Life – leading some to wonder why they are tolerated. The answer is complex, but instructive.
By swelling the online population figures without doing much of anything useful except squatting in place, Pop Bots fill a certain void – particularly in the users online statistic. Although bots were originally only a small part of the online population, over time the ranks of the bots have grown to the point where some estimate that 40 – 50% of the online users are actually permanently AFK bots.
For the Lab, this presents a problem. Residents complain about bots, but a sudden change in bot policy could result in sharply reduced concurrent online users and raise troubling questions in among players already concerned with the direction of the world.
Beyond concern about the direct effect of going cold-turkey on bots, there is a financial upside for the Lab from the growing presence of bot swarms.
By allowing land owners to game the search system with bot-inflated faux avatar traffic, the Lab tacitly encourages other shop owners to buy classified ads in retaliation – and classified ads are purchased from the Lab for sometimes surprising prices. The next time you are tempted to grumble at a box-o-bots, keep in mind that they indirectly help the Lab’s bottom line.
Please join us in congratulating the Pop Bots as they proudly join CopyBot and the always charming and thoughtful Prokofy Neva in the immortal pantheon of avatars who have deeply affected the virtual world for better – or worse.
Graves
Feb 14th, 2009
And here, I was hoping that the fad of posting articles about bots was over with :/
Maggie Darwin
Feb 14th, 2009
I’m delighted that bots “indirectly help” LL’s bottom line the same way burglars improve things for security firms. No, really.
So, “Alphaville” now–LL finally cracked down on the trademark thing. And you couldn’t convince then to buy you out like SLX.
I’m sure your desire to compete directly with Massively was the real motivation. Have fun with that.
Little Lost Linden
Feb 15th, 2009
In my sl travels it appears to me that at least half the sl traffic is bot related.
In 2006, I completely remember traffic levels near 45K on the weekends. 2 to 3 years later and it’s 80k, and I know that a huge percentage of the new traffic is bots that have infiltrated the sims.
Someone really needs to invent an inworld game of detecting bots and posting sims with the most bots and photos etc., similar to the german site that does it, only in English, or at least translated to English so people from the US can participate.
In exchange for not only having the fun of hunting bots, and believe me, it is sometimes fun trying to find them when they are well hidden in a sim (please note, this is the opposite of finding them when you are searching for real people to chat with, socialize with), but also, if someone could reward bot hunters with Lindens, or with name recognition or something, I think those would be the 2 key catalysts for everyone to start noticing bots more.
And that is the problem, too many people don’t know they exist, and don’t realize how many problems bots can cause, and they also don’t understand how deceptive bot use is.
It also wastes a persons time when they are looking for things to do, people to see, etc., and they are constantly being tricked into visiting empty sims filled with bots.
Once you start to realize, that bots, simply, are the equivialent of e-mail SPAM, that is when you realize how bad bots are and why they should be removed or at least reduced, similar to how SPAM filters are in place.
If you get nothing from this rant, please try to burn this image into your mind and hopefully you will start to understand:
SL BOTS = SPAM
SPAM = BAD
It really is just that simple.
LOL
Feb 15th, 2009
@Little Lost Linden
If you really are a linden then you are right on par for the course just figuring this out now that spam and bots are bad. Why don;t you go to the SL forums and the XSTREET LL forums and post your new revelation there. Leave at least this forum free of “The Linden Taint”
Darien Caldwell
Feb 15th, 2009
Actually, it would help LL’s bottom line to get rid of bots. Then the people gaming the system would have to buy classifieds too. And more importantly, real paying customers could get online instead of non-paying bots. Just say no to bots.
Jahar Aabye
Feb 16th, 2009
When you first did the story on the supposed rate of 40-50% of avatars being bots, I wrote a comment pointing out the many statistical errors involved in those studies. They were pretty glaring problems, and anyone who actually believes that 40-50% of online avatars are bots is either statistically challenged or….well, let’s call a spade a spade here, either willfully ignorant or wishing to hype their own personal agendas.
Since one presumes that Prof. Ludlow has, at some point in his academic career, taken some sort of basic statistics courses (c’mon, even the social sciences need to know basic statistical math), I suspect that it is the latter possibility….although obviously the possibility of willful ignorance would not be overly surprising.
Bloodfang Clawtooth
Feb 16th, 2009
Congradulations to you all. I hope to see the same fine work in different lights.
Alazarin
Feb 16th, 2009
If LL were to reclassify bots as NPC’s then we could declare open season on bots and have some fun…. for a while until people get bored with orbiting bots.
Little Lost Linden
Feb 16th, 2009
Jahar Aabye. Do me this favor. The next time you have a few moments, visit a bunch of random locations of your choosing by looking for the busiest sims around your random location, and actually try to do a little bot hunting why don’t you?
Once you do, you will realize that in fact, 40 to 50% bots is probably much closer to the accurate truth than what the lab reports.
This is why we need a site dedicated to bot hunting. It would be great if we could run our own numbers since the Lab doesn’t have a clue how to do it, which means confirmation of each bot farm needs to be done manually by hand.
And I agree, it is not right to be locked out of the grid on busy days because of all the bots and the resources they use.
Somebody needs to come up with a bot filter to work in much the same way as a spam filter does. It would be nice to filter those bots out of the maps that is for sure.
Anya Ristow
Feb 17th, 2009
I disagree that bots help LL’s bottom line. Gaming search and filling the grid with crappy business models makes truly useful and interesting things more difficult to find, and makes the search for them surreal and just plain creepy. Second Life would probably be more successful as a social medium if it weren’t so full of creepy, stupid crap.
There’s a cost to being buried in noise. I propose that the cost is greater than the classifieds revenue, and even greater than the tier the hucksters are paying.
Galatea Gynoid
Feb 19th, 2009
“Jahar Aabye. Do me this favor. The next time you have a few moments, visit a bunch of random locations of your choosing by looking for the busiest sims around your random location, and actually try to do a little bot hunting why don’t you?
Once you do, you will realize that in fact, 40 to 50% bots is probably much closer to the accurate truth than what the lab reports.”
LLL: You’re demonstrating Jahar’s point. Setting aside for the moment that you used the word “random” twice to describe a non-random method when you require actual random selections for a valid study, the method you just described is guaranteed to result in a wildly unrepresentative sample of avatars on the Grid. It would be unsurprising for someone using this method to wildly overestimate the number of bots on the grid, assuming they understood survey methodology poorly enough to mistakenly think this is a valid way to conduct a study. Let me rephrase what you just asked Jahar to do:
“Next time you’re on the grid, select a location, pull up the map and look for the place nearby most likely to be using bots, then go to that location and count the number of real avs vs bots at this location.”
If your estimates of the number of bots out there are not overinflated by at least an order of magnitude, I’d be very surprised…
IntLibber Brautigan
Feb 19th, 2009
The obvious reply to claims that bots help LL’s bottom line is that a bot operator spends between 1000-3000 L$ per month per bot, at a minimum. If LL implemented some simple heuristics that logged off users more frequently who did not move, talk, walk about, use voice, or teleport to other sims, they would greatly reduce the number of bots out there by the simple expedient of requiring that bot makers make their bots more sophisticated, and by extension, more expensive to operate, which would of course result in less bot usage. It would encourage merchants to put more money into classified ads, which would of course increase LL revenues.
LL also needs to learn that the vast majority of their newer users are not significant economic players, even when they are not bots. Most (not all) newer users have significantly less disposable income than older users, whether those newer users are from domestic or new foreign markets. They cannot expect to continue in business while pissing off their most productive, active, and creative customers.
Little Lost Linden
Feb 20th, 2009
Galatea, you misquoted me. If you pull up the sl map and look for many green dots on the map and randomly tp to those locations, you will find that many times, the locations are filled with bots. It really does not get any simpler than that. When you look at any part of the sl map, you see many dots.
Many of those dots are bots. And in my travels, I would argue that 50% of the dots I see on the map, always turn out to be bots.
You however, will not notice the same thing until you actually attempt to locate the bots in the sims. Which, you either have not done, or for some reason, you like to be blinded by what the Lindens tell you about bot traffic.
Jahar Aabye
Feb 20th, 2009
LLL: We understand exactly what you are saying. What you are not understanding very well is how statistical sampling works, and how to form a proper statistical model.
Obviously any location with lots of bots will have lots of green dots. So if you teleport to a location with lots of green dots, there is clearly a higher chance of finding a bot farm than if you teleport to a location with only a few dots. Nobody is disagreeing with you over the actual question of whether teleporting to a high-traffic location is likely to result in finding bots (although it is important to note that there are many high-traffic locations that don’t use bots as well).
However, this doesn’t give you any real sort of statistical sampling, because you are getting a sample of only one specific situation: a region with a high number of avatars.
If you look at the map, these regions are not very common. Thus if you base your statistical sampling mainly on those regions with many avatars, you are badly skewing your data.
The regions with only a few avatars greatly outnumber the number of regions with many avatars. If you are looking to model statistically the spread of avatars on the grid, you would want to be careful to make sure that the sampling of regions that you visit is representative of the grid as a whole. A statistical sampling strategy that seeks out places where the variable that you are testing is most likely to be found is a poor sampling strategy for reasons so obvious that anyone should be able to grasp.
Here’s a better little experiment: Let’s work backwards from your assumption that the grid is 40-50% bots. Let’s say that most of these bots are found on these bot farms with large numbers of bots crammed together. There are currently 70,000 residents logged on. If 40-50% are bots, then this means that there are 28,000-35,000 bots online at the moment. That’s a lot of bots. Even with 100-av bot farms, you would still need 280-350 of these 100-person regions. Pull up the map, do you see 280-350 100-av regions?
Or let’s be more conservative, let’s say that these bot farmers use 50-person bot farms on their regions. Then we’re looking at 560-700 50-person regions. Again, this is far more than you see on the map. The green dots on the map are spread out in a pattern that does not suggest 28,000-35,000 bots on the grid.
Please, when you do finally graduate high school and go off to college, sign up for one of the basic stats courses.
Little Lost Linden
Feb 21st, 2009
Jahar,
If you pick a random location on the sl map, regardless of the number of dots you see (many, or even just a couple), and just go to any of the locations, you will start to see the bots in a vast majority of the places. Even if you do not look for the sims with the larger numbers, and just randomly pick any sims from anywhere on the map at any time of the day, you will start to find that a great percentage of those numbers are due to bot traffic.
You see, I spend quite a bit of time in sl (sad I know), and after a while, you can sort of determine a pattern. If every single time you go somewhere in sl, and you walk around and you notice more bots than humans everywhere you go, that has to tell you that the bot concentration is somewhat high. And yes, you could throw in your example of everywhere you go as being an unfair sample because I of course go to different places than others, however, the fact still remains that there are a great number of bots in many different locations.
And they aren’t always limited to the statistic of a sim with many dots, although, as you point out, a sim with many dots, has one of the highest concentrations of bots. This alone is a major, major problem of deception.
This problem alone causes many, many people to become frustrated with sl. Many, many complaints of sl are related to persons who just want to open up the sl map, dive right into a large group of real people and just start talking.
Well, unfortunately, almost every large group of people on the sl map, is more than likely going to be nothing more than a bot farm.
And we really aren’t even talking about the number of simowners who leave their avatars online 24/7. And those are the sims that might just have 1 dot on them, but many, many sim owners do this. Those are the sim owners who don’t take the time to create bots, but nevertheless, enjoy leaving their own avatar on their own land all the time.
And then, you have campers.
Campers, should truly be listed as bots, as campers are a form of a temporary bot.
When you leave your avatar behind on the sl grid, it shows up on the map and it inflates the traffic numbers.
This should not be tolerated, but at present, it is. You should not be able to inflate the traffic numbers in this manner. Camping should not be allowed, and between campers, and bot farms, and sim owners who leave their avatars behind on their own sims 24/7, you start to see that this extra traffic inflation really starts to make people wonder about the numbers on the grid at any given time.
You start to ask yourself, why is this deception tolerated?
You already know the lab is just a company making money and you know they are leaving the numbers inflated in these many different manners to make more money.
Unfortunately, the consumer in this case, is being spammed. If you could eliminate the extra traffic above that doesn’t belong in the game, you would see that the numbers are overly inflated. Until these problems are addressed, you will never get a fair number for traffic results, but honestly, what does your gut tell you on this one?
If your gut instinct tells you that the number of bots is insignificant, than so be it, that is your choice to make.
I choose to believe from my personal experiences on the grid that the number of bots is massive because it affects my gameplay every single minute I spend inworld. I can not go anywhere without coming across bots, or campers, or simowner left behind avatars. I see them far more than I see real people no matter where I go. This can NOT be a coincidence. It simply can not be.
The reason is bots, campers, and sim-owner left behind avatars.
All three of these items need to be removed from sl before we can get ANY sort of real traffic numbers in the game.
Until this happens, we will always have people doubting the very high numbers of inactive green dots on the maps, and we will always have users getting spammed by bots, campers, and sim-owner left behind avatars. (apologies for not remembering the term for sim-owner left behind avatars).
Lindens, please, please, fix the bot and camper problems with sl. We are tired of being lied to and we are sick of spam. thank you
Jahar Aabye
Feb 21st, 2009
I could point out that my personal experiences have not found very many bots, if any, in SL. But see, I know that personal experience means jack shit, because the plural of “anecdote” is not “data.” I don’t doubt that there are a lot of bots out there, but I have yet to see any credible evidence that the number is anywhere near 40-50%, and the so-called “studies” that have been “published” have been laughable. No p-values listed, which is understandable given that their methodology has been pretty bad. It generally only takes finding one bot farm to skew the study because of how they count avatars. The “studies” have either been set up by people with no understanding of statistical analysis (such as, no offense, yourself), or have been set up deliberately with the intention of producing an artificually inflated number of bots.
Show me the evidence, not your paranoid delusional fantasies.
LOL
Feb 21st, 2009
@Little Lost Linden
I asked you nicely not to fill this forum with the taint of “Linden Lab” now please Shut the Fuck up and shovel your Bullshit on the Linden Blog, XStreetLL or at your office hours where I am sure you can find avatars to kiss your ass so hard they can tell us all what you ate for lunch.
Little Lost Linden
Feb 21st, 2009
Indeed, this is why, as I have pointed out in the past, a worldwide bot hunting site needs to be established. Currently the only good bot hunting site that exists does not have English translation enabled. But let me ask you this. Have you even seen a bot farm yet?
Your comments lead me to believe that you haven’t really seen a 50+ person bot farm yet. If I were to post say…20 or 30 or 40 high population bot farm links, would that be considered proof for you that bots are a problem or not? Or would you just throw that into your statistics analyzer and say it doesn’t mean anything, and stick by your little measly 10-15% bot pop estimate, which in my opinion, is ridiculously low and uninformed.
Because, I have thought of doing just that, starting a site with worldwide translation for bot hunters with the help of this site:
http://www.slinfo.de/phpbb/viewtopic.php?t=16939&postdays=0&postorder=asc&start=240
If you can translate, perhaps you may get a better understanding on the issue. We need to have a site like that one with english translation. Once that occurs, then everyone can see the true bot numbers. Until that happens, we can only guess at the numbers.
Little Lost Linden
Feb 21st, 2009
LOL, this forum is for posting bot information, in fact, it is entitled “Pop Bots – Avatar of the Year – 2008″.
So any bot information that can be found, does in fact, belong right here, in this forum. Why would you think it would not belong in this forum? It would seem to me, that your antics do not belong in this forum, as you seem rather unprofessional to me. That’s just my opinion of you from your rancid postings.
Jahar Aabye
Feb 22nd, 2009
“Your comments lead me to believe that you haven’t really seen a 50+ person bot farm yet. If I were to post say…20 or 30 or 40 high population bot farm links, would that be considered proof for you that bots are a problem or not?”
Please see my earlier comment regarding the number of bot farms required to hold enough bots to comprise 40-50% of the online resident population. With 70,000 residents online, 40-50% would mean 28,000-35,000 bots. That’s 560-700 50-bot farms. So no, showing me 20 or 30 or 40 of those 50-bot farms wouldn’t really impress me much. It might imply that bots are a problem, but it would do nothing to actually demonstrate a 40-50% prevalence.
Are there really 560-700 50-bot farms on the grid? I have yet to see evidence that this is the case. You *really* need to learn more about statistics before you go about any project like this, because otherwise you will just be wasting your time….that is, if you really want to get an accurate count of the prevalence of bots on the grid. If, on the other hand, you simply wish to find some hyperinflated bot count to back up a pre-existing concern….for example, a way to write sensationalistic Herald stories, or a way to try to convince people to switch to an OpenSim-type grid in which you have a financial stake, then by all means, go ahead.
But if you actually want intelligent, rational people to take your 40-50% prevalence seriously, you might want to gather real data. You seem to believe this number, but you don’t seem to understand that many of us value data and evidence over belief.
Little Lost Linden
Feb 22nd, 2009
Oh the 45-50% is not my number, it is the number that many bot hunters believe, and I believe the other bot hunters know the numbers a little better than the lindens so I tend to believe the bot hunters more. I also have seen so many bots that I find it much more difficult to believe the linden numbers.
I do not have a financial stake (I wish I did), I am just a person who has seen so many sl bots that it would make your head spin, and I also am annoyed by the problem and wish to contribute to a solution to the problem. I feel that after you spend a certain amount of time in sl, you start to really get a sense of just how many bots are around you at any given time.
I am also not trying to sensationalize anything, I am just the type of person who receives a piece of e-mail spam and immediately contacts the ISP for resolution, as I believe e-mail spam is a very bad problem, and for sl, I believe the bots problem is also a very bad one that needs to be addressed. It’s really no big deal and I can’t understand why you seem so upset by this type of conversation.
I also do not think it is fair for a company to allow overinflated traffic numbers to go unchecked. It is just dishonest in my opinion and so I can completely understand when I meet people who feel that the bot problem needs to be addressed sooner than later.
Little Lost Linden
Mar 1st, 2009
For anyone who would truly like to start seeing sl bots, there is a simple way now.
Just open your Search window and type in Sex, Strippers, Whore, or just about anything related to sex because a majority of the bot farm owners have these key words listed. Soon you should start to see the bot farms by the hundreds. They are truly everywhere people. Look around.