bot count: 44%

Fascinating data, Anya. Thanks for putting in the effort. I would suggest that those who are discarding your data as meaningless try collecting some of their own across multiple sims instead of just throwing out an example about one sim they happened to be in once as proof that you're wrong. Although you are using a small sample, it's a larger sample than I've seen anyone else use so far.

Your survey was on mainland, which I find interesting. One would think that since mainland land owners have to share the sim with others, they would be less likely to fill it up with bots since they can be AR'd for misuse of sim resources. However, it is probably the best way to get a balanced sample, since mainland is a mix of commercial and residential properties. My main store and residence are on the Corsica continent and this appears to be the case to me from my wanderings.

I wonder if the percentage of bots is higher on commercial sims that are owned by one person. This would probably be more difficult to survey, though, since it would require a lot of hopping around the map.

I see we have had this discussion before And I see that I made the same points before.

In the post where you explain why you think the number of bots varies during the day, it's all about you think this and you think that - guesses not data. It doesn't mean that you are wrong, but it doesn't mean that you right either. Of course I don't know whether your guesses are right or wrong. In fact, my guess is that you are right but in a smaller way than you think, which is why I have said things like "generally" and "pretty much all" rather than "all".

It's my "guess" that bot runners in general run them 24/7, as I do, because they are there to raise the traffic figure, and there's no reason not to raise it 24/7. Many campers probably turn their machines off at night, but others take their places.

However, in this thread, you haven't stated a range of percentages to represent a 24 hour period - you simply stated "bot count: 44%". The statement is true (according to your bot criteria) for what you found in a tiny number of sims during a small period of one day, but that's all. It cannot be considered true for the grid in general. It certainly isn't true for the grid in general.

For comparison, Zee sort of explains LL's bot counting method in a blog comment:
http://blog.secondlife.com/2008/07/08/second-life-virtual-world-expands-35-in-q2/#comment-614159



That will make any comparison tough, since it uses numbers that aren't really available to the public, and it's not mentioned how far from typical an avatar has to be before it's counted as a bot.

Anya you said this is an indication of the direction in which SL is moving...Im not sure i understand that. Can you explain?

Well you certainly put in the time and effort and it makes for an interesting read.For me it confirms what I think of the logged in figures. When I log in ,I look at the number and halve them.I really believe the padding is that high.But anyway nice job.

I *am* actually a parapsychologist and my extensive powers enable me to confirm that, whatever the concurrency, between 43.9% and 44.1% of all logged in avatars were bots at any one time in the last six months.

Pep (I'll let you all know if that figure changes significantly)

Sorry I can't accept the 44% figure.

Did you choose sims that deliberately had a lot of people in them? What about noobs who were afk?

I tend to agree with Phil, your sample is too small to be of relevance. In fact the only way to do this accurately would be to use a bot

Hear hear. People from all over these forums are complaining about the number of bots and LL's honesty or capability of defining the number. Now Anya has tried to come up with at least some data, representative or not, but the effort is not appreciated. As if she's doing some harm. I don't get that.

mainland, sunday, 60000+ logged in.
19 sims. I thru a dart and then took 19 in a row, going west
43 avatars
0 bots

Actually I appreciate Anya's efforts, but the only way to get the information to LL and get something done about the problem is to get accurate data. This means transparency so that the criteria for determining bots is agreed by both residents and LL, and a relevant sample size (10,000 sims + of data from both mainland and estate) is collected over a 24 hour period on both a weekday and a weekend day.

As well as bots, there's all those people logging in more than one alt for whetever reason.

This part confuses me. No matter the percentage, I agree there are likely more bots (though fewer campers) than when I signed up, but what does that have to do with how social the grid is?

These days it is much more difficult. Originally, the bots only wore default avatars, skins, and clothing of newbies, but now many of them have graduated to other skins and clothes.

The biggest difference is that if the bots are being used to increase the traffic numbers, then they will quite often be all clustered together in a box or on a platform very very high up above the area that wants to imply that they have traffic. Even in this case, there is no way to know for sure - it is just very very likely.

At an actual camping spot, I am not really sure how to tell the difference.

If a bot or two are being used in a store to model, you can sometimes tell by their profile or their group tag.



Well....... we already know that is true.





The majority of us already knew that -- LL would never be truthful about that because it would mess up the PR about the number of people that are involved in SL.




Definitely - especially since we already know that 87% of all statistics are made up on the spot.

Regardless of any criticism, your effort to quantify this is the most comprehensive and detailed analysis that anyone has undertaken to date. So, I thank you for your efforts and for sharing it with us.

I don't think anyone has the right to criticize your effort until they can produce a better analysis to refute it.

I guess the big question for me is, how do you decide what is a bot?

Yes, some are obvious. That group of 20 avatars stuffed into a box at 2000m are probably not just afk people. But, I go to a club, start my AV dancing and open up the forums or start sorting my inventory or open up a script I'm working on and have a browser window with the LSL wiki covering my SL window. Someone wanders by and says hello, I don't see it and do not respond. So, now I'm a bot?

Usually I set Busy when I do that but many times I don't. So, how do you know it is a bot?

LOL

Okay, here we go: I'll spank that ass and sing songs for L$5000





next.....

we shall agree to disagree then

It is not enough for me

I don't see how bots cause the grid to be less social. That somehow implies that each bots removes a real person taking away someone that could be social. Bots don't remove people they just add fake concurrency numbers. You will still have the same number of real people to socialize with bots or no bots.

I would, however I have no sure fire way to correctly identify if an avatar is a person AFK or a true "bot"

also if it is a bot, what kind of bot is it, what is it being used for, as those things do make a difference to me

are we talking strictly traffic bots, or all bots?

I would not attempt such a task as I know that my results would be misleading and false as I have no way to confirm anything.

I'm not decrying her efforts. I'm simply pointing out that the results are meaningless if they are applied to the grid as a whole, which is what the original post appeared to do. Whether or not they do any harm is debatable. If the post causes people to believe that 44% of logged in avs at any time are bots, then it's a false rumour, which could be considered harmful.

these numbers could be way off..
there was a time i thought my partner to be had turned into a bot..when i would talk he would only answer with..yes dear no dear..the sex was like i was by myself and he camped in front of the tv for hours watching SL hockey with 9 of his best friends on a daily basis..

Sorry, Phil baby, but it's not meaningless. Anya's survey has enough data points to be statistically significant...about 1% of the sims in SL. This survey is a LOT more reliable than that "10%" figure Linden Lab puts out.

And it's scary. Nearly half of us are bots? God help the grid.

As you point out, bots are on 24/7. The survey was done at peak concurrency, so what this means is that at times of low concurrency, SL consists ALMOST ENTIRELY of bots. The real people only come on on the weekends. Except for us diehard addicts.

I can tell you one way. Land.

Linden Lab bases the creation of land on the number of SL users. But bots don't buy or use land. If half of us are bots, then the land area of SL is twice as large as it needs to be to support the population. No wonder land prices are so low. No wonder newbies ask me "where is everybody"?

We're spread too thin...and that affects how "social" the grid is.

I tend to side with Phil on this one, as I feel her numbers are skewed as well.

I am not saying LL's numbers are any better.

I just do not understand why the OP was even made, there is no solid proof that they were bots, it is just one person's viewpoint and opinion, put out there as though it were fact

That is the part that kinda bugs me.

but...meh, I don't care how many of the dots are bots, they are a part of SL and most likely always will be

I still see real ppl, I still sell items, so ... there are still people out there