Bot Count, The Estates Strike Back: 50%

The second of these three threads started to get very interesting with talk of defining what is counted as a bot, etc. It would be a pity if an organised sweep doesn't happen as a result of the threads. I was hoping it would result from the second thread.

Defining what is and isn't a bot is a problem, as we've seen from the other threads. We do know what Anya counts, and we also know that not everyone wholly agrees with her. The evidence for that is in the threads.

We also know that real-person avs camp, and they are not bots. I'm not going to show any evidence for that, other than Cristalle and some of her friends used to do it, as stated in the thread, Sling stated it, and I used to do it. I don't know if you accept that real-person avs still camp or not, but I do, and therefore, because of Anya's admitted counting of them, her figures must be inflated.

There's no camping system in evidence, but as I've said I don't make a distinction. I'm not out to ban bots, and I don't care if you're running a bot client. I'm looking for avatars that are on the map and in the concurrency statistics but not really *there*. If you are sleeping IRL you aren't really in SL. I'd count you as a bot and I'd be happy with that.

Well, this is all very messy. TBH, I don't really give a damn how many bots are in-world, but I'm very interested in what the concurrency numbers would be if we managed to finally kill off Traffic as any input to Search ranking. So, for my parochial purposes, campers are very definitely the same as bots--but not all bots are the same... I'd really only be interested in those that inflate Traffic, so mostly traffic-/camperbots. I'd like to exclude land searchbots if I happened to stumble on them. Mannequins and greeter bots--hard to tell; they're probably about half Traffic-motivated, I suppose. Same with group-invite bots, inasmuch as two could service the entire grid, so they're likely often traffic-motivated, too. But anyway, that's just my agenda, not necessarily anybody else's. And I don't know how to efficiently collect data to get a better approximation of that, anyway.

But I have a feeling *somebody* is going to have to replicate some part of Anya's results somewhere, just to be a bit scientific about it.

So far, I really can't tell what Elanthius is actually measuring. I'm reasonably okay with the "stationary agent" approach, but at the moment, that raw data shows no bots above 112m, and we just know that finding 113 sims without a single botbox at high altitude would be extraordinarily unlikely.

In that case, you shouldn't be writing thread titles about "Bots count: 52%" and stating things like "52% bots overall" because many people could actually believe it. Now you're saying that you're not counting bots specifically.

You may be, but we would differ. There is a difference between a throwaway bot account and an account that belongs to a real person, that is actually social for large expanses of time and for whom making a few cheap L overnight is a side goal. If this is just a measure of inactivity, what is the point of calling them bots, which is by any conscious forum user's standards an incendiary term?

By framing these people as "bots" you create a false assumption that is rightfully challenged. If you want to call them stationary avatars, you can then make a projection about SL's social activity that doesn't smear people with presumptions arising from the improper use of incendiary terminology.

Is SL a ghost town? Yes. It has been, for years, and has only gotten worse as the landmass has expanded. This isn't anything new. Camping has also been around for years. If you are trying to capture a specific phenomenon, you're using too broad a brush to define it.

Elanthius, what is your bot doing? Is it just transporting an LSL scanner or is it itself doing the scanning?

It occurred to me that your owning 1000 parcels to scan from eliminates two problems with surveying: getting stuck in a building at the landing point (a bot would have a hard time with that) and being ejected by security orbs.

If your bot is scanning, there are at least three types of avatar data available to you. You can get the map data, which only has X/Y coordinates and is only good to within 4m, and you can get this data remotely. You can get mini-map data, which isn't limited to sim boundaries but is limited to 1m accuracy X/Y and 4m accuracy Z, and you can only get it locally. And because it uses 8-bit integers (times 4 meters) it tells you avatars over 1024m are at 0m. Third, you can get accurate coordinates of all avatars within your draw distance.

Do you use any of that info in your scans?

I don't care either, but I *am* curious. One thing that occured to me is that, if the actual number of traffic bots and campers is quite low - say 20% at low concurrently - then LL might be more inclined to do away with traffic rankings, but if it is high - say 90% at low concurrency - they are likely to want to keep the numbers. So, in that sense, lower could be better for getting rid of traffic bots and camping. If we could actually get the numbers, and they are low, then we might have a good chance of persuading LL to get rid of traffic rankings.

They can't or won't get rid of traffic rankings until better mapping functions can be integrated into the new search. Many still do search based on the number of green dots they see on the map because they are looking for places with life, not gamed picks. But of course this is a seperate topic.


I do have to ask the OP - What exactly is the point to knowing how many bots or campers make up the concurrency numbers without there being some sort of agenda or motive behind it?
It seems to me these numbers would be forever changing, and even if you could accuratly count who is a bot and who is a camper, you would need to do this survey many times to create some sort of average.

My thought, too, although I'm not sure the low-concurrency share matters as much as contribution to total user hours and the high-concurrency metric, both of which I believe LL hypes.

IMO almost all campers are bots and some of them are the worse kind of bots because they are using the standard viewer and all the resources that it sucks up rather than a slimmed down bot program. Even if the person is camping for only 90 minutes, they are a bot during that period if they are not interacting with the rest of the virtual world. If someone is AFK camping with their primary AV they are still a bot, IMO. I don't see a reason to distinguish between bot and camper, especially for this survey. The OP has stated numerous times what her agenda is. She was feeling like SL is becoming less social and she wanted to understand why.

Even if you lumped them all into the bot category without making any distinction, that number would still be in flux and would require an average to even be close to being accurate. Doing it a couple times is meaningless.

The term bot becomes meaningless if you expand it that way. Campers existed long before bots did, and to throw camping noobs in with bots is a distortion of the activities, especially since a camping noob IS going to be social at some point in time, for an extended length of time.

Some history...

I've felt for a long time that SL is getting less and less social. Then I had an evening that went like this:

http://greendots.typepad.com/green_dots/2008/07/automated-club.html

I had known about traffic bots and campers, but that someone might use them to simulate club traffic and think that'd actually make the place more compelling just seemed kinda dumb, and that evening in that club, which I used to enjoy, was just really creepy.

So I started making random visits all over the grid, and writing about what I found:

http://greendots.typepad.com/green_dots/about_this_blog/

I'm finding creepiness everywhere I go. But creepiness and "SL feels like a ghost town" aren't very objective.

After making random visits for a while I posted in some bot threads that I thought there were perhaps 40% bots on the grid, and nobody believed me. Of course they might have been right and my gut instinct may have been way off and my experience may have been atypical. So, this current survey. More to convince myself I wasn't imagining it than to convince anyone else.

I've read that view a number of times in this forum, and for you and those who hold it, it's true. But for me it's totally false. In the early times of camping, campers were only people using viewers, whether or not they were sitting at the keyboard while they were camping. Nobody called them bots, because they weren't bots. The word "bots" came about when automated avs were introduced (i.e. "robots" and, to my way of thinking, it should only be used for automated avs. It's one thing to lump real people with bots because of what they are doing at a particular time, but it's another thing to say that they *are* bots during those times. They are not.

I understand now. I also understand how some places, especially clubs can feel that way. My own included, and I know for a fact I do not use bots to fill the place. Sometimes people are just quiet, engaged in IMs with others, tune in to the stream and afk or work in another program. Too many reasons why a busy place can be so anti-social. Just depends.

Oh noess... I'm a bot. That "box" over their shop is most likely their workspace... I don't really think you can count a designer who has their workspace over their shop as a bot. We may not be moving around a lot, but we're working and doing customer service. And yes, I've been on for hours at a time standing in the same skybox... it's where I usually am when logged in. I can tell you from the chatter going on in the Designers in Seclusion group IM's that there is definitely socializing going on at the same time!

What is meaningless is your statement that campers existed long before bots did. If it is even true... what does it matter? It is still happening now and misleading people into thinking that there are more "real people" online and at a certain place than there actually are. Yes some bots may be social or actively doing something at other times... then and only then should they be counted as a person logged into SL.

They have automated their avatar by overiding the default setting which causes you to go (away) when you are AFK. Essentially they are using a very poor version of a bot program. So they are a bot.

Yeah, I know that. I have one, too.

I wouldn't even consider an owner in their store or in their workspace unless their are other indicators. Like other bots. And even then you'd have to not move at all for ten hours before I considered you a bot.

Sometimes I wont log out of SL for days on end, not to generate traffic, traffic count is irrelivant as far as Iam concerned and I cant remember the last time i checked my numbers. My PC is on and allways downloading so I stayed logged into SL and any other online application I happen to be running. So If I go off to bed for 6 hours and off to work for 10 hours and remain logged in without moving am I a bot?

No. That doesn't hold water. There's nothing automated about turning a setting on or off. It no more makes an av automated than setting the time to midday. If anything, it's the other way round - having it set to "automatically" log the av out.

There is no "if it is even true" about it. It's definitely true. And the campers were people, and not bots - just like many campers today.

If it walks like a duck...

I'm looking for avatars that are on the map and in the concurrency numbers but not actually *there*. So I'd count you as a bot and I'd be happy with that.

Of course you'd also have to be on a property that'd benefit from the traffic, and have other bots, or a pay-for-picks sign, or some other indicator that you're gaming search before I'd even notice you.

And if I get it wrong, so what? Nobody dies. And these carefully-constructed edge cases don't change the numbers at all, even in the unlikely event your sim is surveyed.

this comment rubs me the wrong way....

A lot of folks log in and take care of other things, be it building, or working on textures or whatnot

it doesn't mean they are part of any problem or mouthing off on the forums

Phil is doing the same thing everyone else here is doing, putting his thoughts and opinions out there for ppl to see.

If he is a problem then so is everyone else contributing to this thread (including me)

(also if he is a problem because he is logged in and not responding to ppl, then so am I, as I tend to do that myself)