LL - Let us count the bots!

Why?

O.k. Some of us live here, as in it is our virtual escape and a place we care about and even love. The numbers *are not irrelevant*. I am sure that there are those, like me, who first saw the numbers are thought that represented souls - a direct correlation to the possible social dynamic that could be on going with 50,000-60,000 real people online. In December 2007, i had no idea what a bot was and had not even heard of them. So i thought the numbers were actual people.

And the numbers are not just the business of LL. the numbers also reflect a pool of virtual customers, participants, or possible user base that any RL business or SL based business may wish to cater to, and when you are pitching someone a business proposal in RL to get $35k usd for your SL startup, those concurrency numbers matter because people = business, bots do not, and 60,000 people looks more lucrative than 38,000 people and 22,000 bots.

So yes, there are reasons to care about concurrency and there are reasons to care about the high numbers.

Manequins, models and group inviters spring readily to mind.

I'm not a clothes shopper but, from what I read in this forum, people find it very useful to see the clothes on avatars before buying, instead of only a picture. Same with hair, I understand.

There is even more benefit in bots when they are used as models for animated furniture. For instance, I have 6 of them in my store. Four of them are for demonstrating the bed animations for people who are there alone, or who don't want to hop on sex animations in a public place. There's a couple of models where people can operate the bed's menu to see the animations, and there are two other beds, each with a single model on - a male on one and a female on another. People who are there alone can use them to try the animations for themselves. They are all behind a wall so, although it isn't done in private, it is done out of plain sight. I also have a couple on a cuddle sofa, again so that people who are there alone can see the cuddles. All of them are there to demo the animations before deciding whether or not to buy.

The user experience would be lessened without such manequins and models.

You can't pick the high point and the low point and call them averages.

http://taterunino.net/statistical%20graphs.html



I started by stating I don't have enough data to make firm conclusions. Here's what I said...



That's a lot of wiggle room, but my observations (and they are *observations* not *guesses*) indicate there are a lot more bots than LL acknowledges.



Until someone else puts in the same effort I have in actual observation I suggest my observation is better informed than yours.

You don't need to limit it to premiums and concierge only, payment info will do fine as well (although using those criteria plus "LindeX activity in the past X days" would probably yield a large enough sample), just as long as you count per unique RL identity. This gets you rid of most bots.

If you stop there you just shift back from bots to camping though so you need to additionally cap the amount contributed per parcel and per day. If you want to keep traffic as-is, say 120 (2 hours) per parcel and 300 (5 hours) per day.

---

You could still run payment verified bots, but since they're all under the same name the most you'd get out of it is 120/day (as opposed to 1440/avie/day now).

You could entice people to stay and camp but each of them would again also only contribute 120 at most and if they're logged on under an account at the same time the camping alt may not even count at all that day. And if they're willing to camp at your place they're likely camping elsewhere as well and they could have already exceeded their daily total and you could again pay for no increase at all.

Search might become a whole lot more volatile with ranking changing day by day (a new release/review/mention in/of a semi-'popular' store could make it highly 'popular' by the next day) but that might not be a bad thing either. And payment verified account are a large portion of all established accounts so I'd have to think that they're representative enough for all of SL so there's no problem there either.

I'm not sure that there is an significant anti-bot sentiment per se.
In "these threads" we are talking about specific uses of bots, and not about bots in general.

Personally, I'd love to make some time to start working on a bot.
Right now I'm doing a second-best by moving my alt around using a scripted attachment while I test the effects of various weapons on its combat meter.
I'm sure she's love to be able to fight back.

The idea of a bot that gives guided tours (mentioned in another thread), or gives performances would be very interesting too.
There must be a zillion creative ways of using bots.

It's what LL now term "disruptive bots" that are the problem.

I am a bot user. We use an invite bot to control our update group. But now things are going to change.

I just want the concurrency numbers to show who is really alive online - it is our world, our social experience, let us get an idea of what is happening so we can decide for ourselves how much to continue investing here emotionally, socially, financially, all of the above, etc.. If our world is dying a slow death, i want to know. I just don't want some idiotic blog post to appear one day telling us some horrible news 96% of us did not expect that involves declining usage or sign-up numbers.

This is a legitimate use of bots in my opinion.

None of that is any reason for any of us to be interested in the concurrency numbers. When I joined in Decemeber 2006, the concurrency never reached 30,000, and yet there were loads of people to socialise with. Even if there are still only 30,000 real people, there are still loads to socialise with.

If someone is tempted to join SL for the purpose of making money, and they would be put off if there really isn't 70k people in, then they should do a bit more research before starting out on the venture, or at least before sinking a significant amount of money into it.

Whichever way you look at it, the concurrency numbers don't affect us in any way, and they should be of no real interest to us.

I didn't say anything about averages. I asked 40% of what - twice before this post, and I still haven't received an answer. Is it 40% of the average (something that you didn't mention), or 40% of the high 70k concurrency, or 40% of the low 30k concurrency? If you meant the average concurrency, you didn't make it clear.

If you meant the average, then what is the average, because halfway between the high and the low isn't the average for the day.

When I get around to scripting a bot, I won't have any hesitation in floating "BOT" over its head. I would flag it as a bot to the system if there were some way of doing that so that concurrency or any metrics on 'live' activity isn't affected by its presence.

Perhaps if there were such a flag/declaration then LL could outlaw the running of a bot without that flag being set.

A group invite bot is about as "useful" as landmark givers in my opinion. If you want to spare people the effort of having to search for a group the web link to them is more than good enough.
There's a good reason why there is no llSendGroupInvite and every time I get any automated group invite it's AR'ed as the spam it is.

Mannequins and models are just dressed up bots. I fail to see *any* use in them other than to stand there and look pretty.

Unless the bot is constantly changing outfits (so much for the "no grid impact" in that case) it'll be wearing one single outfit out of dozens. That's only "useful" if that happens to be the one you're thinking about buying.

I couldn't care less what some outfit looks like on a bot, it'll have the perfect shape the creator used to make the prims in the first place (tells you nothing about fit), I care about how it looks like on me and that's why you offer *demos*.

Give anyone the choice between being able to buy a demo or see something demo'ed on a bot and I don't think the bot would get all that many votes. It's just not useful.

When you check the stats on your website, are you more interested in how many page hits you've gotten or how many unique visitors you've gotten?

You should be interested in both, don't ya think?

If I am going to be given information to use to decide if I want to invest money into this platform, why am I only given the artifically inflated concurrancy numbers and not the actual number of unique users?

What I think most people object to is the lie inherant in the concurrancy number. It is meaningless and inaccurate to anyone who wants to know how many HUMAN BEINGS there are on the SL platform. To some people, that number is entirely relevant and of great interest, especially in a social network.

Banning bots is unecessary. Bots are tools and tools have their uses.

Keep your faked concurrancy number on the front page. Then include the number of unique users. That way we all have better and more accurate information.

What's so hard about that?

I'm only repeating what I've seen people write in the forum about manequins. I don't shop for clothes or hair myself. It seems your view is different to those, and I'm not judging.

I hadn't thought of using a web link to facilitate group invites. I use a notecrad with instructions, and people only receive it if they click the sign to get it. But I understand, again from this forum, that people do use group-invite bots and, just because there's another way, doesn't mean that the bot way isn't useful. I'm sure that the bot method is by far the best way.

I didnt say they were right or wrong trouble is anyone can make figures do what they want fro themselves

Until I/we get proof i wont believe either side of the camps

Sorry. I believe the number of bots is about 50% of the daily average minimum concurrency (of about 40K) and about 40% of the daily average high concurrency (I rounded to 65K but it's about 67K, and I also rounded the 40%--by the numbers it's about 38% but there's a lot of uncertainty so I rounded). I believe bot usage does vary, for the reasons I stated, but varies far less than human usage, so you get a changing percentage during the day.

In that case, I don't disagree with you - I don't agree either, because I smply don't know.

Not long ago, it occured to me, since bots are 24/7 (generally speaking), and the low concurrency that I was seeing every day was just over 30k, that the number of bots is likely to be a significant percentage of that 30k low.

secondlife:///app/group/UUID-here/about will be a clickable link in chat and pops up the group window (which has the join button if you're not already a member).

Well, the invite bot actually has the name: NBGroupInviteBot

I think it is a good idea too. However the furniture models just have regular alt names with a "N&B Model" group tag.

A bot can be doing multiple things. The mannequin bot can also be the group invite bot and simultaneously be doing other things for its owner. My bot checks the map for traffic info for a project I haven't announced. If I need a mannequin or a group invite bot I don't have to increase the bot population.

Thank you. I'll change my notecard method.

Wrong in so many ways.

First and foremost you should not be speaking for "us" in respect to telling "us" what we need to be interested in. It may not interest YOU. it may not apply to how YOU do your business. But for "some" of "us" it does.

Great idea and please let me know as i will be happy to test them on our sims for you However we do make it clear the ones we use are BOTs, but I would love to Beta test any you make and thats a promise

Mine didn't have a tag until the other day. Now they have "Demo Model" for the tag. Before that, people sometimes walked into where the bed models are and walked straight out again - even saying "sorry" occasionally lol. There are signs but not all signs get read, so I added the tag a couple of days ago.

Alright. I'll rephrase it to...

I don't see any reason why any of us should be interested in the concurrency figures because, as far as I can tell, they have no effect on any of us.

I don't know if she's still doing it, but I assume she is...

Go take a look at Cherry Tokyo's store. There's a bot modeling one of the Kimonos. Arguably only useful for that one Kimono, but it also shows the quality of her work, and makes a more interesting display than a picture on the wall. Seeing it in the flesh is more compelling than seeing it on a picture, and even if she were to provide demos I'd bet more people would request them than if there weren't a mannequin.

On the other hand I've seen mannequins modeling truly awful non-prim outfits.