Jump to content

Are Bots/Fake accounts a ligitimate question?


You are about to reply to a thread that has been inactive for 693 days.

Please take a moment to consider if this thread is worth bumping.

Recommended Posts

6 hours ago, Coffee Pancake said:

Elon is just trying to skip out on the deal because .. shock surprise .. he was never actually going to buy bird site and the whole adventure has just been yet another 100% on brand stock manipulation

He's a troll/influencer/manipulator like that.

In the crypto world, he did the same thing with BTC, C*MMIES and SHIB.

  • Like 2
Link to comment
Share on other sites

23 hours ago, Wallace Wirefly said:

 As one wanders about Second life, which by the way is the largest area in multi-verse, one notices "avatars" just standing around in various poses. Also when one logs in, as everyone needs to do assuming one logs out, you see the statistics of how many are "in-world". How many are really "in-world"? Are there ones who never log out? ( of course except for the dreaded sim restarts). But then there are those who will just be sent to another place to stand. Are there "avatars" created for just that purpose? Spawned every so often to keep the count of those "in-world" a healthy level?  Will the real data come forth? So what is the "real" in-world count for Second Life and the percentage of these "zombies"? Does it matter to the business?.... Will the statistics of Second Life ever appear again? 🤷‍♂️

Don't know about Twitter and it's bot problems but one thing I have always thought was a little strange about the S/L concurrency statistics is that the Minimum Daily Concurrency always seems to hover around the 30K mark +/- about 2k either way. It has been around that mark for well over a decade and with the exception of blips either way, is remarkably consistent without showing any real seasonal changes or general trends up or down. This in spite of there being trend increases and decreases in the Maximum Daily Concurrency. I could understand if the numbers were being inflated somewhat to show such a consistency being that if there was general trend downwards, it likely would result in an economic downturn for S/L as residents lose confidence in its continuance. So certainly S/L could have a similar motivation as Twitter in using bots to offset any visible signs of a downward trend.

  • Haha 1
Link to comment
Share on other sites

18 hours ago, Rowan Amore said:

Personally, I would but everyone has said traffic isn't calculated as it used to be. 

Maybe nobody told the owner. It can be amazing what people who should know often don't know.

In any case, I still quite often encounter what I'd call 'traffic bots', too (there may be other explanations for those bot meetings, like they are rogue AIs, taking enjoyment in plotting virtual world domination "in person", but the only one that comes to my mind without much thinking or fantasizing, when I see those conglomerations, is 'traffic bots', in any case).

Link to comment
Share on other sites

18 hours ago, Sid Nagy said:

I think that the term traffic bots is spot on, when SL is concerned.
Generate attention bots is a bit long isn't it?

Maybe "traffic zombies" is more appropriate. You don't really need a bot program to control an avatar that is just standing idle so unless the owners are really stupid (which they may well be) they probably just create a bunch of alts, log them on and go afk.

  • Thanks 1
Link to comment
Share on other sites

19 hours ago, Love Zhaoying said:

Would you classify those as "traffic bots"? 

That was always the term used for them, although they aren't bots according to LL's description of what bots are.
 

18 hours ago, Rowan Amore said:

Personally, I would but everyone has said traffic isn't calculated as it used to be.

I've never seen anything to suggest that it's changed. Originally, the SL results were ranked solely on traffic. The more traffic a place had, the higher it was in the search results. Many years ago, LL replaced that system with a rented Google search appliance (GSA), but they continued to count traffic in the same way that they'd always done. However, instead of traffic being the sole factor in the rankings, it was accounted for in the GSA as a small factor. People got the idea that it became such a small factor that gaming the results with traffic bots was hardly worthwhile, which, on the whole, was true, although in some circumstances, it could be very effective. That change is why many people wrongly thought that traffic was no longer effective, or wasn't even used.

The GSA was expensive, and LL rented 3 of them, so, after a while, it was replaced by an Apache freebie search system. I've never seen anything about whether or not they incorporated traffic as a factor in the results, but the method they used to do it with the GSA was so simple that there would be no reason not to continue it with the Apache system. That's assuming that the Apache system was a copy of the Google system that was published online, and that all the engines copied.

Whether they include it or not is irrelevant though, because, for a completely different reason, traffic is still very effective in the search rankings, which is why traffic bots are still around. The reason is that most SL users use non-LL viewers, which include the traffic-only legacy search. My understanding from reading in the forum is that the legacy search is widely used. I certainly use it in preference over the web-based search because that one often takes forever to return results. So using bots to manipulate traffic and attract visitors is still effective.

Edited by Phil Deakins
  • Like 2
  • Thanks 1
Link to comment
Share on other sites

17 minutes ago, Love Zhaoying said:

My conclusion from your information is, keep an attic full of Zombie dolls!

It depends on what you want to use the dolls for lol.

LL's description of what a bot is does not describe the usual traffic bots, because they are not controlled by a programme to perform tasks. BUT LL came up with the Scripted Agent status specifically to get rid of traffic manipulation, so I don't think the fact that normal traffic bots are not bots will persuade LL to allow them in the case you described :)

 

Edited by Phil Deakins
Link to comment
Share on other sites

4 minutes ago, Phil Deakins said:

It depends on what you want to use the dolls for lol.

LL's description of what a bot is does not describe the usual traffic bots, because they are not controlled by a programme to perform tasks. BUT LL came up with the Scripted Agent status specifically to get rid of traffic manipulation, so I don't think the fact that normal traffic bots are not bots will persuade LL to allow them in the case you described :)

 

Ok, so if I understand inactive / unattended avatars that are:

Scripted agent = good

Not scripted agent = bad

 

Link to comment
Share on other sites

5 minutes ago, Love Zhaoying said:

Ok, so if I understand inactive / unattended avatars that are:

Scripted agent = good

Not scripted agent = bad

The only complication that comes to mind: a bot not registered as a scripted agent wouldn't be a problem on a parcel not set to show in search.

  • Like 2
  • Thanks 2
Link to comment
Share on other sites

11 minutes ago, Love Zhaoying said:

Ok, so if I understand inactive / unattended avatars that are:

Scripted agent = good

Not scripted agent = bad

 

That's usually correct when the unattended avatar is not just an afk one. E.g. I am logged in for hours each day and afk most of the time. I don't need to register myself as a Scripted Agent.

The 'scripted agent' term means that the agent is being controlled by a programme (script) and not by a human. Even then, LL's description of a bot states that the agent is scripted/programmed to perform tasks, whereas being logged in but being afk, I am not programmed to perform tasks, so I am not a bot and I don't need to register myself as such. That applies regardless of how many avatars I have logged in, and regardless of what programme I use to log them in. Sometimes I have a fair number of them.

Edited by Phil Deakins
Link to comment
Share on other sites

1 hour ago, Meccha Suki said:

In any case, I still quite often encounter what I'd call 'traffic bots', too (there may be other explanations for those bot meetings, like they are rogue AIs, taking enjoyment in plotting virtual world domination "in person", 

Actually you might not be far off the mark with "world domination".  Many of those groups of bots, especially when you see them on Linden land (usually at the bottom of the sea) are playing Tiny Empires, a hud game which is exactly about world domination.

This all makes me wonder what in-world concurrency would look like if there were no bots and no long-term AFK avatars in SL. What if everyone had to complete a captcha to log in, and another every few hours to STAY logged in. I have a hunch that it would be nearer to 20,000 than 60,000.

  • Like 3
  • Haha 1
Link to comment
Share on other sites

9 hours ago, Arielle Popstar said:

Don't know about Twitter and it's bot problems but one thing I have always thought was a little strange about the S/L concurrency statistics is that the Minimum Daily Concurrency always seems to hover around the 30K mark +/- about 2k either way.

If you look at the concurrency graph for any platform or service with hourly figures you will see the same peaks and troughs sat on top of a volume that never changes.

We don't get all the numbers, just aggregate ones. The 30K minimum aren't the same 30K from one second to the next, there is a huge churn of accounts coming and going, just after you work out the difference, that's what's left.

The low concurrency doesn't raise due to timezones .. Homework question ... where in the world do peak home entertainment hours and SL population lows overlap?

 

  • Like 2
Link to comment
Share on other sites

4 hours ago, Coffee Pancake said:

Homework question ... where in the world do peak home entertainment hours and SL population lows overlap?

I'm not sure but I do know where peak home entertainment hours and SL population highs overlap:

image.png.e227d103fb9a8106ed47b938de502186.png

  • Like 1
  • Haha 3
Link to comment
Share on other sites

On 6/9/2022 at 3:31 AM, Love Zhaoying said:

I was kind of riffing on the news that Twitter deleted a massive number of "fake accounts" recently. Whatever "fake accounts" are.

hmmm. they just searched the word fake and deleted all those accounts. I saw something on tv, a docu, about dangers of AI and ad buyers who can pretty much control the world via ad buying what if that is paired with bots. John Doe goes to twitter I want you to run this political ad and I also want to purchase 200,000 bot accounts to push the products to these markets. So are they deleting those? If so would that be deleting evidence? 

Link to comment
Share on other sites

1 hour ago, Paulsian said:

hmmm. they just searched the word fake and deleted all those accounts. I saw something on tv, a docu, about dangers of AI and ad buyers who can pretty much control the world via ad buying what if that is paired with bots. John Doe goes to twitter I want you to run this political ad and I also want to purchase 200,000 bot accounts to push the products to these markets. So are they deleting those? If so would that be deleting evidence? 

Twitter routinely has "bot" purges. They aren't for the most part automated accounts, so the term "bot" is misleading.

Typically there are companies that exist to deliver engagement, picture one person sat in front of racks and racks of phones paid to do whatever is making them money today.

This practice is also commonly used on Twitter to create the illusion of grass root public political reaction, and groups running these schemes are well organized, in some cases state funded and active outside of domestic territories via VPN services. Thousands upon thousands of accounts pushing a daily geopolitical agenda that are deliberately set up and operated to look like real people who are suddenly very concerned about some very specific issue or another.

MAGA was heavily "botted" in this fashion on Twitter and other social media platforms. The running joke was "its the Russians", mainly because the bulk of the people working the "bot" accounts kept Russian office hours (and from time to time they would slip up and leak their real location). You could tell when the manipulation on reddit started like clockwork every day.

This is also why things like Bengasi or HunterBiden will spontaneously trend in the middle of the night when regular Americans are asleep, the goal is to have them in the trending feed first thing in the morning US time, so local political teams can pick up the baton and try and make it the days trend. Twitter sometimes manages to kill this, sometimes the subject matter is close enough to actual domestic political commentary they can't tell the difference and it's allowed to stand.

This same tactic has been used by other special interest groups trying to push political agendas on Twitter. British anti transgender / anti LGBTQ protesters while small in number are able to kick up a huge fight almost every single weekend in JK Rowling's mentions and cause her to trend. This happens on the weekend because unlike "the Russians", these people have day jobs during the week and can't sit online pretending to be a crowd from their actual place of work.

Crypto is another heavily manipulated subject although being techy nerds, they tend to rely more on automation and AI than brute force political influence farms and are easier to spot.

 

It's been a constant arms race between astroturfing influence teams and social media platforms to identity and remove these fake accounts, and unfortunately, the social media companies are losing badly. This is the real reason behind "real identities" being linked to online accounts, they (wrongly) think this is the magic bullet that will restore order and save these big platforms.

 

A good hint that a regular looking Twitter account might be a "bot" is that it's only tends to post about a single issue.

Real people get angry online all the time, but they also post cat pictures and other crap and making up the "fake mundane" posts is something the botters aren't very willing to spent a lot of time (money) on.

  • Like 2
  • Thanks 1
Link to comment
Share on other sites

3 hours ago, Coffee Pancake said:

Twitter routinely has "bot" purges. They aren't for the most part automated accounts, so the term "bot" is misleading.

Typically there are companies that exist to deliver engagement, picture one person sat in front of racks and racks of phones paid to do whatever is making them money today.

This practice is also commonly used on Twitter to create the illusion of grass root public political reaction, and groups running these schemes are well organized, in some cases state funded and active outside of domestic territories via VPN services. Thousands upon thousands of accounts pushing a daily geopolitical agenda that are deliberately set up and operated to look like real people who are suddenly very concerned about some very specific issue or another.

MAGA was heavily "botted" in this fashion on Twitter and other social media platforms. The running joke was "its the Russians", mainly because the bulk of the people working the "bot" accounts kept Russian office hours (and from time to time they would slip up and leak their real location). You could tell when the manipulation on reddit started like clockwork every day.

This is also why things like Bengasi or HunterBiden will spontaneously trend in the middle of the night when regular Americans are asleep, the goal is to have them in the trending feed first thing in the morning US time, so local political teams can pick up the baton and try and make it the days trend. Twitter sometimes manages to kill this, sometimes the subject matter is close enough to actual domestic political commentary they can't tell the difference and it's allowed to stand.

This same tactic has been used by other special interest groups trying to push political agendas on Twitter. British anti transgender / anti LGBTQ protesters while small in number are able to kick up a huge fight almost every single weekend in JK Rowling's mentions and cause her to trend. This happens on the weekend because unlike "the Russians", these people have day jobs during the week and can't sit online pretending to be a crowd from their actual place of work.

Crypto is another heavily manipulated subject although being techy nerds, they tend to rely more on automation and AI than brute force political influence farms and are easier to spot.

 

It's been a constant arms race between astroturfing influence teams and social media platforms to identity and remove these fake accounts, and unfortunately, the social media companies are losing badly. This is the real reason behind "real identities" being linked to online accounts, they (wrongly) think this is the magic bullet that will restore order and save these big platforms.

 

A good hint that a regular looking Twitter account might be a "bot" is that it's only tends to post about a single issue.

Real people get angry online all the time, but they also post cat pictures and other crap and making up the "fake mundane" posts is something the botters aren't very willing to spent a lot of time (money) on.

Yes, this! 
Do we have any equivalent in Second Life? 

  • Like 1
Link to comment
Share on other sites

4 hours ago, Love Zhaoying said:

Do we have any equivalent in Second Life? 

The only thing we have that resembles social media is this very forum, and I haven't noticed it occurring here. Unless it happened here on a very very small scale, it would be noticed and talked about/against.

Edited by Phil Deakins
Link to comment
Share on other sites

6 hours ago, Coffee Pancake said:

A good hint that a regular looking Twitter account might be a "bot" is that it's only tends to post about a single issue.

Real people get angry online all the time, but they also post cat pictures and other crap and making up the "fake mundane" posts is something the botters aren't very willing to spent a lot of time (money) on.

I see this crap all the time on Twitter. It's so freaking annoying, too. All low-effort troll posts and you can completely predict ahead of time not only which hashtags and thread comments are going to attract them, but what they're going to say. Ugh. Plus, the weirdly named hashtags with low numbers of tweets that just so happen to be sitting on the trending bar - if you watch long enough, you can see them climb rapidly from 1,000 tweets up to 10k and beyond.

I'm at the point now where if I see any seriously dumb hashtags on the trending bar, I just block them.

  • Like 2
Link to comment
Share on other sites

You are about to reply to a thread that has been inactive for 693 days.

Please take a moment to consider if this thread is worth bumping.

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
 Share

×
×
  • Create New...