« Where Will the First Amendment Take Place? | Main | SLB5: Not Censorship, Not Discrimination, But Right of Private Company to Promote What it Likes »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Ann Otoole

Oh indeed there needs to be some sort of control here. But control over LL's mismanagement of SL in general.

But that will never happen. This is a dead horse. LL is pushing people out to open grid. LL is competing in the economy with a huge unfair advantage. Ethics has never been a serious priority with LL staff. (who warned Luke about the impending bank ban anyway?)

The list goes on and on.

There is a purpose for bots. NPCs actually. SL needs NPC capability for rp sims. evil mobs, etc.

But whats the point of bots traversing the grid entering property where they are not wanted? Who decided that is OK?

Yes where is this new CEO anyway? It is time to strap him to a chair and demand the truth about what the real plans for sl are. Is LL going to drive out land marketing opportunities? (Bay City)... Is LL going to open LL Mart for clothes shapes, etc.? I.e; is LL going to run out the people that know how to create content? LL certainly doesn't know much about it. Oh wait, there's millions in China willing to work for nothing. No problem.

Yes I agree with you Prok. But there is nothing that can be done. LL doesn't even consider the "pjira" as anything other than a place for people to waste time thinking they are helping when all the while LL decides what they will do, fix, and apparently spend most of their time breaking.

Good effort though. Kudos.

Tammy Nowotny

Ann said: "Is LL going to drive out land marketing opportunities? (Bay City)"

The auction prices for the Bay City land are astounding... one beachfront 1024 m^2 lot is going to go for the equivalent of more than $750 US. Even inland ones are going for more than $200 US. That seems a little whacky given that you can buy a whole sim at auction for about US$1000 or even less... although I suppose the lower tier makes up for the discrepancy somewhat.

Smoke Wijaya

I am a firm believer in the moral neutrality of technology in itself, but during many discussions inworld about bots and connected issues, it is indeed clear that most (if not all) people would like to see them identified.
But tech-people present at those times, despite liking the idea, said all it very likely is not possible. And the possibility apparently has nothing to do with the architecture or anything (I am not a techie myself). I mean, not something LL could change/implement, even when the political will would be there .. like with the demand for more groups for example, it would be possible with a complete overhaul.

Anyway, most nuisances seem to come from those that operate complete armies for traffic rates, suck out camping money or those bot operators that do not revert the mistakes made by humans while wanting to transfer their land and which got scoped by a landbot, for example.

So .. maybe it is more practical for us to focus on those avatars/groups that operate this technology in an immoral way, e.a. taking advantage of human fallibility or sucking up disproportionate amount of resources?

Cherowolf Redgrave

Tammy - pay closer attention to the Auctions. The highest I've seen a Bay City parcel go for so far is L$320k or about $1185 usd. Total madness when you can get a new mainland sim for the same amount. If you look at the high rolling buyers it is an interesting mix of foreign, old school and newbies (if the newbies aren't actually alts). To add to the price insanity the flippers have already descended on the place.

Prokofy - I am not ignoring your post topic. I went to the JIRA to vote and saw that it was closed. Sorry about that.

Prokofy Neva

Technology is not morally neutral, because it is not divorced from humans who wield it, Smoke. You are willing to concede immorality in these flawed humans who suck down camping or grab land mistakenly set and won't give it back -- but you seem to believe that their hammers and sickles are part of that bright future of utopianism that will someday be wielded for a Better World.

They won't. People's infuse tools with their flawed nature.

This is what always gets communism -- which is always fascinated with the industrial and the technical when it isn't obsessed with the pastoral and agricultural -- in trouble.

Even when programmed to function with "artificial intelligence," bots are not independent entities that can "objectively perform good" and be "helpful to science". Indeed, to program a piece of technology to steal camping money does indeed render that technology evil.

I realize there is a whole school of thought from tekkies that constantly yammer about hammers being neutral, and only those pounding being not neutral, but I find this silly to the extreme in the era of widely flexible and calibrated software that performs enormous numbers of functions that humans used to do.

There are examples often discussed in these debates of two pieces of technology that, when used, had far larger consequences than intended: the atom bomb and thalidomide. I fail to see that technology that "unconsciously" commits the evil of killing and maiming humans who were imperfect and did not foresee their evil consequences is somehow "redeemed from evil". Indeed, it's part of the fatal fallacy of the utopian ideal of perfectable humans that they can make neutral technology that can be perfectible, even if organic humans cannot.

Bots are creatures that operate and exist on the command of their makers, who are the few imposing their will on the many, usually without visibility or accountability. They need to be regulated.

The thrust of the JIRA comments seems to be "you can't tell a bot from a human, bots fool the client". Well, I'd like to hear more mature, informed, and investigative opinion on this. I see the usual infantile JIRA- jeerers line-up from the usual suspects whom I simply don't believe because of their bad behaviour.

If a human eye can determine a bot is a bot at 96 meters by their hunched position, I'm not finding it credible that the Linden servers cannot do the same thing. Perhaps I've discovered the bots' Turing test that can't be automated -- the hunching or old fashioned "birthing position" (unfortunately, few women are allowed to get into this position to give birth anymore, which is a shame, and makes for a lot of back labour pain).

But there are a score of other factors involved in manning bots. There are commands that go to them. There are patterns of activity -- or rather, non-activity in that they don't talk, click on stuff, buy stuff, etc.

I don't believe that marking bots takes any overhaul whatsoever, that's absurd. The Lindens marked "No Payment On File" and made a land ban for it as fast as you can say "Jack Rabbit" when the community went into an uproar over free accounts. The Lindens can add an account marker easily -- they can make it say "All People with U.S. Credit Card Addresses" if they like. It's a matter of political will.

As I said, they can sell bot accounts that are marked, encourage compliance, discipline non-compliance. They could even sell such accounts with some kind of extra features perhaps. If those people logging on bots don't comply, they could become liable under the TOS.

There are lots of ways to skin the cat. The idea that you "can't do" something technically NEVER persuades me in SL because I see the closed, rigid little minds at work screeching this. People say you can't run a 3-D streaming world on a giant server farm spread across 3 locations, either.

Prokofy Neva

Cherowolf, the JIRA is opened -- and will be re=opened again and again if need be. So vote if you like.

Khamon Fate

If the servers are programmed to use activity patterns as bot indicators, users will have to make a conscience effort to not match those patterns e.g. not responding to IMs in a timely manner, not moving around within a timed period, issuing keystrokes to the server too quickly. I'd likely get pegged on the latter very often for clicking menus habitually and typing eighty words a minute. Then how will I have the stigma of "bot" removed from my account?

One land-based option they might consider is allowing us to set the dead timer on parcels rather than it begin a global sim parameter. For example, I might choose to boot out anyone who lands in Slate and doesn't move in the first two minutes that they're there.

Tammy Nowotny

I found an easy test for a 'bot... tell them I will give you L$20 if you speak up on open chat within 60 seconds. Since there is no one there, they won't get the L$20. Of course that does also catch folks who are merely AFK or not paying attention. But it's a fun trick to play if you stumble into a nest of camper 'bots, which happened to me the other day. The funny thing was the ones I stumbled upon didn't even have skins... their master was so eager to create them as quickly as possible that she didn't even do the most basic steps needed to make look minimally legitimate.

Drifting OT a little, this points out a larger trend with online scams. Their operators often make themselves more conspicuous by skipping basic elements of legitimate traffic just to save a minimal amount of effort. The classic example would be spam, where the spammers don't even bother to create realistic return addresses, and don't even bother to make the mailserver name appear to match the sender's domain. Or, to cite another one, I used to buy AdWords ads, and one of the signs of click fraud was when the clicker didn't load any of the images on my page in. The reason for this was so they could put in more clicks per hour... but it made the fraudulent clicks mighty obvious.

The botnest I visited had about 20 grey zombies standing around. The person who created the zombies probably saved a few seconds per zombie by not bothering to give them skins or clothing.... but is the time she saves worth the cost of making her scam that much more obvious to random passersby?

Prokofy Neva

Khamon, I wouldn't exaggerate the ability of such a monitoring system to restrict real humans as you imagine.

1. A new system is announced whereby you are asked to register -- and pay for -- your accounts used as bots, which carry markers on them.

2. If you don't comply, you are liable under the TOS.

3. The Lindens rely on a) abuse reports which they examine on the merits and b) spot checking, as they do with casinos or "ageplay," spotting certain patterns. Will they make mistakes? Occasionally, they might. I don't see it as any problem of any significant proportions.

4. Since the ability to make bots only entered in 2007 or whenever it entered, they might reasonably exempt all accounts made prior to that from bot examination. Will some people game the system and sell oldbie accounts to do illicit botting? Sure, but not in any huge numbers.

5. Always, in pursuing a social good, there are tekkies who carry to extremes any piece of "logic". But I think if you have a set of parameters like patterns of activity, age of account, region of activity, etc. you will be able to monitor it reasonable.

6. I continue to remain unpersuaded that the LL servers cannot distinguish from humans and bots.

Khamon, you idea of booting people on a timer doesn't seem reasonable to me, because especially newbies can be frozen and indecisive or just AFK for long periods.

Yes, Tammy, you're pointing out all the obvious markers of bots that we all see. Spot-checking like this on the basis of abuse reports would immediately tell the tale. The question is whether there is an automatic server function that wouldn't cost Linden human administration.

These problems aren't trivial. The 3-D Internet will be riddled with these infestations unless we take a stand now. I am not for awarding to the few the ability to exploit the many without any accountability.

Melissa Yeuxdoux

Just out of curiosity, can you give an example of any technology that you do not consider evil? (I'd guess not, since it's all created by imperfect people.)

People aren't perfectible, nor is technology. But that doesn't mean it's not worth trying to improve them--come to think of it, you ask for such improvements rather a lot--and on the whole, I'd say that things are improving.

Prokofy Neva

Melissa, you're a literalist tekkie. I've never said human beings are improvable, and that you shouldn't try to improve them; I've said they aren't *perfectable* and those with illusions on this score like the Extropians are objectively evil -- because they must always lie and use force to hide this truth (as they did quite classicly by banning me pre-emptively).

Is all technology evil? Perhaps there are technologies that are inherently good, I'd have to think about it. But yes, there *are* technologies that are evil because they have welded into them human intent, which is either imperfect or evil. Is an artifact merely lying there doing nothing retaining its evil if no evil human picks it up to kill with? Well, it has the potential for evil built within it, it can be actualized. Potential evil remains...potential evil, able to be actualized.

Prokofy Neva

*are NOT improvable


I suppose a registration system may help tag accounts that are meant to be used as bots. People who don't register their bot accounts will suffer the inconvenience, when caught, of having to create a bot alt although, if they had to pay a retroactive registration fee (as all bot accounts are supposed to be premium under your system), that might act as a deterrent if LL can actually track down who's creating and operating which accounts with and without a bot client any given login.

It seems to me just too easy though to log the "bot" account in with the normal client regularly enough to "prove" that I actually use it as a person.

Ann Otoole

Doesn't all this rely upon an organization devoid of any ethics having to exercise ethical control?

I mean the Governance team is composed of griefers. Why would they rat out their buddies?

Nothing of a responsible corporate nature will work unless the new CEO cleans house. And that guy is MIA so he must be totally useless.

I just don't see LL having any interest in doing the right thing for sl while for all practical appearances the staff of LL is working to burn SL down. Maybe Rosedale should have been a grown up business person and known who all to fire along with Corey.

Prokofy Neva

Ann, as usual, you are being way too harsh, and exaggerating terribly.

While there might be some griefers getting a pass for old time's sake, or some geeky types on the G team who don't get it about griefing, I think it's unfair to say the G team is "composed of" griefers. That just seems excessive, given that they do fight griefing.

I actually don't think this new CEO is MIA. I think he's probably very present at the job he came to perform -- running Linden Lab -- and he probably appears just as much as he needs to, to those who hired him -- the board.

Philip saw himself presiding over a community building the world -- he was in the world because he saw that as co-extensive to Linden Lab.

The new guy most likely views the world as already built, separate, and not really heavily his concern -- he has the company to worry about.

I think they are probably very different in outlook and temperment, but what they needed there wasn't somebody to cheerlead the world being built -- that was done -- but to run the company on a larger scale.

I do wonder if the staff is purposely or unwittingly burning the world down. I don't get a lot of what they do (monkeying with search, building Bay City, refusing to finalize the work needed to eliminate ad extortionists). But -- I'm done with them. They may be burning or not -- I will try to work around them.

John Lopez

Fighting technology with technology doesn't work. The following problems online show just how hard it is to differentiate:

* the continued torrent of spam (despite anti-spam tools on clients and servers).

* the continued torrent of viruses (despite massive expenditures to the anti-virus industry)

* the failure of CAPTCHA to prevent spam on high profile sites (thanks to redirection of work and broken CAPTCHA algorithms)

* the failure of ticketing agencies to keep the majority of online tickets from going to scalpers.

There are many more examples where automated tools create much annoyance online and there is GREAT WILL TO FIGHT THEM, and yet the fight is a failure.

Here in SL there is no will to fight them and even if there was it would take about ten minutes to filter chat to some third world workers who respond minimally to ensure the bots aren't taken out (as they already do with gold farming bots in MMOs now).

Not that the industry isn't open for a solution to gold farming bots, scalpers, CAPTCHA fixes, viruses or spam. If you have the solution to any of those, just go implement them because you will be very, very rich.

Desmond Shang

I could care less if an avatar is a bot or not.

What bothers me is if the avatar hammers a search function 100,000 times an hour. Or something equally egregious.

It's simple really - regulate the behaviours that are damaging. A responsible bot that runs a mere 100 searches per hour isn't going to bother much.

So what could be regulated? Any given IP address could be limited to X avatars making Y asset requests per hour. After that, cut it off. That kind of thing.


This is totally off topic but I didn't want to go through the horror of logging into The Grid All Hail The Central Grid to give you this URL:


You may know it already; but it's news to me so I'm passing it along.

Tammy Nowotny

John mentioned: "* the failure of ticketing agencies to keep the majority of online tickets from going to scalpers." That's actually not a problem for the agencies themselves. They themselves are scalpers... they buy tickets low and sell them high. I don't have a big problem with that. That's capitalism.

In fact, eBay now owns a service called StubHub which is a portal for scalpers... a ticket seller can put the ticket on consignment with StubHub and the buyer can pick it up at the box office, and not necessarily at the same price he or she would have gotten had she shown up at the box office before the event originally "sold out."

Ann Otoole

Hey Khanon I wonder who the lucky bot was that nabbed this one:
Bay City parcel 4j - Morton (238,217) 992m L$0 May-27-08

Martin Magpie

LL have begun to delete posts from the blog that they do not like. More so any post by certain ppl. Do they honestly believe ppl will just shush? Hardly, after all this is the information age and anyone who can type can blog.

I am starting to resent LL and their constant interference into the affairs of their paying customers. They need to remember that we do not work for them; they work for us.

Martin Magpie

Sorry I got side tracked my point being LL is defining what they want and disregarding what they don't want. Unfortunately it is never for the good of the paying customer it's for the good of the LL bank account. Recent LL actions prove that point in it's entirety.

Yumi Murakami

The problem is that even if you ban bots, you can still have problems with "cyborgs" - real clients that integrate bots.

For example, a client could be written that would automatically buy land sold at less than a given price - the same as a landbot - but would otherwise run the same as the regular Second Life viewer. Does this need to be banned?

Maklin Deckard

Wow, just read the jira entry and voted for it. Gordon and that other dink (that started ever post with Dearie) are quite odious...makes me wish there was a report function like on the main forums.

As to the idea of bots being 'undetectable' that's bullshit of the first order. Unless the bots are behaving totally at random (which rather defeats the POINT of a bot) there is a detectable 'click stream' of actions that landbots do, another that campbots do, etc. This can, even with live human attempts at obfustication, be detected. Case in point, Ultima Online. UO has a rule, no unattended macro-ing (close as they get to bots). I was mining, using a macro to mine and load ore, while manually moving and selecting places to mine (just saving myself 2 clicks per mining attempt and a click drag to inventory). At least 3 times in one week a Gamemaster beamed in and spoke to me, saying things that would not be easily parsed by a script and expecting a response (such as typing a few letters of garbage and asking me to type it back, etc).

No response in a set timeframe, Unattened Macroing. Proper responses (which I gave) resulted in 'Have a great day!' and the GM disappearing. It is quite possible to detect these 'bot patterns' automatically and forward suspected bot names/locations to Lindens, even amid random input to make it 'look' human or even real human input.

Hell, a fast bot test in SL is for a linden to force-port a suspected bot to somewhere only a linden can get you in and out of. If the AV just sits there silently, or continually tries to port out over and over without communicating...bot. If they react with conversation to a present Linden, avatar. Could even automate it...have a board with a randomly shown code as a graphic (like when we post here) and the player has to say the code on a randomly generated channel (also shown in text). Player types it on that channel, gets released...bot sits there after a set time, is locked out pending review as a bot.

If a bot is flagged, per Prokofy's suggestion, then if it is detected, its not ported to the 'inquisition' room. Simple, but 'there is no way to detect' according to the Jira trolls (No way to detect 100% automatically and flawlessly, so per techie thinking 'not 100% code? Can't be done'

Maklin Deckard

"Not that the industry isn't open for a solution to gold farming bots, scalpers, CAPTCHA fixes, viruses or spam. If you have the solution to any of those, just go implement them because you will be very, very rich." - John Lopez

Typical Techie Logic...there must be 'THE SOLUTION'...an incremental decrease, if it is not 100% serves no purpose or (in the case of LL) is an arms race and too much work.

By that logic, since the police aren't 100% effective, why bother? The IRS doesn't catch all fraud, so why bother? This may come as a shock to those of techie / gamer bent, but NOTHING in life is 100%..its not all 0 and 1...but a plethora of fractional improvements in between absolutes of code. Laws don't stop crime....they reduce it to TOLERABLE levels.

As far as code goes, sure, filters do not stop spam, but they reduce it to a manageable level. Antivirus software has not eliminated viruses, but they make computer use possible without daily / weekly wipes and reinstalls. Bans and software monitoring do not catch all gold farmers in MMO's...but again they keep it to a tolerable level. Same with ingame bot checks on SL...they WOULDN'T eliminate bots utterly...but would, when combined with a registration system for bots / account bans for unregistered bot operations, make the presence of bots MORE TOLERABLE to the average players.

Sticking to the techie '0-1 / 100% or nothing' tunnel vision is quite frankly, not only unhelpful but actively aiding and abetting the destructive elements.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name and email address are required. Email address will not be displayed with the comment.)

Blog powered by Typepad