The REAL New Search
|
|
Macphisto Angelus
JAFO
Join date: 21 Oct 2004
Posts: 5,831
|
05-02-2008 16:35
From: Dakota Tebaldi I think this system's vulnerability to abuse is tied into the number of flags necessary for a listing to be x'd. If I read the specifics of the announcement right, it would cost a competitor US$90 to flag you ten times - assuming his normal account is premium (if not, it'll cost him $100). People's traffic bots, not being premium accounts, couldn't be used to flag anything. Would ten flags be enough to axe a listing? Wait.. LL not only wants the residents to police the classifieds, etc for them.. but they are also CHARGING for us to do it? BWAHAHAHAHAHAHAHAHAHAHAHAHA
_____________________
From: Natalie P from SLU Second Life: Where being the super important, extra special person you've always been sure you are (at least when you're drunk) can be a reality! From: Ann Launay I put on my robe and wizard ha...
Oh. Nevermind then.
|
|
Alicia Sautereau
if (!social) hide;
Join date: 20 Feb 2007
Posts: 3,125
|
05-02-2008 16:35
L'o'L™
freebie avi`s less then ** days, how old are traffic bots? how many of them around?
another one for the harasement book in the making...
|
|
Winter Ventura
Eclectic Randomness
Join date: 18 Jul 2006
Posts: 2,579
|
05-02-2008 16:36
I can see it now: From: someone Remove your competition, GURANTEED! For only L$300,000, my bot armies will mercilessly flag any classified placed by the avatar you specify. Our automated system uses advanced scripts SL tools to read the Classified listing sof your target every day, and any ads place by them will be flagged. We guarantee 150 unique flags every day.
_____________________
 ● Inworld Store: http://slurl.eclectic-randomness.com ● Website: http://www.eclectic-randomness.com ● Twitter: @WinterVentura
|
|
Macphisto Angelus
JAFO
Join date: 21 Oct 2004
Posts: 5,831
|
05-02-2008 16:38
*still dying of laughter*
OK... this is likely one of the lamest things to come down the pipe.
_____________________
From: Natalie P from SLU Second Life: Where being the super important, extra special person you've always been sure you are (at least when you're drunk) can be a reality! From: Ann Launay I put on my robe and wizard ha...
Oh. Nevermind then.
|
|
Carl Metropolitan
Registered User
Join date: 7 Jul 2005
Posts: 1,031
|
05-02-2008 16:40
I believe Dakota was talking about the cost for purchasing Premium accounts to flag with; not a cost for flagging individual entries. I don't believe that the original SL Dev Mailing List excerpt mentioned Premium accounts, though. What it said was "Anonymous basic residents or residents whose accounts are less than x days old will not be able to flag content".
However, there's a gem of a good idea in the misunderstanding of Dakota's post. What if there was a nominal L$ cost for flagging? Might that serve to discourage frivolous flagging?
|
|
Kitty Barnett
Registered User
Join date: 10 May 2006
Posts: 5,586
|
05-02-2008 16:41
From: Dakota Tebaldi If I read the specifics of the announcement right, it would cost a competitor US$90 to flag you ten times - assuming his normal account is premium (if not, it'll cost him $100). People's traffic bots, not being premium accounts, couldn't be used to flag anything. Would ten flags be enough to axe a listing? Non-anonymous is likely to be any "payment info" account which are free to create although much easier for LL to identify in case of misuse. |
|
Amity Slade
Registered User
Join date: 14 Feb 2007
Posts: 2,183
|
05-02-2008 16:43
From: Dakota Tebaldi I think this system's vulnerability to abuse is tied into the number of flags necessary for a listing to be x'd. If I read the specifics of the announcement right, it would cost a competitor US$90 to flag you ten times - assuming his normal account is premium (if not, it'll cost him $100). People's traffic bots, not being premium accounts, couldn't be used to flag anything. Would ten flags be enough to axe a listing? If it's ten premium account flags, nothing will ever get fixed. When you limit it to premium accounts, you exclude a huge number of SL users anyway. Now find ten that actually use Search enough to get the same abusive listings, and who actually care enough to take the time to flag it. Making it easier to flag listings does something to overcome a user's ennui when it comes to being a tattle-tale. Requiring ten honest flaggings may be a bit of a high bar. Alternatively, when it comes to gaming reports, ten people with premium accounts can get together for the mutual benefit of all. "Support me on my flaggings, and I'll support you with your flaggings." That already happens with Abuse Reports. Everyone knows that the honestly, sincerity, or magnitude of an abuse report has absolutely no impact upon whether or not Linden Labs deals with it. The only thing that matters is numbers of Abuse Reports. So people routinely ask their friends to assist with Abuse Reports by the friends submitting ARs on the same issue, even though the friends had absolutely no involvement in the first place. Check the forums here. Find a thread about "What do I do about this problem?" or "Can I AR this?" The most frequently given piece of advice is, "Get all of your friends to make ARs as well, because Linden Labs ignores them unless there are multiple ARs on the same incident." I assume the same mentality can be transferred to the listing-abuse-vote policy. That's why instituting a sytem to review quality of ARs, rather than relying on sheer numbers, would greatly benefif the system. The more the system relies on pure formulae and numbers, the easier it is to game. |
|
Amity Slade
Registered User
Join date: 14 Feb 2007
Posts: 2,183
|
05-02-2008 16:51
From: Carl Metropolitan However, there's a gem of a good idea in the misunderstanding of Dakota's post. What if there was a nominal L$ cost for flagging? Might that serve to discourage frivolous flagging? No. The reason is that they are asymetric rewards when it comes to honest flagging and dishonest flagging. For honest flagging, you are relying on sheer goodwill of people volunteering to "help the community" by getting rid of rules-breakers. The honest flagger has nothing to gain, except the potential feeling of "doing the right thing." Arguably, there is a long-term benefit, in that "if everyone does their part," the Search function will be improved. However, that benefit is hard to immediately quantify for someone who is making the decision of whether to report or not. In most cases, the resident will decide it's not worth his or her time. Now, add an actual monetary cost to that reporting, and now that potentially honest reporter is now losing money as well as effort to do something in which he or she has no other thing to personally gain, other than the feeling of "doing the right thing." The dishonest reporters about which we are concerned often do have something monetary to gain by griefing business competitors. Where as the honest reporter is weighing the cost of a nominal feel against the reward of nothing, the dishonest reporter is weighing the nominal fee against potential profit gain by squeezing out competitors. The nominal fee is thus nominal discouragement for the dishonest reporter. The pure griefer who griefs for the sheer fun of it probably isn't discouraged by the nominal fee either. Causing harm to another is a form of entertainment to the pure griefer, and the nominal fee may be a small price of admission. Especially considering that the nominal fee can be easily obtained through a short amount of automated camping. |
|
SuezanneC Baskerville
Forums Rock!
Join date: 22 Dec 2003
Posts: 14,229
|
05-02-2008 16:55
What would happen if the search was dispensed with altogether and people used the embedded browser to do their searching with?
_____________________
-
So long to these forums, the vBulletin forums that used to be at forums.secondlife.com. I will miss them.
I can be found on the web by searching for "SuezanneC Baskerville", or go to
http://www.google.com/profiles/suezanne
-
http://lindenlab.tribe.net/ created on 11/19/03.
Members: Ben, Catherine, Colin, Cory, Dan, Doug, Jim, Philip, Phoenix, Richard,
Robin, and Ryan
-
|
|
Darien Caldwell
Registered User
Join date: 12 Oct 2006
Posts: 3,127
|
05-02-2008 16:56
From: SuezanneC Baskerville What would happen if the search was dispensed with altogether and people used the embedded browser to do their searching with? Where would the search results come from? |
|
Macphisto Angelus
JAFO
Join date: 21 Oct 2004
Posts: 5,831
|
05-02-2008 17:00
From: Carl Metropolitan I believe Dakota was talking about the cost for purchasing Premium accounts to flag with; not a cost for flagging individual entries. I don't believe that the original SL Dev Mailing List excerpt mentioned Premium accounts, though. What it said was "Anonymous basic residents or residents whose accounts are less than x days old will not be able to flag content".
However, there's a gem of a good idea in the misunderstanding of Dakota's post. What if there was a nominal L$ cost for flagging? Might that serve to discourage frivolous flagging? Oh good! *whew* That is one misunderstanding I am glad I had.  OK, not such a lame LL idea now. Dunno Carl. It *could* discourage abuse but would also discourage legit reports by people who don't feel like paying lindens to help right a wrong. If it was a small token amount a person who is out to mess with someone else's business to help their own would pay. They pay campers, high classifieds, etc often so this would not stop them. I really think a penalty for false reports is the answer. I think automatic removal is the wrong answer. After all.. even if something is in the wrong place it isn't going to do too much damage before someone reviews/removes it.
_____________________
From: Natalie P from SLU Second Life: Where being the super important, extra special person you've always been sure you are (at least when you're drunk) can be a reality! From: Ann Launay I put on my robe and wizard ha...
Oh. Nevermind then.
|
|
Winter Ventura
Eclectic Randomness
Join date: 18 Jul 2006
Posts: 2,579
|
05-02-2008 17:26
Just another tool to equip the "Religious Right" with another angle to attack adult content providers...
Pass.
_____________________
● Inworld Store: http://slurl.eclectic-randomness.com ● Website: http://www.eclectic-randomness.com ● Twitter: @WinterVentura
|
|
Phil Deakins
Prim Savers = low prims
Join date: 17 Jan 2007
Posts: 9,537
|
05-02-2008 17:33
From: Dakota Tebaldi I think this system's vulnerability to abuse is tied into the number of flags necessary for a listing to be x'd. If I read the specifics of the announcement right, it would cost a competitor US$90 to flag you ten times - assuming his normal account is premium (if not, it'll cost him $100). People's traffic bots, not being premium accounts, couldn't be used to flag anything. Would ten flags be enough to axe a listing? Who needs bots with accounts when you have a load of friends  |
|
Macphisto Angelus
JAFO
Join date: 21 Oct 2004
Posts: 5,831
|
05-02-2008 17:36
From: Winter Ventura Just another tool to equip the "Religious Right" with another angle to attack adult content providers...
Pass. How so? Mature ads under mature headings = OK. Mature ads anywhere else = Not OK. Same as mature and PG areas on the grid. What does this have to do with the "religious right"? hehe
_____________________
From: Natalie P from SLU Second Life: Where being the super important, extra special person you've always been sure you are (at least when you're drunk) can be a reality! From: Ann Launay I put on my robe and wizard ha...
Oh. Nevermind then.
|
|
Winter Ventura
Eclectic Randomness
Join date: 18 Jul 2006
Posts: 2,579
|
05-02-2008 17:50
From: Macphisto Angelus How so? Mature ads under mature headings = OK.
Mature ads anywhere else = Not OK.
Same as mature and PG areas on the grid. What does this have to do with the "religious right"? hehe I have had "mature ads", marked as "mature" deleted by Linden Labs because "someone" AR'ed them for being "mature" Even though the ads were, in fact, marked as "Mature". Pictures mysteriously being removed, listings "unsubscribing themselves".. you get the idea.. no actual NOTIFICATION from LL, just the ads are quietly disabled again and again. Because Classified ads are listed in people's profiles, regardless of rating, ALL ads are apparrently required to be PG.
_____________________
● Inworld Store: http://slurl.eclectic-randomness.com ● Website: http://www.eclectic-randomness.com ● Twitter: @WinterVentura
|
|
Macphisto Angelus
JAFO
Join date: 21 Oct 2004
Posts: 5,831
|
05-02-2008 17:55
From: Winter Ventura I have had "mature ads", marked as "mature" deleted by Linden Labs because "someone" AR'ed them for being "mature"
Even though the ads were, in fact, marked as "Mature".
Pictures mysteriously being removed, listings "unsubscribing themselves".. you get the idea.. no actual NOTIFICATION from LL, just the ads are quietly disabled again and again. Well that is wrong! Mature is mature. We all know what can be expected from mature ads. I think the Lindens were going against what the spirit of mature is by doing that to you. They also justified frivolous AR's and gave the reporter undue power by doing that.
_____________________
From: Natalie P from SLU Second Life: Where being the super important, extra special person you've always been sure you are (at least when you're drunk) can be a reality! From: Ann Launay I put on my robe and wizard ha...
Oh. Nevermind then.
|
|
Johnnie Carling
Registered User
Join date: 17 Aug 2007
Posts: 174
|
05-02-2008 18:06
More from Jeska From: someone Heya, Thanks for all the great thoughts everyone, very helpful! I wanted to clarify that the Search Flagging described is for search listings -- not inworld content, avatar or group profiles, land sales etc. - it is ONLY for parcel listings, classified ads and events listings. Also, for those who are concerned about abuse/griefing, here is some more detail on the proposed design: * If a listing is flagged as "Mature" by x Residents and is not marked Mature, it will be automatically changed to Mature - it will not be removed. Totally agree that more clarity is needed around what is "mature" (and "prohibited" I will take that back to the Governance Team as feedback. * If a listing is flagged as "Prohibited" by x Residents, it will be automatically taken down and reviewed by the Governance team. Prohibited listings might include listings for gambling or child pornography. * Not all accounts are able to flag search listings, anonymous basic Residents or Residents who are under x days old will not be able to participate in search flagging. * Each account which can flag is only able do so once per search listing and there is a limit to the amount of search listings each account can flag per day. * One account flagging a listing will NOT cause it to be automatically removed * Nominate for Showcase is that exactly - a nomination for review by an editorial board for possible inclusion. * Any classified ads that are taken down will be refunded (prorated for time active), with notification Are there any other restrictions to who can flag or when/how/etc anyone can think of to help both prevent overt abuse while still helping stem grievous abuse of search listings (see: current Event Calendar spam as one example)? Cheers, Jeska |
|
Phil Deakins
Prim Savers = low prims
Join date: 17 Jan 2007
Posts: 9,537
|
05-02-2008 18:10
I don't think that adds anything. It confirms that the taking down will be automatic, and it doesn't say anything about malicious use of the flagging.
|
|
Macphisto Angelus
JAFO
Join date: 21 Oct 2004
Posts: 5,831
|
05-02-2008 18:16
Thank you for posting it.
_____________________
From: Natalie P from SLU Second Life: Where being the super important, extra special person you've always been sure you are (at least when you're drunk) can be a reality! From: Ann Launay I put on my robe and wizard ha...
Oh. Nevermind then.
|
|
Sindy Tsure
Will script for shoes
Join date: 18 Sep 2006
Posts: 4,103
|
05-02-2008 21:26
From: Jeska Linden, via Johnnie Carling If a listing is flagged as "Prohibited" by x Residents, it will be automatically taken down and reviewed by the Governance team. Prohibited listings might include listings for gambling or child pornography. This bit I really don't like. I have an adult place and I take the LL stance on under age people, either RL age or apparent avatar age, being involved-in or in proximity-of 'adult' areas _very_ seriously. I really don't like the idea of people flagging my parcels as Prohibited because they see somebody short-but-adult and having LL automatically shut me down until they get around to sorting it out. Got no problem with them investigating and would happily help them with any info they want, especially if it's a real case of something prohibited, but the word 'automatically' really makes me nervous. If they 'automatically' do anything with this stuff, people will get burned. |
|
Matthew Dowd
Registered User
Join date: 30 Jan 2007
Posts: 1,046
|
05-02-2008 21:37
From: Macphisto Angelus
I really think a penalty for false reports is the answer.
How do you distinguish between a deliberate false report and a genuine mistake? When gamlbing was banned, some Lindens (repeatedly) deleted pay for free games which were not banned! When the whole child pornography issued flared up, some genuinely thought the ban included child avatars (not helped by LL constantly talking of "age play" , and reading the forums it appeared some even felt uncomfortable meeting child avatars in mature areas (thinking a minor was being corrupted - although you can't morally corrupt a pixel image and any such avatars should be an adult in RL). There will be a number of false reports just due to misunderstandings or even cultural differences (some may report things which are prohibited or frowned upon in their community but which globally and within SL are permissible) not out of deliberate malice or intent to grief. A penalty for false reports is likely to result in a number of people being stung for making what they believed to be a genuine report. As such the penalty is more likely to reduce the number of people risking any report (genuine or not). So, no, I don't believe penalties for false reports is the solution to the inherent problems in the proposal from LL. Matthew |
|
Sindy Tsure
Will script for shoes
Join date: 18 Sep 2006
Posts: 4,103
|
05-02-2008 21:44
From: Matthew Dowd A penalty for false reports is likely to result in a number of people being stung for making what they believed to be a genuine report. If you call the police on somebody when they're not doing something illegal, the police will certainly explain to you the error of your ways. Maybe a 'penalty' for false reports is a bit strong but some feedback should be given or the person won't know to stop wasting their time. |
|
Macphisto Angelus
JAFO
Join date: 21 Oct 2004
Posts: 5,831
|
05-02-2008 21:46
Point taken but I would think one accidental report wrongly sent would be seen differently then a series targeting one person by multiple people. One would be seen as a oops, the other a grief. Reading Jeska's second post it is a moot point anyway. From: someone * If a listing is flagged as "Mature" by x Residents and is not marked Mature, it will be automatically changed to Mature - it will not be removed. Totally agree that more clarity is needed around what is "mature" (and "prohibited" I will take that back to the Governance Team as feedback. * If a listing is flagged as "Prohibited" by x Residents, it will be automatically taken down and reviewed by the Governance team. Prohibited listings might include listings for gambling or child pornography. So it would appear that xyz furniture store is not going to be able to just have abc's furniture store removed from classifeds instantly unless under the prohibited flag. The clarification by Jeska changed the dynamics of what I was discussing on the other page. At least for me it did.
_____________________
From: Natalie P from SLU Second Life: Where being the super important, extra special person you've always been sure you are (at least when you're drunk) can be a reality! From: Ann Launay I put on my robe and wizard ha...
Oh. Nevermind then.
|
|
Kitty Barnett
Registered User
Join date: 10 May 2006
Posts: 5,586
|
05-02-2008 21:47
From: Matthew Dowd How do you distinguish between a deliberate false report and a genuine mistake? It makes things more complex, but historical data based on what action was taken based on the flagging. If someone genuinely means no harm, but consistently gets the criteria for a flag wrong, you don't want to punish them, but you don't really want to rely on their input any longer either. If I flag a listing as spam and others do as well but a Linden determines it's not actionable, lower the reliability score by 1. If I flag a listing as spam and a Linden ends up finding it actionable, raise the score by 1. If I flag a listing as spam and it never gets enough flags from other residents to warrant a look, nothing happens. Keeping track of the score would just be an automatic thing based on what action a Linden ends up taking. Once a score drops below a certain number, prevent that person from flagging (or give them illusion of flagging it but don't count it). |
|
Macphisto Angelus
JAFO
Join date: 21 Oct 2004
Posts: 5,831
|
05-02-2008 21:48
From: Sindy Tsure If you call the police on somebody when they're not doing something illegal, the police will certainly explain to you the error of your ways.
Maybe a 'penalty' for false reports is a bit strong but some feedback should be given or the person won't know to stop wasting their time. Agreed, but my penalty thought was for multiple obviously griefing use of the system. Maybe I was not clear on that, sorry if that is the case. My thinking was not in smacking down every accidental report. Just on the ones that could be proven to be grief oriented. Thanks for the oppurtunity to clear that up.
_____________________
From: Natalie P from SLU Second Life: Where being the super important, extra special person you've always been sure you are (at least when you're drunk) can be a reality! From: Ann Launay I put on my robe and wizard ha...
Oh. Nevermind then.
|