<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Reputation in Second Life</title>
	<atom:link href="http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/</link>
	<description>blog of Lex Neva in Second Life</description>
	<lastBuildDate>Thu, 15 Jan 2009 10:49:20 -0500</lastBuildDate>
	<generator>http://wordpress.org/?v=2.8.4</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Tim</title>
		<link>http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/comment-page-1/#comment-5293</link>
		<dc:creator>Tim</dc:creator>
		<pubDate>Thu, 15 Jan 2009 10:49:20 +0000</pubDate>
		<guid isPermaLink="false">http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/#comment-5293</guid>
		<description>Your system basically assumes people are obsessed with checking reputations and want an orderly world. You just don&#039;t get it that what really draws the vast majority of people to the virtual worlds is their relative anarchy. People want to do what they can&#039;t get away with in the real world here because we are all animals inside and always want to eat up smaller guys subconsciously. The day SL is really stupid enough to implement any reputation system to government the virtual world like the &quot;Brave New World&quot; is the day it will lose out its 15 million registered users and fall into obscurity.</description>
		<content:encoded><![CDATA[<p>Your system basically assumes people are obsessed with checking reputations and want an orderly world. You just don&#8217;t get it that what really draws the vast majority of people to the virtual worlds is their relative anarchy. People want to do what they can&#8217;t get away with in the real world here because we are all animals inside and always want to eat up smaller guys subconsciously. The day SL is really stupid enough to implement any reputation system to government the virtual world like the &#8220;Brave New World&#8221; is the day it will lose out its 15 million registered users and fall into obscurity.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Carl</title>
		<link>http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/comment-page-1/#comment-5292</link>
		<dc:creator>Carl</dc:creator>
		<pubDate>Thu, 15 Jan 2009 10:20:28 +0000</pubDate>
		<guid isPermaLink="false">http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/#comment-5292</guid>
		<description>Recent cases of the &quot;Sim Mafia&quot; in the Sims online virtual world who accepts payment of in-world money to gang up on a Sim by bombarding it with negative ratings have emerged. Basically it is collusion to bad mouth about someone, I think your model cannot handle this situation which could be a fairly easy to do attack.</description>
		<content:encoded><![CDATA[<p>Recent cases of the &#8220;Sim Mafia&#8221; in the Sims online virtual world who accepts payment of in-world money to gang up on a Sim by bombarding it with negative ratings have emerged. Basically it is collusion to bad mouth about someone, I think your model cannot handle this situation which could be a fairly easy to do attack.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Lex Neva&#8217;s thoughts &#187; Linden Lab Open Sources SL Client</title>
		<link>http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/comment-page-1/#comment-2293</link>
		<dc:creator>Lex Neva&#8217;s thoughts &#187; Linden Lab Open Sources SL Client</dc:creator>
		<pubDate>Sun, 11 May 2008 02:36:10 +0000</pubDate>
		<guid isPermaLink="false">http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/#comment-2293</guid>
		<description>[...] I have no way of knowing if I can trust the person releasing it. If we had a system like the Reputation System I proposed previously, then it would be much easier for me to make a quick and yet confident [...]</description>
		<content:encoded><![CDATA[<p>[&#8230;] I have no way of knowing if I can trust the person releasing it. If we had a system like the Reputation System I proposed previously, then it would be much easier for me to make a quick and yet confident [&#8230;]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: lex</title>
		<link>http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/comment-page-1/#comment-871</link>
		<dc:creator>lex</dc:creator>
		<pubDate>Fri, 31 Aug 2007 16:25:57 +0000</pubDate>
		<guid isPermaLink="false">http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/#comment-871</guid>
		<description>That&#039;s a good point, Warkirby.  What I described in the essay works fine for my PERSONAL scripts and objects, but it breaks down immediately when I want to sell stuff.  Since I make a fair amount of money in-world every month from sales, I&#039;m sensitive to what you say, and I realize that we&#039;re pretty much stuck.

The only solution to this that I can see would be for LL to vette people running their own sims before allowing them to connect to the main grid... and that strikes me as kinda nasty.  The only reason I can sell things in SL is because LL has a complete and artificial enforcement of IP rights.

If we ever go to the kind of grid that will resemble the WWW, ie, a truly heterogenous grid run by disparate service providers, IP rights will probably go out the window, and every in-world asset will effectively become open source.

Damn.</description>
		<content:encoded><![CDATA[<p>That&#8217;s a good point, Warkirby.  What I described in the essay works fine for my PERSONAL scripts and objects, but it breaks down immediately when I want to sell stuff.  Since I make a fair amount of money in-world every month from sales, I&#8217;m sensitive to what you say, and I realize that we&#8217;re pretty much stuck.</p>
<p>The only solution to this that I can see would be for LL to vette people running their own sims before allowing them to connect to the main grid&#8230; and that strikes me as kinda nasty.  The only reason I can sell things in SL is because LL has a complete and artificial enforcement of IP rights.</p>
<p>If we ever go to the kind of grid that will resemble the WWW, ie, a truly heterogenous grid run by disparate service providers, IP rights will probably go out the window, and every in-world asset will effectively become open source.</p>
<p>Damn.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: lex</title>
		<link>http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/comment-page-1/#comment-870</link>
		<dc:creator>lex</dc:creator>
		<pubDate>Fri, 31 Aug 2007 16:20:51 +0000</pubDate>
		<guid isPermaLink="false">http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/#comment-870</guid>
		<description>Yoyo, you make a good point, and I did realize that as I was writing this.  But keep in mind, the data would not NECESSARILY be O(N^2).  You have to take human behavior into account.  Every person is not going to meet every single other person, and they&#039;re definitely not going to vote on them!  Lots of social networking websites have data that would appear to be O(N^2), and they don&#039;t run out of space.  Maybe this would be the same.</description>
		<content:encoded><![CDATA[<p>Yoyo, you make a good point, and I did realize that as I was writing this.  But keep in mind, the data would not NECESSARILY be O(N^2).  You have to take human behavior into account.  Every person is not going to meet every single other person, and they&#8217;re definitely not going to vote on them!  Lots of social networking websites have data that would appear to be O(N^2), and they don&#8217;t run out of space.  Maybe this would be the same.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: lex</title>
		<link>http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/comment-page-1/#comment-869</link>
		<dc:creator>lex</dc:creator>
		<pubDate>Fri, 31 Aug 2007 16:19:33 +0000</pubDate>
		<guid isPermaLink="false">http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/#comment-869</guid>
		<description>Aestival -- Sorry it&#039;s taken me so long to reply here.

First of all, you raise a good point about the problem of creating a nation of conformists.  The last thing I want is to subtly push everyone to act like everyone else.  In some way, I guess, that IS what I want; I want everyone to act NICELY.  But you&#039;re right... when reputation can be about ANYTHING, like I&#039;ve described, then that does have the danger of leading to conformism.

On to your suggestions.

1. I like this idea.  It&#039;s definitely true that you&#039;re likely to spend more time in the presence of people you like than people you don&#039;t.  Even if you don&#039;t know everyone who&#039;s around you, if someone&#039;s acting out, it&#039;s likely that they&#039;ll drive a percentage of people in their vicinity to leave.  That&#039;s the kind of emergent statistic that will show up when a lot of data is aggregated.

On the downside: this specific metric would be easy to game, by having a group of people make their avatars all sit around in a group, harvesting reputation.  It also could suck for people like me... I tend to hang around in Suffugium a lot, often in my little hidden home, and yet people still like me.

2. If you go back to just one vote per person, than you might well end up with the system we had before, where people just voted each other up sight unseen at parties.

3. I definitely like this idea of making reputation relative.  When I read what you said, I envisioned some kind of &quot;6 degrees of Kevin Bacon&quot; type system, where it tries to find the shortest chain of people to get from you to the person in question.  Then it could tell you whether it thinks you&#039;d like this person, based on how the people you like feel about the people that like them.  It&#039;s possible that this could be fairly hairy to implement, computation-wise... but then again, Orkut, the social networking service, seemed to be able to tell you a chain of friendships that would lead to any person you cared to find on the system.</description>
		<content:encoded><![CDATA[<p>Aestival &#8212; Sorry it&#8217;s taken me so long to reply here.</p>
<p>First of all, you raise a good point about the problem of creating a nation of conformists.  The last thing I want is to subtly push everyone to act like everyone else.  In some way, I guess, that IS what I want; I want everyone to act NICELY.  But you&#8217;re right&#8230; when reputation can be about ANYTHING, like I&#8217;ve described, then that does have the danger of leading to conformism.</p>
<p>On to your suggestions.</p>
<p>1. I like this idea.  It&#8217;s definitely true that you&#8217;re likely to spend more time in the presence of people you like than people you don&#8217;t.  Even if you don&#8217;t know everyone who&#8217;s around you, if someone&#8217;s acting out, it&#8217;s likely that they&#8217;ll drive a percentage of people in their vicinity to leave.  That&#8217;s the kind of emergent statistic that will show up when a lot of data is aggregated.</p>
<p>On the downside: this specific metric would be easy to game, by having a group of people make their avatars all sit around in a group, harvesting reputation.  It also could suck for people like me&#8230; I tend to hang around in Suffugium a lot, often in my little hidden home, and yet people still like me.</p>
<p>2. If you go back to just one vote per person, than you might well end up with the system we had before, where people just voted each other up sight unseen at parties.</p>
<p>3. I definitely like this idea of making reputation relative.  When I read what you said, I envisioned some kind of &#8220;6 degrees of Kevin Bacon&#8221; type system, where it tries to find the shortest chain of people to get from you to the person in question.  Then it could tell you whether it thinks you&#8217;d like this person, based on how the people you like feel about the people that like them.  It&#8217;s possible that this could be fairly hairy to implement, computation-wise&#8230; but then again, Orkut, the social networking service, seemed to be able to tell you a chain of friendships that would lead to any person you cared to find on the system.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: WarKirby Magojiro</title>
		<link>http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/comment-page-1/#comment-866</link>
		<dc:creator>WarKirby Magojiro</dc:creator>
		<pubDate>Fri, 31 Aug 2007 00:10:20 +0000</pubDate>
		<guid isPermaLink="false">http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/#comment-866</guid>
		<description>The biggest problem I see with opensourcing, will be the ease of stealing content. Griefers are tolerable.

Unfortunately, this system isn&#039;t really going to help with that. You may choose not to enter a non public sim. But it just takes someone else to buy a copy of your product, and take it there themselves. Unless you plan to restrict sales to only the trusted elite with high reps, (which would destroy any business), there&#039;s nothing that can be done.

I don&#039;t see hopw any system, technical or social, can prevent that.</description>
		<content:encoded><![CDATA[<p>The biggest problem I see with opensourcing, will be the ease of stealing content. Griefers are tolerable.</p>
<p>Unfortunately, this system isn&#8217;t really going to help with that. You may choose not to enter a non public sim. But it just takes someone else to buy a copy of your product, and take it there themselves. Unless you plan to restrict sales to only the trusted elite with high reps, (which would destroy any business), there&#8217;s nothing that can be done.</p>
<p>I don&#8217;t see hopw any system, technical or social, can prevent that.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Yoyo Bean</title>
		<link>http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/comment-page-1/#comment-579</link>
		<dc:creator>Yoyo Bean</dc:creator>
		<pubDate>Tue, 19 Jun 2007 09:22:52 +0000</pubDate>
		<guid isPermaLink="false">http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/#comment-579</guid>
		<description>There are a few technical issues with the literal suggestion.  For example, storage requirements to keep track of, &quot;invested,&quot; opinions will be O(n^2) at the least, and even LL doesn&#039;t keep O(n^2) data around except for friends lists and the &quot;My Notes&quot; section of profiles.  I have the feeling that both are very sparsely populated, but this reputation system would not be.  Millions of residents makes that a bit of a problem.</description>
		<content:encoded><![CDATA[<p>There are a few technical issues with the literal suggestion.  For example, storage requirements to keep track of, &#8220;invested,&#8221; opinions will be O(n^2) at the least, and even LL doesn&#8217;t keep O(n^2) data around except for friends lists and the &#8220;My Notes&#8221; section of profiles.  I have the feeling that both are very sparsely populated, but this reputation system would not be.  Millions of residents makes that a bit of a problem.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: aestival</title>
		<link>http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/comment-page-1/#comment-31</link>
		<dc:creator>aestival</dc:creator>
		<pubDate>Thu, 01 Mar 2007 00:35:05 +0000</pubDate>
		<guid isPermaLink="false">http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/#comment-31</guid>
		<description>This is a great idea, make people accountable and they might behave!

I like the basic approach of having a social solution to a social problem, and actually I think most of the adoption and utilization of a reputation would work. 

My only worry is similar to Seifert. The system *can* be gamed, just like an election because there is literally a game system in place. It&#039;s the game of who has the most friends who agree with him, also known as Politics or, my favorite middle school variant, Cliques vs Losers. Also, unless I misunderstood the proposal, it&#039;s a negative feedback loop - so it&#039;s easy for the winners to become very strong and the losers to get ground down. The more unpopular I am the less my opinion counts. 

This could be even worse than the passionate people having the upper hand - as you have in a get out the vote scenario. Because the reputation system is based on popularity *and* alignment of opinion, you could have a reputation get trashed simply because they have unpopular opinions on their profile, or they engage in behavior is legit and protected by the terms of service but considered undesirable by some group. As an extreme example think of furries. Most people don&#039;t think about them, some folks hate them, some of them hate non furries. What if folks (furry and not) started rating people based on their amount of fur? Someone who spends part time as a furry might be &quot;outed&quot; in their reputation profile. Ultimately the majority would win and furry folk would not be welcome in SL because their other activities were limited by their communal trashing of reputation. By the system they would be treated like a griefer bloc on a large scale. 

Heck, any subject two folks could dissagree on, could eventually be subject to the global popularity test - ultimately resulting a population that thinks, acts and looks exactly the same about everything! 

Eeep!

Well, why not? It&#039;s actually very similar to the historical reasons minority groups have been persecuted. If you wear funny clothes I can tell you don&#039;t belong in my night club, shouldn&#039;t shop at my grocery store, etc. Only now we&#039;re not just judging on outward signs of conformity and acceptance - every aspect of a person&#039;s behavior and thoughts is publicly available for scrutiny and subject to a popularity test which might be grounds for segregation.

Actually, thinking about it, a more likely outcome to the furry scenario is different cliques adopting different reputation systems - less dystopian than a grid full of clones but just as sucky. The diversity in the grid only hangs out with their clones because those are the ones using the reputation system where their clique wins.

Simply put, my concerns are the way that reputations build on reputations (popularity contest), and the way that opinions or any attribute can become filters for reputations (segregation).

But since I don&#039;t like to poke holes without being constructive, here&#039;s a few random ideas which might help. ^_^

I wonder if there is not a more &#039;natural&#039; way of getting an aggregate reputation on somebody than voting and networks of endorsement? I know these are very popular techniques these days and are very useful for problems like rating website value, product quality or even product review trust worthiness. But maybe reputation requires something different. Maybe it should be based on how much time I spend interacting with someone. Time I spend with you works to give an implicit increase in reputation. Maybe there&#039;s other actions that I can take, or that more importantly, groups of people can take together that act as endorsements. There is probably a lot to explore in this direction but I&#039;m not sure if it could be done in such a way that a malicious person couldn&#039;t run their own clients and servers to spoof good activity...

As another alternative entirely you could try and take the scheme as proposed and simplify it further with the goals of eliminating the popularity and the segregation effect. For example, what happens if everybody rates for whatever reason and completely drop the accusation / justification side of things. You can give one rating to any given individual for whatever reason you want. You can change it whenever you want. Everything is completely anonymous. This more accurately reflects how my opinion of you works anyhow - my opinion is valid and reflective of something no matter how unpopular I am. This makes the ratings less useful for reducing griefing though, and I imagine we start re-treading ground of other reputation systems. How to create an economy of reputation? Well maybe you can&#039;t, maybe reputation doesn&#039;t follow economic laws because it&#039;s not a finite resource...

My third idea is prompted by the failings of the second idea. Namely absolute reputation is essentially a public absolute referendum on any individual. That sucks! Sure it stops griefers but it also stops any kind of &quot;deviant&quot; behavior. So lets take the system and instead of making it absolute lets make it relative and make the effects work only along existing relationships. In other words lets get out of the reputation game and instead get into the trust game. Say I have an infinite amount of trust I&#039;m willing to share, but I don&#039;t share the same amount with everybody. If I trust Lex completely and I trust Torley somewhat, and Lex trusts me somewhat and Torley trusts me very little, then it could make sense that Lex should trust Torley very little and Torley should trust Lex not at all. Lex&#039;s trust of Torley in this case is based on the relationship which passes trust from one person to another - naturally decaying with distance and modified by the level of trust relating each person in the chain. Note that Lex&#039;s automatic trust of Torley is very little because she only has one relationship/endorsement though me, and that this would be true even if Torley was insanely popular and highly trusted by every person who knows Torley. Now I&#039;m not sure that this will also yield a useful way to stop griefers but I think it points towards a system that is not popularity based but based on a network of relationships. In the furry example I started with such a system *could*, *maybe*, lead to cliques where only furries can shop in certain districts and only non furries in others, but I think because it lacks a feedback loop it&#039;s much less likely to escalate that way. 

I dunno - what do you think?</description>
		<content:encoded><![CDATA[<p>This is a great idea, make people accountable and they might behave!</p>
<p>I like the basic approach of having a social solution to a social problem, and actually I think most of the adoption and utilization of a reputation would work. </p>
<p>My only worry is similar to Seifert. The system *can* be gamed, just like an election because there is literally a game system in place. It&#8217;s the game of who has the most friends who agree with him, also known as Politics or, my favorite middle school variant, Cliques vs Losers. Also, unless I misunderstood the proposal, it&#8217;s a negative feedback loop &#8211; so it&#8217;s easy for the winners to become very strong and the losers to get ground down. The more unpopular I am the less my opinion counts. </p>
<p>This could be even worse than the passionate people having the upper hand &#8211; as you have in a get out the vote scenario. Because the reputation system is based on popularity *and* alignment of opinion, you could have a reputation get trashed simply because they have unpopular opinions on their profile, or they engage in behavior is legit and protected by the terms of service but considered undesirable by some group. As an extreme example think of furries. Most people don&#8217;t think about them, some folks hate them, some of them hate non furries. What if folks (furry and not) started rating people based on their amount of fur? Someone who spends part time as a furry might be &#8220;outed&#8221; in their reputation profile. Ultimately the majority would win and furry folk would not be welcome in SL because their other activities were limited by their communal trashing of reputation. By the system they would be treated like a griefer bloc on a large scale. </p>
<p>Heck, any subject two folks could dissagree on, could eventually be subject to the global popularity test &#8211; ultimately resulting a population that thinks, acts and looks exactly the same about everything! </p>
<p>Eeep!</p>
<p>Well, why not? It&#8217;s actually very similar to the historical reasons minority groups have been persecuted. If you wear funny clothes I can tell you don&#8217;t belong in my night club, shouldn&#8217;t shop at my grocery store, etc. Only now we&#8217;re not just judging on outward signs of conformity and acceptance &#8211; every aspect of a person&#8217;s behavior and thoughts is publicly available for scrutiny and subject to a popularity test which might be grounds for segregation.</p>
<p>Actually, thinking about it, a more likely outcome to the furry scenario is different cliques adopting different reputation systems &#8211; less dystopian than a grid full of clones but just as sucky. The diversity in the grid only hangs out with their clones because those are the ones using the reputation system where their clique wins.</p>
<p>Simply put, my concerns are the way that reputations build on reputations (popularity contest), and the way that opinions or any attribute can become filters for reputations (segregation).</p>
<p>But since I don&#8217;t like to poke holes without being constructive, here&#8217;s a few random ideas which might help. ^_^</p>
<p>I wonder if there is not a more &#8216;natural&#8217; way of getting an aggregate reputation on somebody than voting and networks of endorsement? I know these are very popular techniques these days and are very useful for problems like rating website value, product quality or even product review trust worthiness. But maybe reputation requires something different. Maybe it should be based on how much time I spend interacting with someone. Time I spend with you works to give an implicit increase in reputation. Maybe there&#8217;s other actions that I can take, or that more importantly, groups of people can take together that act as endorsements. There is probably a lot to explore in this direction but I&#8217;m not sure if it could be done in such a way that a malicious person couldn&#8217;t run their own clients and servers to spoof good activity&#8230;</p>
<p>As another alternative entirely you could try and take the scheme as proposed and simplify it further with the goals of eliminating the popularity and the segregation effect. For example, what happens if everybody rates for whatever reason and completely drop the accusation / justification side of things. You can give one rating to any given individual for whatever reason you want. You can change it whenever you want. Everything is completely anonymous. This more accurately reflects how my opinion of you works anyhow &#8211; my opinion is valid and reflective of something no matter how unpopular I am. This makes the ratings less useful for reducing griefing though, and I imagine we start re-treading ground of other reputation systems. How to create an economy of reputation? Well maybe you can&#8217;t, maybe reputation doesn&#8217;t follow economic laws because it&#8217;s not a finite resource&#8230;</p>
<p>My third idea is prompted by the failings of the second idea. Namely absolute reputation is essentially a public absolute referendum on any individual. That sucks! Sure it stops griefers but it also stops any kind of &#8220;deviant&#8221; behavior. So lets take the system and instead of making it absolute lets make it relative and make the effects work only along existing relationships. In other words lets get out of the reputation game and instead get into the trust game. Say I have an infinite amount of trust I&#8217;m willing to share, but I don&#8217;t share the same amount with everybody. If I trust Lex completely and I trust Torley somewhat, and Lex trusts me somewhat and Torley trusts me very little, then it could make sense that Lex should trust Torley very little and Torley should trust Lex not at all. Lex&#8217;s trust of Torley in this case is based on the relationship which passes trust from one person to another &#8211; naturally decaying with distance and modified by the level of trust relating each person in the chain. Note that Lex&#8217;s automatic trust of Torley is very little because she only has one relationship/endorsement though me, and that this would be true even if Torley was insanely popular and highly trusted by every person who knows Torley. Now I&#8217;m not sure that this will also yield a useful way to stop griefers but I think it points towards a system that is not popularity based but based on a network of relationships. In the furry example I started with such a system *could*, *maybe*, lead to cliques where only furries can shop in certain districts and only non furries in others, but I think because it lacks a feedback loop it&#8217;s much less likely to escalate that way. </p>
<p>I dunno &#8211; what do you think?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: lex</title>
		<link>http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/comment-page-1/#comment-3</link>
		<dc:creator>lex</dc:creator>
		<pubDate>Tue, 09 Jan 2007 19:27:35 +0000</pubDate>
		<guid isPermaLink="false">http://www.lexneva.name/blog/2006/12/27/reputation-in-second-life/#comment-3</guid>
		<description>These are good points you bring up, and you may have hilighted a key weakness in my proposal.  

However, there&#039;s one key thing that separates the Reputation System from an election as far as getting people to come and vote.  In the real-world election scenario, there a relatively small (as you imply) incentive to come and vote against an issue that a vocal minority is pushing, and that&#039;s the reward of preventing that issue from passing.  In my system, not only do you get the reward of bringing consequences for someone who you think is being a jerk, but you ALSO get a dividend to your reputation for participating, assuming others agree with you.

This also applies to the &quot;bad reviews for bad reasons&quot; issue.  If you believe that someone gave a negative (or positive) review for a bad reason, you have two incentives for stating your opinion on the review: first, you diminish the effectiveness of the review, and second, you could quite possibly see your own reputation go up as a result.  As people become more familiar with the system and realize how important it is to have a high reputation score, my hope is that a high emphasis will be placed on wanting to raise your own reputation score, and that will make you weigh in on more issues, which will keep the system going.</description>
		<content:encoded><![CDATA[<p>These are good points you bring up, and you may have hilighted a key weakness in my proposal.  </p>
<p>However, there&#8217;s one key thing that separates the Reputation System from an election as far as getting people to come and vote.  In the real-world election scenario, there a relatively small (as you imply) incentive to come and vote against an issue that a vocal minority is pushing, and that&#8217;s the reward of preventing that issue from passing.  In my system, not only do you get the reward of bringing consequences for someone who you think is being a jerk, but you ALSO get a dividend to your reputation for participating, assuming others agree with you.</p>
<p>This also applies to the &#8220;bad reviews for bad reasons&#8221; issue.  If you believe that someone gave a negative (or positive) review for a bad reason, you have two incentives for stating your opinion on the review: first, you diminish the effectiveness of the review, and second, you could quite possibly see your own reputation go up as a result.  As people become more familiar with the system and realize how important it is to have a high reputation score, my hope is that a high emphasis will be placed on wanting to raise your own reputation score, and that will make you weigh in on more issues, which will keep the system going.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
