How-to: Fix The Spam Issue (from an online-game-coder's perspective)

[Seltha-Larren]

<p>I ran an online game (MUD) as a hobby for many years (until I got addicted to EQ2), and occasionally the issue of spamming/abuse of chat channels/creating multiple accounts came up.  While Everquest is a vastly bigger codebase compared to a MUD, I think the same principals apply to both games.</p><p>1) We gave the GM's and players in our equivalent of the "guide" program the ability to revoke chat privilages on a per-character and per-account basis.  For our "guide" players, we required that one raised the petition and a second guide confirmed the action (basically to prevent abuse of power).  They could also specify a timeframe for the chat ban (eg 1 hour, 6 hours, 24 hours, etc).  This action was logged for review by a "Dev" next time they came on, to make sure the ban was just.</p><p>This means that someone who starts spamming can be silenced temporarily until they are no-longer drunk, angry, hacked, or whatever it was that set them off.</p><p>2) We gave the GM's the ability to revoke log-in privialges on a per-character, per-account, per-ip, and per-subnet basis.  Again there was a timeframe specified and a log entry for Dev review later.</p><p>The "per-ip" and "per-subnet" bans mean that someone who's being a troublemaker and has already had their character or account banned and comes back with a new account to do it all again will find that they cannot sign up for a new account or log in using their current IP address or current internet provider (depending on the ban type).  This had the potential downside of legitimate players from the same ISP having their service rejected, so we also implemented a whitelist for existing accounts we knew not to be troublemakers, plus blacklisted permanently any proxying services used by the troublemakers.</p><p>3) We ran an automatic spam preventer on the channels.  If the same message appeared more than 3 times (from a history of the last 10 messages sent by the player) or if they sent out more than 10 messages in 10 seconds, they were automatically silenced for 5 minutes.  If after 5 minutes the spamming resumed they were automatically silenced for 15 minutes as well as notifying the GM's/guides for futher review.  Third offense was 60 minutes, and finally if no GM's/guides were available and the spamming resumed a fourth time they were forcibly logged out and their account was locked until a GM was available to review it.</p><p>We didn't automatically ban an ip/subnet, but this could be considered, especially since most people are on DSL/Cable connections these days and the IP's often stay the same as long as the router MAC address doesn't change.</p><p>Due to people running damage parsers these days, you may want to excempt certain channels from being monitored in this way.</p><p>Also, since you already have the "Report as spam" option, you could also tie in the "if the person is reported by 10 people, they are silenced for 5 minutes", etc.</p><p>Finally, since you control the client, you could capture hardware ID's from the machines being used to spam and physically ban the hardware used (eg hard drive, cpu id, network mac address, windows serial number, etc).  From a privacy perspective, store these banned hardware ID's in a table that isn't linked to account details so we don't get all up in arms about 'oh noes, sony knows what my hard drive's serial number is, the world has ended!!'.</p><p>....</p><p>Hopefully these insights are useful in the war against spam, and I haven't been teaching tooo many grandmas on how to suck eggs.</p><p>-- Seltha</p>

[Zarador]

<cite>Seltha-Larren wrote:</cite><blockquote><p>I ran an online game (MUD) as a hobby for many years (until I got addicted to EQ2), and occasionally the issue of spamming/abuse of chat channels/creating multiple accounts came up.  While Everquest is a vastly bigger codebase compared to a MUD, I think the same principals apply to both games.</p><p>1) We gave the GM's and players in our equivalent of the "guide" program the ability to revoke chat privilages on a per-character and per-account basis.  For our "guide" players, we required that one raised the petition and a second guide confirmed the action (basically to prevent abuse of power).  They could also specify a timeframe for the chat ban (eg 1 hour, 6 hours, 24 hours, etc).  This action was logged for review by a "Dev" next time they came on, to make sure the ban was just.</p><p>This means that someone who starts spamming can be silenced temporarily until they are no-longer drunk, angry, hacked, or whatever it was that set them off.</p><p>2) We gave the GM's the ability to revoke log-in privialges on a per-character, per-account, per-ip, and per-subnet basis.  Again there was a timeframe specified and a log entry for Dev review later.</p><p>The "per-ip" and "per-subnet" bans mean that someone who's being a troublemaker and has already had their character or account banned and comes back with a new account to do it all again will find that they cannot sign up for a new account or log in using their current IP address or current internet provider (depending on the ban type).  This had the potential downside of legitimate players from the same ISP having their service rejected, so we also implemented a whitelist for existing accounts we knew not to be troublemakers, plus blacklisted permanently any proxying services used by the troublemakers.</p><p>3) We ran an automatic spam preventer on the channels.  If the same message appeared more than 3 times (from a history of the last 10 messages sent by the player) or if they sent out more than 10 messages in 10 seconds, they were automatically silenced for 5 minutes.  If after 5 minutes the spamming resumed they were automatically silenced for 15 minutes as well as notifying the GM's/guides for futher review.  Third offense was 60 minutes, and finally if no GM's/guides were available and the spamming resumed a fourth time they were forcibly logged out and their account was locked until a GM was available to review it.</p><p>We didn't automatically ban an ip/subnet, but this could be considered, especially since most people are on DSL/Cable connections these days and the IP's often stay the same as long as the router MAC address doesn't change.</p><p>Due to people running damage parsers these days, you may want to excempt certain channels from being monitored in this way.</p><p>Also, since you already have the "Report as spam" option, you could also tie in the "if the person is reported by 10 people, they are silenced for 5 minutes", etc.</p><p>Finally, since you control the client, you could capture hardware ID's from the machines being used to spam and physically ban the hardware used (eg hard drive, cpu id, network mac address, windows serial number, etc).  From a privacy perspective, store these banned hardware ID's in a table that isn't linked to account details so we don't get all up in arms about 'oh noes, sony knows what my hard drive's serial number is, the world has ended!!'.</p><p>....</p><p>Hopefully these insights are useful in the war against spam, and I haven't been teaching tooo many grandmas on how to suck eggs.</p><p>-- Seltha</p></blockquote>Lets see....<ol><li>Volunteers permitted to take actions against paying accounts?  Sorry, I don't like that concept at all. GM's are the only ones permitted to take such actions in game as were the paying customers and their the paid staff.  </li><li>Even worse, players allowed, with a consensual opinion to take actions against other players?  Like that idea even less.</li><li>If I send the same message, 3x in a row, because it records the cue, I'm silenced?  No time frame noted, just don't say the same thing 3x? Must be great if your looking for a group, recruiting for a Guild or trying to sell an item in auction. </li><li>Snooping and gathering Machine information on the client side?  I believe without consent that breaks the privacy act laws in most states.</li><li>Banning an entire location, not just an account? Sounds a bit extreme here.  If I play in a dorm or a military installation or anything of that sort, I can no longer play because some stranger was banned previously?  I'm sure internet cafe's would love that concept.</li><li>If 10 people, (PvP comes to mind) don't like me, I can be banned or silenced before an investigation is conducted by a real employee of the company I'm paying? Bad idea again.</li></ol>So basically....you want us to turn over all our private information to a company to snoop into all of our clients while allowing players to control the community.  No Thanks!!!

[shadowlord12763]

<cite>Seltha-Larren wrote:</cite><blockquote><p>Finally, since you control the client, you could capture hardware ID's from the machines being used to spam and physically ban the hardware used (eg hard drive, cpu id, network mac address, windows serial number, etc).  From a privacy perspective, store these banned hardware ID's in a table that isn't linked to account details so we don't get all up in arms about 'oh noes, sony knows what my hard drive's serial number is, the world has ended!!'.</p></blockquote><p>This would be the best option ever for EQ2 to combat spam. It would totally eliminate the computer that the spammer uses and would never be able to use that computer again in EQ2. This process is almost like WOWs Warden and Scan.dll programs. They snoop your computer  and get this info automatically.</p><p>With the other options the spammer can just use another proxy ISP or reset their ISP to another number and still be able to spam forever without consequences. There are tons of programs that will mask your ISP and even give fake ISP info. Blocking ISPs will never work and is a never ending battle.</p>

[Seltha-Larren]

<p>Zarador, I don't think you read the post properly.  Firstly, I stated what worked for us in our situation.  Sony can take as little or as much from this as they wish.</p><p>re: 1) Because we didn't have a 24/7 GM staff, giving the player "guide" characters (which we trusted as responsible players who wouldn't abuse the power) was a good solution for us.  If Sony don't have 24/7 GM's then perhaps they could consider it too.  If they do, it's not going to be an issue is it.</p><p>re: 2) Players were never given the permission to ban, only GM's had that power.  Responsible "trusted" players (ie ones who had gone through screening and been approved by the Dev staff) were given the ability to silence.  When I say trusted, I don't mean Dev's friends, I mean mature responsible players who wouldn't abuse their powers.</p><p>re: 3) Personally, the amount of time I see people spamming for group or spamming to sell some master that nobody wants or spamming the praises of their guild on every channel in the game IS still spam.  There is a LFG tool for advertising your desire to group.  There is the broker for selling your wares.  There is a recruitment tool for guilds.  There is no reason that people should be spamming the channels for any of those things anymore.</p><p>re: 4) Who said anything about "without consent".  You have to click "I Agree" to the EULA every time you launch EQ2, it could easily be added as a condition of you playing the game you consent to Sony looking at your hardware ID's (anonymously).</p><p>re: 5) Yes, it is an extreme measure.  We had troublemakers coming from the same university campus as some legit players so we did have to flag the legit ones as being "ok" and we had to manually approve new accounts from that site.  It created an admin overhead like I previously stated, but it stopped the troublemakers so the net result was positive.</p><p>re: 6) Given that the 10 man raid force from the opposite team isn't going to see your tell/chat/say anyway (since communication between good/evil is blocked) I don't think this is an issue.  10 is also a number pulled out of a hat, it could be 50, or whatever threshold SOE feel is a good level - they will already know how many "report as spam" clicks they are getting, so if they get 100 reports on average to every 1 spam message sent out, then a good threshold might be 40 clicks.</p><p>re: Shadowlord12763: Yeah, we had to ban proxies, but the ISP thing worked quite well (eg you could do the reverse lookup, eg "x2735982.chicago.msn.com" or similar, and just go "well lets ban new accounts from "*.chicago.msn.com" for 2 hours.  That was usually long enough for the troublemaker to go find somewhere else to cause trouble, and short enough that the likelyhood of a legitimate account starting up from chicago inside of that 2 hour window is fairly slim.  Of course we weren't dealing with the same hundreds-of-thousands of accounts that Sony does, so they are likely to know if a total ban for 2 hours is acceptable, or if it would be better to just automatically bring new accounts from "*.chicago.msn.com" to a GM's attention, where they could watch to see if the first thing they do is /tell <randomperson> r u wan buy plats! or if they start playing properly.</p><p>-- Seltha</p>

[StormCinder]

<cite>Seltha-Larren wrote:</cite><blockquote><p>Zarador, I don't think you read the post properly.  Firstly, I stated what worked for us in our situation.  Sony can take as little or as much from this as they wish.</p><p>re: 1) Because we didn't have a 24/7 GM staff, giving the player "guide" characters (which we trusted as responsible players who wouldn't abuse the power) was a good solution for us.  If Sony don't have 24/7 GM's then perhaps they could consider it too.  If they do, it's not going to be an issue is it.</p><p>re: 2) Players were never given the permission to ban, only GM's had that power.  Responsible "trusted" players (ie ones who had gone through screening and been approved by the Dev staff) were given the ability to silence.  When I say trusted, I don't mean Dev's friends, I mean mature responsible players who wouldn't abuse their powers.</p><p>re: 3) Personally, the amount of time I see people spamming for group or spamming to sell some master that nobody wants or spamming the praises of their guild on every channel in the game IS still spam.  There is a LFG tool for advertising your desire to group.  There is the broker for selling your wares.  There is a recruitment tool for guilds.  There is no reason that people should be spamming the channels for any of those things anymore.</p><p>re: 4) Who said anything about "without consent".  You have to click "I Agree" to the EULA every time you launch EQ2, it could easily be added as a condition of you playing the game you consent to Sony looking at your hardware ID's (anonymously).</p><p>re: 5) Yes, it is an extreme measure.  We had troublemakers coming from the same university campus as some legit players so we did have to flag the legit ones as being "ok" and we had to manually approve new accounts from that site.  It created an admin overhead like I previously stated, but it stopped the troublemakers so the net result was positive.</p><p>re: 6) Given that the 10 man raid force from the opposite team isn't going to see your tell/chat/say anyway (since communication between good/evil is blocked) I don't think this is an issue.  10 is also a number pulled out of a hat, it could be 50, or whatever threshold SOE feel is a good level - they will already know how many "report as spam" clicks they are getting, so if they get 100 reports on average to every 1 spam message sent out, then a good threshold might be 40 clicks.</p><p>re: Shadowlord12763: Yeah, we had to ban proxies, but the ISP thing worked quite well (eg you could do the reverse lookup, eg "x2735982.chicago.msn.com" or similar, and just go "well lets ban new accounts from "*.chicago.msn.com" for 2 hours.  That was usually long enough for the troublemaker to go find somewhere else to cause trouble, and short enough that the likelyhood of a legitimate account starting up from chicago inside of that 2 hour window is fairly slim.  Of course we weren't dealing with the same hundreds-of-thousands of accounts that Sony does, so they are likely to know if a total ban for 2 hours is acceptable, or if it would be better to just automatically bring new accounts from "*.chicago.msn.com" to a GM's attention, where they could watch to see if the first thing they do is /tell <randomperson> r u wan buy plats! or if they start playing properly.</p><p>-- Seltha</p></blockquote><p>While bold to post technical suggestions in a forum that is notoriously trolled by would-be programmers, I too have to take issue with your suggestions.</p><p>SOE does not have 24/7 GMs, so that entire portion of your suggestion is not workable.  There are historically countless other more important justifications for 24/7 GMs and yet they still do not exist.  I doubt something like banning spammers will change their minds.</p><p>Snooping a PC is a compete and utter show-stopper for myself and I'm sure many others.  I believe SOE/Sony is a still a little gunshy about doing like that after the last time they tried it.</p><p>As you said in a previous post, banning groups and then making exceptions for known "good" accounts might work on a small scale, but not when you're dealing with 100s of thousands of accounts.</p><p>Your #3 in the post I quoted is EXTREMELY alarming.  You are unilaterally deciding a new definition (YOUR definition) of spam?  What about the guy who sends out his "joke-of-the-hour" in 1-9 chat...that's annoying too.  Can I classify that as spam?  </p><p>I prefer the current 2-3 spams per week I currently get to the draconian measures you're are suggesting.  You are swatting flies with a sledgehammer.</p><p>I believe Zarador did read your post correctly.  I agree with Zarador.  Sony should take none of your suggestions.</p><p>SC</p>

[KefkaQ]

Lord of the Rings Online has/had a serious problem where gold spammers would sit in the main town and span the OOC channel (the main general chat in that game) Over and over and over again, completely flooding the channel. They implemented a command where you right click the spammer's name, choose 'report as spam', and it automatically puts them on ignore for you while at the same time submitting a spam report to the gms. Gms see 30-100 spam reports come in, go check the logs, and then deal with the account as necessary. It's proven effective. I assume it just routes all those particularly tickets to a special queue, making it easier for them to deal with.

[Chefren]

<cite>shadowlord12763 wrote:</cite><blockquote><cite>Seltha-Larren wrote:</cite><blockquote><p>Finally, since you control the client, you could capture hardware ID's from the machines being used to spam and physically ban the hardware used (eg hard drive, cpu id, network mac address, windows serial number, etc).  From a privacy perspective, store these banned hardware ID's in a table that isn't linked to account details so we don't get all up in arms about 'oh noes, sony knows what my hard drive's serial number is, the world has ended!!'.</p></blockquote><p>This would be the best option ever for EQ2 to combat spam. It would totally eliminate the computer that the spammer uses and would never be able to use that computer again in EQ2. This process is almost like WOWs Warden and Scan.dll programs. They snoop your computer  and get this info automatically.</p></blockquote>So you want EQ2 to become spyware and rely on something you can circumvent with virtualization? No, no, no.

[Armawk]

<cite>KefkaQ wrote:</cite><blockquote>Lord of the Rings Online has/had a serious problem where gold spammers would sit in the main town and span the OOC channel (the main general chat in that game) Over and over and over again, completely flooding the channel. They implemented a command where you right click the spammer's name, choose 'report as spam', and it automatically puts them on ignore for you while at the same time submitting a spam report to the gms. Gms see 30-100 spam reports come in, go check the logs, and then deal with the account as necessary. It's proven effective. I assume it just routes all those particularly tickets to a special queue, making it easier for them to deal with.</blockquote>Sounds alot like the spam report feature in eq2, except for the auto-ignore.

[Ama]

<cite>Seltha-Larren wrote:</cite><blockquote><p>I ran an online game (MUD) as a hobby for many years (until I got addicted to EQ2), and occasionally the issue of spamming/abuse of chat channels/creating multiple accounts came up.  While Everquest is a vastly bigger codebase compared to a MUD, I think the same principals apply to both games.</p><p>1) We gave the GM's and players in our equivalent of the "guide" program the ability to revoke chat privilages on a per-character and per-account basis.  For our "guide" players, we required that one raised the petition and a second guide confirmed the action (basically to prevent abuse of power).  They could also specify a timeframe for the chat ban (eg 1 hour, 6 hours, 24 hours, etc).  This action was logged for review by a "Dev" next time they came on, to make sure the ban was just.</p><p>This means that someone who starts spamming can be silenced temporarily until they are no-longer drunk, angry, hacked, or whatever it was that set them off.</p><p><span style="color: #990000;">This won't do much and right now the anti-spam system is working imho so why add something to what's already working.  Sure 1-2 spams get through, but those are caught relatively quickly.  Plus i'm betting EQ2 does already have a flag system which indicates to GMs/CSRs what acounts are spamming. </span></p><p>2) We gave the GM's the ability to revoke log-in privialges on a per-character, per-account, per-ip, and per-subnet basis.  Again there was a timeframe specified and a log entry for Dev review later.</p><p>The "per-ip" and "per-subnet" bans mean that someone who's being a troublemaker and has already had their character or account banned and comes back with a new account to do it all again will find that they cannot sign up for a new account or log in using their current IP address or current internet provider (depending on the ban type).  This had the potential downside of legitimate players from the same ISP having their service rejected, so we also implemented a whitelist for existing accounts we knew not to be troublemakers, plus blacklisted permanently any proxying services used by the troublemakers.</p><p><span style="color: #990000;">Sadly I see this as more of a problem than a help since plat sellers are getting wise.  There was one article where they said they had IP address listing them in England and such instead of china.  </span></p><p>3) We ran an automatic spam preventer on the channels.  If the same message appeared more than 3 times (from a history of the last 10 messages sent by the player) or if they sent out more than 10 messages in 10 seconds, they were automatically silenced for 5 minutes.  If after 5 minutes the spamming resumed they were automatically silenced for 15 minutes as well as notifying the GM's/guides for futher review.  Third offense was 60 minutes, and finally if no GM's/guides were available and the spamming resumed a fourth time they were forcibly logged out and their account was locked until a GM was available to review it.</p><p><span style="color: #990000;">The spam preventer is pretty much in at the moment. </span></p><p>We didn't automatically ban an ip/subnet, but this could be considered, especially since most people are on DSL/Cable connections these days and the IP's often stay the same as long as the router MAC address doesn't change.</p><p>Due to people running damage parsers these days, you may want to excempt certain channels from being monitored in this way.</p><p>Also, since you already have the "Report as spam" option, you could also tie in the "if the person is reported by 10 people, they are silenced for 5 minutes", etc.</p><p>Finally, since you control the client, you could capture hardware ID's from the machines being used to spam and physically ban the hardware used (eg hard drive, cpu id, network mac address, windows serial number, etc).  From a privacy perspective, store these banned hardware ID's in a table that isn't linked to account details so we don't get all up in arms about 'oh noes, sony knows what my hard drive's serial number is, the world has ended!!'.</p><p><span style="color: #990000;">This starts to fall into the "Ethical" realm when you start banning actual computers.  The main problem is not the fact plat sellers are selling plat, but people are actually buying it.  </span></p><p>....</p><p>Hopefully these insights are useful in the war against spam, and I haven't been teaching tooo many grandmas on how to suck eggs.</p><p>-- Seltha</p></blockquote><p>Personally to me what needs to happen is SoE needs to come out and say "Buying Plat Is Illegal".  After that they need to buy their lawyers some combat gear and start cracking heads on the plat sellers.  These people are not simple buisness people, but are people that can get real nasty harrassing gamers.  They are the ones that farm masters to make the plat then sell that plat for real world currency. </p><p>Another thing is we need to get the FBI involved on some of these cases.  People going to plat seller sites and then having their information stolen.  Hey the people are dumb/stupid for going to the plat site in the first place, but the action of the person running that site isn't a saint.</p>

[Rqron]

<cite>Amana wrote:</cite><blockquote><cite></cite><p>Personally to me what needs to happen is SoE needs to come out and say "Buying Plat Is Illegal".  After that they need to buy their lawyers some combat gear and start cracking heads on the plat sellers.  These people are not simple buisness people, but are people that can get real nasty harrassing gamers.  They are the ones that farm masters to make the plat then sell that plat for real world currency. </p><p>Another thing is we need to get the FBI involved on some of these cases.  People going to plat seller sites and then having their information stolen.  Hey the people are dumb/stupid for going to the plat site in the first place, but the action of the person running that site isn't a saint.</p></blockquote>You are right in one point and I would say it is the most important first step that needs to be taken.SOE NEEDS TO DECLARE PLAT BUYING ILLEGAL. Moreover they have to do it in a way that will drive it home once and all..something like a window you will have to click on every time you log in..in <b>addition</b> to the Eula that nobody reads. A window that declares in big bold letters that everyone buying plat and being found out will be banned for life from ALL SOE games.I can't count how often I see debates where players state that buying plat is not against the Eula and that SOE will do nothing about it. Once it is made clear to everyone then the campaign against the plat farmers can start...everything else will not succeed. It has to start with the education of the player base.J.C

[NiteWolfe]

<cite>Rqron wrote:</cite><blockquote><cite>Amana wrote:</cite><blockquote><cite></cite><p>Personally to me what needs to happen is SoE needs to come out and say "Buying Plat Is Illegal".  After that they need to buy their lawyers some combat gear and start cracking heads on the plat sellers.  These people are not simple buisness people, but are people that can get real nasty harrassing gamers.  They are the ones that farm masters to make the plat then sell that plat for real world currency. </p><p>Another thing is we need to get the FBI involved on some of these cases.  People going to plat seller sites and then having their information stolen.  Hey the people are dumb/stupid for going to the plat site in the first place, but the action of the person running that site isn't a saint.</p></blockquote>You are right in one point and I would say it is the most important first step that needs to be taken.<span style="color: #cc0000;"><b>SOE NEEDS TO DECLARE PLAT BUYING ILLEGAL. </b></span>Moreover they have to do it in a way that will drive it home once and all..something like a window you will have to click on every time you log in..in <b>addition</b> to the Eula that nobody reads. A window that declares in big bold letters that everyone buying plat and being found out will be banned for life from ALL SOE games.I can't count how often I see debates where players state that buying plat is not against the Eula and that SOE will do nothing about it. Once it is made clear to everyone then the campaign against the plat farmers can start...everything else will not succeed. It has to start with the education of the player base.J.C</blockquote> It is NOT ILLEGAL! its just against the soe rules.  ILLEGAL would imply that you could be charged with a crime. This is just flat out  breaking SOE rules not a criminal act!

[Vith]

Or you could just ignore it like most people do.

[Zarador]

<cite>Seltha-Larren wrote:</cite><blockquote><p>Zarador, I don't think you read the post properly.  Firstly, I stated what worked for us in our situation.  Sony can take as little or as much from this as they wish.</p><p>re: 1) Because we didn't have a 24/7 GM staff, giving the player "guide" characters (which we trusted as responsible players who wouldn't abuse the power) was a good solution for us.  If Sony don't have 24/7 GM's then perhaps they could consider it too.  If they do, it's not going to be an issue is it.</p><p>re: 2) Players were never given the permission to ban, only GM's had that power.  Responsible "trusted" players (ie ones who had gone through screening and been approved by the Dev staff) were given the ability to silence.  When I say trusted, I don't mean Dev's friends, I mean mature responsible players who wouldn't abuse their powers.</p><p>re: 3) Personally, the amount of time I see people spamming for group or spamming to sell some master that nobody wants or spamming the praises of their guild on every channel in the game IS still spam.  There is a LFG tool for advertising your desire to group.  There is the broker for selling your wares.  There is a recruitment tool for guilds.  There is no reason that people should be spamming the channels for any of those things anymore.</p><p>re: 4) Who said anything about "without consent".  You have to click "I Agree" to the EULA every time you launch EQ2, it could easily be added as a condition of you playing the game you consent to Sony looking at your hardware ID's (anonymously).</p><p>re: 5) Yes, it is an extreme measure.  We had troublemakers coming from the same university campus as some legit players so we did have to flag the legit ones as being "ok" and we had to manually approve new accounts from that site.  It created an admin overhead like I previously stated, but it stopped the troublemakers so the net result was positive.</p><p>re: 6) Given that the 10 man raid force from the opposite team isn't going to see your tell/chat/say anyway (since communication between good/evil is blocked) I don't think this is an issue.  10 is also a number pulled out of a hat, it could be 50, or whatever threshold SOE feel is a good level - they will already know how many "report as spam" clicks they are getting, so if they get 100 reports on average to every 1 spam message sent out, then a good threshold might be 40 clicks.</p><p>re: Shadowlord12763: Yeah, we had to ban proxies, but the ISP thing worked quite well (eg you could do the reverse lookup, eg "x2735982.chicago.msn.com" or similar, and just go "well lets ban new accounts from "*.chicago.msn.com" for 2 hours.  That was usually long enough for the troublemaker to go find somewhere else to cause trouble, and short enough that the likelyhood of a legitimate account starting up from chicago inside of that 2 hour window is fairly slim.  Of course we weren't dealing with the same hundreds-of-thousands of accounts that Sony does, so they are likely to know if a total ban for 2 hours is acceptable, or if it would be better to just automatically bring new accounts from "*.chicago.msn.com" to a GM's attention, where they could watch to see if the first thing they do is /tell <randomperson> r u wan buy plats! or if they start playing properly.</p><p>-- Seltha</p></blockquote>I think I read your post quite clearly.  <ol><li>While I don't expect a company to jump through hoops for the subscription price that we all pay, I also don't expect the company to rely on volunteer assistance in account matters. Guides have been extremely useful in the past and present, but also rightfully limited in their authority.  People (I cringe using this word) invest in their characters.  many like to feel that any action taken on their account was done responsibly by the company, not their fellow players. </li><li>We all as players have the ability to "Silence" someone, it's called ignore.  I would not like other players, self included, to have the ability to "Silence" others from their side. Even the best of players have "bad" days where they lack judgment.  I can see this power being abused either out of mischief or malice between players.  I can especially see this being a mess for the real powers to be to mop up on the PvP side.</li><li>Your definition of spam oddly circumvents why this who concept would be a bad idea.  Someone has a LFG Macro that they use when they login. After the third use, it's now am offense that gets them silenced?  Someone recruits in the chat channel, they get banned or create a new message every time?  Can't say I like that either. What is a good reason to use the channels? We have class sites? We have information sites, even official ones, so why ask any questions as well? </li><li>You mention that we would all be consenting to be "snooped" by simply accepting the Eula.  I honestly don't think many people are very comfortable with that concept.  Even large credit card companies and clearing houses have been hacked in the past with information about their clients stolen.  If Citi-Corp could not keep such sensitive information private from hackers, I just as soon not provide such information to reduce a few spam messages in a game. </li><li>It's a very extreme measure and one not needed.  I'm going to block a internet cafe, dorm or even military computer recreation center where many players who are deployed, traveling or at school share the same addresses, just so someone does not get unwanted spam?  I've made many friends in online games over the years that are service members on deployment playing with their stateside friends and family.  Yeah, lets block accounts for a few hours from a single domain like as you say "chicago.msn.com" to keep one annoying account from that domain from spamming.  That would last about as long as it takes 100 players from there to cancel their accounts. </li></ol>