Guild Wars 2

The Reason as to why I think we need this
In other mmos that allow addons we the community could code our own spam blockers that were based on keyword and character level methods. Now I understand why there is no need for an addon system in GW2.

The main reason is because Spam is totally out of control its in the mail system that can possibly for some of us fill our collectable mail amount up with spam messages and its in PvP and Pve Instanced worlds. The second reason as to why we need to use a new system is because reporting is simply not effective it is slow and only costs your company recurring money and time.

Examples of cases where I have seen the spam

• In game mail – being filled with spam mail.
• PvP Ranked Instances not just Myst but also inside the match I have seen afk spam.
• Lions Arch flooded with spam on some servers.
• PvE Dungeons.

The Concept for the Solution
We implement an instant chat block when the chat is detected to have multiple lines and contain the main keywords, gold, sale, www., .com, cheap. The system would be pretty easy to implement and is very cost and time effective for the company to do implement.

Each key word would be checked to see if it match’s other key words in the filter such as Gold and the address that would contain a www. and a .com however it would also check for capital letter variants and other such symbols. This method would be applied to both the in-game chat and the mail system.

If a chat message was to contain the following keywords all detected in the single message or over the span of 3 messages multi-line included it would automatically flag the spam system currently set in place that blocks excessive use of the the same message in other words the very system that exists already that stops us from spamming repeated content recruit and trade messages too frequently could be used.

This system if to be put into action would stop the flow of gold spam by up to 90% now that is a based on current spam wording.

The keyword system is both low resource intensive and highly adaptable. If you wanted to take it a step further you could start analyzing the chat and mail for structure patterns and use those algorithms in your detection of spam content but the keyword system if used in its basic form which is to detect common spam keywords it would be very low resource cost quick to implement .

I call this a temporary solution because the sellers will adapt to this over time but for the initial period of use that it would receive it will certainly slow down and decrease the amount of spam that is currently plaguing the chat and mail system currently. It may also have a small con in which it could possibly false flag gold guide link text as spam but if the player chooses his wording carefully it should not be a impacting issue besides you could always turn off the filter for custom channels which would provide a means for players to avoid triggering the spam detection and still share those links. Also its is highly likely that spammers will use a lot of symbols and would further increase the accurate block rate.

You have something to gain from this too Arena Net!
Arena Net you could be saving your selves days worth of ticket report checking for banning seller accounts time which could be better spent doing something more constructive or could just give some of the staff their weekends back and save the company money in the long run.

This solution really cannot make the situation any worse it can only make it better for your customers and for your support staff to deal with.

Conclusion
I believe this system will work it is a proven effective method, It is cost/time effective, it is greatly needed. I hope you do read this thread and take it seriously. I give you all my best regards you have made a terrific game.

A filter system won’t work, gold spammers just annoy players by using x = y+z translating to com some how.

Personally I think there should be a seperate gold spam report button feature attached to chat and mailbox, and a new dynamic and innovating system to support such a feature.

If an account picks up “x” amount of reports, then that account should be muted, unable to communicate with others via chat or mailbox while pending investigation.

Nick Naughty.4690:

A filter system won’t work, gold spammers just annoy players by using x = y+z translating to com some how.

Personally I think there should be a seperate gold spam report button feature attached to chat and mailbox, and a new dynamic and innovating system to support such a feature.

If an account picks up “x” amount of reports, then that account should be muted, unable to communicate with others via chat or mailbox while pending investigation.

You mean like Ascii methods these are blockable as well by blocking multi-line text that is made up primarily of single repeated letters and symbols spaced or apart.

In regards to your reporting function this is abusable if intended to gang up on a particular person, I don’t know how many reports would be needed to pass this action but players could report anyone and mute them for the duration of investigation if they choose to rally up and troll someone. While I support the notion that a report option be present the mute function imo would need very specific set of rules to pass in order to mute.

Maybe a combined keyword check or if the player has continuously triggered the regular spam prevention system before the mute notion would take effect.

LED Head.2439:

Nick Naughty.4690:

A filter system won’t work, gold spammers just annoy players by using x = y+z translating to com some how.

Personally I think there should be a seperate gold spam report button feature attached to chat and mailbox, and a new dynamic and innovating system to support such a feature.

If an account picks up “x” amount of reports, then that account should be muted, unable to communicate with others via chat or mailbox while pending investigation.

You mean like Ascii methods these are blockable as well by blocking multi-line text that is made up primarily of single repeated letters and symbols spaced or apart.

Any filter method will be pointless if they find a way round it, and they would. You could remove chat and mailbox and some how they would find a method.

I say give the some limited power to the players, if an account is gold spamming, then players should report, and after x amount of reports that account should be muted and not heard from untill arenanet investigates what is going on.

On that type of system I could port to lion’s arch and not see gold spam, because the spammer spamming gold spam has been reported 100 times already and is muted. Not forgetting on that type of system the more players the gold spammer effects, the more reports the spammer will get.

Nick Naughty.4690:

LED Head.2439:

Nick Naughty.4690:

A filter system won’t work, gold spammers just annoy players by using x = y+z translating to com some how.

Personally I think there should be a seperate gold spam report button feature attached to chat and mailbox, and a new dynamic and innovating system to support such a feature.

If an account picks up “x” amount of reports, then that account should be muted, unable to communicate with others via chat or mailbox while pending investigation.

You mean like Ascii methods these are blockable as well by blocking multi-line text that is made up primarily of single repeated letters and symbols spaced or apart.

Any filter method will be pointless if they find a way round it, and they would. You could remove chat and mailbox and some how they would find a method.

I say give the some limited power to the players, if an account is gold spamming, then players should report, and after x amount of reports that account should be muted and not heard from until arenanet investigates what is going on.

On that type of system I could port to lion’s arch and not see gold spam, because the spammer spamming gold spam has been reported 100 times already and is muted. Not forgetting on that type of system the more players the gold spammer effects, the more reports the spammer will get.

Ok but investigation is useless they will just return with another account like they already do, secondly reporting slows down the report que process the more times you report a player the longer the time it takes to actually sort the reports out.

I have seen many forms of spam I have literally seen probably every single type of possible mmo spam you can think of, the only guaranteed way to bypass this system is to use physical objects in the world to spell out your spam content. I get where you coming from and I reworded my post as you replied I assume because your quote is missing my edits.

I agree to give some power to the players but any method that requires formal investigation from the support staff is a bad protection method as its not preventing anything its only dealing with the offenders not silencing them before they can spam.

Your solution is to have players go out their way to formally lodge an investigation ticket against a player knowing full well the spammer would have already spammed his messages for at least 30mins prior to actually being muted because not everyone is going to bother to report they are more likely to just mute him for themselves and by the time the instance sees 100 or more people he would have moved on already and switched characters or created a new one.

Any solution that requires interaction is a broken solution when it comes to administration of a widespread and re occurring issue every developer knows that. I can guarantee that my solution will prevent up to 90% of the spam that is text based it can be simple text, ascii style art, spaced out letters anything at all that can be sent with text can be prevented.

My solution is not new it is tried and tested and proven to work in games such as WoW where spam was a MAJOR issue until we made an addon to analyze the patterns in the text and block the chat from sight it could even auto report.

Of course if you wanted to block spaced out text like “w w w . s o m e t h i ng – g o l d . c o m” you would need to do a little more than simple key phrase checks you would need to do paragraph wild card formatting checks these are also pretty low resource cost and easily modified.

If there are logical ways to bypass the system then there are equally effective ways to detect it there is only so much you can do with the chat system before you hit intended limitations. Not to mention this is a temporary concept to alleviate the bulk of common spam. There is no method to post pictures into the chat so you can rule out image spam, and as I have said already you can already easily make an algorithm to detect ascii picture art type spam.

It would also be possible to detect character names that do not form any real words such as “dffklq” but that goes beyond my solution.

If your going to continue to try and argue your point can you at least validate why my solution is invalid, this validation would need to include a a summery of methods that this would not effect that are text based and also where you see the flaws of this method and how they could be exploited.

Simply saying " This wont work, This method is better, your method they can find a workaround" does not justify anything and just borderlines repeating your argument against mine.

If spam hits the filter system, then prevention has already failed, leave the prevention to account security, like secret questions, IP addresses etc, or type in a character name which was used in guild wars.

I say unite and act against unwanted spam, a spammer could pull out 6000 accounts but once they’re all muted, peace and quiet. Not forgetting thats a lot of accounts to be banned.

Nick Naughty.4690:

If spam hits the filter system, then prevention has already failed, leave the prevention to account security, like secret questions, IP addresses etc, or type in a character name which was used in guild wars.

I say unite and act against unwanted spam, a spammer could pull out 6000 accounts but once they’re all muted, peace and quiet. Not forgetting that’s a lot of accounts to be banned.

Actually all chat will pass the filter system and be prevented before its displayed saving packet data to the client as well as interaction being required, also its universal not tied to any account so it would not matter how many accounts the spammers use it will be blocked before the intended audience sees it it serves as both a barrier and a re occurrence prevention system as the sellers would be less likely to target the mmo as frequently knowing how much effort and continual effort they would need to go through to actually efficiently broadcast a message a lot of the smaller time spammers would probably drop out of bothering to try which is exactly what we want, also at the same time it could submit its own report because the filter flagged it as spam and can be investigated if needed but investigation would not be a mandatory requirement to actually prevent future occurrence of visible spam from the seller.

Your accusation of this type of system being a prevention failure is completely invalid for reasons stated above and reasons stated in my other posts about the core functionality of the process.

Also you are using the word prevention out of context, going by your accusation a computer or dedicated hardware based firewall has failed to block the data flow simply because the data has hit the firewall.

That is entirely incorrect of course it is in fact the complete opposite, the mere fact in relation to this thread if spam hit the filter than filter is doing something right as it has detected the content to be possible spam and is phrasing it through the filter rules and if it is doing this than objective is achieved.

Your ideology is only going to encourage the spammers to steal more accounts because the others have been banned it is not preventing anything nor is it actually stopping future occurrences from the same spammer group. Yes they could pull 6000 accounts and have them muted but then what will happen, My guess is that they will try to scam accounts more frequently to pull another 6000 accounts and the process repeats.

I really do not see why you continue to fight against this solution especially when your not even giving any solid reasoning to backup why your solution is better. I have at least to my knowledge invalidated your method in numerous ways now. I have also clarified and given counter validation in numerous scenarios to help my claim.

Please understand that while I respect that you disagree I feel it is just not only incorrect but it is also comes with unwanted side effects that out weigh my solutions side effects.

Some these side effects I have not 100% brought to light but only because they are minor in the big scope of things, such as they continue to try harder using the accounts they already have which slows the account theft rate substantially but as I have already said we can modify the system to counter anything new they throw at us that is done via any kind of chat mechanics in game and as far as I am aware the mechanics of the game so far do not allow exploitable emotes or animations that could display a message using game assets.

You do know how the current report spam feature works don’t you ?

You do know its like 3 clicks away to report someone in chat and 2 clicks for mailbox ?

You also do know that its not character related and infact account ?

You mentioned WoW, how many accounts does that game have which exploits gold ?

Would you like me just to leave it with I don’t agree with your suggestion, because spam finds its way round, and that WoW is a bad example and I wouldn’t take any ideas from that game.

Nick Naughty.4690:

You do know how the current report spam feature works don’t you ?

You do know its like 3 clicks away to report someone in chat and 2 clicks for mailbox ?

You also do know that its not character related and infact account ?

You mentioned WoW, how many accounts does that game have which exploits gold ?

Would you like me just to leave it with I don’t agree with your suggestion, because spam finds its way round, and that WoW is a bad example and I wouldn’t take any ideas from that game.

No validation again, you have not explained to me how WoW is a bad example from what I can see WoW has been the go too example for almost every single mmo game-play format up until now which it lended most of its ideas from earlier generations of mmo WoW is just arguably the mmo that got the best blend for its time which is why it was so successful.

Also what makes you think I am not aware of the report functions available and that its all tied to accounts? I am arguing and might I add with quite detail, why reporting simply is not an end all solution and is not very effective if it was working then these same sellers would not be returning on other accounts to post the same messages. My method does not allow them to use any account to post any such spam content this is far superior to reporting them for doing so.

Might I add that I mention a very specific observed part of the WoW community I need not remind you that WoW is gameplay is very much based on farming the same mobs over and over to get what you want this actually makes it a much bigger target for gold sellers as everything is inflated more due to much higher demand, This system I am talking about works in WoW and they spam a lot more in that game, there is also a much bigger player base which makes it a bigger target for the sellers.

None of this changes the fact that the solution works just as well it can given the fact that after the people started used said anti spam addons that used methods such as this that sellers all resorted to hacking methods that exploited the game mechanics " not create new mechanics" too form picture art using in game assets such as emotions and certain ways of syncing multiple accounts to perform an action so that they could form letters which in tern formed website addresses. They also used hacking methods to float in the air and things like that.

Which I will agree that this method cannot prevent only anti cheating will prevent that which another issue on its own and also sellers can and will do this in this game regardless once a hacking program is made that bypasses mechanics of the game physics. That kind of of stuff is much better suited to reporting than trying to prevent which is where I will agree with your method.

Edit: Define “Exploits Gold” for me please, I am not sure if your asking for seller accounts or people who exploit the gold system in general which is not a great deal considering the overall size of the subscriber base per server but I cannot give exact figures on either. How ever I can tell you I played the game for a solid 8-9 years and the spam only really was controllable by and might I add the players had the power on how much too block because the game supported third party addons so the players made there own third party keyword system that became very popular which I have said in a previous post of mine allows the players to control exactly what was filtered and the formatting checks it did right down to how accurate it should be.

Here I can further validate this by showing you actually what I am talking about and have seen first hand in the game WoW.
Auto Report Addon for WoW
http://www.curse.com/addons/wow/bad-boy

Keyword and phrase based system I mentioned was used by a lot of players
http://www.curse.com/addons/wow/spamthrottle

Those are just to name two of the best available for that game. The user numbers don’t lie.

Now I don’t know how advanced Arena Net will make the anti spam solution of the kind I suggested if any at all but above are two pretty good examples of how to go about doing it. I only suggested the fastest and most light weight method.

At this time I would like to remind you that I am not saying WoW is awesome or any of that I am simply providing to you proof of concept for my suggested idea here which I am confident will work even in this game.

I have to agree that filtering systems can work. As an example, please see the Ultimate Blacklist addon for Runes of Magic (http://www.curse.com/addons/rom/ultimate-blacklist). When I played that game, I always used that addon. I never once had a false positive, and very few “tricks” managed to fool the system. Plus, the filters for any such filter system can be updated as new methods of bypassing them are created by spammers.

You could also make the system slightly more advanced, by requiring that messages meet two filter items instead of just one, or by prioritizing filter results such that certain filters are considered automatic and high priority, while others that are more likely to produce false positives would have to include two or more filter triggers to be flagged and blocked.

You could even set it up so that messages that trigger the blocking system based on filters would also automatically submit a report to the reporting system.

The basic idea behind such a system is to not only combat gold spammers in a reactive way (i.e. they deliver spam, they get reported, an hour or two later they go away), but instead to also actively protect legitimate players from ever having to even see the spam in the first place. As long as the spam is seen, it can be effective advertising, and it is an annoyance. If it is never seen, it isn’t effective, it isn’t annoying, and we can actively combat the problem in a way that reactive reporting simply cannot achieve on its own.

I like your way of thinking, but this is just too complex. All we can do is wait until ANet implements a bot detector similar to what WoW has. In WoW there is next to no goldspam (although I play EU WoW) because they have a bot detector thingy, some sort of ‘warden’ thing which autodetects any bots and destroys them, show no mercy!

I think a report button in mail box and chat would do the trick like others said. After enough reports perma ban them and they have to buy a new game. They wont make enough money each time to afford a new game each ban.
Filters are never the answer which is why they dont have one already its just pointless. The power needs to be with players like a report system. Now maybe an AI system that learns patterns and scans everything could work but its not practical from a time point of view.

Hurhurrderp.2513:

I like your way of thinking, but this is just too complex. All we can do is wait until ANet implements a bot detector similar to what WoW has. In WoW there is next to no goldspam (although I play EU WoW) because they have a bot detector thingy, some sort of ‘warden’ thing which autodetects any bots and destroys them, show no mercy!

“Warden” Is WoWs anti cheating software which was worked on a lot and continually gets better. Though that is about as far as I will agree with your statements.

I saw so much spam in wow it become impossible to really talk at all in any major faction city, It was such a problem in fact that WoW actually implemented the measures I discuss here and the the ways the plugins did it directly into warden sometime after the success of the plugins.

I know this because like I said I played that game for 8 years solid and I saw the whole thing go down around the mid way point of lich king expansion.

I also need to point out this method is extremely simple its so simple to implement in fact that given a chance to mod this game I would write it myself and distribute it. You see this can be as simple or as complex as you want depending on how much effort and time your willing to behind the foundations of the product.

To get a solution like this up and running in its simplest form which can be expanded on later but used just to shut them up for a while would take at the most a weekend for experienced developers to code its only an additional low footprint check on keywords that pass through the chat . They do not even need to give us a interface for it to make this function.

I would very much like to have some sort of limited plugin making ability in future sure limited to the fact its for interface improvement only and really only used to display information better. I don’t see it happening but one can dream right?

If I may jump back to your WoW post for a second the bot detector is also a completely separate module integrated into warden which was actually one of the first things they did.

Furthermore a bot detector in Guild Wars 2 will prevent no such spam they already have a anti spam botting system its the same thing that prevents from spamming my guild recruit message and keeping it the same every 60s in chat. However stopping actual bots will not fix this because you don’t need any program to spam chat all you need to do is stand where u want and setup a macro on your PC and loop it.

The system they have in place already can easily deal with that. Also creating an anti bot program is much much more complex than this solution ever will be even if they give us the control over keywords and trigger rates.

juiceman.2870:

I think a report button in mail box and chat would do the trick like others said. After enough reports perma ban them and they have to buy a new game. They wont make enough money each time to afford a new game each ban.
Filters are never the answer which is why they dont have one already its just pointless. The power needs to be with players like a report system. Now maybe an AI system that learns patterns and scans everything could work but its not practical from a time point of view.

They already have a report button in mail and in game for spammers and you have seen how not effective that is at keeping them away. There will never be any auto ban system for spam so forget that idea, I have made that idea invalid over a number or ways and in a number of posts already. Until someone can completely validate why that is a better solution and completely flaw this one I’m not buying it as reasonable option.

We don’t need any AI system that learns anything all we need is a simple set of rules that can be flagged if used in the chat a certain unwanted way, Its all of a days work to make a very simple prevention keyword system.

Also you do you think these spammers buy the game, maybe some of them do but I can guarantee that most of them steal the accounts and use them for exploiting and spamming, They also steal the accounts to take the gold from them to sell its actually one of the highest reasons accounts are attacked in the first place.

The rest of them probably sell gold in many mmo games where there are free trials which cost them nothing to actually spam in and they use the money they make which could be a fair amount since I know they make a bit of money and would likely just buy an account for GW2 again as needed.

gold spamm is out of control

something must be done

I agree, but when is something going to be done? that is the question.