About COMM spammers.

Perhaps this will be interesting for you. 

We've already stated that at the moment, Niantic makes no effort to quickly block spoofers or bots. However, this post will be about something a bit different, but the overall theme is very similar.

Very few of you probably knew about the fact that the number of spam messages was drastically reduced by the Niantic tech team back in 2018. In 2018, during the period of active spam flood about selling in-game stuff or services, or even about some sex shops, just in order to troll admins, players, and in particular Andrew Krug, Niantic was forced to create a filter that could drastically restrict the flow of messages. As a result, from one of our accounts, we could only send 35 messages for the first hour and 15 for the second hour. In the best-case scenario, one account could send an average of 50 messages for its entire life until it's been blocked. Back then, their solution radically fixed the issue, which made conventional spamming method not cost-effective for us and even for our competitors (who also messed with COMM chat), because it would require a ton of newly registered accounts. While before that, we could use 1-5 accounts to create millions of messages without much effort, now with the new filter it would not make sense to even bother with registering 10-20 accounts for 1000 messages.

At that time, the tech team was actively "fighting" cheaters, at least that's what's been reported on twitter, LoL! Niantic was also creating an illusion of them blocking bots and improving their banhammer. Of course, that's what was demanded the most by the players through reports or posts. In fact, Niantic would be foolish not to act on this when a large number of people all over the world are playing your game. However, nothing stops us from recalling the moment when, after the end has come for Redacted, one of the private bot developers published screenshots and clarified the main functions of the his software. Later on, he also replied to some AMA, which led some users to believe that Ingress has never been an honest game, that any user could always get an edge with free or paid solutions. Basically, the same thing can be observed now, as there are many ways to cheat in the current version of the game. There's no solution for this at the moment, and there will probably never be one. 

We were wondering, how Niantic would react if the spam started flooding again. After all, it's quite problematic to implement it after their changes. We took into account all possible filters added by Niantic after the last spam waves back in 2018 have ended and created a special proprietary algorithm. The basics of it are quite simple, actually: we just need about 150 characters of text and 10 nicknames from the player's database for pinging. Nicknames are selected based on the player activity, it's possible to specifically target active players, or vice versa - not active players who enter the game every 3-6 months. Also, a feature has been added that limits the maximum number of messages per account, which allows us to rationally use all accounts, added to the database, without getting them banned after a short while for sending a ton of spam messages. As a result, each spam account now sends 50 messages every few hours and continues to send them until it's been banned. Ever since this method has been implemented, every player should have received our message at least 3 times, that is considering the fact that our main goal had been to reach the most active players first. That is because the active players are usually the ones who won't hesitate to file reports and create a general awareness of the issue, which puts Niantic under the kind of pressure it can not ignore.

Since there are a lot of players in the game, and a dozen accounts would not have been enough to notify all players, new accounts are being registered automatically. First, Google account needs to be created, and after that, it can be registered with Ingress Prime, using specific methods. Niantic banns some of the accounts the next day, so we create 50-150 of them every day, depending on the mood. Then, all the accounts are added to the database on the server, which makes it possible to regulate the message flow with a few clicks. As a benefit, we no longer need to log in to the account in the future to check its status, which means once the account has been added to the database, the login and password from it are no longer needed. And even if Google had disabled the account, it could still be used through the Intel map. Also, a special IITC plugin has been created to manage accounts and send messages.

For nearly three months of actively spamming COMM, Niantic has not been able to take any action to restrict or block those messages. Many agents have tried to block the spammer nicknames in the game to avoid subsequent message waves. But it turned out to be useless, because each time the spam bot sends a message from a new account, you shall receive the scheduled message anyway, the only solution to that is to disable game notifications in general.

Meanwhile, dozens of topics were actively flooding the official forum, Reddit, Twitter, and other resources, asking Niantic to fix this or to introduce new filters that somehow limit the enormous flow of messages through hundreds of accounts. Agents themselves proposed quite a number of solutions. Popular of those were: block the prefixes "dead" "C7H", prohibit URLs in messages, introduce limits on the radius, or even restrict the ability to send messages up to a certain level. We've read most of the posts and whitelisted some agents on a by-request basis. Seriously, it was quite fun to read :)

At this point, Niantic has yet to create a solution to avoid spam. And most likely there isn't gonna be one. Perhaps, there are more important issues at hand right now, or it may be difficult to restrict this type of multi-account spam. After all, blocking the URL will also restrict other friendly agents from sending links to conversations in telegrams, for example. And restricting messaging ability up to a certain level will prevent new agents from communicating.

We send our regards to @NianticBrian, aka @brianrose. He's the only one out of the Niantic representatives, who commented on the spam problem and promised to fix it, which never happened... Keep it up, bro!

Well. We think that the goal has been achieved, and there's no point in continuing. Good luck and Whack-a-mole! 

Edited: Dmitry Mostov
Last update: 16.08.2020