The May 7 HYP raid threats
We're disclosing all events related to the recent raid threats.
Content warning!
This blog entry will contain evidence screenshots of one or more of the following:
Racial slurs (in a censored form)
Racist language/tone
Reference to real-life tragedies (in a censored form)
Sexual language (in a censored form)
Encouragement of self-harm
Please do NOT read this article if you do not wish to see these types of content. You have been warned!
Three months ago, Hyper Air Group (abbreviated to HYP, a virtual airline alliance in a game called Airlines Manager) administrators were made aware that two members from Pacific NorthWest (abbreviated to PNW, another virtual alliance), as well as external members of a server with a dedicated raiding group, planned and threatened to raid HYP's Discord server, which has Unifier installed.
We have decided to disclose all information we have on this incident, as we believe that it is important that members of the Unifier community are aware of such safety-related incidents. We disclose information to inform, not to harass, so we ask everyone to respect each other and not harass anyone involved in this incident.
This marks the second, and most severe server raid-related incident UnifierHQ has responded to.
Prelude
This incident includes a prelude of extreme harassment from a PNW member we will refer as "Vic". As the raid threats are linked with his extreme behavior towards HYP, we believe that it is important to disclose this part of the story as well.
Where it all began
Around February, three members from PNW asked me if they can transfer over to HYP. According to them, they were not happy that decisions that were made by PNW management without notifying staff or asking them for their opinion. For example, Unifier rooms were briefly hidden from PNW's Discord server without any prior notice.
I'm in no position to comment on PNW's management team, because I don't really know what had happened in there that led to the transfer, and is not really relevant to the case. Despite this, this was a reasonable motive of transfer, so I decided to let them into HYP.
This transfer was heavily condemned by PNW's management, especially because I was not transparent with the transfer. I agree that more transparency would have been beneficial as partners (at the time), and I have taken responsibility for this lapse in judgement.
But this does not justify the harassment that the three members and I have received, which is why I am disclosing them here. Once again, this article is only for disclosure and not for defamation. I want the Unifier community to know a more fuller extent of the events, but not spark additional drama.
"group of slaves"
On April 7th, I got a rather unexpected message in a server Vic and I used to share, until he was banned for his actions. I will refer this server as "Foxtrot" for the rest of the article.
This was an act of defamation, and we asked Foxtrot moderators to delete the message. I didn't really take this case seriously, but still kept documentation of it, as it marked the start of the harassment cases.
For context, a HYP member DMed me about how a PNW member was replying to the HYP member in a rather provocative tone. I asked the PNW member to stop, however they refused to, and most likely contacted Vic. I don't really understand why he said I was getting my "group of slaves" to harass him though, since:
I received zero evidence of HYP members harassing PNW members, thus invalidating the "group" and "harassing" parts, and
I never told anyone else to harass PNW members, so I was not making anyone my "slaves". In fact, I specifically asked HYP members to NOT harass any PNW members, otherwise I'd take action:
Encouragement of suicide
We were also informed that after I announced my retirement from Airlines Manager, Vic indirectly encouraged me taking my own life. Although he does not directly ask for my death, he still does so to an extent.
Racist behavior/use of racial slurs
Racial behavior, as well as use of racial slurs, was a common case in the harassment. These are all self explanatory.
Promoting hate over alternatives
We asked Vic to host his own Unifier instance, since our code is open source anyways. He could've made his own instance with no rules whatsoever, if that's the kind of Unifier he wants.
He refused. This is why I'm so confused as to why PNW accuses me of continuing the drama, when Vic would rather pick harassing me over just hosting his own instance.
Other cases of harassment
There have been multiple other cases of harassment targeted towards others, especially towards Foxtrot's moderators post-ban. In one case, he accused a Foxtrot admin of enjoying CSA without any evidence to base his claims on. For obvious reasons, we aren't gonna put the uncensored version on here.
And in another case, he would call furries (which are, according to a friend of mine, "individuals who have an interest in the design of fictional, anthropomorphic animal characters") as escapees of a WW2 concentration camp in eastern Europe. His anti-furry behavior was also seen on a Unifier room, and we took action to protect our community.
Previous bans from Unifier
Vic has been banned from Unifier multiple times for:
Posting about a fictional plane crash in detail, which is considered a sensitive topic as it is a tragedy
Harassing others outside Unifier community and planning to raid Discord servers, and
Harassing people within Unifier community, including developers, moderators, and members
For the first two cases, we unbanned him after he had apologized. For the third case, we decided we've had enough of his behavior, and I decided to permanently ban him with no appeals possible.
Apparently, these bans also contributed to Vic's hostility towards me and HYP, according to ex-PNW members. But we made it clear from day one that we're meant to be a safe place to chat in, because we want to accommodate for all persons from different backgrounds. This is what Unifier is about, and if people don't want this, they should not use our product. If they continue using it anyway, and use it as a platform to spread hate towards communities, we will take action.
The raid threats
On May 7, we were notified by a member of a server Vic and I used to share that Vic was encouraging a raid against HYP. The owner of the server, "Lilith", agreed to start a raid on HYP's Discord server as long as Vic provided the invite.
Later, Vic started to plan the raid, encouraging raiders to join a server HYP markets their alliance in.
Besides, we've learned that the co-director Kim may have been also involved in encouraging the raid to an extent. As this is an assumption, not a firm statement, made using the timeline of events before and after the screenshot below, and I've received a response from Kim after I wrote this article, I would like to present both sides:
I initially presumed that Kim sent defamatory statements against me, hence the "encouraging the raid". Given the timing of the events and Vic asking Kim to tell the server more about me, this was a reasonable assumption to make in our perspective.
However, Kim denies this and says that he did not do this after learning "what kind of people they were". If this is true, this would clear him from any suspicions that he encouraged the raid, but not clear Vic.
These screenshots were sent in a public channel, and thus should be classified as public information.
Magnitude of potential attack
During our investigation, we've identified over 30 raiders that the perpetrators could have used to attack HYP. This means that at least 90 messages a second could have been sent in the server (this is likely an underestimate - raiders can probably send more than 3 messages a second, but I went for this rate because it seems most probable for any spammer), which is way beyond what Unifier can handle.
If we factor in bot and alternate accounts, the raid magnitude becomes even more severe. Bot accounts would lead to an increased rate of messages through automation, while alternate accounts would extend the raid length by allowing raiders to rejoin using a VPN (and a throwaway phone number for SMS verification).
Responses
Perpetrators' response (from witnesses)
This section includes testimony from third parties that have known Vic.
For the witnesses' privacy and safety, we have redacted any identifiable information from the screenshots. Please do not ask UnifierHQ or HYP to identify these two individuals.
HYP administrators have received a response from the perpetrators after HYP informed allies about the raid threats. However, we have confirmed that the primary claim made in the response is false.
The primary point of the response was that the raid threats were merely a joke and were not to be taken seriously. We've talked with two witnesses to confirm his claims.
Witness A claims that Vic is still willing to raid HYP "to finish [the conflict between HYP and PNW] once and for all":
Witness B, after I sent them the testimony from Witness A, assured us that the threats should not be left ignored, as Vic has a past history of raiding other servers:
With the two testimonies, we were beyond convinced that the raid threats were serious. Thus, we continued investigating the threat and implementing countermeasures.
Perpetrators' response (and how to verify our evidence if they get deleted!)
Vic claims that I wrote about something that "has not happened", despite actual evidence showing that these things did indeed happen. Do note that we're talking about raid threats here, not actual raids, because HYP was not raided (if that's what he means) but it was under threat of being raided.
As for Kim - I've already stated that he's denied my presumption, and I've edited this article so that readers can see his response for that right under the presumption.
Alliance director's response
Through the help of an ally alliance's director, I communicated my concerns to the perpetrators' alliance's director. He dismissed my concerns as being "oversensitive", and it is presumed that he did not take any further action against the raid threats or harassment.
Our response
As of August 25, 2024, Under Attack mode has been released to v1.2 and v2. Safety Alerts are still undergoing testing.
Before the raid threats, we permanently global banned PNW for two reasons:
Continued harassment and disrespect towards HYP staff and members by multiple PNW members, and
Lack of any responsibility from PNW's director or elite staff members by refusing to take any action whatsoever, instead shifting the blame on me for apparently "causing drama" because I banned Vic from Unifier.
As I am one of the administrators at HYP, UnifierHQ was able to take immediate action to limit the effectiveness of the raid by warning all connected servers of the threat. We also recommended HYP to pause invites until they've installed Wick, one of the most effective raid blockers for Discord servers (FYI, we weren't sponsored to say this).
We also initially disclosed a limited range of screenshots HYP and UnifierHQ had in our possession of the raid threats. We kept the rest of the screenshots confidential, as we were unsure if we would be making the right call, especially since the main target would be HYP and not UnifierHQ. But now that things have settled, we (UnifierHQ, HYP representatives, and other victims of the harassment) have agreed that it would be justified to disclose all the evidence we have in our possession.
We've also decided to work on something called Under Attack mode for Unifier v3. This is essentially the same as global banning a server, except that it is to be used by server moderators when a server is being raided or is likely to be raided. This will block all messages from the server being relayed to all other connected servers, until server admins disable this mode once their server is in the clear.
We're also working on a Safety Alerts system (similar to EAS, except for attacks on Discord servers), which alerts all servers if a server is being raided or is under threat of one. We will have four levels:
Raid emergency (red), for ongoing raids or raids that are very likely to happen,
Raid warning (yellow), for raids that are likely to happen or under active planning,
Raid advisory (blue), for raids that are not under active planning but have been mentioned by potential raiders, and
Raid clear (green), once servers are clear from any raids or raid threats.
If the alert level is emergency, warning, or clear, it will ping server moderators so they can take quick action.
Safety Alerts can also be used for threats other than raids, so we can alert you if there's something serious other than raids that are going on.
What if the raid happens?
When we made Unifier, we made it with scenarios like these in mind.
If a server is being raided, you can always block messages from that server from being relayed over to yours. So if HYP Discord is being raided, you can run u!restrict 1097238317881380984, and your server won't receive any messages from that server until you unblock them.
We've also trained our moderators to lock down the Bridge entirely should a raid take place across multiple servers. This will completely delete Unifier Bridge from the host system's memory, so that absolutely nothing can use it until it is re-initialized by a Unifier admin.
How to verify our evidence
We won't be giving you the invite link to the raid group server for your own safety. This is only for those who have already accepted the risk and joined the server, who want to verify the authenticity of our evidence.
The evidence he's created hasn't been deleted from the origin server yet, but if he does, we will be posting the message links. If you want to know why message links would help, let me give you an introduction to Discord Snowflakes (commonly known as IDs).
Snowflakes are not just a random number, they actually consist of many parts:
A Discord epoch, which is like Unix time, except it starts at January 1, 2015, in UTC time
A Worker ID, which is assigned to each worker by their parent process
A Process ID, which is assigned to each server process
A Sequence, which is incremented for every ID generated
But how can IDs be useful if the messages were to be deleted?
Deducing timestamps from messages
So let's take the raid threat message for example. This has the message link discord.com/channels/1187448733546328125/1230922500066770954/1237507944460058744, and from this message link we can deduce that the message:
was sent in a server with ID 1187448733546328125, which is the raid group's Discord server,
was sent in a channel with ID 1230922500066770954, which is their general-lessrules channel, and
has the ID 1237507944460058744.
The first ID corresponds to the server ID, the second to the channel ID, and the third to the message ID.
It's a perfect match with the screenshot I've taken, which is also in CEST (this screenshot was re-taken on August 15, which is part of Daylight Saving Time).
We can actually do the same for the messages that were sent before and after.
The message sent before has the message link discord.com/channels/1187448733546328125/1230922500066770954/1237507927888105562, which means the message:
was sent in a server with ID 1187448733546328125, matching Vic's message's server ID we deduced,
was sent in a channel with ID 1230922500066770954, also matching Vic's message's channel ID we deduced, and
has the message ID 1237507927888105562.
As 1237507927888105562 is smaller than 1237507944460058744, this means that this message was indeed sent before Vic's. Checking this one on snowsta.mp gives us a matching timestamp.
And for the message sent after, this has the message link discord.com/channels/1187448733546328125/1230922500066770954/1237507975191462021. Once again, the server and channel IDs match, and 1237507975191462021 is bigger than 1237507944460058744 proving that the message that was sent after was indeed sent after.
And the message also has a matching timestamp of May 7th, 22:54.
Deducing the "reference" message ID
If you reply to a message on Discord, your message will have a "reference" data. This is used so that Discord knows what message you're replying to, so that it can render a preview of the message like this:
But what if the reference message is deleted? In this case, we cannot see what message the user is replying to.
Unless we dig a little deeper into the message data.
Let's take this message, for example:
This message is from HYP's testing channel I made to test Nevira's custom features made for HYP, and it has the ID 1216832080378331187. The reference message was deleted, and had the ID 1216831985457303615.
We can open the Inspect tool (use your browser for this), then open the Network panel. Then, we'll reload Discord so that we can see the messages that were fetched on load. You'll have to search for "messages" in the search bar.
As the limit variable for this request was 50 (it varies sometimes, for some reason), and the message was sent quite a while back, it would be likely that the message ID may not be present in the request. If the ID is not present there, we'll have to scroll up, until we see the message we want to get (the message with the deleted reply, not the deleted message). This will lead to more messages requests appearing on the results, and we can search the ID there.
But in our case, as not many messages were sent in this channel, we can just look at the first request, and find the message we want.
We can see that:
The message ID is 1216832080378331187, which was the message ID I specified earlier, and
The content is "youre on command-r now!", which is the exact content of the message.
And if we scroll down, we can get what we're looking for.
We found the message ID we wanted: 1216831985457303615.
We can do the same thing with Lilith's message:
To save on time and number of requests, I'll be using the message link of Vic's message, so we can get to the message easier.
We can find the data for their message:
And since Vic's message hasn't been deleted yet as of writing this, we can find his message too.
You can see the index of Vic's message is 4 units higher than Lilith's. This is because:
Vic's message was sent before Lilith's, and Discord lists the data from newest to oldest, and
Three messages were sent between the two messages, explaining the 4-unit difference:
But let's go back to Lilith's message, then look at the referenced_message data.
Remember the ID 1237507944460058744 I mentioned in the Snowflake timestamp method? Well, we have the same here, in the reference data.
And we can see the referenced message's data right below, too, as the message hasn't been deleted. If it were deleted, the value would be null.
The data is the exact same. And not only that, the author of the reference message is 494513176935661579, matching Vic's user ID.
Speaking of his user ID, even if the reference message data is null, you can get his ID easily by looking at the mentions data. Because his server nickname was followed by an @ in the reply preview, this means that Lilith kept mentions on for that reply (it's toggleable), so his user ID would be there.
Closing thoughts
As one of the victims of the absolutely unnecessary harassment, I am saddened at what PNW has become - a once friendly alliance turned incredibly toxic over a game of all things. Although the majority of their members (including staff) are generally friendly and are neutral to or condemn Vic's behavior, the director's indifference to, and Kim's (presumed, but also disputed) endorsement of, Vic's behavior easily overshadow this friendliness.
This also reminds me of the criticism I've received from PNW's director and Vic, where I've been called me a "dictator" and being "worse than [a dictator]". Despite these comments criticizing my moderation style, I stand by my decision and I do not regret any of my actions whatsoever. In fact, the harassment and raid threats actually proved to me even more that Vic's past global bans were completely justified, and that he, the director, and Kim should indefinitely stay out of our community, even if they apologize.
It is important to remember that Unifier is not a platform for harassment. We made Unifier as an alternative to a larger community Discord server, and we want to inherit all its upsides while fixing its shortcomings. These upsides include active moderation, so that Unifier remains a safe and civil place for everyone to chat in. Going completely free of any rules would cause even more chaos, and it would ruin the experience for the majority of us who really enjoy chatting in a comfortable environment.
UnifierHQ will continue to monitor this situation and implement countermeasures where necessary. We will stay in close contact with representatives from HYP, so that we can be notified of future raid threats.
Our condolences go out to those affected by these actions. We're very sorry that this had to happen to you, and we'll do everything we can to make you feel safe in our community, because that is what we do here at UnifierHQ.
Takedown policy for this article
Hyper Air Group representatives, as the targets of these threats and victims of the harassment, have agreed that it is best that everyone knows the full extent of the situation that has developed. Thus, UnifierHQ will not be processing any takedown requests for this article unless specifically requested by the victims or required by law.
However, to prevent misunderstandings, PNW staff can still request takedowns or modifications regardless of the above conditions for sections without supporting evidence, as long as the reason is justified. We recommend submitting counter-evidence should you submit a request.
Last updated
