wellthatsucks
asshole
- Joined
- Apr 19, 2019
- Posts
- 1,779
- Reputation
- 1,779
Avoidant
Shekel Shack Inhabitant
- Joined
- Oct 4, 2019
- Posts
- 1,589
- Reputation
- 2,090
wellthatsucks
asshole
- Joined
- Apr 19, 2019
- Posts
- 1,779
- Reputation
- 1,779
- OP
- #3
Once upon a time I wanted to implement a secure forum system that could be used by groups like the underground drug forums, or the hacking forums. I am rather mixed on helping the pedo forums secure themselves, as I entirely am happy to assist people in looking at such things securely, but some people on such forums are truly fucked up people who need to be in psychiatric care facilities at the soonest possible time, but that isn't the point of this thread anyway.
However, although I still enjoy using drugs and hope to someday pick back up using things like LSD, Ketamine, and so on; however, not to the same extent as I used to --- I simply am not really that associated with the drug community any longer, and whereas once I felt a strong duty to contribute, today my focus is not so much on it.
Today, I find it much more appealing to implement something like a distributed and secure academic journal / research platform, which actually has many similarities to how a secure forum system would be, but is with some differences in the focus of it.
One of the things that I would want to implement into such a platform, and something that would be much more out of place in an underground forum oriented system, is a more comprehensive style of polling system. The polling systems on forums such as ours are okay for quickly getting an idea regarding a single question, but they have numerous issues. First and foremost, they are really just designed for asking a single question at a time, rather than being more comprehensive surveying solutions. Secondly, they are rather insecure in numerous ways, including the relative ease with which someone can have alts and such things, or the loss of anonymity that comes from not letting people have alts (thereby forcing them to vote with their primary account, which is essentially their identity, whereas an alt is almost like a mask for them, especially if coupled with something like Tor).
http://pornharmsresearch.com/wp-content/...tasies.pdf
Explains a traditional technique for conducting sensitive research (i.e., research that potentially asks people to incriminate themselves, which is important for scientific understanding, but which can only be accomplished when the people being surveyed feel safe in doing so, as otherwise they will give socially desirable answers and the research will be biased because of this).
The goals of such a system are twofold: One, restrict voting such that each participant can only take the survey once, to prevent malicious people skewing the results of the research by cheating and taking the survey multiple times; two, allow the takers to maintain their anonymity, such that a survey cannot be linked to the taker of it.
One technique (I thought the one used in this paper, but it looks like they might have used a slightly different one) is for the professor to print out a bunch of random identifier strings: one for each student. Then, they place them in unmarked envelopes and pass them out during class. Each student receives one random identifier, but the student to identifier link is randomized by first shuffling the unmarked envelopes in front of the class and then passing them out, or letting students randomly pick one by going up and grabbing it. This step is to restrict the survey such that only students in the class can take it, and also such that they can only take it once. Each student also receives a copy of the survey in the same manner, which they fill out in private, and then take to a drop off center in an unmarked envelope, with their random number where their name would have otherwise been. The survey center removes from the set of random numbers each one that is processed: If there is a survey with an invalid number on it, or a survey that has been submitted with the same number twice, it is discarded rather than tallied. Upon dropping off the survey, the students oftentimes can get a generic card or something to demonstrate that they actually dropped something off at the survey center, which is required for credit. After tallying the surveys, they are physically destroyed and only the raw data is kept.
There are numerous issues with this technique; however, it is better than nothing. The primary issue is that it is weak to forensic linking of surveys to students, for even if students don't put their names on such a survey, they will almost certainly oftentimes have them covered with their fingerprints anyway, such that they ultimately may as well have their name on it if someone really wants to figure out who submitted which survey. Also, if only one student ends up actually completing it, and only one student turns in a card for credit, well, that student obviously is the one from which all the raw data came.
This system makes it a bit more anonymous and secure to conduct such research, which is good for people who may otherwise inadvertently incriminate themselves by admitting to doing something like illegally having fucked a 15 year old once or something, and also is good for science because it allows more accurate results to be gathered.
I can implement a similar system using technology. Presuming that there is still a physical class that rendezvouses at a central location (e.g., the classroom), what can be done is that as before unmarked envelopes can be passed around with each having a unique number in it; however, this isn't actually even required. Each student can receive a number that is linked to the student even. A Tor hidden service .onion URL can be publicly displayed to the class by writing it on a whiteboard or something, with the purpose of publicly displaying it being to ensure that each student is given the same .onion (so they can be confident that they are not being tricked into using a unique server that could uniquely identify them --- in other words, to demonstrate that there is a crowd to hide in). Coupled with the .onion, there also must be a hash of a public signature key that is used, for the same reason as the .onion must be publicly displayed. Students would then be able to use something like a live booted OS (e.g., booted from a USB memory stick) that boots directly into the pertinent software (i.e., a GUI that hides the OS backing it), which they could download just from something like github after being made aware of the site to download it from.
The software would take the .onion and hash of the public key first, and then connect to that server and start the protocol, which is a two step process that entails a registration step and a temporally delayed surveying step. There ought to be an enforced delay between the registration period (spanning, say, 24 or 48 hours) and the surveying period (which activates after the registration period). This is to prevent temporal linkability, as students during the registration period can use the number they received, which is even linkable to them (perhaps their student ID number) in order to participate in a blind signature protocol such as this:
https://pdfs.semanticscholar.org/e58a/17...de244e.pdf
Which I already have implemented in Rust,
https://crates.io/crates/blindsign
So, there are two major types of blind signature scheme. The first is commonly used by time stamping servers: You want the server to be able to authenticate that it was presented a piece of information at a given time, but you don't want it to know what it is signing. So, this is really simple: You hash what you want to be timestamped with a "one way function" that allows easy reproducible transformations in one direction f(x) = a but which is hard or impossible to reverse (determining the value of x from the output a). Then you have the time stamping server sign the hash of the document, without ever sending it the document. Then you can just couple the signature you receive on the hash of the document with the document, and in the future the timestamp can be authenticated just by seeing if the signature on the hash of the document validates with the public key that the server had during a given time period (sign with private key, validate signature with public key --- and the timestamp server keeps a list to a certain degree of granularity [say, hours] of the corresponding public keys it had during a time period, so the authenticator looks at the purported timestamp on the document, queries the timestamp server for the corresponding public key for that time period, and then authenticates the signature on the hash, which can itself be derived by hashing the document with the one way function).
However, that type of blind signature scheme is less interesting to me, but is a good first basis to understand this next type of blind signature scheme, which is called a Chaumian blind signature, or an unlinkable blind signature. In this case, as before, you want the signer to be able to sign the document without being able to see what it is, but you also want to be able to take the received signature and to show it to the signer who created it, such that they can verify it is a valid signature that they made, but such that they cannot determine which one pf the signatures they made with that key that it actually is. So, in the first sort of blind signature scheme, you get a signature on a hash of a document, but if you were to send this signed hash back to the server, if it were keeping track of such things, it could say exactly at which instance and for who it made this signature, for it saw the actual signature it made of the hash prior to sending it to you (it just didn't see the document because it saw the hash of it instead --- so in the first type of blind signature scheme, the signer is blind to what they are signing, but in the Chaumian blind signature scheme the signer can, in addition to not being able to see the document they signed, also not see the ultimate signature that they made on the document).
So, I will spare the intricate technical details, but essentially what happens in the case of a Chaumian blind signature is the same as in the timestamp style schemes, other than for that the hash of the message to be signed is first blinded, the signature is made on the blinded hash, and then the received signature is unblinded by the recipient of it, such that the unblinded signature is valid on the unblinded hash, but the signer never saw the unblinded hash, and the unblinding step on the signature randomizes it such that the signer can still validate it with his public key as having been made on the unblinded hash, but has never seen the unblinded hash or signature before.
So, upon connecting to the server for my proposed surveying protocol, the public key of the server is received and authenticated against the provided hash that is publicly presented in the classroom (to ensure that an attacker doesn't simply use a different public key for each survey taker in order to distinguish their surveys), and then upon providing the student ID number (again, can be linkable to the student), the blind signature protocol is engaged in such that the student in return for the student ID that is linkable to him receives a blind signed token that allows for participation in the surveying stage of the protocol (as compared to the registration phase). Because the student ID number is linkable to the student, credit for registration can be provided to the student at this time; because the student ID is used to initiate a blind signature protocol that results in the student receiving a token to take the survey, the token cannot be linked to any given student; however, because the student ID can be removed from the list upon one completion of the blind signature protocol, the student cannot take the survey more than one time.
So, after the registration period completes, the surveying period begins. Again, you need to isolate these periods temporally, for otherwise students may simply take the survey immediately upon registration, which allows for implicit temporal linkability between registration and survey completion. Also, having a registration period and surveying period temporally isolated allows for building up a crowd such as would happen with a time-delay-style mixing system.
During the surveying period, a new Tor circuit is built to the hidden service (providing network level unlinkability between registration and surveying sessions), and the survey is taken via Tor in the confines of a live booted Linux OS backed interface (reduce the risk of malware that could have infected the student's Windows system, as well as have a standardized system everyone uses like Tails such that discrepancies are reduced toward preventing linkability). To be truly anonymous the survey needs to be multiple choice without the ability for students to input text, as otherwise there is the risk of stylometric based linkability. In any case, after completing the survey, the software can submit the answers along with the blind signed token from the registration step, which can be validated and then added to a set of signatures such that it cannot be used to submit a second survey (each submissions token is checked against previously submitted tokens to ensure uniqueness).
If there is a class between the registration period and the survey taking period, people can have the list of students who registered to take the survey publicly displayed, so that students have a little bit more confidence about the size and other characteristics of the crowd they are going to blend into. Of course, nothing stops an attacker from falsely signaling that more students received credit than really did, but there are technical and social protections toward reducing the risk of this. Social is perhaps the best, for the professor can simply ask that those who registered for the survey come up and tell him or something, to create a visually analyzable crowd that can increase the confidence of each survey taker.
There is unlinkability provided on the network layer by Tor, there is a crowd built up to blend into and the crowd can be to some extent confirmed, the survey is protected from cheaters taking it multiple times to skew the results, there is the ability to give credit to those who register to take the survey, and there is unlinkability on the application layer by the blind signature scheme.
So, this technique has advantages and disadvantages as compared to the traditional pen and paper style of doing such things, but I don't want to overly analyze the differences between them in this thread. One benefit is that there is not the issue with leaving things like fingerprints on the survey, or even DNA and such things for that matter. The big issue is that computers are notoriously insecure, so there is the risk of an application layer exploit deanonymizing survey takers essentially, but this risk can be reduced by the standardized system being just a bare bones and heavily secured platform booted into from a live USB.
So, that is just one example of something that I would like to integrate into a secure research and journal publication style platform; essentially, a built in system for assisting in conducting sensitive research. Oftentimes it is best to have a physical rendezvous location to initiate such things, but similar systems with slightly different properties could also be made for surveying online communities such as ours; however, I will not propose such a system in this thread, having already rather exhausted myself typing out the previously proposed system.
Other examples include something quite similar to a forum, but rather than making posts using bbcode and such, it could have a built in LaTeX style markup language for formatting posts as actual academic papers. Indeed, such a platform could be used even for trivial posts as are common on forums such as ours, and the system can be thought of as similar to a forum system, but with the intention of being more suited for people who commonly make posts such as this one.
I could try to implement secure decentralized metric type systems as well, to be able to keep track of things like H-indices, by which I mean the number of times someone has a "post" quoted essentially, or even a federated type of indexing system that can keep track of the number of times a given "forum" has posts from it cited on another, with federated forums being able to register with a central index that can put various restrictions on membership (for example, nobody gives a fuck if this community cites something or not, so we would not be allowed on any reputable indexes).
Essentially, I want to make an open source decentralized (semi-centralized with federated trackers) academic research platform, for both writing and publishing academic literature, as well as for conducting sensitive sociological type research.
FatJattMofo
Superior to Gigachad
- Joined
- May 29, 2019
- Posts
- 20,978
- Reputation
- 19,461
D
Avoidant
Shekel Shack Inhabitant
- Joined
- Oct 4, 2019
- Posts
- 1,589
- Reputation
- 2,090
AvoidantSearch
[Theory] Intension and Integration
- Thread starterwellthatsucks
- Start date3 minutes ago
[IMG alt="wellthatsucks"]https://looksmax.org/data/avatars/l/1/1518.jpg?1573419661[/IMG]
wellthatsucks
There is nothing to be gained here
JoinedApr 19, 2019Messages1,651
3 minutes ago
+1 Reply
Report
[IMG alt="Avoidant"]https://looksmax.org/data/avatars/l/3/3401.jpg?1572216008[/IMG]
Avoidant
Shekel Shack Inhabitant
JoinedOct 4, 2019Messages1,192
2 minutes ago
Reply
Report Edit
[IMG alt="wellthatsucks"]https://looksmax.org/data/avatars/l/1/1518.jpg?1573419661[/IMG]
wellthatsucks
There is nothing to be gained here
JoinedApr 19, 2019Messages1,651
A moment ago
- New
- Add bookmark
- #3
Once upon a time I wanted to implement a secure forum system that could be used by groups like the underground drug forums, or the hacking forums. I am rather mixed on helping the pedo forums secure themselves, as I entirely am happy to assist people in looking at such things securely, but some people on such forums are truly fucked up people who need to be in psychiatric care facilities at the soonest possible time, but that isn't the point of this thread anyway.
However, although I still enjoy using drugs and hope to someday pick back up using things like LSD, Ketamine, and so on; however, not to the same extent as I used to --- I simply am not really that associated with the drug community any longer, and whereas once I felt a strong duty to contribute, today my focus is not so much on it.
Today, I find it much more appealing to implement something like a distributed and secure academic journal / research platform, which actually has many similarities to how a secure forum system would be, but is with some differences in the focus of it.
One of the things that I would want to implement into such a platform, and something that would be much more out of place in an underground forum oriented system, is a more comprehensive style of polling system. The polling systems on forums such as ours are okay for quickly getting an idea regarding a single question, but they have numerous issues. First and foremost, they are really just designed for asking a single question at a time, rather than being more comprehensive surveying solutions. Secondly, they are rather insecure in numerous ways, including the relative ease with which someone can have alts and such things, or the loss of anonymity that comes from not letting people have alts (thereby forcing them to vote with their primary account, which is essentially their identity, whereas an alt is almost like a mask for them, especially if coupled with something like Tor).
http://pornharmsresearch.com/wp-content/...tasies.pdf
Explains a traditional technique for conducting sensitive research (i.e., research that potentially asks people to incriminate themselves, which is important for scientific understanding, but which can only be accomplished when the people being surveyed feel safe in doing so, as otherwise they will give socially desirable answers and the research will be biased because of this).
The goals of such a system are twofold: One, restrict voting such that each participant can only take the survey once, to prevent malicious people skewing the results of the research by cheating and taking the survey multiple times; two, allow the takers to maintain their anonymity, such that a survey cannot be linked to the taker of it.
One technique (I thought the one used in this paper, but it looks like they might have used a slightly different one) is for the professor to print out a bunch of random identifier strings: one for each student. Then, they place them in unmarked envelopes and pass them out during class. Each student receives one random identifier, but the student to identifier link is randomized by first shuffling the unmarked envelopes in front of the class and then passing them out, or letting students randomly pick one by going up and grabbing it. This step is to restrict the survey such that only students in the class can take it, and also such that they can only take it once. Each student also receives a copy of the survey in the same manner, which they fill out in private, and then take to a drop off center in an unmarked envelope, with their random number where their name would have otherwise been. The survey center removes from the set of random numbers each one that is processed: If there is a survey with an invalid number on it, or a survey that has been submitted with the same number twice, it is discarded rather than tallied. Upon dropping off the survey, the students oftentimes can get a generic card or something to demonstrate that they actually dropped something off at the survey center, which is required for credit. After tallying the surveys, they are physically destroyed and only the raw data is kept.
There are numerous issues with this technique; however, it is better than nothing. The primary issue is that it is weak to forensic linking of surveys to students, for even if students don't put their names on such a survey, they will almost certainly oftentimes have them covered with their fingerprints anyway, such that they ultimately may as well have their name on it if someone really wants to figure out who submitted which survey. Also, if only one student ends up actually completing it, and only one student turns in a card for credit, well, that student obviously is the one from which all the raw data came.
This system makes it a bit more anonymous and secure to conduct such research, which is good for people who may otherwise inadvertently incriminate themselves by admitting to doing something like illegally having fucked a 15 year old once or something, and also is good for science because it allows more accurate results to be gathered.
I can implement a similar system using technology. Presuming that there is still a physical class that rendezvouses at a central location (e.g., the classroom), what can be done is that as before unmarked envelopes can be passed around with each having a unique number in it; however, this isn't actually even required. Each student can receive a number that is linked to the student even. A Tor hidden service .onion URL can be publicly displayed to the class by writing it on a whiteboard or something, with the purpose of publicly displaying it being to ensure that each student is given the same .onion (so they can be confident that they are not being tricked into using a unique server that could uniquely identify them --- in other words, to demonstrate that there is a crowd to hide in). Coupled with the .onion, there also must be a hash of a public signature key that is used, for the same reason as the .onion must be publicly displayed. Students would then be able to use something like a live booted OS (e.g., booted from a USB memory stick) that boots directly into the pertinent software (i.e., a GUI that hides the OS backing it), which they could download just from something like github after being made aware of the site to download it from.
The software would take the .onion and hash of the public key first, and then connect to that server and start the protocol, which is a two step process that entails a registration step and a temporally delayed surveying step. There ought to be an enforced delay between the registration period (spanning, say, 24 or 48 hours) and the surveying period (which activates after the registration period). This is to prevent temporal linkability, as students during the registration period can use the number they received, which is even linkable to them (perhaps their student ID number) in order to participate in a blind signature protocol such as this:
https://pdfs.semanticscholar.org/e58a/17...de244e.pdf
Which I already have implemented in Rust,
https://crates.io/crates/blindsign
So, there are two major types of blind signature scheme. The first is commonly used by time stamping servers: You want the server to be able to authenticate that it was presented a piece of information at a given time, but you don't want it to know what it is signing. So, this is really simple: You hash what you want to be timestamped with a "one way function" that allows easy reproducible transformations in one direction f(x) = a but which is hard or impossible to reverse (determining the value of x from the output a). Then you have the time stamping server sign the hash of the document, without ever sending it the document. Then you can just couple the signature you receive on the hash of the document with the document, and in the future the timestamp can be authenticated just by seeing if the signature on the hash of the document validates with the public key that the server had during a given time period (sign with private key, validate signature with public key --- and the timestamp server keeps a list to a certain degree of granularity [say, hours] of the corresponding public keys it had during a time period, so the authenticator looks at the purported timestamp on the document, queries the timestamp server for the corresponding public key for that time period, and then authenticates the signature on the hash, which can itself be derived by hashing the document with the one way function).
However, that type of blind signature scheme is less interesting to me, but is a good first basis to understand this next type of blind signature scheme, which is called a Chaumian blind signature, or an unlinkable blind signature. In this case, as before, you want the signer to be able to sign the document without being able to see what it is, but you also want to be able to take the received signature and to show it to the signer who created it, such that they can verify it is a valid signature that they made, but such that they cannot determine which one pf the signatures they made with that key that it actually is. So, in the first sort of blind signature scheme, you get a signature on a hash of a document, but if you were to send this signed hash back to the server, if it were keeping track of such things, it could say exactly at which instance and for who it made this signature, for it saw the actual signature it made of the hash prior to sending it to you (it just didn't see the document because it saw the hash of it instead --- so in the first type of blind signature scheme, the signer is blind to what they are signing, but in the Chaumian blind signature scheme the signer can, in addition to not being able to see the document they signed, also not see the ultimate signature that they made on the document).
So, I will spare the intricate technical details, but essentially what happens in the case of a Chaumian blind signature is the same as in the timestamp style schemes, other than for that the hash of the message to be signed is first blinded, the signature is made on the blinded hash, and then the received signature is unblinded by the recipient of it, such that the unblinded signature is valid on the unblinded hash, but the signer never saw the unblinded hash, and the unblinding step on the signature randomizes it such that the signer can still validate it with his public key as having been made on the unblinded hash, but has never seen the unblinded hash or signature before.
So, upon connecting to the server for my proposed surveying protocol, the public key of the server is received and authenticated against the provided hash that is publicly presented in the classroom (to ensure that an attacker doesn't simply use a different public key for each survey taker in order to distinguish their surveys), and then upon providing the student ID number (again, can be linkable to the student), the blind signature protocol is engaged in such that the student in return for the student ID that is linkable to him receives a blind signed token that allows for participation in the surveying stage of the protocol (as compared to the registration phase). Because the student ID number is linkable to the student, credit for registration can be provided to the student at this time; because the student ID is used to initiate a blind signature protocol that results in the student receiving a token to take the survey, the token cannot be linked to any given student; however, because the student ID can be removed from the list upon one completion of the blind signature protocol, the student cannot take the survey more than one time.
So, after the registration period completes, the surveying period begins. Again, you need to isolate these periods temporally, for otherwise students may simply take the survey immediately upon registration, which allows for implicit temporal linkability between registration and survey completion. Also, having a registration period and surveying period temporally isolated allows for building up a crowd such as would happen with a time-delay-style mixing system.
During the surveying period, a new Tor circuit is built to the hidden service (providing network level unlinkability between registration and surveying sessions), and the survey is taken via Tor in the confines of a live booted Linux OS backed interface (reduce the risk of malware that could have infected the student's Windows system, as well as have a standardized system everyone uses like Tails such that discrepancies are reduced toward preventing linkability). To be truly anonymous the survey needs to be multiple choice without the ability for students to input text, as otherwise there is the risk of stylometric based linkability. In any case, after completing the survey, the software can submit the answers along with the blind signed token from the registration step, which can be validated and then added to a set of signatures such that it cannot be used to submit a second survey (each submissions token is checked against previously submitted tokens to ensure uniqueness).
If there is a class between the registration period and the survey taking period, people can have the list of students who registered to take the survey publicly displayed, so that students have a little bit more confidence about the size and other characteristics of the crowd they are going to blend into. Of course, nothing stops an attacker from falsely signaling that more students received credit than really did, but there are technical and social protections toward reducing the risk of this. Social is perhaps the best, for the professor can simply ask that those who registered for the survey come up and tell him or something, to create a visually analyzable crowd that can increase the confidence of each survey taker.
There is unlinkability provided on the network layer by Tor, there is a crowd built up to blend into and the crowd can be to some extent confirmed, the survey is protected from cheaters taking it multiple times to skew the results, there is the ability to give credit to those who register to take the survey, and there is unlinkability on the application layer by the blind signature scheme.
So, this technique has advantages and disadvantages as compared to the traditional pen and paper style of doing such things, but I don't want to overly analyze the differences between them in this thread. One benefit is that there is not the issue with leaving things like fingerprints on the survey, or even DNA and such things for that matter. The big issue is that computers are notoriously insecure, so there is the risk of an application layer exploit deanonymizing survey takers essentially, but this risk can be reduced by the standardized system being just a bare bones and heavily secured platform booted into from a live USB.
So, that is just one example of something that I would like to integrate into a secure research and journal publication style platform; essentially, a built in system for assisting in conducting sensitive research. Oftentimes it is best to have a physical rendezvous location to initiate such things, but similar systems with slightly different properties could also be made for surveying online communities such as ours; however, I will not propose such a system in this thread, having already rather exhausted myself typing out the previously proposed system.
Other examples include something quite similar to a forum, but rather than making posts using bbcode and such, it could have a built in LaTeX style markup language for formatting posts as actual academic papers. Indeed, such a platform could be used even for trivial posts as are common on forums such as ours, and the system can be thought of as similar to a forum system, but with the intention of being more suited for people who commonly make posts such as this one.
I could try to implement secure decentralized metric type systems as well, to be able to keep track of things like H-indices, by which I mean the number of times someone has a "post" quoted essentially, or even a federated type of indexing system that can keep track of the number of times a given "forum" has posts from it cited on another, with federated forums being able to register with a central index that can put various restrictions on membership (for example, nobody gives a fuck if this community cites something or not, so we would not be allowed on any reputable indexes).
Essentially, I want to make an open source decentralized (semi-centralized with federated trackers) academic research platform, for both writing and publishing academic literature, as well as for conducting sensitive sociological type research.
+1 Reply
Report
[IMG alt="Avoidant"]https://looksmax.org/data/avatars/m/3/3401.jpg?1572216008[/IMG]
Remove formatting
BoldItalicUnderlineStrike-through
Text colorFont familyFont size
Insert linkInsert imageInsert videoSmiliesInsert
AlignmentListInsert table
UndoRedo
DraftsToggle BB code
Write your reply...
Post reply Preview
Attach files
Viewing This Thread (Looksmaxers: 5, Bluepilled: 1)
- Avoidant
- wellthatsucks
- Petsmart
- FatJattMofo
- Zuvay
AscendenceImminent
red flesh and white bones
- Joined
- Aug 18, 2019
- Posts
- 974
- Reputation
- 935
shut the fuck up schizoid
wellthatsucks
asshole
- Joined
- Apr 19, 2019
- Posts
- 1,779
- Reputation
- 1,779
- OP
- #7
Cocksucker who are u
D
Deleted member 1774
Kraken
- Joined
- May 18, 2019
- Posts
- 7,958
- Reputation
- 10,154
Similar threads
siniurn
