Privacy, Anonymity and Traceability
To put is simple and short, privacy is a space that one enjoys without anyone barging into it uninvited. Privacy has been legally accepted as a fundamental right by most constitutional and legal frameworks across the world. It would be treated so in this post too.
Privacy is a right that comes with the responsibility of not misusing it. It is a doubled edged sword and therefore, it is a right that must be subject to justice. Some do not agree with such definitions and instead fight for unfettered privacy. They are myopic and see only few enemies of privacy — the least dangerous of them all in fact, and are blind to the more dangerous enemies of privacy. Their solutions build defences against these enemies but inadvertently build highways for the more dangerous ones. Such strategy for privacy is a suicide mission.
Anonymity
Privacy isn't anonymity. For an individual to assert his personal space, claim privacy and enjoy it, he must be identified in the social space. Identification is a prerequisite to privacy. Therefore, anonymity isn't privacy.
Anonymity too is a double edge sword — we need it, but we don't need it either. In cases such as crime, anonymity is dangerous; while in cases such as whistle blowing and investigative tips where the system cannot protect the subjects, anonymity is necessary. This makes it impossible to categorically make anonymity a right or a criminal act. Furthermore, to claim any right, one's identity is a prerequisite.
Some say that anonymity is required for free speech. But anonymity and free speech are mutually exclusive outcomes. The existence of free speech can be validated only when a person speaks without a mask asserting his identity. How is masking one's identity and speaking (an act that can be done with no consequences even in societies with no free speech) an exercise of free speech? In a society that truly exercises free speech, one need not stay anonymous to speak his mind. And if one stays anonymous for fear of persecution or attacks owing to his speech, then either he is too paranoid or there is indeed no free speech.
True anonymity on the Internet is doubtful because identification is a requisite for communication, and the fundamental goal of Internet is to facilitate communication — not storage. However, even if true anonymity is achievable, I doubt whether society will allow it on the long run because of the ensuing crimes. To the masses, it is an easy decision to give up anonymity than be a victim of it.
Traceability
Traceability is a double edged sword too. The society needs it in certain cases but does not in other cases. Therefore, we cannot ban traceability nor use it on all. The key is to regulate the use of traceability using an independent third party like the judiciary system.
Most of the general questions on traceability can be found on the statement Meta has made on traceability. Deconstructing this statement serves as a good essay for traceability.
Meta claims,
Traceability requires messaging services to store information that can be used to ascertain the content of people’s messages, thereby breaking the very guarantees that end-to-end encryption provides.
This is false. Encryption and traceability are two different features — encryption is the scrambling of the contents of a message using a cipher while traceability is the ability to identify who send a message — encrypted or not. Traceability cannot be used to ascertain the content of people's messages.
Secondly, if the government wants to trace a message, it is not obvious that the content of that message is already known. How else will they approach tech companies otherwise?
Even if you consider the meaningless act of just randomly picking an encrypted message and asking it to be traced, there is no need to break encryption because traceability is about who sent it rather than what was sent.
In order to trace even one message, services would have to trace every message. That’s because there is no way to predict which message a government would want to investigate in the future.
This is true, but this does not compromise end-to-end encryption. It just demands some tiny bit of additional storage which you have a lot.
In doing so, a government that chooses to mandate traceability is effectively mandating a new form of mass surveillance.
It becomes surveillance only if the government asks tech companies to track a particular person's or organisation's communications. Traceability is used in the context of tracking messages, not people.
The funny thing is that Meta is already doing this surveillance to find ways to monetise WhatsApp already.
Companies would be collecting more information about their users at a time when people want companies to have less information about them... Companies would be collecting more information about their users at a time when people want companies to have less information about them.
Meta can store only the information that is relevant to social justice and stop collecting every other information, and the information collected will be low. Users won't complain either.
Traceability forces private companies to turn over the names of people who shared something even if they did not create it, shared it out of concern, or sent it to check its accuracy. Through such an approach, innocent people could get caught up in investigations, or even go to jail, for sharing content that later becomes problematic in the eyes of a government, even if they did not mean any harm by sharing it in the first place.
The very reason why governments want traceability is to identify the source, not the players in the middle. Given this case, the threat to someone who did not create the content, shared out of concern or sent it to check accuracy is improbable. These threats, in fact, aren't technically non-existent per se, especially in states where the judiciary is weak. Therefore, their mitigation will be discussed after making four points to Meta.
- Meta should relieve itself of the trouble of deciding who is innocent and who goes to jail. This responsibility is up to the judiciary after due process called a trial.
- The law decides what is a crime and what is not; not Meta. Sure, legislation is fallible, but it's certainly more trustworthy than a team of Meta management. It is not up to Meta to decide the law.
- It is not fair that Facebook sees only one side of the coin. People can share something for ulterior motives too, such as the intention of spreading fake news or propaganda. These motives must be considered too while making a decision on traceability.
- It is not up to Facebook to decide what was the intension of person when he or she shared a post. That is the prosecutions job and it's not decided, but established with due process.
Now, on mitigating the possibility of threats to innocent people, and also considering point number 3, Meta can address these because it it they who own the information. All they need to do is to ensure that only the source is exchanged to the authorities; not the rest.
The threat that anything someone writes can be traced back to them takes away people’s privacy and would have a chilling effect on what people say even in private settings, violating universally recognized principles of free expression and human rights.
Completely wrong! Traceability is not a violation of privacy, it is a violation of anonymity. As explained in the beginning of this post, to enjoy and assert privacy, identity of the subject is conducive.
Nor is traceability a violation of free expression and human rights. If anything, traceability is a validation of free speech and expression; for if one is to express himself untraceably for whatever reason, what freedom does he have, enjoy and boast about? Nothing. Only if he expresses himself in a system where he can be traced, can we judge free speech to exist or not.
On human rights, the fact is that Meta will be committing human right violations if traceability is not implemented. When a crime is committed using social media, and in those crimes where traceability could have given justice to the victim, the refusal to implement traceability even after much deliberations infers that the organisation denied justice to victims; and denial of justice is a human right violation. The fact that traceability was asked to implemented but denied adds to the gravity of their violation.
The above statement is a perspective to prove that human right violations can happen on both sides. In such case, the sensible thing for us to do is eliminate the possibility of violations where it cannot be controlled and allow it where it can be controlled, and then regulate it to ensure zero violations.
Let me explain: it is easier to control and regulate government than the criminals. We know who the government is, where is it and how it functions. So we can chase them, fight them and even dethrone them. Whereas, we do not even know who the cyber criminals are let alone have access to them, fight them or kill them. Nor can law enforcement catch them all in the absence of traceability.
So it is sensible to design an infrastructure that—
- ensures prosecution of criminals, thereby preventing denial of justice and human right violations of the victims,
- has a risk of government misuse, and
- is regulated by a independent body to prevent the misuse.
It is not the perfect way, but it is the only way.
***
Facebook further quotes few sentences from UN rapporteur's report to the UN that says the provision reverses the burden of proof. The 'provision' in context sounds like traceability, but it may be wrong to assume that. This is so, because the report was 'expressing concern with the traceability provision in a bill at the Brazilian Senate'. So, to truly understand what 'provision' means, the bill must be brought in to context.
Facebook's use of the quote was indeed cunning, no doubt, because traceability, in its general meaning, does not reverse the burden of proof. Burden of proof is a legal doctrine and is independent of the tools and techniques used in the investigation. Almost all judicial systems follow the tenet that a person is not guilty until proven so. So it is a stupid presumption to make that burden of proof is on the individual if the traceability is implemented. It is on the prosecution to prove that the individual who shared something did it with the knowledge that it was fake.
The funny thing about that argument is that it beats itself. If the burden of proof is on the individual then it will be applicable in every case because it is a legal tenet. Irrespective of whether or not traceability is implemented or the messages are encrypted, the law enforcement agencies can simply arrest you on the charges of disinformation. Now the burden of proof is on you to prove that otherwise.
The point is that traceability does not cause the reversal of burden of proof. It is the law and the spirit of justice that decide with whom the burden of proof is.
***
Meta continues, questioning whether traceability will work? Their example is ridiculous.
If you simply downloaded an image and shared it, took a screenshot and resent it, or sent an article on WhatsApp that someone emailed you, you would be determined to be the originator of that content.
Not really. That person is the originator of the message, but not of the content if it has been downloaded shared it. In this case, the resource where the image was downloaded from will be the originators of the content. After all, the goal of finding the source of a message is to find the source of the actual information. To use their tree analogy, when you follow the branches, you will reach the trunk of the tree and the root. And if the root gets its information from the soil, the soil will be the culprit.
At another point, someone might copy and paste the same piece of content and send it along to others in an entirely different circumstance. Think of this like a tree with many branches -- looking at just one branch doesn’t tell you how many other branches there.
The limitation that looking at one branch does not tell you about others branches should not be a reason not to look at one branch. It is not logical to say that the police should not patrol a particular area for law and order simply because they cannot patrol all areas.
If multiple people have engaged in disinformation, thus creating several branches, all of them should be brought to justice. The law isn't stubborn that only one guilty person will be convicted.
In a typical law enforcement request, a government requests technology companies provide account information about a known individual’s account. With traceability, a government would provide a technology company a piece of content and ask who sent it first.
And this is a problem how? When there is a homicide and the murderer is not known but his biological traces are found, the investigation will focus on identifying the murderer from these traces. When an account of interest is known, the government will ask for its details by giving the account details. When it is a content that is of interest and that originator is unknown, the government will obviously ask who sent it.
***
Every organisation, company or privacy advocates is requested to think of the health of Internet from society's perspective. Internet, by itself, do no have health. The health of the Internet is its impact on the society. Every Internet causes that users fought — like net neutrality or censorship — was because they had some form of social impact. Now it is time that we grow from impact to impacts. It is difficult to make judgement when there are impacts as opposed to a impact, primarily because solutions posses unintended consequences. So what we must do is find a sweet spot.
The government generally is not the enemy of the people. And if some government at some point becomes so, let us not forget that they are not the only enemy. Let us not design products by only addressing a part of the problem.
As I stated earlier, traceability is a double edged sword. So, the tools must be used with oversight. The court, in my opinion is the appropriate party to oversee the use of traceability. They must allow traceability on the basis of truth. Let the government bring reasonable proof as to what the post violates and if the court is convinced, let it ask the organisation to trace that message.
One can argue that the court makes judgement based on the laws that are made by the government itself. Therefore, what independent power does the court have. It is a very good argument. This is where the relevance of India's Supreme Court ruling on privacy comes. Privacy is a fundamental right in India and it is a constitutional guarantee. Therefore, we cannot have laws that infringe individual privacy without proper reasoning. So, even if the court bases its judgement on laws enacted by the government itself, because we cannot and will not have privacy infringing laws, the judiciary is well equipped to act as the inter-mediatory.
And by 'proper reasoning', I mean that we cannot and will not have laws that allow the government to conduct mass surveillance of the citizenry, or tap into individual conversations or snoop private lives of individuals as they like. However, we can and must have laws that will allow law enforcement agencies to follow or track certain people of interest to society and justice. All our constitutional rights are subjected to certain conditions. And I believe that must be the case; how else are we going to curb crime?
The problem we face today is that people are generally more worried about the government than the criminals. And most governments have earned this response from their citizenry. If you believe that government is your enemy, do not assume that they are your only enemy. Do not forget that there are cyber criminals out there who can make your life miserable - perhaps, more miserable than the government can, or intends to. Let us also remind ourselves that the big data corporations are no safer. We must be sensible. I cannot stress enough on this: do not inadvertently build a highway for one while attempting to guard yourself from the other.
Data Ownership
Privacy is a consequence of data ownership — ownership will full control of one's personal data.
There are two kinds of data: one that pertains to the business association, and the other that does not. The former belongs to both parties, to be used under agreed terms. But the latter belongs only to the user.
Take for instance an e-commerce business. Data such as the user's name, email address, postal address and phone number are data pertaining to the business association between the user and the retailer. Without these data, the association and transaction is not possible. Such data belong to both parties, to be used under agreed terms.
However, data such as browsing history, medical history, health status, details of family members (including name, email address and phone number), trauma history, etc. are data that can be used to further the e-commerce business with the user. However, such data does not pertain to the business association between the user and retailer. They therefore, belong to the user only.
True data ownership is that which gives users complete control over their data, such that they can decide what they want to do with it. If they monetise it, they must profit from it. If, on the other hand, they keep it private, then privacy comes as an inalienable consequence.