In this post we would like to explain our thinking on the various forms of privacy and how we have implemented them to support you and keep you safe from the bullies, spammers and scammers who have plagued the internet for decades. This is not written in legalese, but it can get a little technical in places. But stick at it or come back to it later when you have time. We want you to be fully aware of how we treat your privacy in principle and practice. We will not be founders or marketing hacks who say one thing about your privacy, and lawyers who say something else.
The key concepts for understanding how Tuvens treats privacy are:
- Technical Privacy – the way your data is stored and encrypted so that what you think is private is genuinely private
- Managed Privacy – the account & content settings that you have or will have on Tuvens to keep your data safe and secure
- Visibility – the contextual controls that you have or will have to control the visibility of your profile and content within the various spaces that you engage in and those you do not wish to engage in
- Reach – the capacity you have to communicate and influence other users on the platform based upon your reputation and your direct relationships with other people using the platform
The first thing to note here is that Tuvens is private by design and a privacy first platform. We aim to preserve and promote your right to privacy as a universal human right and the basis of a free democracy. Your data, including anything you submit to Tuvens voluntarily, is your own, for you to export or delete as you choose.
Having said that, implementing all of the privacy functions we have in the pipeline is a matter of transparency, as we aim to make you aware of the privacy implications of what you submit to us, and what we are doing to balance the utility of the application with the levels and types of security in place.
For example, we will design our systems with end-to-end encryption as a principle, so that we design a cradle-to-grave secure lifecycle for the management of your data, including securely destroying said data when we no longer need it for the normal function for which we stored it in the first place. We will not engage in building shadow profiles or amassing pools of data to share with advertisers and marketers without your knowledge. But that does not mean that your direct messages to other users will be “private”. For technical reasons as well as concerns we have about customer support they will not be end-to-end encrypted, at least not in the beginning, and although other users will not be able to read them you should treat them as public. We will be able to read your direct messages (like Twitter and Facebook can) but we will keep tight and audited controls over who has access to them.
We are a small unfunded team and we regrettably ask you to have some faith in our intentions in this regard. We will not have all of the resources of the big fish to deliver on our intentions in day one, and since we feel that venture capitalists are part of the problem in this domain we won’t be taking any cash with strings attached that would pressure us into compromising on those principles.
With that out of the way, the concept of privacy can be further subdivided into technical and managed privacy. In other words, ‘privacy’ typically has a technical meaning and a subjective meaning. What can feel private, like browsing the web at home, is often not private in a technical sense. The difference between technical and subjective privacy is the threats that the subjective person is attending to. Large complex long term threats are not easy to comprehend, but the fear that a former friend will screenshot a message you have sent is easier to comprehend. Is a private message on an encrypted messaging app private? Well, technically yes, but not if they screenshot it and post it publicly on the web, or into another group message. On the other hand, when posting into a social network in which only people you have actively approved as ‘friends’ can see your content it is easy to perceive this as a closed circle, even though the social media company may in fact take the legal position that you have no right to privacy over any content you publish on their platforms.
Tuvens takes technical privacy very seriously, not just complying with regulations such as the UK’s Data Protection Act 1998 and The Privacy and Electronic Communications (EC Directive) Regulations 20013 and GDPR in the EU, and COPPA in the US, but by prioritising privacy in the architecture and design, and adopting as a principle privacy by design, with end-to-end encryption and a transparent or even open sourced codebase. We are also, crucially, building a revenue model that does not incentivise us to be invasive.
Technical privacy means being proactive about security rather than reactive, anticipating security events and preventing them from happening rather than trying to weather the media blow back from our failures. We will make mistakes, but not by design. Where other platforms move fast and cut corners on privacy, we move slowly and carefully, so we don’t break shit.
A key distinction between managed privacy and technical privacy is that with the former there is no action required by the user to protect their privacy, it is secure by default, where as managed privacy is about your choices and preferences.
Technical privacy is not bolted on, its not a specific feature, it is part of the process of developing and securing every feature from exploits and human errors.
If everything is going right, you will never notice technical privacy. You only experience technical privacy when something has gone wrong, which we will endeavour to ensure never happens, as a matter of respect for you as a human being and not a resource to be exploited.
But Privacy also includes the granular privacy controls available to all users. Managed Privacy is the baseline settings that allow the user to know definitively who can and cannot see something that they have voluntarily submitted about themselves, including their content or personal information. These are standard functions we see on most social networks and media applications. For example, on Instagram I can enable people who I follow and who follow me to see my profile, which is akin to a Friend connection, or I can make my profile totally public. I can block another user, or hide my profile from search results. We may set a default that is more open for your convenience, and flag it as clearly as we can.
Today social media apps are designed around individual connections, and managed privacy is a binary concept. Either a user can see your profile, your content, or message you, or they cannot. An ephemeral message is sent to a user, or it is not. If you post into a group everyone in that group can see your content. If you post to your news feed you set the privacy, and falling into or out of that privacy setting is a binary function. We are either close friends or not, this post is public or private, you are in this group or not. There is no grey area. A post may not reach a specific user, but that is not the same as saying it is not visible to them.
But in the real world, while there exist boundaries, such as walls and doors, in public spaces visibility has a gradient, of more to less, and some people are less private than others depending not only on the subject matter, but on their general levels of openness and extraversion. Even in the reality of social media usage, a private message may be screenshot and shared outside your privacy controls.
Visibility is the grey area between public and private that cannot be accounted for by managed privacy.
To return to the concept of subjective privacy, both managed privacy and visibility relate to the practical perception user’s have of who can see them and their content on the platform. But while managed privacy is completely under their control, visibility is only partially under their control.
Take, for example, a private conversation one has in a public space, like a bar or restaurant. How private is this if the person at the next table can hear it? Or what about a conversation in public with a pseudonymous account? If a person’s identity can be deduced from the public content and later doxxed, was it ever private in the first place?
Put another way, managed privacy tells you who you are sharing your content with, but visibility pertains to who actually sees it.
Visibility really relates to how much a user trusts the people they share content with, even in a public space. How comfortable people are posting into a group with their real identity, or at least pseudonymously with a known “screen name”, is a product of the integrity of trust networks. On Tuvens, as the reputation of users within spaces increases they can see more content from other users in that space. If a user’s reputation decreases they may lose access to conversations or events that they could see before, even if they have commented on them before.
Put simply, if one person says to another “you should dance with them, they’re really good to dance with!” there is no implication that you should let them drive you home or babysit your children, but there is an implication that they are not going to overstep implicit boundaries within the social dance context. Thats what a trust network is, a contextual relationship between people based upon their reputation within a peer group.
No app can give you total control of the visibility of your content, but we can help to control the visibility as with tools layered on top of managed privacy. An example of a common feature for controlling visibility is ephemeral content, like self-destructing comments or images, or content that is only visible while the receiver has their finger on the screen.
In contrast to privacy and visibility, reach is not a measure of who can see a person’s content, but how widely their content is published and seen, or how widely their content is propagated. On Tuvens this is a product of their rank within Spaces and how many people Follow them. People with a higher reputation within a Space will have higher reach when posting into that Space, but they can also choose to increase or decrease the visibility. Some people will post into the Space, but until they reach a minimum threshold of reputation/rank their Post or Event will not display in search results. The UI will clearly prompt the user to increase their reputation, through Follows, Endorsements and References, to increase their Reach within the relevant Spaces.
In short, Tuvens takes the principled position of facilitating Freedom of Speech, but not Freedom of Reach. This position reduces the technical and financial burden on Tuvens for identifying and removing accounts engaged in bad or disruptive behaviour, and keeps users safe from spam, scams and trolls.
Both privacy, visibility and reach are all aspects of ‘personae’. A persona is the way you portray yourself within a specific community and your reputation therein.
People can set their visibility for each of their personae differently, depending upon how comfortable they are with that community. This means, just like in real life, they may be very public in salsa and less public in philosophy, where the subject matters (the Interest) are differently sensitive.
Equally, a person’s tango persona can have much more reach than their salsa persona. When posting to spaces a person does not choose which persona to use, their persona is determined by the interest the space is related to, so a person cannot set their reach, only their visibility or privacy can be set by them unilaterally.
The final thing that users cannot hide is the contents of their personae, assuming their profile is visible. If someone can see one of your personae, as in if they land on your profile page, they can see all of your personae, but not necessarily all of the content you have added with it.
The Spanish Flu pandemic that lasted from February 1918 to April 1920 killed between 20 to 100 million people and infected up to one third of the global population at the time, or 500 million people. Depending upon the estimates, this was the deadliest pandemic since the black death of the 1300s. Nobody was vaccinated, and it never went away, yet the number of deaths from this novel influenza virus in 1921 was effectively zero.
As surely as pandemics happen, and will happen again, it is also sure that pandemics end.
Covid-19 is a tragedy on many dimensions. To save lives we have all made incredible sacrifices on behalf the vulnerable, the elderly and those with significant co-morbidities who are deeply concerned about becoming infected. Vaccination technology has advanced rapidly to meet the need of the crisis, which will save many lives through the long tail of the pandemic, should it continue to spread into 2022. What we know is that, like the Spanish Flu, Covid-19 will become endemic, something we live with as a normal coronavirus or rhinovirus, and in the rare cases where vaccination fails to prevent serious illness doctors will have robust treatment protocols. Covid will be, and arguably is already at the time you are reading this, a known quantity.
Unfortunately a moment which should have galvanised society together in a united front has become politically charged and nerves are fraying. A life without purpose is a life without meaning, and human beings are communal by nature, we find our purpose in each other, like rivers coming into confluence to form larger rivers destined for the sea. What can seem trivial or frivolous to some people on the outside, like dancing or sport or basic human contact, is essential to the formation of healthy relationships and communities. Whatever your political persuasion, if we emerge from this pandemic treating each other as carriers of infectious diseases, or in some cases even carriers of dangerous misinformation, rather than human beings with unique purpose and potential, we will emerge a poorer and less humane society.
Tuvens is a platform for facilitating the organisation and coordination of in-person events and online gatherings, but we do not organise any events ourselves and we don’t advocate that people create events that are prohibited in their legal jurisdiction. We recommend that all of our partner organisers to work within their local legal and moral constraints. Every place is different, every culture sets its own priorities and ways of navigating the virus. This is not a competition, nor an opportunity for recrimination and settling political scores.
Tuvens will endeavour to suppress misinformation, about covid-19 or anything else, not by infringing on freedom of speech but by targeting the root causes of misinformation, which is the attention-based business model prevalent in social media that is based upon feeding us more and more information that affirms our existing beliefs.
The way to fight covid is to create an environment that fosters confluence and community and helps people come to shared values that are not externally imposed. Group cohesion and coordination are a necessary foundation to deliver on the solutions created by scientific advisors, whether they be testing (PCR or Antigen), contact tracing or vaccination. Without group cohesion every single one of those measures will fail.
And the political discord created and caused by ad funded social media, through the spread of misinformation and targeted confirmation bias, directly inhibits the capacity of governments and health systems to meet the enormous challenges of covid-19.
Where your local authority recommends mandatory mask wearing, we recommend following those guidelines. Where your local authority recommends limited numbers or tighter protocols for managing the spread amongst guests & participants, we recommend following those guidelines.
But if we can do our little bit here at Tuvens to help suppress the virus and save lives, it will be to bring people together, to engender more trust and less acrimony and recrimination.
If you wish to discuss covid safety amongst your own community, please be respectful of each other’s varying opinions and levels of risk tolerance, assume good faith, and remember that you are talking to human beings with diverse priorities and values. Disagreements are healthy and desirable, even on core values. A community that cannot critically assess its own core values has a name, its called a cult.
But do not take anything you read anywhere on Tuvens.com as authoritative public health advice. Organizers should always refer to their local authorities and public health officials for restrictions and recommendations. There you will ideally find information about:
- Social distancing
- Contact tracing
- PCR testing
- Rapid antigen testing
In the meantime, Tuvens aims to support curate online gatherings so people can stay as close as possible.
Building a system that puts communities in charge
If you’re reading this and it sounds familiar, you might be right. We read Substack’s moderation policy and found that it is so aligned with our own views that we were inspired to use it as a basis for our own policy. Chris, Hamish, and Jairaj, we hope you will take this as the sincerest form of flattery 🙂
As we set out to grow Tuvens into a social network with global reach that is also deeply embedded in local grassroots communities, there is considerable onus on us to make our position on content moderation and censorship as clear and early as possible, and explain how we got there.
It’s a complicated issue with no perfect solutions. We could say, for example, that we are advocates of free speech, but the uncomfortable truth is that this aspiration requires short and medium term compromises. It is obvious that saying things that are clearly illegal should be suppressed, but some people also say things within the law that we are uncomfortable facilitating. On the other hand, discernment doesn’t scale, and no human moderators or censors, however independent, have the resources to regulate contextually meaningful speech. A word in one context, uttered by one person, is not the same as that of another in another context. Machine learning results in a kafkaesque experience for users who have to navigate constraints that they are not allowed to fully understand, and it is often expensive for developers to reverse engineer the patterns used by AI systems to make their opaque judgments – and thats before you get into the fraught terrain of the debate around algorithmic bias. Then there are the policies themselves, blunt instruments at the best of times, which cannot account for the nuance of context, and which become shifting sands in a politically charged environment, sometimes selectively applied.
When you are faced with two bad options, it is necessary to get to the root of the problem, to forge a third way, and this article covers our thinking to date. In short:
- How the Tuvens model puts integrated communities in charge
- How our company’s beliefs and values inform our approach to content moderation
- Why we promote quality content & events on the platform
Tuvens is different from social media platforms.
In conversations about content moderation on the internet, there will be a tendency to consider Tuvens in the same category as Facebook, Instagram, YouTube, and Twitter, all of which also host “user-generated content.” But there are fundamental differences between Tuvens and these platforms that have major implications and go way beyond the superficial similarities in the UI and basic functionality.
Difference 1: In fact, we don’t consider ourselves a social media platform.
The term ‘social media’ was coined by the founders of MySpace when they were pitching to sell their burgeoning platform to News Corps. The pitch was simple, that MySpace is the perfect media company, where the consumers produce the content, and the profiteers just add ads. Before this time the standard term was ‘social network’. Once Facebook adopted this business model the dye was cast. Therefore, without ads Tuvens is a good ol’ social network, and we aim to bring back the feelings of community and safety that those early social networks engendered.
A lot of people will suppose that we started Tuvens to be the next big thing in social media. But what we’re actually trying to do is subvert the power of the attention economy. We want integrated communities, not engagement-motivated platforms, to ultimately be in control. We think this path offers a better future for our partners specifically, and for culture generally.
Difference 2: On Tuvens, users are in full control of what they see.
Today’s dominant social media platforms dictate to a large extent what you see, pushing content to people in news feeds. The content that appears in these feeds is filtered and ordered by algorithms that have been designed to maximize engagement. For billions of people, these engagement-optimized feeds have replaced newspapers, magazines, and TV news channels in being the main deciders of how timely information finds its way into our brains.
But with Tuvens, users choose what they see. A user makes a conscious decision about which other people and organizations to follow, about which subjects, and which ones to support with money. If you want to follow a salsa teacher, why would you also be interested in their politics? If you want to follow an author, why would you be interested in their dictates about public health policy?
Difference 3: On Tuvens our partners – organizers, artists, creators, thought leaders and experts – are paid directly by users.
All of today’s big social media companies make their money from advertising, which means they compete to dominate your attention. For these companies, no metric matters more than engagement, which is why the world now has autoplaying videos, trending tabs, and clickbait. It also means that these platforms succeed by amplifying irresistible content, which is often sensational material and conflict-driven exchanges.
Tuvens’ key metric is not engagement. Our key metric is partner revenue. We make money only when Tuvens partners make money, by taking a 10% fee on the revenue they make from subscriptions, with heavy discounts for partners who join as ambassadors. With subscriptions, partners must seek and reward the ongoing trust of their communities. And Tuvens only gets paid if partners feel like they’re getting ongoing value. Our entire business depends on holding partners’ trust, which is exemplified by how easy it is for a partner to leave the platform. With Tuvens, as a proud open subscription platform, partners own their content, mailing list, and payment relationships – and they can export it all with the click of a few buttons.
When engagement is the holy metric, trustworthiness doesn’t matter. What matters more than anything else is whether or not the user is stirred. The content and behaviors that keep people coming back – the rage-clicks, the hate-reads, the pile-ons, the conspiracy theories – help sustain giant businesses. In that commercial context a tweak to an algorithm or a new regulation wouldn’t change things for the better. The only option was to change the entire business model.
Tuvens is a real alternative to this status quo. We created our business model to be deeply aligned with the interests of users, even to the point where subscriptions are a key part of our method of amplifying content, since a monthly subscriber that is tied to a specific interest is a key signal that this partner is adding value to their community.
Tuvens is not apolitical
None of these views are neutral. Many technology companies strive to make their platforms apolitical, but we think such a goal is impossible to achieve. As the founders of Tuvens, our beliefs are fundamental to how we have been building the platform. Our personal politics, while differing in specifics, are liberal in the general sense. We favor civil liberties, believe in democracy, and are against authoritarianism of all kinds. We also hold a set of core beliefs that are reflected in every aspect of the company:
We believe that subscriptions are better than advertising.
We believe in letting people choose who to trust, not having click-maximizing algorithms choose for them.
We believe that the prevailing social media ecosystem is in disrepair and that the internet can be used to build something better.
We believe that hosting a broad range of views is good for democracy.
We believe in freedom of association and in free speech – and we do not believe those things can be safely compromised – but with caveats:
- If we as a platform are facilitating speech then we must qualify that freedom of speech is not the same as freedom of reach. Everyone has a right to free speech, but nobody has a right to an audience.
- It is not our right, not even on Tuvens itself, to decide who has a right to an audience or not outside the explicit rules we set for the platform. We do not have the right to insert our personal politics from an ivory tower into a contextually contingent public discourse.
These beliefs inform how we have designed Tuvens, which is why, for instance, we don’t support advertising in the product despite many calls to do so, and it’s why we will never use algorithms that optimize for engagement for its own sake. However, we believe that our design of the product and the incentive structure we have built into it are the ultimate expression of our views. We do not seek to impose our views in the form of censorship or through appointing ourselves as the judges of truth or morality.
All things in moderation – including moderation
From the start, we have set out to encourage a broad range of expression on Tuvens. In most cases, we don’t think that censoring content is helpful, and in fact it often backfires. Heavy-handed censorship can draw more attention to content than it otherwise would have enjoyed, and at the same time it can give the content creators a martyr complex that they can trade off for future gain. We prefer a contest of ideas. We believe dissent and debate is important. We celebrate nonconformity.
None of these are new ideas, of course. The fights for a free press and free speech have been fought for centuries. But, increasingly, there are questions about how to handle questions of free speech when the internet can spread damaging ideas faster, and when vast conspiracy theories are allowed to take root via social media persuasion.
We are aware of the history here, from the emergence of the penny press to the propaganda wars of the early 20th century, and of how initial hopes about the internet’s ability to promote healthy and productive discourse have been disappointed. Look around you: the internet is broken. But we are not convinced that the solution lies in more censorship; nor do we think the problem is that almost anyone can publish anywhere on the internet. The major issue, we think, is that business models based on engagement have created a class of wildly successful media products that distort online discourse. It is increasingly difficult to participate in reasonable discussions on these platforms, or to understand what reasonable peers believe about any politically contentious subject, when we are incentivised to self-censor for fear of base accusations, projection and public recrimination.
There are no doubt some people, alarmed by the events of recent history, who will argue that Tuvens should put free speech concerns behind a need to cultivate a more controlled community that can guarantee safe spaces to all involved. Some people will argue that we should cultivate communities that fall within a narrow window of a specific conception of respectability; that we should embrace the role of moral police (as long as it conforms with their views).
We appreciate that there are reasonable arguments to be made on all sides of these questions. We just disagree with those who would seek to tightly constrain the bounds of acceptable discourse. We think the principles of free speech can not only survive the internet, but that they can help us survive as a society that now must live with all the good and bad that the internet brings. We welcome competition from anyone who thinks we’re wrong about this. Anyone can attempt to recreate the software platform we’ve made and we make it easy for participants and organizers to opt out at any time. We are happy to compete with “Tuvens but with more controls on speech,” just as we are happy to compete with “Tuvens but with advertising.”
With that in mind, we commit to keeping Tuvens wide open as a platform, accepting of views from across the political spectrum. We will resist public pressure to suppress voices that loud objectors deem unacceptable. If you browse Tuvens’s spaces in a year’s time, we hope you’ll see organizers and artists from the left and the right, the populist and the elite, the low-brow and the high-brow, the secular and the faithful, the activist and the academic. We will endeavour to foster this range and strongly believe that diverse and heterodox integrated communities are stronger, more resilient, more equitable and more civilised.
Ultimately, we think the best content moderators on Tuvens will be the people who have earned their influence within their own communities: the partners themselves. On our platform, each space is its own dominion, with people who have gathered there through common interests. And users, in turn, choose which other people to follow, boost, and subscribe to and which communities to participate in. As the platform, we cannot presume to understand the particularities of any given community or to know what’s best for it. We think it’s better that the organizers and trusted members of those communities set the tone and maintain the desired standard, to amplify and suppress content that is not aligned with the shared and emergent values of the space, and we will continue to build tools to help them to do that. Such an approach allows for more understanding and nuance than moderation via blunt enforcement from a global administrator.
Where communities become isolated, or siloed, due to some form of legitimate dissidence or an otherwise intolerable set of values by the majority, where people create multiple discrete accounts just to engage in taboo or hateful discourse, we must consider this on a case by case basis. Since there was a time when homosexuality was marginalised and excluded from society, a righteous ban hammer can be wielded against the vulnerable as much as the hateful, and we don’t always have the tools to tell the difference. Where possible we will create incentives for bridge building and greater integration, so that people don’t feel afraid to be who they are, that they can express their moral instincts in an environment that nurtures nuance and fosters a spirit of diversity and shared humanity.
Of course, there are limits. We will not allow porn on Tuvens, for example, or spam. We do not allow doxxing or harassment. We will have content guidelines (which will evolve as Tuvens grows) with narrowly construed prohibitions with which users and partners must comply. But these guidelines are designed to protect the viability of the platform at the extremes, not act as a filter through which we see the world. There will always be many users on Tuvens with whom we strongly disagree, and we will err on the side of respecting their right to express themselves, and the right of other people to decide for themselves who they wish to associate with.
At the same time, while we take a hands-off approach with who may use the platform, we will continue to take an active approach in helping, nurturing and promoting exciting and thriving communities that we identify as groups we would like to actively support. We are doing this by improving discovery on the platform and building programs, such as ambassadorship, to share the wealth and value generated by the system with those users who contribute the most to its growth, integrity and health. Our partnerships team will also be built to work with high-revenue and high-potential organizers & influencers. We do these things because they help Tuvens’ business – a partner’s financial success is our financial success – which in turn means we can make larger investments in the overall health of the platform and the level of support we can offer to communities generally.
Through this mix of philosophies and measures, we hope that Tuvens’ approach to content moderation improves on the status quo and allows a diversity of voices to flourish while letting users retain full agency.
Communities are in charge. Users can opt in and out of spaces as they wish, and they are in control of what they see. Partners can choose to leave the platform at any time while retaining ownership of their content, mailing list, and payment relationships.
Tuvens holds liberal ideals on matters of free speech and free association. We will continue to encourage a broad range of expression from viewpoints across the political spectrum. Our content guidelines will evolve over time, but the prohibitions will remain focused and with a strong presumption of protecting that freedom.
We will support quality content & activities being created and promoted on Tuvens however we can, including by helping users more easily find those people and organizations who are held in the highest regard from and contribute the most to their communities.
Thank you for reading, there’s a lot more to come.