web analytics

Kind Words is not a meditation game, but a kindness game for which meditation is the optimal strategy.

This article was originally published on Wonk Bridge.

Kind Words is a game without winners and a social network without identities. The only objective is to show kindness to strangers, and its social functions and gameplay are the opposite of the most insidious features of modern social networks that make them simultaneously anti-social, depressing and impossible to quit. For this reason, I consider it a prototype for a type of system that we might build to foster broader, more profound and more positive-sum interactions between people.

Opening: Highs without Subsequent Lows

The thinker Naval Ravikant identified a pleasing set, “highs that don’t lead to subsequent lows”:

This, I compare to social media, which contains many of the facets that ought to make it a high without a subsequent low: community, conversation, information, reading; but, I argue, these media in their current configuration deliver sorbet highs and pudding lows. They encourage us to misunderstand difficult people, ignore neutral people, flatter those we’re fond of, and hate ourselves.

And they make us come back for more. One might ask, can we create social media that would be less insidious? This, of course, raises the follow-up question: would such media gain any ground or, like Louis Theroux said of Scientology, are the insidious parts what we actually like?

Kind Words appears to have created something akin to a social network, but which is also a high without a subsequent low. This, I theorize, it does through a combination of thoughtful mood and emotional queues, careful user interface choices, alongside a social design that lends itself to unimpeded generosity. I believe that Kind Words, thus, has prototyped some of the features that we need to implement in our networks to promote better relations among individual people and as a society.

Further, the philosopher Daniel Schmachtenberger observed how a great many of our conflicts as a species result from our having differential degrees of affinity with different groups of people: we care for our family more than we do others, we find it easier to go to war with people from different countries, we smear those far from us on the political spectrum.

Schmachtenberger suggests that Buddhism may hold something of a solution to this: Metta meditation, the act of directing good will first to those for which it is easy to do so, then steadily increasing the circle to more and more challenging people, possibly even to a whole city or the whole world. My thesis is that Kind Words offers this sort of training, through its gameplay. This is, firstly, why it is HAL and, secondly, why social structures like this show us how we might build new systems (on-line and off-line) that foster more collaboration and interactions of a more positive-sum nature.

Explanation of Gameplay

Kind Words is a platform for sending kindness to real people, via writing. In this respect, it shares a great deal with social networks, such as Facebook. However, the available actions are fewer, but more direct and personal.

There are three main actions:

Public posts:

  • One can write messages of kindness, which everyone on the platform can see.
  • One can write about one’s uncertainties and struggles via posts in the same format, which, again, everyone can see.

Private responses:

  • One can respond to other people’s problems and challenges by writing them kind messages, which only they can see.


  • One can send a choice from a selection of odd, charming stickers, either alongside one’s response or as a sort of “thank you” for the kindness one receives.

Visually, one’s avatar sits in a room, writing, while the messages and stickers are delivered by a charming deer, who also introduces new tracks from the game’s lovely soundtrack. There is a little more to it, but those are the broad strokes.

Hereafter I will describe what to me seems the heart of the game: the subtle interactions and design choices that make Kind Words work, and which earn each one of its creators the title assumed by Buckminster Fuller and Ted Nelson: systems humanist.

Emotional Valence

The emotional valence of the whole game is, perhaps, the first thing that the player will notice. The colors are warm, slightly muted, the resolution and detail is a little low-fi: this gives the game a natural, reassuring feel, like a combination of autumn leaves and candle-light.

The presence of the deer is particularly calming: your first mission, upon entering the game, is to write the postmaster deer a letter of encouragement: the creature encourages benign action on the part of the user. To be anything less than cordial, encouraging and magnanimous, and thus to implicate the deer in delivery feels like swearing in front of some combination of a child and a sweet elderly relative.

Of course, then there’s the music: as I mentioned in the introduction, I can’t do it justice. The score (available on YouTube and BandCamp for those that want to give it a listen) has a similar character to the deer: it is fairly uncomplicated, sweet, a little nostalgic. It is, however, never mawkish, never boring or derivative. These factors: the deer, the colors and the music, make for a synthesis that almost immediately puts one in a benign and creative mindset: so much so that my wife and I sometimes activate the game in our shared office to enjoy these aspects of it while doing other things.

The True Social Medium

As I mentioned above, Kind Words is a social network, but contains features in precise opposition to the most insidious aspects of social media.

Infinite Scroll

Firstly, there’s no infinite scroll. This is one of the most depressing aspects of Facebook, Twitter and YouTube: one scrolls and scrolls, looking for that one thing that will provide the entertainment or interest that one is seeking. The problem isn’t that the available content is functionally infinite: the scope and variety of material online and on social media means that there’s more to challenge us in ways that we didn’t expect.

Rather, the problem is that with no end to the content, no pagination, or no need to go somewhere else to find other things, one easily slips into a passive mode of engagement, driven by fear of missing out on something of value below the fold. Like a debased currency, the content that one witnesses plummets in value the more you scroll, as the ratio of real effort and engagement to content gets more off. Eventually, you get to the equivalent of having an unnecessary serving of food at the all-you-can-eat buffet, because it’s free.

As we do so, most of us are barely conscious, instead we’re in an odd state of disengagement (not really reading, not really watching or remembering) combined with overstimulation (in just a few minutes on Facebook, say, one might see more human faces than one of your ancestors might see in a year).

We browse, we never find what we’re looking for, and when we finally tear ourselves away, we’re sadder than when we started.

I think I’m right in saying that Kind Words has no scrolling. There are many lists that the user will need to access: of requests for help from other users, one’s own requests, the messages in one’s inbox, one’s stickers, all of which are navigated with next and previous buttons. The grass isn’t greener below the fold, because there is no below the fold. Meanwhile, there are no images: no images in messages or requests, or profile images, so no Facebook-style overstimulation.


Meanwhile, there are no read receipts on messages, and no notifications when you’re in the middle of something. Facebook, of course, is famous for its little red integers on the message button, say, that give you a little dopamine hit and make it almost impossible to carry on what you’re doing until you’ve seen who had what to say. There was a time, say, two years ago, when I and others confirmed that the notification system on Facebook was reliably faulty: phantom notifications would come up, only to reveal that there was nothing new. This only made the platform more addictive, in responding to a given notification, one was actually gambling on whether it was real or not.

Kind Words has nothing of this. You write your messages asking for help: you don’t know who read them, how many people read them, and what they think. You write your responses and are as kind as you can be, but you don’t know whether the recipient read them, what they thought, and they can’t reply. There’s no Facebook-style waiting while holding your breath for the read-receipt that tells you that they’ve read the message, or, worse, once they’re read the message, for their reply. Nor is there the tantalizing plunge of Twitter, when, after tweeting, one waits to see how many likes and retweets one earns.

Kind Words pegs the satisfaction that one gets against the work that you do to spread kindness, rather than recognition. I derive my enjoyment of the game from the creativity of responding to people’s requests through writing, and of the feeling that I may have done something to help or bring someone happiness.

The closest thing that Kind Words has to anything like a trophy is the stickers, which users can send as thanks for kind words. Because they are trophy-like, I enjoy the stickers less than the other functions, but they are charming. Meanwhile, unless you only answer one request, you don’t know who sent them. Thus, game makes it very hard to keep score, so there’s never the feeling of “where’s my sticker?”

This imbues the game with a sense of selfless virtue for its own sake, best expressed by the Tao Te Ching:

When success is achieved, he seeks no recognition.

Because he does not claim for the credit, hence shall not lose it.

No Conversation

I am, of course, pro conversation: my point, however, is that Kind Words has created a well-functioning, socially engaged community, but with no chat, public posts, comments, or any back and forth at all. Thus, one can’t write a message and hope for a reward in the form of a reply, nor can a seemingly benign social post turn ugly when an unwanted connection shows up: again, the highest reward in Kind Words is the satisfaction one gets from doing one’s best.

Obviously, it’s neither possible nor desirable for all social media to be structured like this, but the fact that Kind Words made this structure work raises the question of what other highly social but non-reciprocal structures we could create: Twitter but if you didn’t know who liked your tweets or who follows whom; Instagram but when you like a photo, the user whose photo you liked doesn’t know which photo it was and who liked it.

The Algorithm

Of course, Kind Words must have some automated systems to ensure the game’s proper function. Algorithm, is used both as a trashcan for all manner of things we don’t understand, and as a scapegoat onto which to cast our fears about tech. Algorithms need not be scary or recondite, however: they can be simple, complex, and seek any purpose a human might seek. With that in mind, the Kind Words algorithm appears to be benign and equitable in its design: I’ve never had a request for kindness go unanswered, nor has any one request received an unreadably large volume of responses.

It would appear that the system modulates the visibility of requests so as to smooth out the volume of responses we get. Compare this to the Facebook and Google algorithms, which further promote material that is already successful, leading to power-law distributed success, often called the Matthew Principle. As with wealth, networks, success, the more you have, the greater your ability to get more: this is not inherently unethical but there are some undesirable features.

One is the fact that a random event can confer a small advantage which, compounded, results in huge profits. Another is that a large incumbent is hard to unseat by a worthy newcomer. Finally, effects like this can incentivize lowest-common-denominator behaviour: on Twitter, say, one will generate orders of magnitude greater attention by appealing to tribal loyalties and with clickbait language, through the resulting retweets.

Kind Words saves us from the lowest-common-denominator effect because no matter how hard you try, any public post will likely get no more than four replies. Furthermore, on Kind Words, your reply will be seen by one person, and other people will only come into contact with it if the recipient copies it somewhere else or shows someone their screen: so you don’t have to optimize it for sharing, you can make it as complex, challenging and as ecumenical as you will.

Fundamentally, I’m not saying that everything should be redistributive like Kind Words, only,

  1. That this game seems to have created a redistribution algorithm that actually works
  2. It seems worth experimenting with this sort of structure so as to raise the tenor of our conversations.


I feel as though I have to apologize for this part: Kind Words is moderated. Perhaps it is as though any problem insoluble by design and systems, leaving us to fall back on human intervention, represents something of a failure. Perhaps. That said, without popping the hood, it’s impossible to tell the extent to which Kind Words is held together by an army of moderators, or whether this measure is a minor corrective. One can say, however, that as a user the experience is unsullied by trolls and abusers.

The True Video Game

I’m not really a video game guy; they’re fun, but I fear them for the extent to which they so perfectly stimulate and manipulate the mind. Kind Words subverts this and other patterns that I dislike about games, but which I thought were essentially part of what video games are.

Underwhelming Dopamine Quests

Video games are compelling for a number of reasons, one could divide these features into cheap and expensive: expensive reasons are intrigue, puzzles, action, plot; cheap reasons are those features that have the player do things that take longer or more effort than they assumed, and which deliver a reward, but less than anticipated, keeping you reaching for satisfaction. This is the underwhelming dopamine quest: it’s instrumental in keeping one in front of the computer or console, until one finds that it’s dark outside and, to return to Naval’s aphorism, that one feels a sense of shame for how long one played. Games figured out how to do this first, of course, then social media caught on and gave us their dopamine quest in the form of the infinite scrolling news feed, as mentioned earlier.

Kind Words offers no cheap rewards, the only objective is to project kindness to others. You do so by sending writing and stickers. One walks away from the game, therefore, with a sense of fulfillment, not shame.

Real Objectives

Another of the reasons that I rarely play games is that the rewards themselves are, usually, virtual: to complete a level, defeat a boss, etc. This was always unsatisfactory to me, in that I wanted to walk a way having actually created something new and lasting, and feel guilty otherwise. This is a personal foible, not a value judgement. That said, if you feel something similar, Kind Words is for you.


Finally, Kind Words is profound because it is a rare example of a positive-sum game. Some background: positive-sum, zero-sum, etc. are terms of game theory. Zero-sum games are those that have only one winner and to the extent that someone wins, someone else must lose: football (either sort), snap, Mario Kart, etc. are examples of zero sum games.

While diverting, I find this sort of game miserable precisely because one wins at the expense of others, and because all the most interesting and stimulating interactions that our species has developed exist in the positive-sum realm, wherein all parties can benefit: capitalism (wherein the seller benefits in that they prefer the money and the buyer because they want the goods, and where ecosystems of companies [sometimes in competition] profit more than if there were fewer of them) or the arts (wherein, say, two musicians play together for the first time and bring forth new ideas, benefiting both and at the expense of none).

This is why I’m so confused that most societies induct school children into competitive sports, which seems to get people into the zero-sum mindset that has more in common with war than it does with the fantastically generative civil societies that emerge when people learn that they have more to gain by collaborating.

Clearly, two players enjoying a winner-takes-all video game match are engaged in a positive-sum dynamic, to the extent that they’re both enjoying themselves. I, perhaps quite idiosyncratically, just prefer things when they are positive-sum on all levels.

Suffice it to say that I love Kind Words for the fact that you benefit to the extent that you show kindness and request kindness honestly: it is a microcosm of the pattern without which our species would not have moved an iota beyond the predator-prey-nature-struggle in which one lives but goes nowhere.


These facets, in combination, allow Kind Words to do something very interesting: increase a person’s affinity for strangers.

At risk of oversimplifying, perhaps the next frontier in interpersonal relationships will be humanity gaining the ability to feel an sense of kinship towards an arbitrarily large group of people. Put it this way, if you’re a normal person, you probably feel a very strong bond with your family, a powerful yet less intense one with your friends, still less towards casual acquaintances or perhaps your colleagues at work, and so on.

Ultimately, it’s possible to go so far away from you that you either don’t care about a given person, perhaps even to the extent that you would kill them or sanction their death. This state of affairs, which is of course quite natural, makes it possible for us, psychologically, to ignore the homeless person in the street, to do business with warlords, even to flatten cities during war: these actions would be impossible if we could project the level of affinity that we normally feel toward those closest to us, around the world.

I couldn’t claim that some universal brotherhood would solve all our issues; just looking at the beastly way in which some siblings treat each other is enough to eliminate that possibility. However, the ability to broaden our circle of whom we ought to wish well ought to improve the way we relate to each other, at lest a little.

I heard this idea first expressed by Daniel Schmachtenberger in conversation with Eric Weinstein on The Portal podcast. His recommendation is that we use the tradition of Metta meditation to do so.

There are variations, but generally Metta works thus: you begin by sending good wishes to someone whom it is easy to do so (some versions begin with the self, some begin with some other individual whom it is easy to love), then to someone about whom one feels neutral, then to a person for whom it is difficult, then even to large groups of people.

This is the practice of developing the ability to send kindness and well-wishes to anyone one chooses, not just those for whom it is easy due to accidents of birth, political affiliation, nationality and so on. There’s nothing wrong with loving one’s friends, but there are more novel benefits to be gained from loving those one has never met, or those who one is tempted to hate: this includes the fact that one is much less likely to be unkind or boorish by accident if one is practices wishing difficult people well.

In essence, the claim here is that it is difficult to act with a global ethic because our affections are provincial, and that we can broaden our affections by practicing the art of showing kindness to those who are not close to us.

Does that remind you of something? Kind Words, hopefully. This game is the practice of sending kindness specifically to people we do not know: if we met them they might be our friends, forgettable, or they might repulse us, but on Kind Words we know only that they’re human and that they write; it is Metta meditation, except taken a step further, with the kindness actually expressed to the relevant people. It strips away all temptations to make it about us: personal aggrandisement and popularity are impossible, the available actions in the game and the rewards are practically the same, the value of each interaction is prevented from being debased by oversupply.

Meditation, especially as expressed through practices like Metta, is an old patch on our ancient operating system. Computers, as Ted Nelson taught us, have no innate character, they scale whatever particular aspects of ourselves, intentionally or otherwise; created cynically or naively, games and social media scale those traits which give us tribal thinking and egoism. Created intelligently and in a spirit of humanism, games and social media can scale the kindness, simplicity and humbleness required to solve the civilisation-level problems that face us.

But I have three precious things which I prize and hold fast. The first is gentleness; the second is economy; and the third is shrinking from taking precedence of others.

With that gentleness I can be bold; with that economy I can be liberal; shrinking from taking precedence of others, I can become a vessel of the highest honour. Now-a-days they give up gentleness and are all for being bold; economy, and are all for being liberal; the hindmost place, and seek only to be foremost; — (of all which the end is) death.

— Tao Te Ching


  1. I mention the lovely Kind Words Soundtrack by Clark Aboud briefly but do not do it justice.
  2. Being on Steam, Linux users like me aren’t well-catered for: Steam supports only Ubuntu Linux, so I (using Fedora Linux alongside 1.2 million others) have to run an Ubuntu virtual machine. Kind Words is good enough to justify the trouble, but the trouble is totally avoidable.

Leave a Reply