web analytics

This is my proposal for a set of principles to facilitate interoperable conversations.

Taking inspiration from the Internet pioneer Jon Postel and extending his Law, I have created some principles which, I hope, will provide for more robust conversations between people of divergent backgrounds and viewpoints.

Here they are:

  1. When speaking: be clear.
  2. When listening: be charitable
  3. When speaking: try not to cause offense.
  4. When listening: don’t take offense.
  5. When good-faith errors occur: be charitable.
  6. When bad-faith errors occur: treat them like good-faith errors.

Explanation and Discussion

I am proposing that we take steps to make our conversations and modes of communication work better among people of different mindsets. This, I hope, will avoid much of the unnecessary offence-taking and confusion, and facilitate more sharing between people who see the world differently.

Recently, I’ve seen a great number of conversations become unproductive or even get nasty, but not necessarily due to any real enmity or disagreement between the participants. My hypothesis is that this is happening due to the increasing Balkanization of media, disciplines and conversations, which is lowering our ability do understand people that don’t think like us.

Though the current situation consists of multiple incompatible narratives, I am not advocating for a single, shared narrative. I am advocating for some very simple shared principles that provide sufficient alignment to communicate as one, alongside sufficient freedom to coordinate on any scale.


After reading up on the history of the Internet (ARPANET, TCP, Jon Postel, etc), I thought that it would be interesting to look at conversation as though discussing technical communications protocols.

Interoperability is defined thus:

Interoperability is a characteristic of a product or system, whose interfaces are completely understood, to work with other products or systems, at present or in the future, in either implementation or access, without any restrictions.

Interoperability in our systems is immensely productive: with a few exceptions, you can open most files on most media on most computers and get somewhere. Decades ago, we were faced with a plethora of incompatible file systems and file types, meaning a file saved on a disk on one computer might not open on another.

Large technology companies were once very reactionary in this regard, attempting to make their systems closed and accessible only by their products.

Open source and free software projects like GNU, Linux, etc. are almost interoperable by definition: they make all the aspects of their systems available so others can create workable complements. Meanwhile, the large closed source tech corporations have become more open recently, both with pressure from open source projects and because interoperability is actually rather profitable.

To return to the point, and in plain terms, interoperability can be thought of as a common agreement as to where to find things, what they are called, and what they are supposed to do.

USB works because there’s an agreed standard as to which pin in the port corresponds to a given function: if a company created devices that swapped the function of two pins and didn’t tell anyone that they had done so, they would be unpopular and their products wouldn’t work.

People do this all the time in thinking and talking; if your computer program is flawed or your product non-interoperable, it will simply not work; if your ideas are internally inconsistent and your definitions off, you can do a great deal: including building cults, religions and corporations.

Fundamentally, I am making these recommendations for the sake of progress and social sense-making, not because it seems profitable. Rather, there appears to be a certain advantage to corporations and religions generating non-interoperable narratives in the minds of their clients — the person whose thinking is incompatible with others not from their tribe will find it hard to leave, not unlike if all your files and software work only with Apple.

Some Examples of Non-Interoperable Conversations Going Wrong

It seems apt to give some examples of how conversations can go wrong. I’ve deliberately chosen political examples, but nothing contemporary so as to avoid pushing buttons unnecessarily. Both examples also address the definitions of words: this is not necessary either, narratives, interpretations, premises, etc. can all posses or lack interoperability.


I had a lovely conversation with some friends regarding socialism. However, we burned about 30 minutes spinning our wheels because we weren’t agreed on definitions. The disagreement went along the lines of:

Participant 1: I advocate socialism.

Participant 2: The common ownership of the means of production? That would be a disaster.

Participant 1: I don’t mean like that, I mean like Nordic countries.

Did you see what happened? The dictionary definition of socialism is “a political and economic theory of social organization which advocates that the means of production, distribution, and exchange should be owned or regulated by the community as a whole.”

Many people, meanwhile, call the system of government in Nordic countries “socialism,” in that those nations advocate for the social good; socialism more as an attitude rather than an economic system.

Fundamentally, someone might say they want socialism because they think that a strong social safety net is a good idea, but their conversational partner might be dumbfounded, thinking that the speaker is advocating for the forced purchase of the country’s mines, utilities, public transport, farms, etc.

Misunderstandings like this risk sending conversations off the rails because people are so easy to offend and because misunderstandings cause frustration.

To return to the USB analogy: imagine trying to access the “strong social safety net” pin on the plug (labeled “socialism”), but finding that it was actually the “communal ownership of the means of production” pin.

Libertarian Socialism

Please forgive the preponderance of socialism, here, but note that the failure of communication here is to do with the word “Libertarian”. A friend of mine took a political quiz, which spat out the orientation: “Libertarian Socialist” in response to their answers.

They responded to say that “Libertarian” didn’t describe them, as they didn’t advocate deregulation and were pro the welfare state.

Rather, the political system to which Libertarian Socialism refers has little to do with Libertarians as in the political party, or for that matter Ron Paul or Gary Johnson. It is a variant of socialism that advocates against the use of force by a centralized state or vanguard party, and instead for decentralized and local control by communes or syndicates.

Libertarian the word, for that matter, is used in many other non-Ron Paul contexts, such as Libertarian free will.

You might see, dear reader, that this error could have been solved in at least two ways: 1. The quiz could have more clear in explaining what the result meant, and warning against comparisons with a well-known political believe of a similar name; 2. My friend could have looked up the term so as to clarify.

What Happens when Things Go Wrong

So far as I can tell, most of the time one of three things happen when things go wrong.

  1. Confusion: people just can’t understand each other and share ideas.
  2. Offense: people get offended or angry, which usually puts as stop to communication or at least hampers it.
  3. Escalation: the misunderstanding is characterized as a sign of a particular party’s malintent. Sometimes people do genuinely cause misunderstandings on purpose, but escalation is usually unnecessary.

My Proposal

This leads me to my proposal, as outlined at the start of this piece: we should modify Internet pioneer Jon Postel’s Law, “Be conservative in what you send, be liberal in what you accept.”

Practically, in the terms of the Internet, this means that devices should transmit messages that adhere to the protocols strictly, but should accept messages whose message is clear but which are not compliant.

To modify this for normal conversation, we might say: speak in complete adherence with definitions and assumptions that you are justified in presuming that your conversational partners know (if in doubt, define); listen with an open mind, take reasonable logical jumps to interpret speech that is not totally clear and, if in doubt, ask.

This is to assume the shunning of logical fallacies, personal attacks, etc. I’d like to extend this into six separate principles, although intuitively I feel that the latter four can be derived from the first two:


1. When speaking: be clear.

Use only definitions and assumptions that are commonly accepted, known to your audience, or define what your argument requires.

Example: Clarify that in “Libertarian Socialism” the modifier libertarian refers only to an emphasis on freedom and decentralization, not to the political movement.

2. When listening: be charitable.

Before assuming foul play or extreme ideas on the part of the speaker, consider other more charitable ways of interpreting their words, or ask for clarification.

Example: Might “Libertarian” in this context be descriptive, rather than denotative?


3. When speaking: try not to cause offense.

“A gentleman never insults anyone unintentionally.”

Example: Libertarian can be an insult, so it might make sense to use another word or to clarify, so as to avoid unnecessary offense.

4. When listening: don’t take offense.

Just don’t take offence.

Example: If Libertarian is merely descriptive, there’s no point in being offended because one isn’t being targeted for an insult; if Libertarian were meant to cause offense, why give them the satisfaction?

It’s tempting to go deep down a tangent on this one, but I’ll save it for another post. Put it this way: if the next time that something that would have offended you failed to offend because you had the self-mastery to resist, would you have lost anything? We can maintain our sense of justice and decency without the anger and clouded-thought that comes from taking offense.


5. When good-faith errors occur: be charitable.

It is very easy to castigate speakers or listeners for not understanding, or even for getting offended due to a misunderstanding. This is common when discussing subjects that relate to morality or politics. Mistakes and misunderstandings don’t necessarily imply foul play on the part of either participant. It makes better sense to forgive mistakes and, again, clarify where possible.

Example: the term Libertarian may be confusing, and people may even get offended when described thus, but don’t assume deliberate obfuscation or strategic offense-taking unless there’s very good reason.

6. When bad-faith errors occur: treat them like good-faith errors.

Be vigilant in identifying bad-faith actors and actions, but address them as one would good-faith errors.

Example: Young-Earth creationists are especially bad for bad-faith errors of communication. “Evolution is just a theory.” they say, having been told on multiple occasions that in this context, “theory” refers to the highest available level of certainty, quite distinct from what most people mean when they say, “in theory.”

However, all bad-faith errors should be treated like good-faith errors: a false positive is like calling an honest person a liar (usually a conversation-stopper). Meanwhile, many people that act as though in bad faith are in fact innocently repeating arguments they they have not interrogated: many young-Earth creationists are this way.

Nonetheless, real bad-faith actors are the spammers or botnets of conversation. They degrade communication, cause discord, and waste time. However, the accusation of bad faith action is the nuclear option: we should avoid unnecessary proliferation. I am, thus, open minded and curious as to how to deal with it.

My best idea is to firstly to kill bad actors with kindness, applying the rules of clarity and charity, and if necessary refer them to the rules: “Your speech is not clear.” “Your listening is not charitable.” Beyond this, and for more prolific bad actors, I am open to suggestion.

Conclusion: What We Stand to Gain

As I mentioned in the body of this post, the incentives are not all on our side on this one: locking people into non-interoperable structures of thought and definition can be a successful strategy in business, politics and beyond.

That said, conversation is the way we move forward and innovate. Moreover, some of the most profound insights are found when worldviews and disciplines collide. As such, I hope this system allows us to cross wider gulfs of worldview and mindset.

As mentioned in the discussion, my fear is that the constituent mindsets present, say, in the USA, become so Balkanized that they risk losing the ability to coordinate on a national scale. It is particularly tragic, for example, to see people united by goals and ethics divided by definitions and narratives.

Nor is a unity of mindset desirable: such a scenario (and the desire to achieve it) is both authoritarian and fragile.

Rather, and as Nassim Taleb would recommend, the best configuration appears to be fractal: fractals contain infinite variation and depth, but a common structure permeates; in a collective mindset, this would correspond to very simple universal norms and values that pertain mainly to communication itself, with successively diversifying viewpoints of immense variety.

This structure would facilitate collective action and sensemaking, would be antifragile, and would promote the diversity of thought and perspective that powers science, thought and innovation.

2 thoughts on “Conversational Interoperability”

Leave a Reply