Of late I have been pondering whether there may be viable business models in providing support for some of the more abstruse forms of exchange that can exist between sentient entities, with a focus on trade that is not widely accepted as being trade at this time. Some thinking out loud will happen in this post on the way to perhaps nowhere in particular.

For the purposes of clarifying my own thoughts, I'm going to create a taxonomy for a few broad types of trade. When I talk about trade I mean the dictionary definition, or at least that is where I'm starting: a trade is an exchange that occurs between any arbitrary number of entities. Typically trade refers to property transfers these days, whether that is currency for groceries or gold for oil, but I'm going to expand my usage here to include any actions on the part of the trading partners, not just assignment of property rights. If you publicly announce that you will view a specific painting if each of fifty people you have never met give $10 to their favorite charity, and this comes to pass, that is still a trade.

Trade occurs because those involved derive value: they consider themselves better off afterwards than before. Value is subjective and a slippery thing to pin down when looking at the trades made by other people. It can include how the participant feels, expectations of long-term gain even when the short term outcome is a loss of value, and even more nebulous line items. Perceived value can vary widely by state of mind, time of day, duration of the trade, and for any number of other reasons.

The Vanilla Standard: Causal Negotiated Trade

Most trades are causal, and more importantly, negotiated. By causal, I mean that a trade proceeds by cause and effect, moving forward in time, as participants take actions and react to the past actions of others. Trade is negotiated because it involves multiple exchanges of information on the way to completion: the participants communicate with one another directly and deliberately to establish the items or actions that will be traded.

A simple example is that I place an ad to sell a car, someone answers it, we haggle, the car is sold. That is causal, negotiated trade. So is walking into a market to buy an apple at the set price: intent and desire is communicated in both directions as a part of conducting the trade.

Causal Non-Negotiated Trade

Can trade be non-negotiated? Can parties trade without communicating to define the trade? Yes, and an (at least partial) example can be found in cultures with strong traditions involving gifts and obligations: recipients don't have a say in what is received, and often don't choose the timing either, but there is an evolved system of reciprocity over time. You might think of this as a unilateral approach to trade, with each party acting in isolation until handing over a gift.

That doesn't mean gifting is simple in practice: consider the complexities of the gift economy of software development in which code can be published and used - even actually exchanged - by people who never meet and never talk to one another. For a more direct example, consider the sort of gift exchanges theorized by sociologists and archaeologists to have taken place in prehistoric tribal societies.

Non-negotiated trade involves a process of modeling. The gift-giver doesn't communicate with the recipient but rather models the recipient in an attempt to analyze wants, desires, and reactions so as to decide on a suitable gift.

I have been vague on what constitutes communication about the trade, however. Arguably the culture of the open source software gift community is a form of diffuse communication between parties, even though there is often no direct communication between people who make use of one another's published code. Similar for prehistoric societies, there must have been some form of communication associated with gift-giving ceremonies even if the gifts were already determined in advance. The point is that there is no direct negotiation regarding the nature of the traded goods, services, or activities.

Causal Non-Negotiated Trade, Defined in the Past

Trade between the past and the future is commonplace: a great deal of culture is built upon the transmission of ideals and desires beyond the lives of the originators, and we can view much of this as a form of causal non-negotiated trade. It is non-negotiated because the terms of the trade are set out in the past, and the future participants don't have any say in that - they only choose when and whether or not they wish to take the trade by undertaking the actions defined by the past trading partners. At the originating end a potential participant specifies the acceptable trade criteria, and at the other end a potential participant makes a yes or no call on whether to undertake such a trade.

Consider two trading partners that we shall call Past Foo and Future Bar. At some point in the past, Past Foo invented and commercialized an important therapy for a rare pediatric disease. As a part of the public record of these events, Past Foo stated that he desired that anyone who benefited from his work visit his statue and tithe to a medical research foundation. Future Bar would likely not have survived to adulthood but for the work of Past Foo. He has access to the historical record, and determines to make the trade with Past Foo: he does so by following the terms laid out by Past Foo.

Wills with conditions can also be considered a form of causal non-negotiated trade, where the trading partners are specified exactly rather than loosely.

There are also further examples of causal non-negotiated trade operating on a shorter time frame: consider that some individual online content creators post wish lists to encourage supporters to send specific gifts, and these gifts can arrive from supporters who have never communicated with the creator. These are causal non-negotiated trades: an individual provided a service in the (near) past and requested specific actions from one or more people in the (near) future.

Causal Non-Negotiated Trade, Defined in the Future

For actions to be considered a trade, do all parties have to agree that they are in fact participating in that trade? For example, we can outline something that looks a lot like the trade between Past Foo and Future Bar but is defined at its future end rather than its past end.

Let us suppose that instead of stating that he wanted beneficiaries of his work to undertake certain actions, Past Foo made no requests: he advanced the state of the art and created benefits in the usual way for contemporary rewards, and asked for nothing from the future. Nonetheless, Past Foo gave many interviews on his views and was the subject of a detailed biography. Future Bar, who benefited greatly from the work of Past Foo, has access to these historical records. He determines to make a trade by taking actions that he believes Past Foo would have desired: providing financial support for ongoing medical research in the same field.

Thus we have a non-negotiated trade wherein the two sides take place at different times, and which is defined and arguably initiated by the future trading party rather than the past trading party. This can occur even if Past Foo and Future Bar are not contemporary: Past Foo might have been dead for decades at the time Future Bar creates the trade.

Isolated (or Acausal) Trade

As you might have noticed these examples are moving in the direction of less communication and contact between trade participants. Can trade take place when the trading parties never in fact interact or communicate in any way? Hypothetically yes. All parties to the trade must be able to infer the existence of at least some of the others, model their desires, and then act out their own part of the trade. This would run much along the lines of the example above of a trade initiated at its future end, but instead of looking at historical fact each trading party models the likelihood of existence and desires of the other trading partners.

In that part of the futurist community interested in strong artificial intelligence this is known as acausal trade. If there is by definition no interaction between parties to the trade then it doesn't matter when or where these entities exist in relation to one another: past or future, inside or outside one other's light cones, or perhaps in other parts of the hypothetical multiverse based on the Many Worlds interpretation of quantum mechanics. All they need is the ability to infer the probabilities of one another's existence and participation.

What actions would be exchanged between isolated trading partners? What could they do that would provide value to one another given that no action will have any effect on the other trading partner's timeline? Value is subjective, so one might argue that an altruistic entity would look to trade acts that do good, and derive value from such acts even if taking place at a distantly removed location in time and space. In the case of the futurist AI community, acausal trade centers around entities who agree to build the processing power to emulate one other's minds, and who each find it valuable that such copies of themselves exist.

The challenge for isolated trade, and what makes it - I think - more of an interesting idea than a practical activity at this time, is the need to understand the odds of the modeled trading partners' existence. To take a trade, you have to believe that you are obtaining a decent expectation value from it: when that value is (or was, or will be) provided by an entity that may or may not exist (or have existed, or come into existence), that existential uncertainty reduces the expectation value of the trade. But by how much? I don't think that it is possible at present to quantify the odds of existence for suitable isolated trade partners in any useful model, or even to create such a model - though it may be the case that something can be done with a few of the better chronicled portions of recent history.

Present impossibilities become future technologies, however. I would expect that given enough processing power even this sort of speculative modeling and assessment of probabilities will become tractable - for at least historical scenarios. Our understanding of cosmology is not yet comprehensive enough to do more than speculate about other options.

A Dust Hypothesis For Trade

You'll notice that in the example of non-negotiated trade wherein the trade is defined in the future, I removed all intent to trade from the past participant. Moving another step along the line of reduced intent, can we call it trade when actions undertaken by various entities provide value to one another, yet all such entities are unaware of one another and have no intent to trade? Further, is it a trade if these actions would only have value to another party if they are known?

To return to the altruists, who derive value from knowing that good is done somewhere by someone: if two altruists each donate to charity a thousand years removed from one another, and the future altruist knows nothing of the past altruist, can we arbitrarily declare this to be a trade? The difference between this and an acausal trade are the lack of intent and modeling on the part of those involved.

There are further awkward questions to encounter if proceeding down this road. Is it a trade if the entities are not sufficiently aware and intelligent to understand the concept of trading? Does trade have to involve volition rather than unconscious, unintelligent, or involuntary actions? Consider the evolved symbiosis between stomach bacteria and we humans. Are we trading for mutual benefit with our bacteria? Were Medieval people, who had no knowledge of such things, still making the same trades? Can we even call it trade when the other side isn't sentient?

The Dust Hypothesis is slightly disturbing in much the same way as the Many Worlds interpretation of quantum mechanics, especially for people who make little distinction between the self and a copy of the self, but it bears some semblance to where one ends up when thinking about trade in this way.

But Back to Business Models

Metaphysics is metaphysics, but business is business. At some point in considerations one realizes that one has fallen off the path and into the water: for the purposes of making a living it doesn't seem to matter all that much whether or not we can link the universe with a web of non-negotiated trades between sentient and non-sentient entities.

Practical concerns suggest that we should draw the line at a definition of trade that requires at least one volitional party, since selling services to volitional parties is the objective of this line of thinking. The timeline of the world, and maybe the universe, is still linked by a web of trade, but we only have to concern ourselves with trades in which we can access one of the volitional parties involved so as to propose a traditional, vanilla causal negotiated exchange: here is a service or product that will help you with your non-negotiated and possibly isolated trades, please consider paying me for it.

Some potential classes of service or product follow, in no particular order, and meandering between near term and long term types of causal non-negotiated trade.

Improve Distant Future Discovery of Present Trade Participants

There are a range of ways in which we might increase the likelihood of an individual being discovered and accurately modeled by desired types of future trading party, with an emphasis on events that are at least a few decades away, possibly more. This boils down to creating a larger footprint in records most likely to last a long time, and tying that footprint to (a) keys to identify the individual, (b) what the individual is offering in trade: specific actions taken, deeds done, and so forth, and (c) optionally, what the individual would like from the future.

This overlaps with the desire most people have to leave a mark on the world, some record that they existed - a transmission of what they consider to be important. So such a service has a broader potential client base than just those individuals who seek to establish long-delayed trade with the future. That would be a small to non-existent demographic at this point. (Though one could always argue that any monument to the self is exactly that: perhaps vague, but an attempt to obtain some future behavior in exchange for past action - even if as simple as provoking certain thoughts based on the words on a gravestone).

It seems a fair bet to assume that future individuals will have a much greater capacity to model trading partners in the past, especially those from the information era. If, for example, we set up a secure time vault for corrosion-resistant tablets containing participating individuals' social security numbers, email addresses, phone numbers, a selection of IP addresses, and trade details, it is entirely plausible that an entity a century or two from now could build corroboration and a fair model of that individual, working from inherited copies of the enormous phone and internet data vaults presently under construction. This strategy piggy-backs on the efforts of big business, government intelligence agencies, and other parties that are recording a large fraction of all human interactions - few of these efforts are in any way benign, but all of that data is likely to survive for quite some time.

Unless you specifically declare your position, then it is very possible you will be misinterpreted, or overlooked in favor of people who did declare their trades or generate a lot of data - the famous and the foresightful.

The Bitcoin block chain, or that of whatever turns out to be its more broadly adopted successor, is another possibly interesting place to place important messages for the future. This data is also likely to be preserved for posterity.

Improve Present Discovery of Near Term Past Trade Participants

People who are looking for specific types of past actions to reward in the present can be assisted through marketplaces, search, historical data, analysis, and modeling. This doesn't have to start with the urge to trade: on the short term this includes the Google scenario of looking for desired forms of content created in the near past and then deciding to reward the creator with a donation. So there is a swathe of very mainstream economic activity taking place under this heading.

Better Remuneration Systems for Near Term Past Trade Participants

If you look at gift economics such as the open source software community, there is little in the way of what you see in the content creation ecosystem - people setting up wish lists and donation mechanisms. One could try to make trading from future to past more direct and easy through any one of a variety of types of reward or micropayment systems, and ways to reliably send resources to an individual based upon their public profile without that individual having to first take any action. Examples of this sort of thing can be seen in the bitcoin ecosystem (tipping bots on Reddit, for example), and in projects like Gittip.

Improve Discovery of Distant Past Trade Participants

Many present groups and individuals have at least some interest in participating in non-negotiated culture trade with the past: conservative societies, anachronists, luddites, recreationists, and so forth. This is rarely explicitly framed in these terms, but there may be a market for informational resources for those who wish to adopt the behaviors desired for the future by past individuals, and who think that the efforts of those past individuals provide value. Think of publications and databases for presently popular past eras, with a focus on what specific individuals wanted - either generally, or specifically from future individuals.

Here we run into the limits of modeling, and would essentially be mining the historical record: presenting information on specific times and regions in the past with a different slant to the usual historical record. We would be looking at biographical materials, wills, and other materials in which individuals presented enough information about themselves to know what they did and what they wanted. There may be some overlap here with the general market for historical analysis and repackaging of primary research materials.

Coming to No Specific Conclusion

Evidently this is something I'm still mulling over, with numerous other possibilities. Thus this piece trails off into an end without much in the way of a conclusion. Clearly a lot of people are working on ways to improve some forms of short-term past to future and future to past causal non-negotiated trade. Gift economies and the value they provide are broadly recognized.

Longer term causal non-negotiated trade, between centuries or generations or lifetimes, is poorly recognized I think, despite the fact that a lot of people engage in it in various ways. One can imagine a business that wraps past and future non-negotiated trade discovery services under a single heading - "Reward the Past, Guide the Future" or similar - and builds its market by starting with groups that seek to emulate or maintain specific aspects of the past.

So thinking continues, perhaps to no specific endpoint, as warned at the outset.