altruism genes, natural selection, and how this relates to bitcoin

Recently I’ve heard a radio discussion about how “gene of altruism” survives natural selection. That discussion made me think about how modern network peer-2-peer applications should be designed to “survive” in their own natural selection.

At first when you think about altruism vs. natural selection - seems like altruism has no chances, as the altruistic individual is likely to leave no descendants, thus his genes will not propagate.

So how do the altruism gene survives?

A positive factor for altruism gene survival is sacrificing to someone with the same or similar gene, so that while some altruistic individuals die, their genes survive effectively because of coordinated efforts of multiple individuals.

Think of human body as a conglomerate of cells - every cell is extremely altruistic, as all functionally different cells are sharing same genetic code and working towards same goal, and only few germ cells are making it towards the reproduction.
Honey bees are going even further - numerous individuals share same genetic code and work selflessly towards survival of the whole hive.

DNA code defining actions of an individual on this level of abstraction is similar to a software program code defining actions of an electronic device.
Multiple individuals with similar DNA acting together towards same goal then are similar to peer2peer software installed on multiple devices across the network. Software defines actions of individual devices and interfaces and interactions between devices, and “sacrifice” of one device for sake of another is just a specific routine in the interactions protocol.

So if a program (DNA or software) allows recognition of compatible nodes and defines interaction protocol - network of nodes creates distributed system of a higher level, and this system as a whole is bringing much more profit to every node comparing to unlinked nodes working by themselves.

However, in cases where individuals are not sharing same genetic code (source code) and are not completely compatible - various problems arise.

There was an example In that discussion - certain unicellular creatures under harsh conditions can combine themselves into sporocarp, a fungi-like multicellular quasi-organism, consisting of stipe and cap. And, what’s important - cells that are getting into stipe are losing ability to reproduce, while cap dwellers keep this ability. Stripe is critical for survival of the whole organism, but getting into the it is completely unprofitable for a single cell (remember - it’s a quasiorganism, a temporary formation of otherwise competing unicellulars).
Eventually some of such unicellulars developing ability to always get into the cap, and survive at the expense of others. However, after a while all altruists in this “community” extinct and the cooperation ability is lost completely and has to be reinvented from scratch in next generations.
Result - complex and effective distributed system cannot be formed of nodes with different source code.

Society

Similar approach can actually be applied to the human society.
Consider these DNA differences and natural selection pressure - and origins of the stratified classful society structure becomes quite obvious, also it’s also obvious that a complex, effective and fair for all society cannot be based on nonaugmented human beings. If individual acts only for his own reasonably selfish interests - the total benefit for society is likely to be negligible. If the individual acts for the sake of whole society - it’s likely for him to be unsuccessful in personal reproduction and his altruistic genes are likely to be lost.

We can and should consider not only DNA, but education, culture, legal restrictions and other conditionings applied to every members of any society. These mental and social conditionings are acting similar to DNA, effectively being transferred from parents to kids and from older existing members of particular society to the newborns.

However human is not reliable in this matter because:
- there is a need for rearing and education process, lengthy, expensive, error-prone and with unpredictable final result, and it has to be applied to each individual separately;
- a human can and will eventually forget conditionings applied to him;
- there are underlying DNA-defined contradicting altruistic and egoistic programmes that interfere with social conditioning and each other and often provide weird results;
- information processing capacities of an average human are quite limited.

So in the end of the day - structuring and labor division do exist in the world, but it’s complexity and effectiveness is limited by the capacity of single nodes of the executing network - human beings.

Through the history humanity consequently increased effectiveness of universal society codes of conduct. Laws were created and enforced, centralized education and state propaganda were introduced, commandments were saved on external devices (books). Eventually codes evolved and overall effectiveness increased. However the basic limitation is still there - all possible society codes of conduct are executed by weak mortal humans with all possible side effects.

Solution

Now we have our beloved computing devices entering the scene. It allows effective saving, quick and reliable replication, unified and fast execution and potentially unlimited complexity of “codes of conduct“, all of which is impossible for humans.

So if the altruistic behaviour is implemented in the software solutions and protocols - the whole distributed software system can reliably expect every it’s member to act in agreement towards mutual benefit.

The obvious conclusion - to improve certain social interactions it should be implemented in some kind of p2p software, and this software should use an unified protocol that will imply and enforce altruistic approach for every node of the network.

“Natural selection survival” of the software

Okay, let’s think what software should look like to succeed in the competition. Survival and replication of the software depends on how good it fits into the environment it lives in. The “environment” here means actual software users, humans owning devices that run programs. Users decide to install or delete the software based on the usefulness they feel.

Requirements applied to software
A p2p software with unified protocol is likely to be more effective than competing non-p2p software and is likely to be chosen by users.

It is important though that decision making has to be hardcoded into software and not delegated to users. If it’s given away to users - we again have all sorts of unwanted side effects and system falls apart.

It’s also important that software should be able to “mutate”, to adjust and improve itself to fit ever changing user requirements. An open source software allows rapid parallel multiple mutations, allowing effective and quick improvement. Also every mutation has to prove itself useful through network adoption, and useless mutations will not survive the competition.

Torrent protocol proved itself to be more useful than anything else for file distribution and downloads, and it won the evolution race.

Requirements applied to user
User has to donate some part of it’s PC resources and data input to the network.
Obviously user would prefer to consume useful resources and not give back anything, so the system has to either apply restrictions or provide incentives for the user to give back.

Torrents, especially at the early days have been issuing strict seeding requirements to prevent leechers from ruining the system.

Bitcoin implements a mining mechanism that is monetarily rewarded.

So in general the system should rely on these principles:
- there are some resources created by the system that allow to solve certain real world problems
- the system provides these resources to the users in return of some work of the sake of the system

So the effective system will be:
- distributed;
- with no privileged nodes;
- with one unified protocol;
- insensitive to node destruction;
- working out of the box;
- without any competition between the nodes, only with cooperation;
- opensourced;
- enforcing or incentivising user altruism.

Examples
Obvious examples are torrents, bitcoin and partially skype (before it was bought by M$).

Torrents for free gave us download speeds that otherwise would require massive infrastructure investments.

Bitcoin allows secure, fast, reliable and cheap transactions and majorly outperformes conventional currencies in these areas.

Skype set new standards of international call costs.

However these systems are obviously imperfect by itself and are far not enough.

Skype is well known to be not fully decentralised, and unfortunately now it tends to centralize further.

Bitcoin is often accused to have no real value. Some may argue that fiat currencies have no real value also, but there is an obvious demand for such functionality, and it’s just yet not implemented in any currency.

What systems might be useful?
In fact almost all of the existing interactions within the society are still centralized, so there are plenty of places to go. And in most of these places p2p implementation will bring more benefit to it’s nodes than a centralized one, and of course it will bring less benefits to the currently existing privileged central authorities.

For example - baking with real values for the currency. In the real world we have various ledgers, databases and accounts of property, cars, all possible goods in fact. If all these real world values will be hashed and attributed to the blockchain, and any transactions with these real values will be always accompanied by the opposing money transactions - it will make cryptocurrency effectively backed by all real world values that sit in the blockchain.

Another example - a secure and reliable storage of socially important information. We have Google Cache, we have Wikipedia, we have Internet Archive, but all these are more-less centralized services, vulnerable to legal attacks, human errors, natural or artificial disasters.

It’s still long way to go, but eventually it might incredibly rise overall society effectiveness. And we already see results.