You know how ants leave pheromones to let other ants know where to go? Well as Dana-Edwards has pointed out, this is kind of how Steem works. Your votes create a neural network of sorts, leading other steamians to the best content.
So, it isn't rocket science to move this idea to code.
First you allow people to put up very simple rules or pieces of logic. From this they can create a computer-like device.
With these simple building-pieces they have a ‘computer’, which they can now program or instruct to do anything.
Electronic principles of logic gates and circuits could be the inspiration for initial coding. Maybe slightly more advanced programmability and fuzziness could be available in each of these pieces or instructions.
Anyone can add code or rules, but based on votes only the best stuff becomes obvious.
So voting is like Steem. When you vote for a block of code, each edge and node receives so much of the vote. (I want to call edges ‘stems' or ‘lines' for some reason.) The larger the conglomeration of 'code' you vote for, the more influence you exert, but with less focus. The smaller the code block, the more focused the vote becomes.
Voting for specific code strengthens pathways. Using code could strengthen pathways as well, using the same or different units.
The formula for voting influence is P = SQRT( n ) / n, where each piece getting up-voted is P, and n is the number of pieces.
Anyone using this large computer chooses what code to use. What does he choose? It depends on what he wants. He could basically run the program himself, all the code being online. He could use an interpreter of some sort with customizable settings.
The user may not choose every piece of code by hand. He may have to trust other people's opinions or copy their settings with a few tweaks.
Voting may have ‘colors’ so that different categories could be represented. Based on the color, a different behavior or function is manifested.
You get out your interpreter. You want to run a cool program someone suggested, on this Smart Commons. You copy and past some settings, and then run the program. The settings may be specific nodes and branches to use, or may be rules of how to choose pathways.
In the second case, your 'program' may run even better than expected or may not function quite right, as it used to.
Here's a vision of what could happen. A bunch of people start adding nodes representing very simple logic. It's a game at first with tiny helloworld type results.
Then some intelligent people work together and build a cool evolving computer brain type thing, which takes commands.
Or someone makes an actual computer that takes specific commands. It could even have a compiler.
The point is a simple computer would be built which then could help interpret other pathways or code people add, to simplify things for more ordinary people.
So instead of one choosing a version or branch of code to run from Github, the system is constantly evolving real-time.
I'm not sure what this most simplest of logic units would be, but it might involve simple bit-math. It could be slightly more evolved than electricity logic. (It may be based on the Borg… just kidding.)
I should go into this more, but it should be very simple, possibly having blockchain, network or broadcast type capabilities.
This may mean having pulses or waves that push execution of operations. Both bitcoin and computers depend on a kind of heart beat which ensures safe completion of execution-blocks.
The above paragraph assumes that there is some online version of constantly running code, but my initial idea is to have a fluid set of rules only. All execution of rules would be done by individual users. Of course, as they get results they could be re-inserted into the 'computer' and voted on.
This idea is to simply allow anyone to put up code or logic and have people vote on the most useful pathways. Then each user chooses how to use it or what to run.
