Space to Think

The complexity of Multi-Master replication does not stop at looking up what's been new in other replica sites. We could make the data that was created on one node only editable via interacting with that node, but all created by every node replicated to all other nodes.

Yes, I think, currently, the BDR project is a way to go, for example, clickup.com uses it. However, need to dig into it, it would be great to have it decentralized.

// 疑问：狗是通过什么感受器来判断的？ //

Right, so, what human features are dogs sensitive to, and how to extract those features? Now, you're talking of how AI could replace dogs...

// If people look for computational complexity in coins when buying them, then all coins are like NFTs.

Well, computations themselves would be fungible, while the coins in the old sense would not.

Currency wallets "have" the amount it claims to have, because machines believe in the balance computed one particular way. If you introduce that other (more correct) way of computing balance, it would change the game.

If people look for computational complexity in coins when buying them, then all coins are like NFTs.

I read an article that said performance with regard to instruction selection is 10% of total performance.

The rest is cache behaviour. If you get a cache miss, you slow everything down. Computers should be able to analyse our code and work out how to arrange it so that the data is always in the cache. By simply placing data together which belong together.

I should be able to host a variety of services with low maintenance effort by bringing my servers with me.

I should be able to download a program, enter an AWS access key and secret key and get creation of servers in the cloud for me. And use the same program to manage the servers. This program should show me the resource use of the servers and let me create user accounts for various services.

I found this thread about semantic sysadministration quite interesting: https://ttm.sh/dVy.md

Got to be kidding. This is really an issue, how people can be the first-class systems of the web (web sytizens)? How can individuals be as reliable as large enterprise systems, and as clear, interactive and interoperable as standard-compliant APIs, yet as free as text and binary files?

I think we need a general purpose resource scheduler. An organisation could have one for jobs. Allocates people to teams, to jobs.

It is interesting to consider, that pretty much any application is just an arrangement of related objects of types with modification rules. One could imagine having something like "Application-Independent-Related Objects" (AIRO), that in combination behaves like an application, but are not an application, but rather are free objects with their independent identities and locations in address spaces.

This idea is a bit modified. The original idea was to freeze by pressure alone, because water freezes at +36.5'C under pressure of approx. 1GPa, directly into Ice VI, but this idea was criticized, because the body undergoing such pressure would heat up, and that may cause unwanted chemical changes in the body, even if water stays solid at much higher temperatures with even higher pressures.

// competitors will see who they are buying from //

if your business depends solely on your supplier secrecy, then your business is one step away from being a commodity.

Competition is a good thing.

Destroying other people who want to do the same thing as you is also destroying competition. And companies hate legitimate competition. They would rather be a monopoly. Monopolies are really inefficient.

I worry with a transparent marketplace, a company would startup and hire people to pick up all the work and give nobody else a chance to pick up the work.

I think society needs to escape this idea as competition as a means of destroying your competition. it's good when there are multiple people offering the same services. So they can compete on price, quality and branding. But this idea of destroying your competition and creating moats is really harmful. We should have a society where people don't fear being made irrelevant.

Yahoo SM (competitor of Google AdWords) was once a transparent market with "allows you to see who you are bidding against and what they are bidding, so you know exactly where you will rank, and how much you will pay". :)

So, what you're proposing, is some kind of open database or aggregation of market orders, right? It definitely relates to your idea of Wantsfiles and Wants Manifestos.

A problem with that is people's privacy: people don't always want their orders be visible due to multiple reasons:

• competitors will see who they are buying from
• ideologically different will despise each others (while they are friendly to each other not knowing)

So, how do you resolve that? Should people have one big open market, where they know that what they share there is public, and then they can have encrypted orders optionally, or how else would you propose?

Think about it -- in fact, every search query is a market order, and I think, "Categories" are "Queries" :) Obviously, Google has been getting lots of queries, and if every "Query" is a category, it has been hierarchically grouping them to fill these search orders (market demand) with link excerpts (market supply).

Thinking that way, we can see that every request-response (or client-server) system can be viewed as order-filling apparatus. Since the cost of fulfilling an order is usually computational-time, and human-time, and machining-time, you could measure the cost of said orders as such. For example:

• ask a friend to a party -> (human-only)
• make a cup with ready dumb coffee machine -> (machine-only)
• buy bitcoin -> (computational-only)
• buy a product on Amazon -> (computational, human, machine)

It is easy to imagine that, given such open knowledge about world's orders, it would be possible to figure out where to make money, but it is also important to make a distinction between "compute-money" and "human-money" and "machine-money", because human brain (with current BCI capabilities) will never mine the amount of bitcoin that a machine can, and a computer (with current I/O capabilities) will never go for a walk with a friend with intentions to go doing something together, as humans can.

It may be surprising, but most fiat money (call it "human-defined money"), can be made by simply making meaningful friendships, and while Facebook may be trying to fill those orders (with everyone's interaction behavior histories as standing demand orders, and daily feeds trying to fill those orders as supply orders), the most money-making orders is being made within the BS (banking sector, where the fiat money originates) and B2B supply-demand marketplaces.

Imagine if banking sector would have a social network from data like on BoardEx (and from other financio-political databases), and political needs (or "demand orders") of leaders to have certain securities (say "policies") were on the market. I think that would be one of those places with greatest bids, and opportunities in making fiat money. However, indeed, while information about policies is quite open, the downstream orders with specific work demands are less so, because they are often on various B2B marketplaces with less transparency.