this was the debate circa 2014/2015. Should the bitcoin network consist of many low fee transactions or few high fee transactions. Right now the incentive for miners to secure the network is produced with the block reward, but when the block reward runs out this incentive will be from fees alone. In order to provide miners with the same revenue as today when the block reward runs out, the average fee per tx will need to be close to $200. If you increase the block size to 100mb and use block compression tech like xthinner to make blocks essentially 1mb (like today), then the average fee needed is only $2.

>use block compression tech like xthinner to make blocks essentially 1mb (like today), then the average fee needed is only $2.

Claiming that xthinner can reduce 100mb blocks down to 1mb is dishonest. It might reduce the network transfer down to 1mb (assuming you already the txes), but on-disk storage would stay the same.

https://news.ycombinator.com/item?id=25688004

A single transaction costs more than the space to store the entire chain many times over.

block propagation was the main reason for being opposed to a block size increase, storage costs were debunked time and time again as being irrelevant.

Why many low fee transactions were not chosen?

You'll get many different responses to this question depending on who you ask. If you want to read into further i recommend you read the great scaling debate[1]. Its pretty long but does a fantastic job of summarizing the history.

In short the most popular reason for not having a block size increase (to allow many low fee transactions) was that it would increase the cost of running a full node, in turn centralising the network. More transactions would mean nodes would need a bigger hard drive and a better cpu to process the transactions. The small block camp wanted people to be able to run nodes on a raspberry pi. This theory is misguided in my opinion though, they chose to sacrifice cheap transactions to keep node running costs cheap.

[1]https://medium.com/hackernoon/the-great-bitcoin-scaling-deba...

They (the devs that control the repo that almost all miners run code from) chose to stifle the base layer to profit from higher layers using 'goat herders should be able to run a full node' as the excuse.

>to profit from higher layers

How are they profiting?

the liquid network is owned by blockstream and they get the fees for transactions on it.

Because Bitcoin works in a way that means miners, collectively, control the network.

They voted with their hash power against Bitcoin Cash when it forked, which was the “increase the limit” fork aiming at low fees and high number of transactions.

Those miners can now enjoy high fees as a result.

In my experience, I have seen literally 0 devs that are trying to increase miner profits. No one is worried that miners aren't getting paid enough.

The miners control the network, longest chain and all that.

They also have lots of money to pay for devs.

Foolish to think a bunch of open source devs are controlling the protocol in defiance of miners who would rather they do something else.

This is the real reason for the block size cap.

That economic analysis is clearly facile. If it worked, miners would want there to be 1 transaction per block.

Miners want to maximize transaction fees. So more is better. But there is a massive tradeoff, being every transaction must be stored to verify later.