It's one of many reasons to keep blocks small. Ability to send over many kinds of restricted bandwidth mediums, including the Tor network is part of it. Other concerns are bandwidth caps/monthly limits, storage costs, validation times (which affect negatively affect miners).
Some other hardforks of Bitcoin which promoted themselves for having bigger blocks, have had examples where block validation times have exceeded the average block mining time of 10 minutes. Obviously not suitable for running on a Raspberry Pi.
Bitcoin depends on its network being widespread and diverse. Any increase in the requirements to run node will create centralization pressure which does not help. The goals of big blockers are to have just one or a few big data centres hosting bitcoin and everyone else connecting to them. We already have PayPal for that.
> Some other hardforks of Bitcoin which promoted themselves for having bigger blocks, have had examples where block validation times have exceeded the average block mining time of 10 minutes.
Such a ridiculous counter argument.
The only fork that had a long propagation time was when BSV produced > 100 MB blocks. Note that I say propagation time and not validation time caused by them not propagating transactions before including them in the block. Which everyone told them was a bad idea as the software wasn't optimized for it yet.
But using it as a counter argument against moderately larger blocks than 1 MB, like 2 or 8 MB, just stupid. Instead we should just accept insanely large fees and wait for LN which may never be a suitable replacement.
> The goals of big blockers are to have just one or a few big data centres hosting bitcoin and everyone else connecting to them.
No. The goal of big blockers is to scale on-chain and to allow people to run full nodes if they want to. Not to have it work on the crappiest hardware you can and kill adoption. It's a realization that most people won't run full nodes as they want to use their phones as wallets.
SPV wallets provide enough security for mobile phones with only having to trust that the POW system works. In reality only exchanges and other businesses have a need to run full nodes.
I could for example easily run a full node with full 32 MB blocks today with my current hardware. I would just have to upgrade my storage occasionally for a minor cost. It would amount to 1.7 TB per year which costs less than $60.
> We already have PayPal for that.
Even if only two data centers hosts full nodes it would still be better than PayPal. Better update your anti big block propaganda to use actual facts.
