The art of money consists in looking not merely at the ability to run a full node on a Raspberry Pi, but looking at the economic and network effects of any act or policy; it consists in tracing the consequences of that policy not merely for one group but for all groups.
Consider the case where the transaction capacity of a money system is artificially constrained on the grounds that it is most essential that the poor be able to run a full node on the network at a "reasonable" cost. The artificially limited capacity leads to higher and unreliable transaction costs as people struggle to outbid their neighbors for space in the network.
A poorer person may now find themself unable to afford to use the money system, or forced to wait lengthy periods until demand for transactions has again fallen below the limited network capacity. An attempt to artificially ease the cost of running a node on the network left them unable to actually use the network. If they run a full node, they suffer the burden of validating the transactions of others while being unable to make a transaction themself.
Additionally, if we ration the transaction capacity on one money system, and the public cannot get enough of it, it will turn to some substitute.
If the transaction capacity of a money system is artificially constrained, the number of users is also artificially constrained. The usefulness of a money system depends on the network effects of widespread acceptance, thus artificially limiting the number of users necessarily limits the utility of the system as a whole.