You're missing a few options.
- None of the above
- All of the above
- Some of the above
- Some other possibilities
For example, what if my vision doesn't include "Supermarket shopping", but does include "durable goods" such as:
- Automobiles
- Furniture
- Household Appliances
- High end electronics
- Firearms
- Jewelry
Or what if my vision includes business-to-business transactions, especially things like:
- Farm Equipment
- Construction Equipment
- Wholesale
- Business Services
Note that if the on-chain volume isn't high enough, then the typical person (even someone "saving" or engaging in "international remittance" won't be able to participate. If huge multinational corporations are using bitcoin for international remittance of millions, or even billions of dollars worth of bitcoins per transaction, then there isn't any individual that is going to be able to afford the transaction fees necessary to compete with such transactions.
Thanks for mentioning more points
The current block size debate to its root need to have an estimation about what kind of transaction will be on the blockchain in future. If majority of users are going to use blockchain to do low value high speed transaction, then the block size must be scaled to several GB, which cause significant technical difficulties for the infrastructure. However, if most of the transactions are high value low speed transactions, then it becomes much easier to handle
An interesting comparison is Fedwire system, where Federal Reserve banks sending money between its member banks. Average value per transfer is 6.55 million dollars in 2014, and it made 135 million transactions during 2014, which translates into 4.28 transactions per second, similar to bitcoin right now
And the number of Fedwire transactions per day just increased 2.4x since 1988, so the annual growth is only a couple of percent
http://www.federalreserve.gov/paymentsystems/fedfunds_ann.htmSo if you want to build bitcoin to be a system similar to Fedwire, then current bitcoin is already fine, no big changes are needed for the foreseeable future
However, even Jeff Garzik don't think bitcoin development will take this approach, in his BIP 100 he said:
"In a minority view, bitcoin is a commodity to which an inconsequential payment network is attached. These users view bitcoin as purely as a commodity whose storage cost is far less than the equivalent value of storage. To these users, increasing transaction volume on the network, scaling to support millions of users is simply folly. As the system approaches the speed limit, transaction fees rise. This user base generates little transaction volume, and higher transaction fees are still offset by the value gained from the network’s protection of their commodity. These users likely feel inaction, keeping the speed limit intact, is the most rational step for their situation.
Conversely, a prevailing view is that this speed limit presents a severe constraint on growth and adoption. The speed limit, in the face of user base growth, will lead to higher fees, ultimately reaching an equilibrium when the system capacity of 7 transactions/second is reached. This will cause a ripple effect where blockchain-centered business plans are shelved, or never started in the first place, for lack of perceived scalability. Higher fees then drive a preference for centralized services which can aggregate those fees across multiple bitcoin users, or strike bulk rate deals. The use of bitcoin becomes a rather exclusive club"
So you see, his view is that the low value high frequency trading is important for user base growth. My observation is, for new users it is important to have a frictionless experience, but later on they would still treat bitcoin as commodity instead of daily transaction medium, because most of the people prefer to do those with inflative fiat money. Those people who constantly do small bitcoin transactions are either buying drugs or gambling, mostly illegal in every country's law