God that whole process sounds painful.
In the Uk we would just transfer the money across with a bank transfer as all banks support instant transfers which have no fees (including between banks).
the point is not to be a system that beats a single service for sending GBP from A to B in the UK. the point is a system that has a variety of uses and applications accessible by anybody, regardless of which country the user resides in.
even in countries with good banking like UK, many e-commerce companies will still choose Stripe VISA and PayPal to reach customers outside the UK and integrate a payment processor API. an online shop could offer crypto as a payment mechanism if they wanted to avoid using these companies and if it met their needs.
... so suggest that they use crypto, which for many, comes with an onerous fee too, to be timely. Your "buy from the shop" example better not be urgent, otherwise you're paying gas for someone to verify it, or it sits in the queue.
though eventually as these networks scale most users will be transacting and interacting with L2 rather than L1. fees in L2 are in cents on the dollar, transaction finality is often very quick, and security assumptions are typically good enough for most use cases.
https://www.useweb3.xyz/gas?source=ethgas.watch&referrer=eth...
The UK example just shows that a decentralised blockchain approach isn’t required to meet the use case you are describing - in fact it’s worse for the described use case in almost every metric.
Having a different, worse alternative isn’t exactly something to shout about.
whether it is better or worse for each of those depends who you ask and what their goals are, and also in what year you ask. ten years ago it was inconceivable to do these things with crypto, ten years from now it may be that these systems will continue to improve in terms of fees, privacy, scalability and user experience. consider the current cohort of users to be beta testers who are taking on additional risk and technical burden while the system continues to improve.
