False equivalence. Cars are much older technology than blockchain. In 1908, cars were much slower than they are today. In 1808, horse carriages were much slower. You get the point. The speed at which I expect changes to process differs over time. In some future time, we may all be in self-driving cars that are able to safely exceed 200MPH and we'll think back to our 45MPH street speed limits and laugh. Just like how in 2008, the typical home network had a 5Mbps uplink to the internet, and today it's (depending on the source) at least 135Mbps, with the availability of >1Gbps depending on your ISP and region.
I was using the term "speed" as a concept, and you were using it as a constant. I referred to "speed" as "fast" or "slow." Which is to say, the subjective judgement on what is fast. Which is a decent way to default to judging speed unless the white paper specifies a threshold of transactions per second to be considered "fast." You thought of speed as a constant unit of measurement for speed, which makes no sense unless they have a target for their transactions per second. The problem with that is, the number of transactions per second that a payment solution needs to process is pretty much only going to increase over time, and thus, the goal to be considered "fast" does actually change over time. You're just potentially thinking in too short of a time range to see much drift in that subjective judgement.
I think you're mistaking my intention too. My intention is to say that Bitcoin's perceived speed with handling transactions then, is perhaps slow by today's standards. The goal you gave:
> digital cash with and low fee transactions without the used of a trusted third party.
is at least 2/3rd of the way complete. You have digital cash, with high transaction fees, and no trusted third party. The solution that blockchain designers/engineers seem to have come up with is referred to as "layer 2 rollups" where a bunch of transactions are processed quickly by a trusted third party, bundled together, and then enforced in one big transaction on the actual network in intervals. This promises to be faster, and possibly cost less in transaction fees, but then under-delivers on the third goal of not using a trusted third party. But it is apparently the best way that blockchain engineers have thought up to compete with the transaction processing speed of entities like VISA, at least today. While it's not ideal to trust that third party, presumably you have a choice to opt in or out of the layer 2 network, and enforce your transaction on the slower, more traditional blockchain layer 1. And at least with the layer 2 network, it eventually gets trued up on the blockchain with each of the rollup intervals. It's maybe up for an argument on whether trusting the layer 2 third party is better than, or the same as, trusting VISA. I would potentially argue that it's better, with the caveat of admitting that I am not a blockchain expert. I just potentially know more than the average crypto enthusiast that trades BTC and Doge on institutions like Coinbase.
I will note that the goal you gave said absolutely nothing about speed though.