> how do we know till start developing it?

Because even the most optimistic assumptions about it, are already beaten by Falcon 9.

So if you can't even make it work on paper, how are you gone make it in real live?

> are we condemned to live with refining 1950 designs forever?

I mean you are stretching the definition of 'refining'. Starship is a just as much an upgrade over 1950 rockets then Skylon.

In fact, Skylon is actually how many people in the 1950 imagine the future, while Starship is the reality checked version of what will actually work.

> is that the best we can do?

If you want to really go invest in the future design, how about an air breathing nuclear turbofan. But I think that might cause some opposition.

You also need to consider that money is limited. Every $ going into air breathing engine isn't going into something that has potentially much greater effect. The investment done into Sabre engine would have yield far better results if it has been invested in reusable FFSC. Or something like a closed cycle expander areospike. Or many other things that that have far greater potential then Sabre.

The reality is that Sabre and Skylon are a product of a British team, a team of people that basically spend 20 years designing a bunch of paper rocket, having little experience what so with actual rocket flight operation and they came out with a concept they was so over-the-top that they thought they could get some serious research money finally.