Or, to put it bluntly, if you are not killing people, you are not pushing hard enough.
I may be a weird person, but since when strapping yourself to a tank, the size of a building, filled with high explosives and riding it to orbit could be expected to be a safe thing?
If astronauts wanted safety, they'd be accountants.
The problem is that Nasa don't seem to have a reason for having astronauts there - other than without the astronauts there would be no interest and no Nasa.
You could make it safer by replacing the Shuttle with an Apollo era capsule and sticking it on the front of every commercial ArianeV / Atlas IV launch. Wouldn't be any point in having the man there but it would be safer manned spaceflight.
There is not much reason to have astronauts up there other than to have astronauts up there. You can do a lot of science with unmanned probes, you can launch satellites with unmanned rockets.
OTOH, you can't repair the Hubble without meatware close to it. And no machine can tell you an inspiring story of adventure and exploration.
And that's what NASA should focus on. Returning scientific data is important, of course, but stories of exploration are what inspire us.
OTOH Hubble cost 3x as much to build because it had to be serviceable by the Shuttle and it's scientific use was greatly compromised by the orbit it had to be in.
Each service mission cost (in real $) as much as building a non-shuttle HST.
Compared to the Herschel space telescope I don't see Hubble as a great advert for manned space flight.
You don't need a shuttle to repair a Hubble. Hubble's orbit was limited to what a shuttle could reach and the cost of repairing it compounded by the operational costs of the shuttle.
You could send a repair crew on an expendable vehicle or, even better, launch a tow that could bring the satellite being serviced to the ISS and reinsert it into any desired orbit. The Hubble service missions greatly extended its useful life. I doubt Herschel will last as long.
Hubble was in a particularly bad orbit specifically because of the shuttle but any manned servicing would still require it to be in LEO, there is no way you are servicing something at the Lagrange point.
A lot of Hubble's cost and problems were also the result of Nasa in the 80s - things have got a lot better.
The point is more that a series of disposable satellites, especially sharing common parts, is much cheaper than a single upgradable one. Otherwise we would make other satellites (GPS, comms etc) shuttle rated.
> there is no way you are servicing something at the Lagrange point.
It depends on what Lagrange point we are talking about. It can be easier than a trip to the Moon or more or less like a trip to Mars. Anyway, you could tow it to a service orbit humans can reach and then boost it back to its original (or any other convenient) orbit. Herschel is pretty far to receive a human crew, but it can be towed to a more manageable orbit.
I like the idea of cheap satellites with high commonality, but I also like being able to upgrade and repurpose them. After all, the worst part of the work - launching them - was already done, and the cheaper each satellite gets, the more the launch weights in its overall costs.
Not allowed to fly cryogens on the shuttle as a safety feature.
You also aren't allowed to carry propellant or use propellant on serviceable satellites, that was one of the design problems with the HST + shuttle.
It's not a shuttle specific thing it's the problem of anything man rated - you can't put anything dangerous onboard.
I may be a weird person, but since when strapping yourself to a tank, the size of a building, filled with high explosives and riding it to orbit could be expected to be a safe thing?
If astronauts wanted safety, they'd be accountants.