Shuttle costs

Hi Robert,

Perhaps I'm wrong, but as a computer programmer for some 36 years or
more I get the impression you don't really understand what the problem
is. It isn't the programming as such, so much as who decides which or
whose programs to use, and the mistakes they make. In other words it's
program management decisions that are at fault rather than incompetence
at the actual programming.

Ed Minchau is not really right re software. The largest cost is that of
keeping the launch station going so long as you have very few launches.
Even apart from that, so far, fuel is the largest cost, next does come
not so much software as software faults. When he put his point, I
decided to check with a friend who knows a little more about the costs
side than I do, Duncan Lunan, space author, former president of ASTRA,
and present Treasurer of our Uk-based Space Settlers' Society. Here is
Duncan's reply:

Dear Andy,

Contradiction: he rightly says the cost is people, then says half of it
is software. As critics like Pournelle point out, the Thor missile was
developed in (as I recall) 18 months by a team of 10 people and took
just 8 people to launch in an hour or two. Its Delta derivative takes
10,000 people to launch, you'll be lucky to get a launch in less than
three years and it costs far, far more - which I've always understood to
be wages, not software. If the original software is any good you don't
need new software every time: where software gets very expensive is
when you don't realise things have changed. The first Ariane V was
lost with the four Cluster satellites because they assumed Ariane IV
software would suffice for some aspects of the flight programme and it
didn't.

Best,
Duncan.

You get my point? Not the software, but the failure of choice of it is
what was at fault.

Best wishes, Andy.

Robert Reed wrote:

> NASA should then use open source software. And I am willing to
> contribute code to them.
>
> --- In spacesettlers@y..., Ian Woollard wrote:
> > Ed Minchau wrote:
> >
> > > The largest single cost of any of NASA's launches has nothing to
> do
> > > with fuel or materials (although I agree with you Al, better
> > > materials will drive launch costs down); fully half of the cost
> of a
> > > launch is...software.
> >
> > I find this very difficult to believe. Are you seriously telling me
> > that NASA has spent tens of billions on Space Shuttle software?
> >
> > (ISS I can buy; that's a different matter, there's a lot
> > of strange inter-related software/hardware on that one.)
> >
> > Do you have a cite or URL to back this one up by any chance?
> >
> > > Better, more reliable, _re-useable_ software is the key to
> reducing
> > > space launch costs to acceptable levels.
> >
> > Open source (GPL) rocketry software anyone?
> >
> > Or just launch a LOT. I mean software doesn't really cost anything
> per
> > copy does it?
> >
> > > :) ed
> >
> > --
> > - Ian Woollard (ian.woollard@t...)
> >
> > "Is a planetary surface the right place for an expanding
> technological
> > civilization?"
> > - Gerard O'Neill
>
[Click for Details]
Click for Details