Re: The joy of FORTRAN

On Wed, 02 Oct 2024 17:29:31 GMT, Charlie Gibbs wrote:

On 2024-10-02, The Natural Philosopher <tnp@invalid.invalid> wrote:
 

In practice I work both ends to the middle. Write what is obvious in
spec or code first, and then see what problems remain unsolved.....

 
Stan Kelly-Bootle, when contrasting top-down and bottom-up development,
proposed the opposite of your method.  He called it "middle-out".

“Rapid prototyping” I think is the term for what I do: talk to the
customer, get some initial notes, go away for a couple of weeks, come back
with some initially working code that I/they can actually run, and then
they say “yes, that’s the kind of thing I want, can you add this” or “no,
that’s not quite right, can you change this”. And then I go away for
another week or two, and repeat the cycle until we get a useful
production-quality result.

This way, we always have something concrete in front of us as a reference
point, instead of having to try to conduct discussions in the abstract.
And the customer feels they are actually getting something for their
money, each time they pay my bill.