Scripting Languages?

Karl Glazebrook, Anglo-Australian Observatory (KGB@aaoepp2.aao.GOV.AU)
Mon, 09 Oct 1995 06:10:01 +0100

Hi FADS readers,

Here are my comments on the draft proposal: this all sounds like
excellent sense to me, clearly plug-n-play standards for software
and UIs are the way to go.

However there appears to me to be a major area which hasn't been
mentioned: scripting. There have always been two approaches
to writing new applications. The first is to code it all in C or
F77. The second is to fit together exisiting applications in
scripts - the classical example is iraf CL programming. This is
often the easy route - since most of the execution time is
spent in the sub-applications programming in a more cosy high-level
language costs you little and enables faster development.

Having done extensive programming in CL myself, I don't feel it
has much of a future on the timescales we are talking about. I hope
people won't take offence when I say it has problems as a language
which one has to work around. I don't know much about other
packages execept that a lot of them rely on the shell for scripting
applications which is clearly no good if we want to pass complicated
data objects between our applications.

It seems to me that the way to go is to adopt standard scripting languages
such as perl, tcl or guile (the talked-about and much-hyped
GNU scripting language). Clearly a lot of effort has been
put into the design of these languages, one only has to compare the
error diagnostics produced by perl, for example, compared to CL.
(Again no offence intended, CL has served me as a useful tool). I
am sure astronomers do not want to expend effort in language design
when there are so many people in the general community already doing so.

So my solution would be to provide hooks into the applications from
languages such as perl or tcl or their successors. Moreover this would
be facilitated as these languages already have hooks for C/C++ and
have the ability to pass pointers to complex data structures around.

And of course there must be the ability to treat applications-written-
as-scripts as bona-fide applications in there own right.

Karl Glazebrook

---    [Anglo-Australian Observatory]
----> pubs:
----> pgperl: