The programming languages superorganism (Re: A big boost for Python?)
Boris Borcic
David M. Beazley <d...@asator.Lanl.GOV> a écrit dans l'article <DMB.97Ju...@asator.Lanl.GOV>...
>
> As a followup, I might add that my "report" on Tcl's interest in CORBA
> comes from discussions during the CORBA BOF at the recent Tcl workshop
> (so take that for what it's worth). Interestingly enough, the theme
> of discussion at the BOF could be paraphrased as :
>
> "It seems crystal-clear that the *right* scripting language is Tcl.
> How do we convince OMG of this?"
>
It's been a while I've been wanting to make a case that Python *could* and *should* gobble Tcl.
Since the standard portable GUI for Python is Tk, and that this comes with Tcl, already there
is something to say on, well, the politeness of wanting to hide at all costs the glue language used
in tkinter.
What I would like and suggest is for Python to open a space for Tcl code in its syntax by admitting
double braces {{....}} as delimiters for in-lined Tcl. While Tk is what makes Tcl most interesting, there
is no denying that Tcl also binds to other interesting and useful package that would become more
accessible to Pythoners with that innovation. And there is no denying, either, that Tcl, whatever its
defects, is built around a computation model that has an elegance generally typical of Python.
Besides, I would like to argue that the time has come for (and that's also what CORBA is more or less
about, isn't it ?) alternate "marginal" free language such as Python and so many others to take the
evolutionary step forward towards integration in a "syntactical super-organism". While CORBA and
similar efforts push towards integration through object interfaces (where Python is indeed particularly
strong), I surmise that this may not be sufficient to oppose the imperial march of VB, Java, etc... since
it won't give our "marginal" languages direct visibility. The source is hidden...
So there is an other route to explore : that of syntactical integration; this means, as proposed above,
for Python syntax to spend some of its huge "syntactic whitespace" for giving a clean "syntactical room"
to inline source code in other languages. And the primary candidate is of course Tcl... with Tk
usage a first case for orienting discussion.
Integration of Tcl in Python syntax could perhaps go further than just in-lining. It is the case
(at least to my eyes) that hiding Tcl has motivated important evolutions in Python (e.g. named
arguments etc). If Tcl source-code *strings* make it back to Python source code, these
python features should be possible to "retro-fit" as special properties or operations on the new
string-like object whose constants would allow to hold Tcl code.
To give some acuteness to the superorganism idea, a metaphor : view Tcl/Tk as an organelle
for the Python "cell", like a mitochondria or, more appropriately, a chloroplast.
Regards,
Boris Borcic
--
"I think as I think, therefore I am who I am"