On 4/17/13 10:52 AM, glen wrote:
It's not entirely clear to me where "type" fits (at least not the
specific sense of "type" we use in programming).
Starlogo TNG illustrates types.
http://education.mit.edu/starlogo-tng/tutorial-pics/slblocks.jpeg
That `say' and `play sound' have the same connector shapes, suggesting
they are the same. Without additional information, one might guess they
are just character string names? But a language utterance and a digital
sound sample have different meanings even if their implementation is the
same. A strong typing approach would discriminate the meanings and
they'd have distinct connector shapes. Here there is at least some
discrimination -- one cannot connect a `set score' to a `boing'.
Dynamic typing says that the shapes don't matter. That does make it
easy to hook together things, but not necessarily easier to really make
progress. It just makes it seem that way because there is no scolding
about carelessness. Unix pipelines are like this -- sure it's easy to
connect program streams together, but just because one has a set of
programs that read and write (e.g. tabular data) doesn't mean that a
downstream program knows what is coming at it.
Marcus
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com