Re: Questions about Common Lisp

On Fri, 20 Mar 2009 20:33:19 -0700, stephen@xxxxxxxxxxxxxxxxx (Stephen
J. Bevan) wrote:

George Neuner <gneuner2@xxxxxxxxxxx> writes:
The term "dynamically typed" is meaningless. but it is frequently used
as a lay synonym for "latently typed".

Indeed but then is the "typed" in "latently typed" really any more
meaningful than when it appears in "dynamically typed"?

Yes. It means the types are not manifest in the source but are known
implicitly from and carried by the data ("carried by" meaning data is
tagged - languages with manifest types may also tag data). Both
manifest and latent typing can be analyzed statically.

"dynamic typing" is meaningless because it implies that types are
fluid and may change. While this can actually happen in Lisp because
object classes can be redefined at runtime, that facility is not used
by the vast majority of programs.

Either "type" means something related to type theory or it doesn't.

"type" in all of these terms - even the lay terms - always means
something wrt type theory. Languages are classified by whether their
type systems are strong or weak, manifest or latent.

If it does then how, when to all apparances such languages are untyped?

As I learned it, an "untyped" language is one in which data exists
only as raw words - ie. bits - and operators determine how their
operands are used.

If it does not, then using the term surely muddied the waters given
the existing meaning of the word which applies perfectly well to
programming langauges and which dates back to before there were any
stored program computers.

I agree that existing terminology should be used whenever possible ...
but that just isn't the way things happen.

George
.

Relevant Pages

  • Re: SEM: Standardized path weight greater than 1?
    ... If we are talking about loadings (arrows go from a latent variable to ... a manifest one), a loading above 1 is surely erratic - it's like if ... I hope you are talking about a regression coefficient. ...
    (sci.stat.edu)
  • Hane hair - French passport
    ... Following up on the suggestion that French passports might shed some ... Hair is described as Brown on Manifest ... These same two children have Brown eyes on the Manifest. ... noir" meaning "dark eyes". ...
    (soc.genealogy.jewish)
  • Re: SEM: Standardized path weight greater than 1?
    ... If we are talking about loadings (arrows go from a latent variable to ... a manifest one), a loading above 1 is surely erratic - it's like if ... you had a correlation coefficient above 1. ...
    (sci.stat.edu)