Hallo!

On Mon, 2 Aug 2010 13:18:06 +0200, BPJ wrote:

>  2010-08-01 22:36, Jörg Rhiemeier skrev:
>  >  Indeed not.  The use of the word "language" in expressions such
>  >  as _programming language_ is only a metaphor,
>
>  Not entirely: a human writing or reading a program
>  does process the the _programming language_ and
>  most likely, at least for higher level programming
>  languages, does this using the same faculties as
>  they use for processing human languages, as far
>  as I can tell, and I never saw any claims to the
>  contrary.

It would certainly be too strong a statement to claim that programming
languages have *nothing* to do with the kind of languages human beings
speak to each other.  It depends on how strong a definition you use for
the word "language".

Yet, there is a difference in purpose.  Languages in the traditional
sense are used to communicate ideas between human beings.  Programming
languages are used to direct a computer - a deterministic machine which,
for all practical purposes, lacks free will and is as stupid as your
doorstep.  Programming a computer may look and feel like talking to a
person (or, perhaps even more, a dog) in order to have him do something
you wish to get done; but it is actually more like adjusting the gears
of a mechanical contraption.

>        Moreover a programming language, is also
>  a human language, since it is designed by humans
>  as a tool for humans in instructing a machine.
>  A language, thus: a rather unnaturalistic conlang,
>  so the divide isn't _programming language_ vs.
>  _human language_ or even _programming language_
>  vs. _natlang_, but _non-naturalistic language_ vs.
>  _naturalistic language_.  Of course the average
>  computer scientist will not realize these
>  distinctions, because they aren't a conlanger.

Certainly, if a programming language is a "language", it is a "human
language" as it is used and has been designed by human beings.
Of course, no programming language I know of is a naturalistic conlang,
as those would not offer any advantages for that purpose over natlangs.

The reason why programming languages have been invented rather than
telling computers what to do in plain English, is that computers have
very different language-processing capabilities than human beings.

Programming languages are a sort of "middle ground" between the
capabilities of human beings (communicating with spoken and written
languages) and those of computers (twiddling bits).  Someone once said
humorously: "Humans do not understand machine code.  Computers don't
understand human languages.  So we use programming languages - which
neither side understands."

However, the term "human language" is commonly used in the meaning of
"language which humans use to communicate with each other", and is
often used in contrast against "programming language".

OnMon, 2 Aug 2010 12:34:01 -0400, Micah Johnston wrote:

>  Programming languages are languages in the same sense as natural languages
>  are, albeit with a different purpose, but at least in my experience, the
>  brain is not used in the same way as it is for natural languages. I doubt
>  you could find a programmer who could fluently speak an entirely new program
>  in a programming language. To write in a programming language, you have to
>  think about the entire structure to make sure it fits together and has no
>  errors, while natural languages are optimized for being spoken linearly,
>  only thinking about little bits at a time, and little errors are acceptable.

Indeed, there is a difference between speaking a language such as
English or Japanese, and programming a computer.  The main difference
is that when doing the latter, you "speak" to a machine which may be
blindingly fast and inhumanly precise in processing the data fed into
it, but actually understands nothing, and will gladly follow any
instructions you give it as long as those instructions comply with the
grammar of the programming language, even if those instructions utterly
fail to make any sense.

As you say, programming requires much more planning overhead than talking.
Sure, if you are writing a longer text such as a book, you need some degree
of planning; but short messages can be delivered fluently at the speed of
thought.  This is hardly the case even with small programs.

OnMon, 2 Aug 2010 20:41:04 -0400, Jim Henry wrote:

>  On Mon, Aug 2, 2010 at 7:18 AM, BPJ<[log in to unmask]>  wrote:
>
>  >  A language, thus: a rather unnaturalistic conlang,
>  >  so the divide isn't _programming language_ vs.
>  >  _human language_ or even _programming language_
>  >  vs. _natlang_, but _non-naturalistic language_ vs.
>  >  _naturalistic language_.  Of course the average
>  >  computer scientist will not realize these
>  >  distinctions, because they aren't a conlanger.
>
>  And maybe "general-purpose language" vs. "special-purpose language",
>  the former including all natlangs and many conlangs (perhaps most of
>  the well-developed ones?), the latter including all programming
>  languages, many engelangs, and a subset of artlangs (perhaps including
>  the sketchy ones used as supporting material for a book or film but
>  not fully developed as works of art in their own right -- e.g., Vulcan
>  in contrast to Klingon, or GRRM's Dothraki vs. David Peterson's
>  Dothraki).

Yes.  Many engelangs, at least, serve more restricted purposes than
natlangs.  Rick Morneau's Latejami, for instance, is meant to be used
as an intermediate language in a computer-based automatical translation
system; while he has attached a phonology and orthography such that
it *could* be used by human beings in spoken or written form, that
is not really its purpose.  Many engelangs are meant as experiments
and not intended to ever leave the laboratory for which they are
constructed.

--
... brought to you by the Weeping Elf
http://www.joerg-rhiemeier.de/Conlang/index.html