Consider
The statement:

1. All programming languages have things in common with the English language. if P is a list of ALL Programing Languages and E is the English Language then each Programming Language in the list P will have at least one element in common with the English Language
                
Vs

2. All languages have elements in common such as the component symbols, words or terms that have defined meaning and the rules of syntax or grammar that tell you how to combine the words into sentences to communicate.

if L is a list of ALL Languages then some elements that are common to each can be identified as

component symbols, words or terms that have defined meaning
rules of syntax or grammar that tell you how to combine the words into a sentence to communicate

        Compare the information presented


Donna
[EMAIL PROTECTED]



On 30-Jun-06, at 8:06 AM, Mark D. Niemiec wrote:

dly <[EMAIL PROTECTED]> wrote:

There are no risks in meta meta land except not being perfectly clear.

I have no problems with meta^n language, but it tends to make
many people's eyeballs spin in their sockets (and perhaps cause
them to spit out large quantities of pea soup).
It is unclear whether this is the result of metalanguage--several other possibility come to mind

In a metalanguage for J, there should be a controlled vocabulary with
sufficient words to express any technical sentence about J.
Words should be clear, simple and easily recognized.  You should
choose one word for one meaning. In this way the metalanguage can be
learned and used effectively by the speakers of any language, tho
they may initially need a translation.

Of course.
Agreement


In my example, the word 'English' has one meaning to English-speakers
in the context: 'The language in which we are speaking' (as opposed to
'the language spoken by machines') - as such, the term is clearly relative,
as there is no other reference to Anglo-Saxon language or culture
anywhere else in the J specification to indicate that it is a static reference. (Or, or that matter, referring to linquistic features that are unique to
English other than any other language).
In argument this is called a straw man argument. You have omitted my reasons for suggesting a substitution of a more general statement in the first paragraph. I already agreed that you could use English in this sense tho I would suggest you speak about plain English (see the link in my first post) for an example of plain English vs the English language.)

Later in the suggested revised post I do include the reason for
choosing the terms that have analogous meaning in English using
mostly the same wording as the original text.


J choses these words precisely because the programming model
has a strong connection with the English language and others:

Alphabet - The character set from which words are formed
Word - The fundamental unit from which the language is built up
Noun - A thing, an object being manipulated
Verb - An action performed by or on a thing
Adverb - Something modifying or qualifying an action or thing
Conjunction - Something combining two actions or things into a new one
Copula - Like a verb, identifies one thing with another 
Gerund - A noun that encapsulates the meaning of a verb
Pronoun - A word whose meaning varies from time to time,
  but serves as a placeholder for a previously-defined noun
Sentence - A meaningful, self-contained sequence of words

In all cases, the J meaning is substantially the same as the English one.
Agreement

(One could concievably use the term Adjective when referring to
Adverbs that apply only to nouns, or the term Infinitive rather than Gerund).
Contradiction with the statement that one term and one meaning should be chosen unless you are referring to the option of choosing a different meta language than the J terms.



To summarize I suggest:

1.  stating what it is that all languages have in common rather than
what J has in common with English.
2.  make a distinction between the object language J and the meta
language (possibly referred to as J terms since that was already in
the text).

Terms such as 'function', 'subroutine', 'noun', 'verb' are always metalanguage,
since the languages themselves rarely actually use most such terms
(except perhaps in declarations, such as 'var' in Pascal or 'function' in JavaScript)

You have successfully argued to refute your own point
You do not seem to refer at all to my points
If these things are set out carefully it can improve discussions in
all the forums because people will use the same simple meta language
to describe J.  It can improve programming specs for upgrading and
maintaining J.  It can improve the learning curve.  Otherwise we can
have interesting conversations trying to figure out what one another
means.

The whole point of using a different metalanguage to describe J is
because
does the following actually describe the whole point? Is it clear? Is it a refutation? Is it correct? Does it bring up the pea soup thing?
1) the syntax model is more sophisticated than that of most
programming languages, so the normal metalanguage is inadequate -
it becomes unwieldy when one must invent new terms 'function' vs
'meta-function' for example), 2) an English-like metalanguage
(as I mentioned above) is much more suited to describing these
features of J.

Consider the J block:
  if. x do.
    blah =: 42
  else.
    blah =: +
  end.

If I ask you what 'blah' is in J, in terms of English-based J metalanguage,
I could describe is as a pronoun or proverb; however, it would be
difficult to explain in terms of traditional programming metalanguage,
since most languages do not permit dynamically-assigned functions,
let alone symbols that can be functions and/or variables depending
on the flow of execution.
proverb |ˈprävˌərb| |ˌprɑˈvərb| |ˌprɒvəːb|
noun
a short pithy saying in general use, stating a general truth or piece of advice. See note at saying . ORIGIN Middle English : from Old French proverbe, from Latin proverbium, from pro- ‘(put) forth’ + verbum ‘word.’
I think blah is not considered to be a proverb let alone BLAH BLAH BLAH


-- Mark D. Niemiec <[EMAIL PROTECTED]>

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to