NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/lang/c-part2/15455

Internet Message Format | 1996-08-05 | 4.8 KB

Path: keats.ugrad.cs.ubc.ca!not-for-mail
From: [e-mail] (Kazimir Kylheku)
Newsgroups: comp.lang.c
Subject: Re: Dangling pointer?
Date: 18 Apr 1996 14:46:43 -0700
Organization: Computer Science, University of B.C., Vancouver, B.C., Canada
Message-ID: <[e-mail]>
References: <4l0r4b$[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]>
NNTP-Posting-Host: keats.ugrad.cs.ubc.ca

In article <[e-mail]>,
Lawrence Kirby  <[e-mail]> wrote:
>In article <[e-mail]>
>           [e-mail] "Kazimir Kylheku" writes:
>
>>In article <[e-mail]>,
>>frank. <[e-mail]> wrote:
>>>In article <4l0r4b$[e-mail]>, [e-mail] (chen) wrote:
>>>
>>>>     What is "dangling pointer",can someone give me a defination and example?
>>>> Please e-mail me [e-mail]
>>>>
>>>
>>>It's the complement of a split infinitive. :)
>>
>>Actually it isn't. A there is nothing wrong with putting words between the "to"
>>and the verb.
>
>It depends.
>
>>In fact, it's the only place an adjective can go:
>
>???
>
>>        "I want to quickly get this message out"
>
>Quick is an adjective, quickly is an adverb.

Ah yes, of course.

>>is perfectly grammatical.
>
>It depends on the grammar rules you are using.
>
>>It would be ungrammatical to put the "quickly" in
>>front of the "to" or after the "get". You could rewrite the sentence as "I want
>>to get this message out quickly", but that is a transformation.
>
>In what sense? Certainly it's putting the word in a different place but you'd
>have to do that with any of the alternatives. It certainly shows that quickly
>*can* go somewhere else.

In the sense of transformational syntax. It's just a hypothesis, though. I'm
not sure if the movement of ``quickly'' to the end of the sentence is governed
purely by context-free phase-structure rules, or by transformations (I'd have
to check an up-to-date reference like _Introducting Transformational Grammar_
by Jamal Ouhalla).

But saying or writing ``I want quickly to get this message out'', or ``I want
to get quickly this message out'', is not something a native speaker of English
would recognize as grammatical, or utter (except as a mistake).

Yet people would accept ``...to quickly get...'' without a second thought.

>>Linguists take
>>the view that the syntax of the language is defined by how the speaker's use
>>it, not by the utterly empirically inadequate rules of grammar would have them
>>speak or write.
>
>The rules for prose, or for that matter poetry can have significant
>differences to those of spoken language. They tend to be more formal, or
>perhaps spoken langage is just much looser in the application of the rules.
>French has different verb forms for written narrative.

Yes. Spoken language doesn't have consciously accesible rules. Linguists can
only invent theories about what the (unobservable) rules are and test these
against empirical observation to try to uncover what it means when someone
knows a language. In linguistics, even stating that something is an adjective,
or a noun phrase is considered a theoretical claim that must be backed up by
some sort of evidence that these categories are somehow reflected in the
observed language.

On the other hand, prescriptive rules for how one ought to speak or write are
a different matter altogether. There is no justifiable reason to not split an
infinitive (as I just did, what do you know!), other than someone's arbitrary
idea on what constitutes ``good'' writing. It is somewhat like if someone told
you not to write i[a] in C, or to not write int const a because the const
really _ought_ to be written first, and the integral expression really ought to
be the thing in the square brackets, because it is ``proper'' style.

>>Most grammar texts don't even recognize the recursively
>>generative nature of sentence structure. So the next time someone criticizes
>>your infinitive, tell them to split!
>
>It depends on what you're writing. Usenet tends to be more interactive,
>almost like a spoken language so I'd agree with you in that context.

If you want to address certain kinds of audiences, you benefit from following
these rules for various ``levels'' of speaking or writing: colloquial,
informal, formal, and so on. I've heard left-wing linguist types refer to these
as ``socio-economically preferred dialects''. They clearly have their place,
but are useless as a guide to the syntax of a language.

>So you're talking about a pointer that doesn't hold a meaningful value.

Right. I've heard this referred to before as a dangling pointer. Though
sometimes the word is also used to denote syntactic ambiguities (``dangling
else'' in an if/then).