Oracle FAQ Your Portal to the Oracle Knowledge Grid

Home -> Community -> Usenet -> comp.databases.theory -> Re: The word "symbol"

Re: The word "symbol"

From: VC <>
Date: Thu, 11 Aug 2005 21:34:35 -0400
Message-ID: <>


I sincerely appreciate your effort and apologize in case you've not been offended by my style.

Please see in-line:

"David Cressey" <> wrote in message news:THQKe.4379$
>A few days ago, VC commented on my use of the word "symbol" saying that I
> was inventing new terminology. I'm trying to restrain the urge to rant,
> and just give a sober reply.
> There is a book on my shelves, thanks to Joe Celko, who mailed it to me
> for
> a reason I can't remember. The book is "Data Theory" by Peter C. Jones
> and
> Paul E. Jones Jr. [Prentice Hall]. It seems to be some kind of graduate
> level textbook.
> I'm going to quote from Chapter 1, Section 1.1, the Introduction.
> Quote,
> In Section 1.2, Physical Detection of Symbols, we study the symbols that
> can
> be used to express data. Although our primary interest in this book will
> be
> the representation of ideas, we begin with symbols because symbols are
> tangible, and are therefore more accessible to analysis than ideas. We
> will
> later be able to apply conclusions about symbols to ideas by analogy. We
> conclude that the basic mechanism for managing symbols is a "detector",
> which is a process that defines when two physical things are the same
> symbol.
> End quote.

I honestly do not understand :

a. what is meant by "symbols can be used to express data"
b. how "symbols are more accessible to analysis than ideas"
c. what is meant by "two physical thimgs are the same symbol"

Assuming (a) means a symbol can be used to *name* a piece of data, why not just say so ? E.g. in a first-order logic a function symbol gives a name to a function or a relation. If (a) means something else, then what is it ?

Assuming the symbol in (b) is a name for thing/idea/entity, what does it mean to analyze a name ?

To put it bluntly, (c) appears to be simply nonsensical. How a symbol can be two physical things at the same time ? What's that supposed to mean ? The paragraph you've quoted says that a symbol is tangible. In order to be tangible, the sybmbol has to be a physical thing. How one physical thing can be two physical things at the same time ? I do not know, may be I am lacking the context in which the authors use the "symbol".

In short, my objections to the word (as used in our previous exchange) were:

  1. The definition is ambiguos. The word was used to refer to a thing name(1) and at the same time to internal value representation (2). The ambiguity alone is, or should be, lethal for something that's supposed to be used in a narrow and hopefully precise technical sense.
  2. The word itself is redundant because there are perfectly good and established terms for the things we talked about, such as a constant(1) or a name(1) and internal data representation(2).

> I regard the above as sufficient demonstration that the concept behind the
> word "symbol" is foundational for the theory of data and, by extension,
> to
> the theory of databases. I not only expect it to be understood in its
> common usage in the Englisgh language, but also I expect it to be
> understood in a fairly precise way, as one of the terms that helps define
> our discipline.

Please see above.

> I think anyone who is unfamiliar with the term "symbol" as it relates to
> the description of data lacks breath in his or her experience.

So what's your precise and unambiguous definition of the symbol, not necesserily formal, "as it relates to the description of data" ?

>Some of you
> may think I've given in to ranting, but I think I'm being quite
> restrained.
Received on Thu Aug 11 2005 - 20:34:35 CDT

Original text of this message