Oracle FAQ Your Portal to the Oracle Knowledge Grid

Home -> Community -> Usenet -> comp.databases.theory -> Re: Grammatical Inconsistencies

Re: Grammatical Inconsistencies

From: Senny <>
Date: Thu, 22 Apr 2004 04:02:27 GMT
Message-ID: <ntHhc.1310$>

WARNING: Linguistic pedantry follows.

I see your point and to an extent agree with you. However, technically, there is not necessarily a grammatical inconsistency here; specifically, what we're dealing with is an interesting property of English morphology and semantics.

For some reason, perhaps phonological efficiency, English has excelled in its ability to overload the lexical use of a word-form. As a prime example, take the word-form _jack_. This form corresponds to numerous lexemes, some nominal, some verbal, some adjectival. (Some humorous chain e-mails love to point out the same property of the word-form _fuck_.)

Apparently, the word 'join' entered English as a verb from Old French 'joindre' ("to join"). One day, somebody (perhaps a mathemetician) decided to use the form _join_ as a noun to mean "the process of joining sets" or "the set resulting from the joining of sets". This decision, as strange and inconvenient as it may seem, is completely normal in English. The form of the first nominal meaning might as well have been _junction_ _joining_, but these terms, if ever used, never caught on. The form of the second nominal meaning could have been _joint_ or _joined_, but no, we have just _join_.

So, to slightly alter your statement, one could say that there is a morphological inconsistency, since 'intersection' and 'projection' are nouns morphologically derived from verbs, while 'join' is a noun that overloads a base verb form, and 'union' is a verb that overloads a base noun form.

The case of the form _union_ is so messy, I don't even want to touch it.

Why does this matter? It matters because terminology should be as consistent and unambiguous as possible. In our field, we have quite a problem. We don't have much of a real terminology, but instead a fuzzy subset of colloquial English. I am also uncomfortable with this situation, but I am not really sure what can be done about it. See the discusssion of the c.d.theory glossary for more about this problem.


James L. Ryan wrote:

> Does anyone else find themselves somewhat uncomfortable that, in the
> context of relational databases, we say such as "Join" and "Project", but
> we also say such as "Intersection" and "Union"? Why not "Junction" and
> "Projection" or "Intersect" and "Unite"?
> -- James L. Ryan -- TaliesinSoft
Received on Wed Apr 21 2004 - 23:02:27 CDT

Original text of this message