Re: Terminology question

<pamelafluente_at_libero.it> wrote in message news:1157465192.318720.308470_at_b28g2000cwb.googlegroups.com...
> >
> > In some cases, the romance languages are quite precise in describing
things.
>
> Really? I got exactly the same feeling about English.
>
As an English speaker (well, actually, Amurrican) I suggest that English masters the appearance of precision better than the reality of precision.

>
> Actually for computer things, we borrow all terminology from USA.
> So original Italian terminology does not really exists and they
> are all translation from English. I use English for menu and labels.
>

It's not clear to me that database terminology ought to be derived from "computer things", even though today's databases depend on today's computers. The state of terminology in the world of computer technology is far from clear. In part that's due to rapid evolution. In part it's due to people's willingness to invent new terminology instead of finding existing terminology already in use.

Let's take the simple word, "object". In common parlance, this word is so vague that it can be used to refer to almost anything. In computer parlance, the term is taken to be a precise technical term, but as a technical term it suffers from too many mutually contradictory definitions.

Don't be misled by Bob Badour's responses. I would suggest that "connection to a database" is far from "new terminology", and that it has appeared thousands of times in database oriented literature, and that Bob is well aware of that fact, and is quite familiar with the term, even if he avoids using it. We have an old saying in English that "familiarity breeds contempt". That may be what's at work here. Certainly, contempt is at the root of a large fraction of Bob's contributions to this newsgroup.

Also contrary to the impression Bob has given, discussions about the meaning of commonly used terms are quite frequent in the comp.databases.theory newsgroup. So much so, that there is a glossary of terms that participants have chosen to define for our own purposes here. mAsterdam has been so kind as to be the keeper of the newsgroup glossary, and might be willing to provide a copy of it.

Having said that, it remains true that the advice you get from this newsgroup could easily be irrelevant or even counterproductive if your goal is to communicate with a wide and diverse audience of other application programmers that will be making use of your objects. The precision of communication in this newsgroup is often at odds with the mainstream of the computer industry.

> I guess I could simply say:
>
> Connection or [Connection Info] to denote the object the user re-uses
> to open a connection to a database
>
> "Connection & Query" to denote the object which stores the information
> to retrieve a bunch of data
>
> Here "Query" is not very appropriate as it could make ususally think
> about some pure SQL, while instead the object has also the possibility
> to embed custom programs (written in any language, not only PL/SQL or
> TS/SQL) to perform transformations on the fly (that is while extracting
> data).
>
> So a generical record (x1,...,xn) could be transformed into another
> record (y1,...,ym) of "derived variables". The transforming "function"
> (if we want to it so) is provided by the user through code and embedded
> into the object to which I am trying to give an "appropriate" name.
>
> Perhaps I could say "Connection and ETL" but this sound pretentious and
> incorrect.
>
> My "intuitive" idea was to imagine that object as a kind "logical"
> source of data, where here the word "logical" has nothing to do with
> the math meaning, but uses to contrast with "real" data (we use it in
> that fashon: don't know if it's the same in English). Perhaps s synonym
> could be "virtual". But "virtual" is also an abused word.
>

> Hmmm it quite hard for me to get some English expression that would fit
> nicely these simple concept.
>
> Probably I will just end up with "Connection & Query"
>
> -P
>

Actually, I liked "data source" best of all the terms you proposed. The only way in which that term might be misleading is if your "object" enables other objects to store data as well as retrieve data. Received on Tue Sep 05 2006 - 21:34:13 CEST