Re: cdt glossary

From: mAsterdam <mAsterdam_at_vrijdag.org>
Date: Mon, 18 Jul 2005 11:36:05 +0200
Message-ID: <42db7809$0$61586$e4fe514c@news.xs4all.nl>

dawn wrote:
> David Cressey wrote:

>>mAsterdam wrote:
>>
>>>The current (0.0.4) glossary entry,
>>>
>>>
>>>>[Table/Row/Column] (SQL-DBMS)
>>>>Table: A collection of columns (the table header) and rows (the body).
>>>>Row: A collection of values, conforming to the table header columns.
>>>>
>>>>One table may contain data about one entity,
>>>>about several entities, about one or several
>>>>relationships or any combination.
>>>>A column can be seen as the attribute of the
>>>>entity/one of the entities/relationships
>>>>about which the table is concerned.
>>>
>>>, says nothing about the rows being ordered or not.
>>>Should it?
>>
>>The above is correct, in my opinion.

>
> I guess it depends on what profession we are in. If we are in a
> professional narrowly defined as relational database theory, then I'm
> OK with definitions of this nature. If we want to play in the
> profession of software development, then it would be more helpful if we
> qualify our terms when using them more specifically like this.

This is about the interests of this group: I think 'software development' (albeit with a strong data-flavour) is very much on topic.

> I would
> say that this definition is about tables defined to rdbms tools or
> tables related to relational databases. Instead, this would be much
> better definition for "SQL Table" or "Relational Table", I would think.

What do you propose? The current heading, >>>>[Table/Row/Column] (SQL-DBMS)
does a good job of narrowing the frame of reference down to SQL/DBMS, no? (A generic/broad 'TABLE' entry is welcome and would stress this narrowing down).

"SQL Table" would suggest an ANSI definition (so 'header' and 'body' are out), "Relational Table" - I don't think many people are of the opinion that such a beats exists - here is a nice challenge: describe "Relational Table" in such a way that most people here would agree.

> The software development industry certainly uses the term table to
> refer to a number of representations of data that do not align with the
> above definition. There are some characteristics of all (at least
> most) uses of the term, however. I would think we would want the term
> "table" to be defined to include implementations of tables in a variety
> of languages and tools to avoid miscommunication.
>
> Off the top of my head (rather than researching it right now), most
> uses of the word table would have the reader visualizing a
> two-dimensional matrix of values. Some would include a header for such
> values while others would have a separate object that is a table
> header.

Again, a broader 'TABLE' entry is welcome.

>>In particular, the word "collection" is, in this context, more useful than
>>either "list" or "array".
>>
>>The question of "ordered" goes beyond the definition,  IMO,  even though
>>it's an interesting one.

>
> Agreed. Cheers! --dawn

That's two against being clearer on the specifics of the involved collections in this glossary entry.

I'll just wait a few more weeks - maybe there will be other suggestions/opinions on this. Received on Mon Jul 18 2005 - 04:36:05 CDT

Re: cdt glossary - TABLE