Talk:Database schema

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Hierarchical DB Schema?[edit]

"Hierarchical" is nowhere mentioned. Comparing and contrasting "relational" with "hierarchical" would help to reveal what "schema" means to readers who do not know. One schema won in the ongoing evolution of technology, and the other lost. Why? What was it about their schemas?
Jerry-VA (talk) 16:51, 1 January 2014 (UTC)[reply]

Tables and fields[edit]

It is my understanding that the most fundamental purpose of a schema is to define tables and the fields of tables. The (current) introduction emphasizes many other things, such as relationships, but says very little about defining the fields of tables. Certainly "fields of tables" might be specific to relational databases but it is not far from what a XML schema does, correct? Sam Tomato (talk) 23:34, 3 February 2015 (UTC)[reply]

Confusing[edit]

Is this article about schemas, or "schema integrations"? 130.226.142.243 (talk) 09:20, 30 September 2015 (UTC)[reply]
This entire thing is proof that a person can take a simple topic, and find ridiculously complex ways of explaining it (Maths? Really???). "Schemas are, in effect, sub-containers in a database, that are used to hold separate object definitions and relations. The default schema is dbo." Is there a need to make it more complex? - Ziv (talk) 14:27, 08 July 2017 (UTC)[reply]

I'm sure it IS more complex than that, but not as complex as the article wording. a schema is a *logical* construct like 'entity', the problem comes in that some dbs like MS have chosen to implement the logical construct with a physical expression that doesn't exactly do what a logical schema does, whereas other dbs have traditionally implemented physical USER and DATABASE that together cover roughly what a schema does, and are now trying to implement schema in the way ANSI suggests ( I THINK: I wouldn't have come here if I knew the answers, I'm a humble practitioner).

In order to contribute to understanding I think the article needs to lead from theory and concept to real implementations, which by the way the article https://www.w3resource.com/sql/sql-basic/create-schema.php does a little better, although both are lightweight on real supporting theory and implementation detail. Interestingly, the problematic preamble words in both articles are identical. Hmmm.

The problem with the intro words in both articles is that they are quasi-mathematical nonsense. 'A database schema of a database system is its structure described in a formal language supported by the database management system.' No, that's not what a database schema is, that would be how it would be described in the DBMS, if a DBMS ever supported a 'formal language.' What it IS would be its conceptual and mathematical sense, of which we are left begging.

The second paragraph of the intro is unique in many ways. Anything can be "seen at any instant of time as a mathematical object" but saying so doesn't itself add much. "The notion of a database schema plays the same role as the notion of theory in predicate calculus" a potentially interesting statement but it's not going anywhere, and I fancy it's nonsense. "In a relational database, the schema defines the <objects>" well no it doesn't define them, does it? it is just a conceptual object container for them surely? And all this talk of 'integrity constraints' as if we are clear on whether these are the implemented database constraints or some conceptual construct. It seems to me that at least one big point about a schema is that it is a conceptual object that is the parent of other objects who inherit properties from it, including properties of authority.

And so on and so on.

But what do I know? This article needs serious attention from an expert who understands the conceptual and mathematical background to the various ANSI drafts, is also familiar with the implications for real implementations, and can turn a phrase that can help both an academic and industrial audience.

By the way, although the reviewer says there are already too many see alsos, these are also in the same area: https://en.wikipedia.org/wiki/SQL:2011 https://en.wikipedia.org/wiki/ISO/IEC_9075 https://en.wikipedia.org/wiki/ANSI-SPARC_Architecture https://en.wikipedia.org/wiki/View_model

I guess one challenge here is either to turn all these see alsos into references that support the body of article or else not mention them at all. Atconsul (talk) —Preceding undated comment added 08:36, 14 September 2017 (UTC)[reply]

Merger proposal: merge with Database model[edit]

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.
To not merge (for now), given the absence of support and at least one grounds for objection, with stale discussion. Klbrain (talk) 11:25, 23 November 2023 (UTC)[reply]

I propose merging the Database model into Database schema. The latter got more views. I think the content is basically teh same for two articles and a merger would not cause any article-size or weighting problems in Bar. Any objections? AXONOV (talk) 08:59, 3 June 2023 (UTC)[reply]

I am not sure there is much overlap, to be honest. Database model covers various, fundamentally different ways of conceptualising databases, while Database schema discusses the organisation of a single database, typically under the specific relational model. Felix QW (talk) 20:22, 18 October 2023 (UTC)[reply]
Well that's might be definitely true. Schema might be a specific representation of a conceptual model. There are differences. Let's see if there are more opinions. AXONOV (talk) 05:37, 24 October 2023 (UTC)[reply]
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.