Talk:Object-relational mapping

From Wikipedia, the free encyclopedia

SQL isn't relational, and most of the mentions of 'relational' here aren't correct but refer to SQL actually. So I am correcting the article and will move it. Leandrod 19:58, 21 September 2005 (UTC)

I don't know what you mean by "SQL isn't relational" SQL is a language designed to act create, modify and select data from relational databases. In any case the whole rest of the world refers to what you've decided to call "Object-SQL mapping" by the term "Object-Relational Mapping" I refer you to the google score of 248 result for "object sql mapping" and 380,000 for "object relational mapping" with similar results with and without a dash. If the article needs to be clarified and improved please do so, but unless I hear back from you or others, I will move this article back to object-relational mapping, which, I strongly believe is where the vast majority of people expect to find it. Charles (Kznf) 01:16, 15 October 2005 (UTC)

SQL manipulates Relational databases, and therefore everybody refers to it as "Object-relational mapping". The correct title, I believe, should be Object-relational mapping. Everybody refer to it with that name anyway, so I believe this is misnamed. Remember that the underlying database engines operate with relations, while OOP programming language operate with objects. Mapping Objects to relational database fields merely uses SQL as a means, because allmost every db engine on the planet uses SQL as its query language. 80.212.120.4 20:07, 22 November 2005 (UTC)

This is vandalism it will be cleaned up Real Soon Now. Look at the history of the article and pich a version before User:Leandrod replaced relational by SQL. —R. Koot 23:10, 22 November 2005 (UTC)

I've changed the references to Object-SQL mapping to Object-Relational. While SQL isn't purely relational, this has nothing to do with O/RM - SQL is just a means used to perform the mapping to the relational database - there are O/RMs for non-SQL databases too. Object-Relational is also the term used pretty much everywhere (including much of this article) - Wikipedia should be describing the term, not redefining it. I've also removed the discussion about purely relational DBMSs - it doesn't belong here, and a lot of it seems wrong. Impedance mismatch, for instance, has nothing to do with how relational SQL is - it would still exist with a purely relational language too (In fact, it would probably be worse, since OO has some concepts, such as object identity independant of state, that go against a purely relational model.) Brian McErlean 01:18, 22 December 2005 (UTC)

Contents

[edit] Cache

The author of this article seems to be slagging on the Cache database a bit. However I the Cache's claims to multidimensional database are on the level as far as I know.

From their website:

>>Innovative Database. Guaranteed Performance.

CACHÉ is a multidimensional database that uniquely combines robust objects and robust SQL, thus eliminating object-relational mapping. Caché enables rapid Web application development, extraordinary transaction processing speed, massive scalability, and real-time queries against transactional data - with minimal maintenance requirements.

Cache is based on the MUMPS programming language and database. I beleive their claims to true multidimensional database cabability to be more true than the author of this post-relational database article claims. Comments, Suggestions?

65.213.63.17 20:17, 19 July 2006 (UTC)Ira Carmel

[edit] XForms

Do any of these tools support or help generate XForms XML? --JonathanFreed 14:53, 29 March 2006 (UTC)

[edit] Firestar lawsuit

News broke today that a company called "Firestar Software" is suing Red Hat, owners of the JBoss software, over infringement of a patent which appears to cover ORM. I need to think through whether/how to add that info to the article, but wanted to add a note about it here to remind myself and anyone else who wants to work on writing it up.

An article about the suit is here: http://www.infoq.com/news/RedHat-Sued-Due-to-Hibernate-3-O Ubernostrum 15:26, 30 June 2006 (UTC)

[edit] ObjC version of EOF, where?

Could someone point me to the ObjC version of EOF that is mentioned in the article. I have been under the impression that ObjC version was killed soon after Apple bought NeXT and Java version is the only one there is.

[edit] Object-relational modelling

Is there a difference between this and Object-relationship modeling? If not, should these two pages be merged? Agreed. This article should take over Object-relationship modeling.

I've made the Object-relationship modeling page redirect here. There wasn't really any content in there that isn't already in this article (and written in a better style) so I didn't add anything to this article. Mutant 17:54, 27 February 2007 (UTC)

[edit] Links

There are far too many links. Please limit to only authoritative and well-organized sources. --Froggienation 20:30, 12 February 2007 (UTC)

[edit] Critique

It appears that the critique is violating NPOV, and seems very much like a personal opinion. It states as fact that ORM is a non-problem, and fails to explain or cite the fundamentals of data management, or explain why they are considered to stand in opposition to ORM. No citation is provided for the failure of Object Databases.

Also, it seems biased towards Java, where this should be an (OO) language-neutral discussion; it talks about JDBC specifically.

I'm considering editing it, but I think I'd be happier if the original editor made it less personal and provided some citations.

Racecondition 17:00, 31 January 2007 (UTC)

Agree, removal of strong emotive language with no references would be a good ida. But not by myself, seeing as I don't know much about the topic and came looking to learn more.

Agree. In fact, I think the whole section should be removed. It's completely baseless, and unreferenced. There should be a "criticism" section in this article, but I don't think there's a single factoid in that section worth keeping. Mutant 17:16, 9 February 2007 (UTC)
I read the Critique and immediately came here to remark on the NPOV issues. But, needing a definition of ORM, I thought it was valid criticism, if emotively worded. ('Utter failure'! I can't imagine using this in any form, written or conversational, without provoking an argument...) We should point out the objective facts underlying this:
1. it's easy to convert a fetched database record into a containing object or structure; (although the problems - missed by the criticism - are that code is repetitive, and must correspond with the data structure in the DB and therefore becomes a source for inconsistency errors);
2. relational databases hold the lion's share of the market - not that I have figures, and a citation would be required, but SQL server, Oracle, and MySQL together (and I'm not trying very hard here) must hold most of the market by money spent, users served and developer hours. Or it might be easier to find stats by required job skills?
The second paragraph ('impedance mismatch') has a core of truth; however, while I can see the relevance of the analogy, there's no guarantee that every reader would follow it. --Froggienation 20:30, 12 February 2007 (UTC)

I removed the first paragraph as soon as I read it, it's so obviously biased and specifically bases it's primary point on JDBC, which is Java specific when the article not language specific. I left the second paragraph intact since it wasn't an outright rant like the first paragraph. --mikeal

"The correct mapping in the relational model is between object and type." Are you sure? Class and type maybe...

What does "type" mean in this context? It isn't declared in the article. —Preceding unsigned comment added by 81.207.148.46 (talk) 08:57, 29 December 2007 (UTC)