Talk:Database/Archive 1

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Archive 1 Archive 2

Merging dbms and database?

I think that these two should be merged. The overlap is already very big and the difference that there strictly is can be explained in a small paragraph. Anyone have a strong opinion on this? -- Jan Hidders 16:04, 23 Nov 2004 (UTC)

Yes. A database and a DBMS (Database Managements System) are two very different things. A database, as I am sure many readers will be aware, is a particular collection of data and a DBMS is the (usually automated and computerised) system for managing any (compatible/normailsed) collection of data, any given dataset or database. Paul Beardsell 09:00, 13 Dec 2004 (UTC)

Database is to DBMS what book is to librarian. Database is to DBMS what an art collection is to a gallery, what postage stamps are to album, etc etc Paul Beardsell 09:00, 13 Dec 2004 (UTC)

A database is a set of data files that can be edited by a data base management system (DBMS). This is similar to the relationship that a document file has with a word processor. The document file can be edited with the word processor (if it is in the right format). A word processor cannot be edited by a document file.

As defined, databases are collections of data, not information. A knowledge base, often used to describe help file systems and FAQ lists, is a collection of files that have been pre-selected and edited for an audience based on pre-assessed use or need. Databases contain information in raw forms (data) like ages, times, or prices that, by themselves, are not meaningful to an observer. Knowledge bases are made up of information (processed data) gathered or compiled by an outside agent (human or software).

Disagree on merging them, but there is a lot of material in database that belongs in dbms. This should be about content and purpose, and talk about pre-digital databases, etc., not focus on technology. Lfstevens (talk) 00:24, 6 July 2010 (UTC)

No merging. The DB is "what needed to handle applications" from modelling point of view. The DBMS is "how to solve it with a SW system" (that runs on computers). --Comps (talk) 03:23, 6 May 2011 (UTC)

Database application

I was redirected to "Database" from "Database application". Is that right? It's not consistent with the description of the latter on Database management system. -- anon 19:43 23 Apr 2005 (UTC)

There's a link from the term in the Database management system article to this article. Neither article says much about the low-end database applications that most of us are familiar with. Oicumayberight (talk) 02:31, 7 February 2009 (UTC)
That's largely because the word is one of those technical terms that has slipped into popular [mis]use; it's not really clear whether an encyclopedia article should try to defend it as a technical term in its original sense or acknowledge the fact that it has escaped into popular culture as a term that can mean almost any ragbag of information. Mhkay (talk) 23:51, 7 February 2009 (UTC)
I would just like to see an article on the low-end database applications. You would think with the popularity of FileMaker and MS Access that there would have been one by now. But I guess the limitations of those applications overshadow their popularity. Or maybe the experts on databases are just too disgusted with the low-end usage to write about it. Oicumayberight (talk) 02:34, 8 February 2009 (UTC)

Database State

I don't believe that the term "Database State" should be included in this entry because it is not commonly used at all. I say this as a privacy advocate and a person in the process of organizing a legal symposium on just this topic. That said, the inclusion of such scarcely used terms (really... just run a Google on it) does not lend itself to a better understanding of what a "database" is. Just an opinion.

I agree completely. This seems to be a political statement made by an anon user at (addition made 05:12, 17 February 2005 if anyone is interested). That user also created a Database State entry as a redirect to Mass_surveillance. This does not belong here. Perhaps it belongs on that page. I'm going to remove it. --Derekian 28 June 2005 18:28 (UTC)

Flat Model

Does the section on "Flat Model" really belong here? OK, lots of data is stored in spreadsheets, CSV files, and the like - but that doesn't make it a database.

Mhkay 21:08, 16 September 2005 (UTC)

Yes, it belongs. An example is a database of bibliographic information. A single table can suffice for many such databases. 17:57, 26 September 2005 (UTC)

So where does the concept come from? Do you find it in database textbooks nowadays? Most of the google hits on '"flat model" database' are copies of this article. I'm not saying it's not a valid concept, I'm just suggesting that it's not a part of the accepted taxonomy, which is what we should be describing here. Mhkay 14:34, 14 October 2005 (UTC)

You can find the concept defined by Googling <define: "flat file">. Flat files have been around forever. The earliest databases were flat files. They predate CODASYL. I don't know about textbooks, or taxonomy. In the 1970s I worked with a flat file database which had 85,000 records and was accessible online throughout the US. Probably, no one knows how many flat file applications exist today.
This article is about database, not DBMS. With any relational DBMS, users can easily create flat file databases, and I suspect many of them are being used for simple applications today. AnonUser 00:49, 15 October 2005 (UTC)

Flat model is appropriate. They don't just predate CODASYL, they predate Jesus Christ by several thousand years. Any table of rows and columns containing information is a flat file database, even if it's just a census containing two pieces of information, the city and the number of people in the city. - Mugs 19:37, 6 March 2007 (UTC)

The Encyclopedia Britannica lists flat as one of the key database models. The consensus, as I read it here, is that the flat model should be included. Do we agree? --Girl2k (talk) 05:00, 20 January 2008 (UTC)

Associative Database

There is nothing about associative databases. I've read a few articles on the net that explained the underlying theory but I really want to know how it's put into practice. 20:57, 18 October 2005 (UTC)

ACID property may favor a particular model of database design

 Atomicity - command batching via a monitor (which wraps access to the database).
 Consistency - apply invariants to database tables. 
 Isolation - threaded execution.  
 Durability - transactions are permanent (can't be redone or undone, thus a logfilesystem would disobey Durability). 

This leads me to believe that either the explanation of the ACID property on the wikipedia page is inappropriate, the concept of the ACID property is shallow, I don't know what I'm talking about, or this is a simple set of rhetoric that was created by paid evangelists to support a particular vendors database design, such as ORACLE. For instance can you name object oriented databases that satisfy the ACID property, or is the ACID property unfairly biased toward Functional Languages and Relational Databases.

I think you are confusing your terms a bit. A database being object-oriented is not related to its being ACID compliant. If there is a question in the above paragraph, please restate it. Turnstep 23:05, 1 November 2005 (UTC)
ORACLE is a DBMS, not a database. - Mugs 19:40, 6 March 2007 (UTC)

LDAP and hierarchical models

Might be worthwhile mentioning the hierarchical model. And the model used by LDAP.


Decent article, but pretty light on sourcing. I may need to tag it to encourage editors to cite sources, if no editors step up to the plate. Remember, sourcing is important, even if you *know* what your writing (i.e. you're an expert in your field), otherwise you might get your edits flagged as Original Research and possibly removed. Thanks. --NightMonkey 00:49, 27 August 2006 (UTC)

Unencyclopedic bits

I think the list of brands is unencyclopedic, contributes nothing to the article, and is a waste of space. I think it should go and isn't even worth keeping as a separate "list of" article. We already have Category:Database management systems.

Likewise, the "See also" section is way too long. There's already a couple navigational templates at the bottom of the article which do more or less the same thing.

Any dissent? --Craig Stuntz 19:24, 30 August 2006 (UTC)

Craig: I agree: it's unencyclopedic, and it just keeps growing. It's like Wikipedia has become the yellow pages. --Girl2k (talk) 15:00, 4 April 2008 (UTC)

Normal Forms

Added section on normal forms, as understanding normal forms is absolutely vital to proper database development and design. - Mugs 19:21, 6 March 2007 (UTC)

I sourced the normal form section. Not sure of the right format, so I borrowed from a couple of examples I found at nuclear weapon and uranium. - Mugs 19:33, 6 March 2007 (UTC)

Belucci ??

...And exactly why is this article named "Belucci" ??? 16:49, 2 April 2007 (UTC)

Sorted. Back at the right name. Some vandal moved the page as their idea of a joke. I mean, April Fools was yesterday, guys. Cheers, Moreschi Request a recording? 19:46, 2 April 2007 (UTC)

too large?

the "database" wikipedia page appears to be too large created a "database models" page and summarised database models section, advise that it is summarised further

A link to Database_model should be placed and all the discussion about models should be moved/merged to/with respective model. Or if the summary is judged to be good should be moved to a new page (.e.g Database_models_summary) and link to it.


I am not saying anything bad about them. But can you give a little more detail on databases??? But man, what person would spend so much time on the internet gathering data, and half of it isnt even correct. —The preceding unsigned comment was added by (talk) 12:55, 15 May 2007 (UTC).


MS SQL Server

Is it relevant to label Microsoft SQL Server as "derive from Sybase"? That was almost 20 years ago, and the modern product is almost entirely rewritten and quite different. DonPMitchell 01:28, 8 September 2007 (UTC)

I removed it. Specific historical information like that is more appropriate in a history section of that specific database implementation than in a general article about databases. ChorizoLasagna (talk) 00:39, 28 March 2008 (UTC)

"In <subject domain>"

Maybe this is just my personal opinion, but starting off an article with "In <whatever subject domain>, a <subject of this article> is ..." is not really nice. Maybe it is still encyclopedic, but seems just a little bit, not meaning to be pejorative, but sort of immature. Anyway, I changed it in the opening sentence of this article. (Maybe I should consult the Manual of Style - I'll do that). Peashy 13:28, 9 October 2007 (UTC)


It seems to me that this paragraph doesn't really belong in this article.

In the United Kingdom legislation protecting the public from unauthorized disclosure of personal information held on databases falls under the Office of the Information Commissioner. United Kingdom based organizations holding personal data in electronic format (databases for example) are required to register with the Data Commissioner.

This article describes databases in a technical sense, but that paragraph is about legal requirements for database security. That's a different subject altogether. Colonies Chris 19:36, 15 October 2007 (UTC)

It does seem out of place. It does not help me understand databases any better that much is for sure. -- Sam Barsoom (talk) 17:08, 16 November 2007 (UTC)

when using "database" to include dbms ...

In the intro paragraph the sentence: "When the context is ambiguous, however, many database administrators and programmers use the term database to cover both meanings."

seems to me it should be when NOT ambiguous ... Fholson 14:01, 4 January 2008 (UTC)

History of the Term

The Oxford English Dictionary lists 1962 as the first use of this term, thusly: "Technical Memo. (System Development Corp., Calif.) TM-WD-16/007/00. i. 5 A ‘data base’ is a collection of entries containing item information that can vary in its storage media and in the characteristics of its entries and items." Does this trump the citation given in the current article, which dates it in November 1963? I don't know. --Girl2k (talk) 05:04, 20 January 2008 (UTC)

Database Development Platform

Regarding the last section, what is a "database development platform" and why is this section here? Could a lead-in be added? --Girl2k (talk) 05:11, 20 January 2008 (UTC)

Database Models

Could a specific, real world example of each model type be added? Network model has one, but not the others. --Girl2k (talk) 05:13, 20 January 2008 (UTC)

Here, here. A prose definition is a nice start, but a concrete example would do wonders for the article. IMHO, most folks learn better when abstraction is paired with applications. A sample would show what a database is (and give meaning to "tables", "records", and whatnot). Maybe explain how a card catalog is a DB; say what parts correspond. (talk) 21:06, 5 March 2009 (UTC)
Are there any other articles about examples of databases like the Suppliers and Parts database article? (talk) 21:10, 5 March 2009 (UTC)

Database as Cultural Form

Anonymous user writes: "I think this section is weird and should be deleted." I've moved this from within the piece itself. —Preceding unsigned comment added by Girl2k (talkcontribs) 16:22, 12 February 2008 (UTC)

As a cultural form?

This section of the article is certainly not elaborated. I do not see how a gigabytes-sized collection of data about customer purchases or business locations is a literary genre akin to the epic. Are these literary critics referring to something else? —Preceding unsigned comment added by (talk) 20:15, 25 January 2008 (UTC)

Yes, of course, they are referring to something else! They are referring to databases that are composed of, for example, the entirety of everything Walt Whitman every said or wrote or read, etc. I didn't want to go on too long in the entry. If people are interested, they could read the citations--at least that's what I thought. The short paragraph (in a long entry) is meant to point out that databases aren't only used for commercial purposes. Some people think they are a new form for expressing things. In an encyclopedia I thought that might be worth mentioning.--Girl2k (talk) 15:43, 4 April 2008 (UTC)

Then I think there might be a better way of saying it. I think it would be good to draw attention to the fact that databases are used for corpora of literary texts (or images, sound recordings, etc) without this talk of "cultural forms" which feels rather out of place in an article that in other respects is technical rather than sociological. Mhkay (talk) 21:21, 5 April 2008 (UTC)

I took out the reference to the literary critic, which seemed to be the problem for you. Hope it helps. I don't think an encyclopedia article should be "technical" to the exclusion of all else. In fact, shouldn't they be well rounded and complete rather than partial? --Girl2k (talk) 18:29, 7 April 2008 (UTC)

Sorry, but this "lit crit" theory on databases as cultural form falls somewhere between POV pushing and something else. Wikipedia is not a soapbox to expound theories. See [WP:SOAP]. If Lev Manovich becomes notable enough to get a Wikipedia article then this this section can be moved there. Removing it and pasting it here for future reference. (talk) 19:32, 9 June 2008 (UTC)

== Database as Cultural Form ==

Although originally a computer technology, the database, according to media theorist Lev Manovich, is becoming a new cultural form in its own right and a genre of new media Lev Manovich: Database as a Genre of New Media Lev Manovich (2001), The Language of New Media, Cambridge, MA: MIT Press . A cultural form is one of many ways that people represent the world—art and literature, for example. As contemporary culture is gradually computerized, Manovich argues, traditional cultural forms are being replaced with new ones that derive from the computer. He calls this transcoding. The database is the computer age’s key form of cultural expression, as narrative was to the modern age via cinema. In this analysis, he is using "database" metaphorically. See also database cinema.

Katherine Hayles has argued, in response, that narrative and database are not in opposition but rather are natural symbionts. N. Katherine Hayles ((2007): 1603-08.), Narrative and Database: Natural Symbionts., PMLA 122:5  Check date values in: |date= (help)

Lev Manovich has his own Wikipedia page. I'm reinstating. There's nothing POV about including this in the piece, though there may be something POV about EXCLUDING it. --Girl2k (talk) 05:13, 20 January 2010 (UTC)

"Relational model" first paragraph

Does anyone mind if I delete the first paragraph under the "Relational model" heading? It doesn't seem (to me) to fit in with the rest of the section. Its description of "relation" pretty much conflicts with the following paragraph. But perhap[s I'm missing something. Northernhenge (talk) 16:45, 25 February 2008 (UTC)

It's certainly a bit messy. Apart from anything else, the section has a high overlap with the "relational model" section of Database models, and has no reference to the free-standing article Relational Model.
I think the paragraph is hinting at the existence of a mathematical relational model which is rather different from the tables-and-columns model used in engineering practice. But is would be better to state this clearly and refer the reader to a different article for the mathematical treatment.Mhkay (talk) 08:43, 27 February 2008 (UTC)

I've now deleted it and added a reference to Relational Model. Northernhenge (talk) 23:44, 16 March 2008 (UTC)

Schools database

Someone dumped the following at the end of article:

In relation to a "Schools Database" Sites, such as Wikipedia, Google, Yahoo, MSN, and other search engines are a much better choice for several reasons.

A.) School Database's are expensive to keep up with. And for the most part no one goes to the sites in the first place. On the other hand, Search Engines are Free of charge to use indefinatly. Schools could save money to go to other things they need more like building repair and textbook purchases.

B.) School Databases have no appeal to todays youth. Kids are not going to want to go look up information on some old "School Database". Computer Users grew up with google, yahoo, msn. Its like an addiction that kids will not easily break out of.

In Short, Google\Wikipedia Rocks. School Databases, Do not.

It might have some merit somewhere, but not where it was and not in the nonwiki format. I have removed it from the article. If someone wants to look after it . . . sorry but I have no time or real understanding of the above issue. Peet Ern (talk) 00:21, 1 April 2008 (UTC)

You will have to find sources for all of that.Jasper Deng (talk) 02:39, 9 February 2011 (UTC)

Remove List of Brands?

Does anyone want to make the argument for why this list of database brands should be in an encyclopedia article about databases? --Girl2k (talk) 18:34, 7 April 2008 (UTC)

I've added a Main Article link to Category:Database management systems. I agree the list should be removed from this article. It is useful to have a list of all the DBMS products that have Wikipedia entries, as otherwise I'm not sure how you could search for them easily, but this is not the correct place. The Category page does need quite a bit of work though. Maybe the list in this article should be moved to a "work in progress" category on the Category page??? --Northernhenge (talk) 08:52, 8 April 2008 (UTC)

I feel that it's one thing to have a table comparing each database server, but another to have a list of links that just go to the main article. Perhaps we should do that here? Psychcf (talk) 00:53, 17 April 2008 (UTC)

Not sure I understand what's being suggested ... Why not just do it! --Girl2k (talk) 16:25, 17 April 2008 (UTC)

singular or plural ?

"...and changes often demand a major programming efforts." —Preceding unsigned comment added by (talk) 03:57, 23 May 2008 (UTC)


According to the article text a "database" is: integrated collection of logically related records or files which consolidates records previously stored in separate files into a common pool of data records that provides data for many applications. A database is a collection of information that is organized so that it can easily be accessed, managed, and updated.

This is not a good definition: an Excel worksheet is not a database, yet arguably it fits the above description.

E.g. Fabian Pascal notes than databases are not "storage systems":

They are organized collections of data from which mechanized logical inferences are made.[1] [emphasis mine]

An Excel worksheet lets you manage the structured data within, but you can't make automated inferences from it: you can't get it to answer an ad hoc question such as "Which customer who did not buy any DVDs last month bought most books the same month?" It is a crucial difference. Amazingly (or not, as F. Pascal would probably say), many sources I've seen fail to make this distinction. GregorB (talk) 19:04, 12 August 2009 (UTC)

You defined a database as " integrated collection of logically related records or files consolidated into a common pool that provides data for one or more multiple uses." However, I feel that this is incomplete. A database is a collection of logically related records, files or fields. A field is a singular piece of information. —Preceding unsigned comment added by RazorXX8 (talkcontribs) 22:22, 27 October 2009 (UTC)

Yes, it's a lousy definition. I'm afraid that coming back to this article after a year or more, I'm depressed about how its quality seems to have got worse rather than better. The problem seems to be that people are coming to it from too many perspectives: for trained computer scientists, database means one thing, but for home PC users it means something rather different. Given the number of people editing, some very childishly, it's hard to see what to do about it. Mhkay (talk) 23:43, 18 November 2009 (UTC)

A database is a particular way of organizing data and their interconnections. Although the awareness of the term raised during the information era, the first sentence of a definition should be as if data processing machines are yet to exist. Also a database is a way of organizing structured data, otherwise a file system is more appropriate. For example, in an archive of portraits, a non-visual search for those who wear a hat is possible only if a database of structured information about the portraits is available. What about changing the first paragraph in something similar to this:
A database is an organized collection of one or more logically related data sets that are normally structured and associated with natural or virtual entities. It serves to users in extracting inferences for particular educational, scientific, entertaining or any other commercial or social purposes. The type of representation defines a database model or schema implemented in a database management system. One way of classifying databases involves the type of content or content semantics, for example: bibliographic, statistical, medical, dictionary. Other classification methods start from examining database models or database architectures: see below. As of 2009[update] the relational model occurs most commonly. Other models such as the hierarchical model and the network model use a more explicit representation of relationships. Sstadao (talk) 18:25, 28 December 2009 (UTC)

"Although the awareness of the term raised during the information era". I'm afraid that statement typifies an awful lot of the nonsense in this article. It simply isn't true. The word database initially described only large electronic databases, it later became diluted so that for some people it means almost any collection of data regardless of size and regardless of whether it's on a computer or not. The question is whether this article should about databases in the original, formal, and scientific sense, or databases in the sense understood by ill-informed schoolchildren. At the moment, the schoolchildren are winning. Mhkay (talk) 00:11, 22 March 2010 (UTC)

Need some links to database languages

For someone like me who came here hoping to find links to database languages I could use, there is nothing. In fact I've found nothing in Wikipedia that lists database languages. I'd like to at least put in links to xbase and sql, so that there is some connection down from the abstract to the practical, but there seems no apperopriate place in the article to include them. And, how about someone doing a Comparison of database languages? (talk) 10:20, 27 September 2009 (UTC)

You mean query languages? GregorB (talk) 13:45, 27 October 2009 (UTC)

"Post-relational" is not neutral

The term "post-relational" is pejorative towards relational databases and therefore non-neutral. It implies that the proposed alternatives are somehow superior to and/or a replacement of relational databases when in fact they are merely alternatives. Furthermore the term is a neologism at best, enjoying only extremely limited use in unreliable sources. -- samj inout 03:10, 5 January 2010 (UTC)

I've never heard anything so ridiculous. The relational model is nearly 40 years old: to suggest that there is no such thing as "post-relational" is to imply that the world has stood still for 40 years, which is clearly nonsense. "post" means "after", not "better", and you can argue whether developments subsequent to the relational model were an improvement, but you can't deny that they happened. Also, the term is quite widely used. Mhkay (talk) 08:50, 7 January 2010 (UTC)
It's not a case of nothing being better than relational. The section title simply isn't accurate. The section should be named "Hybrid Relational Databases", since what's described is hybrids of relational/non-relational DBs.Jberkus (talk) 00:49, 9 January 2010 (UTC)
The accurate term for a database model, that applies object-orientation to relational design, is "object-oriented enhanced relational Database". The accurate term for a database model, that applies relational design to an object-oriented database, is "relational enhanced object-oriented Database". This way the relations database stays a relational database, and the object oriented database stays object-oriented. The addition is of course an enhancement, otherwise it wouldn't have been implemented, so there's no discussion about that.
These 2 things are generally totally different, so we should better not throw them in the same bucket. Cari-Kira (talk) 01:56, 16 January 2010 (UTC)

You are —Preceding unsigned comment added by (talk) 14:18, 9 March 2010 (UTC)


WikiProject Guild of Copy Editors
WikiProject icon A version of this article was copy edited by a member of the Guild of Copy Editors. The Guild welcomes all editors with a good grasp of English and Wikipedia's policies and guidelines to help in the drive to improve articles. Visit our project page if you're interested in joining! If you have questions, please direct them to our talk page.

Substantially shortened this, although I think much of the technology-related content belongs in Database management systems instead. Feedback welcomed. Lfstevens (talk) 23:30, 6 July 2010 (UTC)

What is a database.

"A database consists of an organized collection of data for one or more uses, typically in digital form.One way of classifying databases involves the type of their contents, for example: bibliographic, document-text, statistical. Digital databases are managed using database management systems, which store database contents, allowing data creation and maintenance, and search and other access"

I am sorry but I am not wild about this description.

I mean I would expect something a little bit more basic. possibly fundamental?

launching into the esoteric heady heights of clasifying databases in the second line is a bit on the nose.

Can we backpedal a bit here.

and for heavens sake sentance one is at odds with itself. It is not logical. digital form is not a user.

I would like to try to pin down in a sentance or two what a database is, but I am too shy. Darn it.

Robfor (talk) 09:39, 14 August 2010 (UTC)

Unfortunately this page has been edited by many people with wildly varying perspectives on the subject; some of them seem to have only the level of knowledge of someone doing their first IT course. There are also valid differences of view on the scope of the article and the intended audience. You need to get agreement on these matters before you can come up with a good definition. Frankly, the whole article and not just the lede needs rewriting. In my view what distinguishes a database from any old random bucket of data is that it has all or most of the following qualities: (a) it has a design that defines the permitted constructs that may appear in the database and their meaning, that is, the information that they represent; (b) it serves a community of users with a variety of information needs; (c) it is managed to deliver quality of service, for example a defined level of availability and performance, resilience, etc.; (d) it is accessible via a query language that allows searching and/or inferencing.

Maybe this an indication that the article is trying to cover logically distinct concepts that should properly be separated. If we define database-1 to be a collection of data with a certain conceptual integrity and set of semantics that need to be preserved, and which need to be understood by anyone wishing to make meaningful use of the data, then much of the article seems to be about something else. Indeed, it seems to be about a database-2, which we might define as a collection of data stored on a computer and managed by a piece of software known as a database management system (DBMS). It makes no sense to speak of a hierarchical, network, relational, object based, or post-relational (a term coined by InterSystems for their product) database-1, but it does make sense to speak of a database-2 as falling into one of these categories. In a sense, the situation here is comparable to mathematical logic: propositional logic, predicate calculus, modal logic and so on are different mathematical approaches to formalizing logic in the more general sense. They vary in expressiveness, completeness and computability and are useful in different situations. But none of them is logic in the more basic sense of the term, they are mathematical formalisms that we find useful in investigating or applying logic in its more basic sense. The same can be said of various database implementations, or of modeling languages techniques such as entity-relationship (ER) diagrams. Greg Woodhouse (talk) 20:54, 9 November 2011 (UTC)

No, it's not a problem of multiple concepts, it's a problem of multiple perspectives. It's the same problem as faced by people trying to write an article on measles or depression or volcanoes, where the popular viewpoint and the specialist viewpoint on the subject come to it from radically different angles. Here we've got the viewpoint of the theoreticians, the viewpoint of the engineers, the viewpoint of the users, and the viewpoint of the schoolchildren who are taught about the things at age 14 and believe what their teachers tell them. Mhkay (talk) 21:09, 9 November 2011 (UTC)
Since databases were introduced in the computer science literature, the term has broadened in two ways. Firstly, it is used to refer to collections of data on computer that don't have these properties - for example, a simple spreadsheet. Secondly, it is used to refer to collections of data that aren't even on computer. In my view these usages are incorrect and should be out of scope of the article, though they should perhaps be mentioned.
The definition should ideally be of the form "A database is an X which has properties Y and Z". I think "X" here is "collection of data", and Y, Z should be the kind of properties I outline above. Describing subtypes of database has no place in the definition, though it is often the first thing to follow the definition. Mhkay (talk) 21:46, 14 August 2010 (UTC)

relation-valued attribute redirects here... why?

The topic "Relation-valued attribute" redirects to this page. But there doesn't seem to be any mention of relation-valued attributes in this page at all. This probably is an error. Courier (talk) 23:28, 9 September 2010 (UTC)

Relation-valued attribute began its short life as a "stub" whose sole content was "See also Database". It was put out of its misery by being redirected here. At present the only Wikipedia articles that contain the term are First normal form and Relational model. If you know whether either of those or any other article would be an appropriate redirect target, feel free to fix it (or let me know if you need a hand). Adrian J. Hunter(talkcontribs) 05:17, 10 September 2010 (UTC)civic set o

In my opinion, that is an improper redirect for a number of reasons. A relation is simply the formal term in relational database theory for what we typically think of as a table. It is important to note that a relation is an instance of a table (i.e., one containing a specific set of tuples or rows). In a real database, relations (tables) are typically modified over time, and the result is an entirely new relation. Thus, if you have a DBMS with a PERSON table or an ADMISSION table (say, in an electronic health record system), then ADMISSION doesn't refer to an instance of a relation (which is a value, not a variable) but something that can take on various relations as values. So, to be precise named tables should be called relation variables. Attributes normally correspond to what we think of as columns in tables. A good example might be DOB (date of birth) in the PERSON table. Current RDBMS offerings do not generally support relation valued attributes though, in principle, they could. I'm baffled by the redirection of relation valued attribute to database. That simply makes no sense, at least if we interpret database to mean a collection of tables managed by a single DBMS, or to a collection of interrelated information. Greg Woodhouse (talk) 20:12, 9 November 2011 (UTC)

Merger proposal

It has been proposed that Database administrator be merged into Database. This has been tagged for some time, but not formally discussed. The following is from a discussion on Notability of Database Administrator:

The notability of this article is being questioned, with someone arguing that at the moment it's just a definition (albeit cited) and not deserving of its own article. Can someone with a firm grasp of Wikipedia's notability guidelines please advise? Jwoodger (talk) 00:40, 9 September 2009 (UTC)

I do have a firm grasp of the guidelines, thank you. Wikipedia is not a dictionary Please show how this topic, which the article currently gives only a definition of the job title, can expand beyond this while also meeting WP:N. At best, I could see this being cleaned up, reduced, and merged into database. -- AnmaFinotera (talk · contribs) 00:45, 9 September 2009 (UTC)
I apologize if I sounded combative in my comment above - I think you are correct though, merging this (at least the cited definition) intodatabase seems like an appropriate action. Jwoodger (talk) 01:41, 9 September 2009 (UTC)
No problem. I've reverted your last edit, though, as the lack of citation does not effect a merge discussion. -- AnmaFinotera (talk ·contribs) 01:53, 9 September 2009 (UTC)
  • Support - Database administrator is not substantial enough to warrant an article on its own, and there are questions as to its notibility. What more information there was before has been removed, and there is no indication that more is likely to be added. Gandaliter (talk) 14:10, 20 January 2011 (UTC)
  • Support - You may merge it into the Database article as a separate "Administration" section. Sae1962 (talk) 09:21, 17 March 2011 (UTC)
  • Oppose - Umm, that's like saying 'mechanic' should be merged into 'car', or 'secretary' with 'schedule'. A DBA is most definitely more than a subset of a description of what he works with. Consider: List the properties of a 'good' database, and the properties of a 'good' DBA. Is there ANY overlap? Of course not. They're aren't related enough the describe using the same terms. As far as material... Suffice to say that "Database Administrator" is like "mechanic". There are many specialties within that profession and any one article, or even book, is unlikely to be able to cover all of them, and yet the uninformed tend to expect that anyone that professes to be a DBA somehow must know everything about each facet of each DBMS. Not convinced? Take a look at "BI" some day and you'll see that it's a specialty in itself, and could easily be broken down into several more. To put it another way, these articles are (or should be) describing two entirely different entities, one of which can talk. Wclemmer (talk) 10:42, 21 April 2011 (UTC)
  • Oppose - Two separate, distinct subjects. DBA may be mentioned here, but extremely shortly. Whether it deserves an article is a different subject. It can easily be made an article if functions and skills are elaborated, but definitely not here. --Comps (talk) 03:31, 6 May 2011 (UTC)
Rejected. Tag removed. DBA may be mentioned in a section here, but with the entire list of requirements and qualifications it deserves a separate article. --Comps (talk) 10:35, 9 May 2011 (UTC)

Comment - the article COULD be a decent seperate article (it currently is not). It would be nice if things like average salary or work hours were mentioned. Something actually about people who are DBAs. Jason Quinn (talk) 06:17, 16 June 2011 (UTC)

You are welcome to add the article, if you think it is needed. If you just suggest such, it will either happen or not. The only way for you to make sure it happens is to write it in a way that meets Wikipedia criteria... You can start a stub by generating a red link for "Database administrator" in the current article and follow it to put content in it. Comps (talk) 13:17, 18 June 2011 (UTC)
Correction: The article already exists. Just improve it and add whatever you think suitable. Comps (talk) 13:28, 18 June 2011 (UTC)

Database administrator (DBA) is a role, comparable to programmer or architect. You wouldn't merge the articles for programmer and computer language, would you? This is a nearly exact parallel. In practice, the DBA has responsibility for data that spans products and projects (unlike a programmer who tends to be focused on specific projects). The DBA also plays an important role in the physical implementation of models developed by data models and/or programmers, and is concerned with optimization of maintenance of a database to a greater extent than are developers. Another difference is that DBAs typically have (and require) greater knowledge of the peculiarities of a particular DBMS and system(s) under which it runs. In practice, there is overlap here, and one person may play both the role of DBA and modeler (or even developer), but DBA is an important job classification deserving its own article. Greg Woodhouse (talk) 20:35, 9 November 2011 (UTC)

Some suggestions to improve the article

This article is very important, an entrance into the DB subject, and with many readers (see daily statistics). It is almost 10 years old and in poor condition. After reading the mixture that it comprises I do not feel it serves its purpose, properly explaining what a database is. In fact it is confused and confusing. I also see that a substantial part of the last contributions (I have not gone too deep backwards) is reverting others. This is a testimony of confusion, and that people desperately try to improve.

I propose to adopt the following:

First principle: Describe what a database is, not how it is implemented. How it is implemented is the part of DBMS. Can be done relatively in short, referring to the main article. This means that Storage, Transactions, Concurrency control, Indexing, Replication go into the DBMS section at a very high-level (no too many details).

Major structure suggested:

1. Overview. On DBs purpose and usage.

2. Database types. Enlarging usage. Improving categorization. The current mixes types on different dimensions, including implementation (go down). Actually requirements for different utilization.

3. Data models (the term architecture can be applied here to the models, NOT to the DBMS architecture), and how they fit to model the worlds we need to model. Emphasis on the most common: Relational, but good coverage of the old and emerging useful models.

4. Important issues like Availability, Security, etc. (user point of view)

5. Implementation. How to satisfy the requirement: DBMSs, their types, and main components.

Existing paper paragraphs can be used as a start, but a lot of work is needed, including with the references, to make them more effective.

Pls provide feedback, thoughts.

--Comps (talk) 01:05, 6 May 2011 (UTC)

I have written a new lead section for the article and re-organized its sections according to the proposal above. Both the lead section and the section reorg can provide guidance for the way I think this article should go. A long way ahead. Pls join. --Comps (talk) 14:12, 9 May 2011 (UTC)
I made another section reorg to an order that looks more logical to me (again, no change to previous existing sections). Added place-holders to refining sections. The DBMS section should be reduced to necessary minimum with Database management system as main article (no Wikipedia redundancy is needed). A lot of work is needed, and people are invited to contribute. --Comps (talk) 16:10, 10 May 2011 (UTC)
I have downgraded the quality classes of this article to "start"
Plan above with some enlargement largely executed with a new, expanded lead section, place-holders for sections, respective Main articles links, and respective request for expansion tags, occasionally with a short (sentence or two) elaboration. No old text has been change yet, but can be improved considerably. Please expand (the many) sections and improve text.--Comps (talk) 11:54, 12 May 2011 (UTC)
Well done for taking this on, the article really needs it. I've tried to build on the work you did on the opening sections (a) by improving the English (e.g. cutting out the unnecessary abbreviation DB) and (b) by sticking to the main points and getting them across more forcefully, largely by removing distracting asides. The section is still too long for a lede, but hopefully that will sort itself out as the rest of the article improves. Mhkay (talk) 22:57, 12 May 2011 (UTC)
Thanks, and thanks for the edit. I hastily reverted it due to some elements that initially annoyed me, but I saw the value after digesting and rethinking it, and then re-reverted. Your version is much more readable and emphasizes some important points. Pls let me know if you have any comment on the article's new structure. Similarly I would like to have others comments and reach consensus on it. This is in fact the main article in the Database category and deserves high attention. I hope that many more join its editing. --Comps (talk) 03:38, 13 May 2011 (UTC)
Regarding the lead section length: I find the the current length appropriate. It is within the guidelines, and looks to me quite effective, especially after your edit. My view is (after feedback from others) that the lead section should give a good idea about the article's content, and quite naturally a long article with wide coverage results in a relatively long lead. --Comps (talk) 15:04, 13 May 2011 (UTC)

Another suggestion: Someone correct the spelling error in para 2: "A successful general-purpose DBMS is designed in a way that is can satisfy as many as possible different applications and application designers." Merligren (talk) 18:40, 5 June 2011 (UTC)

If you are sure about your suggestion, it is in general better to correct, rather than suggest. Especially with minor or editorial corrections. With major or massive corrections one needs to be very careful... --Comps (talk) 16:50, 28 June 2011 (UTC)

The restructuring of the article has been completed to a great extent, but quite a lot of work is still needed (many sections need expansion). This is not a job for one or two people, and more people knowledgeable in the area are encouraged to contribute (also improve related articles). Average daily viewing of this article is quite high: ~4.3K, which shows considerable interes. --Comps (talk) 16:50, 28 June 2011 (UTC)


Since the restructuring start on 6 May 2011 the article's size grew up (mostly by original text (but not ideas...) I have added) from 27.K to 64.8K. This article is, and should be, an index to major database related subjects (articles in Wikipedia), by links and by "Main article" section tags. --Comps (talk) 17:19, 28 June 2011 (UTC)


The article is developing well to maturity. Its size now is 89K +.
I have upgraded the article's quality from "Start", where I put it in the restructuring beginning, to "C".
More help is still needed. --Comps (talk) 23:26, 26 July 2011 (UTC)

Feedback on the article is requested

Especially from those who rate the article. --Comps (talk) 17:53, 28 July 2011 (UTC)
I.e., specific comments that may help to improve the article. --Comps (talk) 22:22, 3 August 2011 (UTC)


My restructuring and section writing and revising has been mostly completed, and (briefly) well covers the subject to my taste. The article can be viewed as an elaborated index for the subject, and its size now is 92,555 bytes (starting at 27K on May 6). Some sections are still empty and need expansion. Few old sections need cleanup and possibly rewrite. Referencing is still needed, as well as old references revision (which I have partly done). I hope that more people contribute. --Comps (talk) 14:13, 2 August 2011 (UTC)

Network model removed

The reference to Network model (database) was removed by this edit. I see no reason for that, and there's no discussion on the topic here. Should it be brought back? Nawroth (talk) 09:13, 24 May 2011 (UTC)

No reason for removal. Possible vandalism. Reverted. Comps (talk) 17:42, 24 May 2011 (UTC)

Tag in doubt

Moved from inappropriate placement in the middle of the headers section: "the following tag" refers to the {{backwardscopyvio}} tag. Chris Cunningham (user:thumperward) - talk 10:18, 4 November 2011 (UTC)

The following tag is in doubt since most of the Database article's content is new and original since May 2011 (later than 2010, the year of the publication referenced). Please be specific regarding the identical text (if any; only old text here from 2010 or earlier is possible to be copied into the referenced publication), so it can be checked, and validity of tag verified. --Comps (talk) 16:13, 20 June 2011 (UTC)

See VDM Publishing (Betascript Publishing). This tag is probably justified, if this book (as many others by VDM) consists of copying Wikipedia articles or parts without meeting proper copyright conditions and disclaimers, as requested by Wikipedia. Comps (talk) 16:06, 24 June 2011 (UTC) --Comps (talk) 16:08, 24 June 2011 (UTC)


database design — Preceding unsigned comment added by (talk) 08:31, 6 July 2011 (UTC)

??? --Comps (talk) 15:04, 6 July 2011 (UTC)

Vandalism is no replacement for discussion

This article is being vandalized regularly. It probably has strong connection to the large viewing of it (~4K daily; see page statistics). It seems that some people use vandalism to express dissatisfaction of, or frustration from some aspect or aspects of the article (which not all of it is probably easy to read). This is a very nonconstructive way to communicate it. A much more effective way is by a discussion on the discussion page here, where any concern can be addressed and resolved/explained. People who are not interested in all the details can get an idea about the subject from the lead section in the beginning. --Comps (talk) 15:22, 20 September 2011 (UTC)


Undid revision 456661234 by (talk)Reverted. You have used old version and harmed article. See discusion! Examples include the most mature DBMS. Not an open source article

Which section this above comment refer to? Here is no discussion about Postgres. Can one show why more mature is MS-Sql versus [[[Postgres]]? HoW looks Ms package, when compared in context of serious computing, to robustness of the othervise "discarded" here. Look supercomputing postgres versus [Microsot SQL Serwer]: like 8:5. However considering matures (eg.. critical bug patches) of MSSQLS is even more problematic. So please prove claim or explain what the comment mean or to which already consented section, the above comment refer. Substitution "or to and" will be not considered a bad intent. (talk) 12:38, 21 October 2011 (UTC)
The issue was not Postgres. If strongly important to you it can be added to the list without all details about levels of openness. I would put it in the open source part after MySQL. __Comps (talk) 12:56, 21 October 2011 (UTC)

User damages the article

User has made several changes in this article (Database): Most have been inappropriate, and at the best unimportant but not damaging. He has never given any explanation for his changes. For his last edit he used his older version (which had been reverted by me), and as a result wiped all later work. I hope he did it by mistake only.

This user damages the article with his behavior.

It seems that he is pushing some open source agenda here, with terminology like "fully open," "partially open" (defined in some reference he provides), and "not open." He puts the "fully open" prods in the beginning

re:|He puts the "fully open,"| <objection this is insinuation imagined by adversary (the strike out word is falsely qouted You should say sorry and to show good will write "false quoting is a shame" 3 times in your user page (you can it later delete:) (talk) 13:17, 21 October 2011 (UTC)
Sorry for the word "fully" but this is exactly what you mean by "open..." --Comps (talk) 13:38, 21 October 2011 (UTC)

of a DBMS examples list, which are important but less mature products. The open source aspects are secondary here at this point, and not deserve such emphasis in the lead section. The subject and its impact on the economy of databases may deserve mentioning in the History section.

--Comps (talk) 12:39, 21 October 2011 (UTC)

Will be satisfactory if listing of Microsoft product take place in front on the sentence? There is no open source agenda only basic logic. If there are DB epitheted as as open source other for sake of neutrality and logic have to be epitheted convergently, Either take the openness of source criterion off completely, as not important, or use criterion of openness of sources to all. Try not talk about users, and make an effort to comment the above section. (talk) 13:01, 21 October 2011 (UTC)
No. Not satisfactory, and it is not in the front. The area has some history, which is also reflected to some extent in the list. It is not absolute, and by different criteria you get different orders and different lists. Definitely I would not use openness as a major criterion when trying to give a short representing list of DBMS prods, and not qualify there levels of openness. Regarding user vs. subject matter: I think that you have caused damage (see above), and it is important to have your attention to this, to avoid a repetition. --Comps (talk) 13:32, 21 October 2011 (UTC)
Ok s.word accepted. Now you can continue in discussion. Look at new edits. It is shorter and add new terms about data flow. The words about ~data slavery by database were rather not true. However from sales dept. point of view is quite convenient to persuade customers they can't move out data out dbase to another DBMS. If it will be true what is worth a DBMS which can not export and import any data at will of database admin ? (talk) 13:50, 21 October 2011 (UTC)
The point made was regarding the database (itself, not the data) non-portability among DBMSs. I have overlooked this, but now sentence is correct. See also section on database migration. --Comps (talk) 15:03, 21 October 2011 (UTC)
Still incorrect but less ugly.
DBMSs are regularly ranked by sales. For open source DBMSs this is not a good measure, and thus the separate category in the list. For open source DBMSs it is typically by licenses. Is PostgreSQL the second? --Comps (talk) 09:20, 22 October 2011 (UTC)
Re "DBMSs are regularly ranked by sales." <but the sentence do not push salesperson POV do you see: "well-known and highly utilized products" .There is nothing about sales POV. Are you about to push for it? Also such comparison is erroneous see Division by zero. (talk) 06:28, 23 October 2011 (UTC)
Sales is one of the common criteria for "highly utilized," with proper weight to the buyer that can afford the higher price. Most mature products are still "closed," and thus no explicit mentioning of "closed source" is needed. It is clear from the sentence which is closed and which is open without explicit "closed source" in the sentence.
Please use your high-energy for helping the article where help is needed (e.g., expanding sections where tagged, and writing empty sections). --Comps (talk) 21:20, 23 October 2011 (UTC)
Sales is one of the common criteria for "highly utilized," with proper weight to the buyer that can afford the higher price. Most mature products are still "closed," and thus no explicit mentioning of "closed source" is needed. It is clear from the sentence which is closed and which is open without explicit "closed source" in the sentence.
Please use your high-energy for helping the article where help is needed (e.g., expanding sections where tagged, and writing empty sections). --Comps (talk) 21:20, 23 October 2011 (UTC)
"use your high-energy" < Will it be prone to encourage in coediting commercial advertising for free? Until the art.summary is transformed from add to informative this may not happen.
"Sales is" < Your comparison criteria exhibit math flaw; if you cant understand why you cant calculate division by zero then seek someone help. The above wiki link seem to be sufficient learning aid.
" Most mature products are still "closed,"[citation needed] and thus no explicit mentioning of "closed source" is needed." < When url1 publicized exploit was patched? in Feb 2 or May 12 url2. See also how many matured futures reveal just [url3 one person] in OracleDB in one year.
"It is clear from the sentence which is closed and which is open without explicit "closed source" in the sentence" < not relay if you know it you don't need it but the article is silent. See [url4]["open software" on] ("database">300) oracle open source and [url5 IBM] open source search result lean into believing they have open source. Also You just reacted on it before - sign it were not only obvious but also probably not true. (talk) 08:08, 24 October 2011 (UTC)

The article tile is about database. You shape it in direction of merge to DBMS. What about adding 1942 definition: "databse is colection of documents with one or more futures in common"[url7]. You also put database when should be "database binary file/device " thus narrowing the art summary to computer application of which commercial aspects seem to be major criterion. (talk) 08:08, 24 October 2011 (UTC) ps. Your spam filter do not allow publish related 7 urls so you have to gues them. url0=[ this] url1=[] url2=[] url3=[ url4=[,or.r_gc.r_pw.,cf.osb&fp=dc20b2672385f007&biw=1400&bih=794 url5=[,or.r_gc.r_pw.,cf.osb&fp=dc20b2672385f007&biw=1400&bih=794&pf=p&pdl=3000] url7=[]

So what do you want? What do you suggest here? --Comps (talk) 15:41, 24 October 2011 (UTC)
The suggestions are exactly put in the edits. Don't you read before you change the article , or you just play 'idono'? (talk) —Preceding undated comment added 17:40, 24 October 2011 (UTC).
Re 'idono': It is a fact that you have not done a single edit properly. Both in content and in shape. Most of them were irrelevant and unnecessary. Unneeded cleanup work after each of them. --Comps (talk) 20:28, 24 October 2011 (UTC)

User has knowledge and good ideas. I suggest you propose major changes via discussion, to avoid many article unnecessary edits and reverts, and for best results; not always we are aware of all nuances. --Comps (talk) 11:03, 26 October 2011 (UTC)

Your improper edits to Database

  • This section was removed from User talk: by and put here. Please do not remove from your talk page. Thanks --Comps (talk) 13:45, 31 October 2011 (UTC)

You have recently edited Database multiple times, and most edits have been improper (for many reasons; see History and discussion there) and reverted. Pls propose edits in the discussion page, without destroying carefully written text. --Comps (talk) 18:27, 27 October 2011 (UTC)

What are the other reasons ? Do not let guess ! (talk) 13:09, 28 October 2011 (UTC)
Go over History. You have no respect to others work. --Comps (talk) 14:37, 28 October 2011 (UTC)

List of popular DBMSs in lead section - How to generate it?

1. Leave it as originally, ranked by sales for the mature products, and by licenses for open source, or change to alphabetical order (as proposed by Alphabetical is completely arbitrary, and will change the lists completely (products with A at the beginning...).

2. Add the term closed source (link redirects to Proprietary software) explicitly (as proposed by It is well known that all these mature prods are proprietary.

Please comment.

--Comps (talk) 15:33, 24 October 2011 (UTC)

How you get this "(ranked by license)" ? (talk) 17:37, 24 October 2011 (UTC)
I think it should be left as is, being ranked by sales. After all that determins the true popularity of a product.Millertime246 (talk) 20:36, 24 October 2011 (UTC)

Comps how do you see marking your's own text, by yourself as questionable, by flagging {fact} to it. Do you regard it as smart trick or otherwise ? (talk) 10:13, 27 October 2011 (UTC)

Fact has been correct for some time. The tag is a reminder to provide a reference, by me or others. Have you thought of such possibility at all? No? How come? Smart trick? You are a real pain. You have again unnecessarily and awkwardly edited the article after the essence of your contribution (+ needed addition) has been incorporated by me. In the time I wasted on your nonsense and correcting your improper edits I could expand 10 sections in the article. You are worse than many wiki vandals who can be corrected in a single revert. Please stop this, even if you have a lot of fun, and help in needed expansion (if you can?), or via discussion, without smashing carefully-written-texts in the article. You bring more damage than help. --Comps (talk) 18:15, 27 October 2011 (UTC)
In thew future to draw incorrect articles use scratch box not the main article. In scratch box you can put as many as you like memory help devices including your vision as above. (talk) 15:02, 29 October 2011 (UTC)
There's a difference between being incorrect and being incomplete. Comps has being doing excellent work on this article, which is a thankless task given (a) the amount of random vandalism, and (b) the opinionated single-issue distortions made by people like yourself. Please show some sense of perspective. The article is about databases as a concept, and the names of specific products are there only as examples to help peoples understanding. It doesn't matter what order they are in, and it doesn't matter what the licensing terms for different products are. Oh, and I find it very hard to take seriously anyone who chooses to remain anonymous. Mhkay (talk) 18:15, 29 October 2011 (UTC)
From a point of view quite excellent, one represent the sentence : "Two developments of note are the Object database and the XML database. The vendors of relational databases have fought off competition from these newer models by extending the capabilities of their own products to support a wider variety of data types."
99% of the article is about only one kind of database the old kind of structured tableDB. The all different kind of databases are cornered in one above sentence ("cited" from art) Nothing more abut object database. However the word *object* is mentioned ~30x (as a semantic distraction?) All the art.database is kind of 'vendor fought' to advertise only one of databases: relational database; thus the article represent narrower point of view. (and should be marked POV) The commercial ranking as knowledge criterion make this POV even more ugly. []
Some (little) text is an untouched leftover from before I started on May 6 or so. This is one of the old pieces. Here you may be able to shine, if you can do better than before... I do not mean here that text I have written cannot be touched. But any change needs to be an improvement, not change to something incorrect or inaccurate, or with a language problem, or just a change with no logical reason.
"99%" is completely incorrect. Tables are only mentioned in association with relational. All the rest is general, and I have no clue where you take this from. I hope you do not intend to change 99% of the article. I'm glad to get criticism on the article, but it should make sense.
I'm not sure what you refer to with "object". "Database object" is a generic term used in the academic literature for defined unit of data, with (beyond bits) or without internal structure. Have you counted also such "objects"? --Comps (talk) 08:29, 2 November 2011 (UTC)

You added ref but there is no link or page number. No author or title to search. is this Sheridan in VOL. 163, NO. 5 - April 11, 2011? Could user Comps put proper citation for sake of [WP:WER] ? (talk) —Preceding undated comment added 05:06, 2 November 2011 (UTC).

Indeed needed for better pinpointing in the Fortune booklet. Could user add the missing info since has already checked, instead of asking me or anybody else to do it? --Comps (talk) 08:29, 2 November 2011 (UTC)
(197.87.."has already checked")< your ~"mind reading/guessing" is erroneous.
Above response is not an answer for a question or confirmation of fulfilling a WP:WER task. Again what is exactly the source (if any) you used for the data? (Internet search not found) (talk) 17:29, 2 November 2011 (UTC)
You mention an author name and a specific Fortune issue ("Sheridan in VOL. 163, NO. 5 - April 11, 2011?"). I guess you got it in a dream or some vision and thus not a reliable source. The ref is accurate, and should be augmented for completion. Still not everything is on the Web for free: usually you need to subscribe to be able to to dig in a publication's history, or visit a library that carries it. -- Comps (talk) 15:41, 3 November 2011 (UTC)
Please try understand: the mark '?' commonly mean a question. There was a question. Already a week elapsed, you posted ~7 reply, on the subject, and none solved the problem. You want the to fill the bio, data for you, but how one may know what kind of source you had been reading? Why to flip like fish on shore. Again: what were the page, title and authors of the ref you put too art ? (talk) 10:47, 5 November 2011 (UTC)

The 'bag' of examples

The section "Database type example" do not seem to have any order. Is any goal to list this unordered set of examples? There are examples distinguished by structure interlaced with differenced by utility and mixed with divisions by intended sector of marketing. As first step (to clean this out) the examples has to be moved to end of art. (talk) 05:23, 2 November 2011 (UTC)

Really? The order is alphabetic, since no other better logical order seems to exist: The types span many independent dimensions. Alphabetic is an order that helps a reader to find in a list a given term (if looking for a specific database type). The location in the article's beginning provides a good introduction to broaden the reader's view about databases, and later see (in the rest of the article) the common-denominator that puts all these different dimensions in one framework. The list is quite exhaustive, of the major database types in the literature. Few (from article's older versions) may be dubious, but add more dimensions. --Comps (talk) 08:55, 2 November 2011 (UTC)
Are you really sure for you "no other better logical order seems to exist" ? (talk) —Preceding undated comment added 15:27, 2 November 2011 (UTC).
I wrote "no other better logical order seems to exist". If you have a suggestion, pls write it in the Suggestions section below to be evaluated by readers. But no brute-force change in the article itself, pls. --Comps (talk) 15:54, 3 November 2011 (UTC)
What "brute force" you talking about? The =example section= was not touched by If it was otherwise you can easily prove it by dif. You just overloading for long the art by redundant content copied from other arts. (talk) 22:56, 4 November 2011 (UTC)
I have asked here to avoid this in the future, which you have done multiple times in the past in the lead section. Re "You just overloading for long the art by redundant content copied from other arts" I guess you mean that some text is taken from main articles(very little; it is mostly original, or written by me in the main articles; typically edited for shortening when needed). I see nothing wrong with having some minimal text in a section that has main articles. This is done for article completeness, and is a common practice all over Wikipedia. Zillions of sections exist with main articles, and with some content overlap. Or you prefer that the Database article is just paragraph names with main article links. Does not make sense to me. --Comps (talk) 23:50, 4 November 2011 (UTC)

Really 2

Comps say "Ignorance; even no-data database is too complex due to specialized default data-structures; even harder if proprietary"

Above sounding like oracle response is obviously not true. Simple databases do not need any DBMS and can be facilitated as flat file. So lets put his word "specialized database". (talk) 19:06, 2 November 2011 (UTC)
Ignorance is accurate. A database cannot be a single flat-file, and it seems that you have not understood the basic message of the entire Database article (not even the lead section).
"A database cannot be a single flat-file"! == Flat file database (talk) 04:09, 3 November 2011 (UTC)
This is an anachronistic oxymoron. The moment it becomes a database then by the Database article's definition it cannot be a single flat file anymore: Quite a lot of additional DBMS data with structure is needed to support needed functionality, which is inefficient in a flat file (-- Comps (talk) 16:00, 3 November 2011 (UTC)). Some toys call themselves "database". None of the mentioned DBMS in the article is such; also no place for such in the DB type example list. --Comps (talk) 04:39, 3 November 2011 (UTC)
"by the Database article's definition" < The explanation do not have logic cohesion. Since the article in question is used as argument for thesis. Circular fallacy. The to be proved (content of article) is forged as prove. (talk) 17:31, 4 November 2011 (UTC)
You continue with your unacceptable practices of massive article changes that are hard to track and with no sufficient explanation, and I'll ask the Wikipedia staff to stop you, if you do not stop to sabotage the article and comply with orderly work. --Comps (talk) 20:19, 2 November 2011 (UTC), Pls discuss your suggestions in the special section below, before you destroy carefully written text in the article. --Comps (talk) 20:38, 2 November 2011 (UTC)
Databases can be flat files. My opinion is that the definition should be as broad as possible in order to avoid things like this.Jasper Deng (talk) 23:51, 4 November 2011 (UTC)
Broad is usually good, but too broad is dangerous and can include undesirable objects, like flat-file-database here. This is the main point of the article: It starts with a broad definition that can includes almost any data collection, and than narrows the def by requiring a database to meet a set of requirements, over which a broad consensus exists. This distinguishes databases from any other data collections also of toy prods that call themselves database. The term database has been used too lightly, which misses its point. The article makes this very clear. Pls see in lead section, and then The database concept in History. --Comps (talk) 01:00, 5 November 2011 (UTC)
I mean, we cannot be too specific here. Databasing is not something in CS that can be very firmly defined.Jasper Deng (talk) 04:45, 5 November 2011 (UTC)
Agree with Jasper Deng.
re:Comps - : "is dangerous can includes almost any data collection" AND (what Comps omit) "organized for one or more purposes". < (it was a kind of distraction by fragmentation fallacy). Database is not a random signal stream, unorganized data acquisition storage or stream (but organized collection of acquisited data is a database [u2][u3]). The "usually in digital form" phrase is also problematic distraction. Database originally were and still can can be on plain paper or on media storage which physical nature is analog... We do not going here to debate philosophical issues (whats analog:digital over QCD down to Planck dim.) and the more cohesive wording were proposed here < were it wrong -if so what-and-why? (talk) (sum with links will be 33965)
The current def is very loose and well reflects what a database is. It is defined by requirements, while every req is very flexible by itself, and can be interpreted and implemented over a wide range. No need to loosen further, to my opinion, without the danger of defining all kinds of strange UFO as DBMSs and their data collections as databases. If not all reqs are met to some degree, it is not a database for me. For example, a sophisticated excellent file sys is not a DBMS and its file (not necessarily flat) collection is not a database. A sophisticated email sys is not a DBMS, and its email files are not a database. I'm not talking here about CS defs (I'll be surprised if you find this one in a DB textbook; I know quite many of them). This is the practical essence of a DBMS and a database, and the current major message of the article. I wonder what an alternative def you have in mind, such that does not generate absurd databases. Without a concrete alternative def all this discussion is philosophical and hypothetical. When you have such (if you find one that look to you reasonable), it can be moved to the Suggestions section. --Comps (talk) 10:04, 5 November 2011 (UTC)
re: strange UFO as DBMSs < you have tendency to dispute yourself created imaginary objects, and tendency to avoid answers on the questions.
Lets do simpler try. Comps look at the title of the art. What do you see:
*A Database ?
The impression is user Comps do see DBMS instead of Database. (why?) $This is not the same entity. Most lame prove/explanation: there is one art for database another for DBMS. In fact the art database is now overladed by redundant content and should be substantially thinned. Most of DBMS stuff should go to DBMS art. 22:05, 5 November 2011 (UTC) — Preceding unsigned comment added by (talk)

Special section for suggestions of User:

Do not destroy carefully written text in the article with no discussion. You may suggest specific replacement text. Please wait with changes until a decision is reached here by majority. Please list all your suggestions for change in this section below. -- Comps (talk) 20:41, 2 November 2011 (UTC)

What you have done so far, massive changes without discussion or without discussion-resolution is not vandalism by the Wikipedia definition, but from my perspective more devastating. Pls refrain from doing it. --Comps (talk) 06:39, 3 November 2011 (UTC)

Also, when you get to a change, pls do every change to a single text piece (sentence, or a group of) in a separate edit, with clear explanation, so the logic for change is understood. This is not harder than one massive change (of several unconnected parts), and makes it easier to justify and others to track. --Comps (talk) 15:15, 4 November 2011 (UTC)

The article may help in education. Do user Comps have special privilege to being personally tutored over talk pages, "so the logic for change is understood" by user Comps? If not why need explanation of edits as they are. Over the discussion are numerous factual problems reinserted by user Comps and pinpointed out around. User comps did you put single edit of in question whilst wholesale discredit it. The diff utility is sufficient to study the edits of and will be redundant (and highly inefficient) to present it here for user Comps to read it. If one put particular edit of and point to ones concern about it then will explain the reasons. Otherwise not, and it is perceived ans normal editing. (talk) 17:34, 4 November 2011 (UTC)
Pls keep in proportion. I'm trying to protect the article from incorrect, inaccurate, and superfluous info. I took it on myself at this point, and thus to do this I would like to understand any change done, and make my opinion on it clear, if I oppose. The only way is by discussion, preferably before the change. If you put new text (pls expand where needed), I'll do the same with your text. If just language, and no fact change, pls go ahead, if can be improved.
"User comps did you put single edit of in question whilst wholesale discredit it": I reinserted the following elements of yours:
proprietary (for your closed source)
Market share (rather sales; you wanted explicit reason for DBMS example order) and ref (incomplete yet, with request for completion; to be done by me or other; pls stop repeatedly nudging; also today)
copies -> instances
PostgreSQL (for your Postgres) and SQLite added to DBMS list.
"many millions" for your "hundreds of millions" (with no ref); I added Adaption in SQLite to support it.
I may have forgotten something. All nice but secondary in the article, and the article could live without, to my opinion. I hated to waste on it a week-work or so, rather than adding more urgent material (to my opinion) to the article. Pls help with such rather than unnecessarily digging in and substantially modifying existing text (what I have seen so far from you; my own opinion) --Comps (talk) 16:42, 5 November 2011 (UTC)

People with understanding in the subject are requested to comment here to guide the article in the right and best direction. Please also add any new suggestion (with a new number). --Comps (talk) 06:25, 3 November 2011 (UTC)

Suggestion 1: Change "SQLite supports many millions of instances" -> "SQLite supports hundreds millions of instances" (phrased by Comps)

No, since no direct reference is given for "hundreds millions". The Adaption section reference that I have added there hints for millions, but "hundreds of millions" requires some counting, and I'm not sure is true. It may be true, but requires some research, or a reliable reference. Not obvious from Adaption. Many millions is sufficiently good until reference is given, even if hundreds of millions is true. --Comps (talk) 04:27, 3 November 2011 (UTC)
Sqlite in Firefox alone account for 400 millions. (talk) 17:34, 4 November 2011 (UTC)
First, we need to define an instance.Jasper Deng (talk) 21:42, 4 November 2011 (UTC)
In general it seem to be solved some 2500 years ago by Plato. For art subject scope: A single copy of a running program. The Firefox example (hundreds of millions) is thus lowest bond to account for widespread, in range ~x100*10**6, usage of Sqlite. (talk) 23:11, 4 November 2011 (UTC)
& for sake of lame idono ref 182 should EOT. — Preceding unsigned comment added by (talk) 23:22, 4 November 2011 (UTC)
I like your "instances" (the right term; originally "copies"), and I myself returned it to article after was lost in a revert.
Re Firefox, you can add: "... e.g., inside over 400 million Firefox instances[ref 182] and more popular products" if you are sure it is now in most existing versions. --Comps (talk) 00:38, 5 November 2011 (UTC)

2: Lead section cannot have examples. (phrased by Comps)

I disagree: Lead section can use any reasonable means to make it well understood, as long as it meets Wikipedia guidelines. This is especially true for Database, which is an introductory and entry article to the database subject. --Comps (talk) 04:27, 3 November 2011 (UTC)

3: Just read slowly. Don't claim you can't :"Massive changes that are hard to track by the Wiki diff." (talk) 03:56, 3 November 2011 (UTC) < this entry arrangement was modified by Comps who claim "others" changing somehow it expression (bad&regrettable). Original form here [You are really something: I just put it under a number to have some order in the section. "bad&regrettable" - out of proportion and time wasting to deal with your whining about nonsense. --Comps (talk) 00:09, 5 November 2011 (UTC)]

I have read slowly. You cannot remove or massively change text without discussion. You cannot do it for matters that have not yet been resolved in discussion. --Comps (talk) 04:27, 3 November 2011 (UTC)
So show by dif an edit you consider "massively change text without discussion". It may be good point to reasonable argumentation. (talk) 21:30, 5 November 2011 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── Comps, you really need to control your desire to arbitrate the development of this article. That includes issuing decrees to other editors as to how to go about their business. Additionally, it would be greatly advantageous if you would try to follow the convention of placing new material on the talk page in chronological order, rather than randomly intersecting it with old material. Chris Cunningham (user:thumperward) - talk 15:25, 4 November 2011 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── Chris, I'm definitely not trying to control the process or a person. I'm trying to protect the article. My "decrees" to a specific editor (not editors) is an attempt to generate constructive interaction. I do not mind a complete change, if it is a change with correct facts, without a destruction of facts I consider important. In this case I want a discussion about the change. What we have seen here is an unusual case (which I have never encountered in my 5.5 years here) of clear destruction of important info, and the insertion of incorrect, and inaccurate material with misconceptions, without discussion, and even without comments and explanation. His potentially useful material I have incorporated myself in the article, with expansion where needed. I started in this article, initially a very bad article, on May 6, and made it a pretty good one, to my professional opinion (though a lot of work is still needed). I believe it can give a good, detailed idea about databases to a large audience. I have been asking for comments and help along the process, and got some from one person, with some encouragement. I have interpreted silence as agreement with what I do. Regarding inserting random pieces, most of the article balk is my writing and rewriting and rewriting (with iterative improvements), and this is the way I have worked since 2006 on a group of articles, unaware of any recommended process, and with no any objection from anybody. Unfortunately, this area lacks people like the many you may find on other articles, e.g., in Physics and Math, who have constructive interaction and produce excellent articles. --Comps (talk) 16:28, 4 November 2011 (UTC)

re: "His potentially useful material I have incorporated myself in the article," <<<<<<<< Are you aware so screeching one contribution and coping it under your credentials may exhibit symptoms of plagiarism? (talk) 09:56, 5 November 2011 (UTC)
99.90, this is not a productive line of discussion. Any contributions to Wikipedia are implicitly licensed for reuse and reformatting of this type, and the page history records all of the specifics regarding who originally wrote what. Chris Cunningham (user:thumperward) - talk 13:00, 5 November 2011 (UTC)
Agree when the text is modified over as all usually edit. Here lay particular case. Reverting an edit and coping the reverted content as own is somehow dishonest way of coediting. Do revert mean the edit was 'good' or 'not good'? So editor who revert text expressing: the edit was "not valuable", by coping the words or idea express otherwise: "it was good". This perceived conflicting expressions exhibit "lack of integrity", which lay in core of definition of dishonesty. If plagiarism is a dishonesty in context of editing and if pointed lack of integrity show dishonesty conjugating A and B should be easy. This alone will be sufficient to express regret, but on the top of such behavior such editor rant ethnic (enjoyably dosed) slur unsupported by any facts. Instead of continue this somehow tee rolling thread can some one pressure user Comps to provide the data in his < ref> he refusing to reveal. (see above) (talk) 21:30, 5 November 2011 (UTC)
Plagiarism and dishonesty: I had enough of you and your libel. As usual you are out of proportion, delusional and abusive. I used single words (meaning that you plagiarized the dictionary...) or changed texts only. I always wrote "returned" in the ed comment, to credit your idea, not copied any text, so stop talking about plagiarism and dishonesty. Shame on you! Your attitude and baseless accusations are unacceptable, and I wonder until when you are allowed in this forum. The following summary of your contributions inserted by me is copied from this section above, in case you have not seen or cannot find it:
"User comps did you put single edit of in question whilst wholesale discredit it": I [Comps] reinserted the following elements of yours:
proprietary (for your closed source)
Market share (rather sales; you wanted explicit reason for DBMS example order) and ref (incomplete yet, with request for completion; to be done by me or other; pls stop repeatedly nudging; also today) [This is Comps' original text to support the list order; probably did not like it, to conclude the open-source market share]
copies -> instances
PostgreSQL (for your Postgres) and SQLite added to DBMS list. [Actually I'm not sure if I ever removed it; can be checked]
"many millions" for your "hundreds of millions" (with no ref); I added Adaption in SQLite to support it.
I may have forgotten something. All nice but secondary in the article, and the article could live without, to my opinion. I hated to waste on it a week-work or so, rather than adding more urgent material (to my opinion) to the article. Pls help with such rather than unnecessarily digging in and substantially modifying existing text (what I have seen so far from you; my own opinion) --Comps (talk) 16:42, 5 November 2011 (UTC)
(End quote)
I did most of these changes reluctantly just to accommodate you. I never thought it was important to the article, beyond anecdotal, as you see above, and now I'm "dishonest", stealing your "great ideas".
You already have accused me and an administrator of favoring MS, Oracle, and big corporations. You started your participation here with your "open source" agenda, trying to put open source DBMSs at the top of the DBMS example list. Shame on you! I do not see any reason to continue tolerate all your crap here. I strongly ask administrators to take measures to stop User: from continuing his libel, abuse and other unacceptable behavior. --Comps (talk) 01:49, 6 November 2011 (UTC)
I think both sides here need a shot of WP:AGF and a large bout of civility. Both sides can be blocked if necessary.Jasper Deng (talk) 02:20, 6 November 2011 (UTC)
There is a whole difference between marginal "incivility" and Libel. Especially that for the first I have extended my patience above and beyond with good will, subject to repeated provocations, mockery and unbased innuendos, have my requests to discuss and summarize changes (in ed summary) ignored by the other party for long time, and wasted unprecedented amount of time (you know all the details). For the first you may give a slap on the hand, but the second should be completely forbidden and stopped. --Comps (talk) 03:28, 6 November 2011 (UTC)
Two wrongs don't make a right. Libel is a legal term, FYI, which sort of makes Wikipedia nervous at times. I think patience is best.Jasper Deng (talk) 03:32, 6 November 2011 (UTC)
I have made no legal threat. But I have pointed to the factual serious situation, and I want awareness of it. Any possible repetition of such occurrence should be prevented. --Comps (talk) 04:10, 6 November 2011 (UTC)
I knew you didn't. My point was that such a strong term is not going to help this situation.Jasper Deng (talk) 04:13, 6 November 2011 (UTC)
It is a serious situation, black on white, and should be taken seriously. Not a kids' game anymore. You cannot leave a person out of control. --Comps (talk) 05:03, 6 November 2011 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── Nobody is "out of control" here. The two of you need to calm down and stop making personal attacks on each other. In particular, any future attacks which may be perceived as legal threats will result in indefinite blocks, per our policy on such things, as it is impossible for editors to work collaboratively when the threat of potential legal action hangs over them. Chris Cunningham (user:thumperward) - talk 11:28, 6 November 2011 (UTC)

Well, User: has just recently repeated a false accusation in User talk: (bottom). --Comps (talk) 09:43, 8 November 2011 (UTC)

Rm SELFREF, please?

Please remove the part of the Overview section that reads "Most (brief) sections are backed-up by Wikipedia main articles linked in the sections, which provide more thorough descriptions of the respective subjects. These articles may point to additional Wikipedia articles for further refinement of the subjects and extensive coverage of the database and related areas. Thus this article should be viewed as an introduction and entry to the wide and diverse database area, and also may be viewed as a high-level index for it. For this reason not every fact mentioned is directly referenced here, but rather relies on references in the linked respective detailed articles.", per WP:SELFREF. Thanks.--SarekOfVulcan (talk) 17:25, 3 November 2011 (UTC)

Will it be satisfactory if the two explicit "Wikipedia" occurrences are removed? If "Wikipedia" is removed nothing is left with self-ref, so why also the rest? I do not quite see a problem in generically mentioning main articles and articles in an article that is an introductory one. -- Comps (talk) 21:15, 3 November 2011 (UTC)
 Done. I've removed the entire "overview" section as a blatant self-reference. Articles do not need wordy content overviews; the table of contents suffices to explain what the article covers, and the use of a summary style is implied by the addition of per-section hatnotes. Chris Cunningham (user:thumperward) - talk 10:21, 4 November 2011 (UTC)