Template talk:Databases

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Computing (Rated Template-class)
WikiProject icon This template is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
 Template  This template does not require a rating on the project's quality scale.
 

This template needs some help.[edit]

It links to List of relational database management systems twice. I tried to fix that, but my change was reverted without explanation. Can anyone explain why? Indeed, the article has been added to lots of topics; it would've been a good idea to review the template before publishing it so far and wide.

Here are some other problems; I'd fix them, but the reversion of my previous edit and the large "do not change" comment are a little to intimidating:

  • There's no seperator between "Relational Model" and "Relational Algebra" in the "Concepts" section.
  • The template links to "DB2", which is a disambiguation page.
  • The link to "ODBC" redirects.
  • "Referential integrity" is missing.
  • "Relational database management system" and "Distributed database management system" seem like they belong under "Types of implementations" instead of "Concepts".
  • Seems odd to abbreviate "DBMS" when "database management systems" is spelled-out elsehwere.
  • A couple of links in the "SQL Topics" area go to articles whih are tiny; one of them (UNION) is up for merging.
  • The "Components" section seems a little inconsistent. "ODBC" isn't a database component. Some of the things listed are objects in a database's relational space, while others are features of its implementation.

Let me know how this project is proceeding. --Mikeblas 02:44, 1 June 2006 (UTC)

"Surrogate key" is missin', too. -- Mikeblas 02:45, 1 June 2006 (UTC)

Components and objects split[edit]

With no objections, I've split "Components" to "Objects", and put "objects" under the implementation line. "Referential Integrity" and "Surrogate Key" still aren't in the template.--Mikeblas 21:06, 5 June 2006 (UTC)

no objections, except maybe for "JDBC" as its a type of ODBC and relates more on the java side. --larsinio (poke)(prod) 21:11, 5 June 2006 (UTC)

More Missing[edit]

I've added "partition". How should "Materialized view" be added? With a dash, after "View"? -- Mikeblas 14:11, 6 June 2006 (UTC)

Should logical schema be linked in the "Concepts" list"? -- Mikeblas 14:33, 6 June 2006 (UTC)
what about result set? -- Mikeblas 19:55, 6 June 2006 (UTC)

Is the Database Product listed includes everything.Because I found Cognos Database also exits.

Adding Comparison of SQL syntax[edit]

I want to add it after all other comparisons. Any objections? --GreyCat

Seems like a good idea to me. Might wanna fix the formatting, like
Comparisons: relational | object-relational | SQL Syntax
instead of just adding another pipe.
--Mikeblas 20:13, 8 June 2006 (UTC)
I see that Larsinio has made another change without discussing it here. I guess that big comment in the source isn't too effective! What I was thinking of is shown at Template:Databases/Comparisons. -- Mikeblas 00:52, 9 June 2006 (UTC)
The reason i made the change the way I did was to keep the SQL related items together. --larsinio (poke)(prod) 13:35, 9 June 2006 (UTC)
Why not keep the implementation-level items together? -- Mikeblas 17:53, 9 June 2006 (UTC)
Fair enough. Make the change its not a big deal. In the future try not to create subpages with the templates. as there will eventually need to be a WP:TFD for the subpages . --larsinio (poke)(prod) 18:41, 9 June 2006 (UTC)

What databases to include in the template[edit]

The current list of database software implementations is: Caché | db4o | dBASE | Firebird | Helix database | DB2 | Informix | Ingres | InterBase | Microsoft SQL Server | MySQL | OpenLink Virtuoso | Oracle | PostgreSQL | SQLite | Sybase IQ | Sybase | Teradata | Visual FoxPro

Are all of these notable enought to be listed? Cache, db4o, Helix, and Virtuoso seem to have too little market or mindshare to be worthwhile including in a global template. The template shouldn't try to be an exhaustive list of all DB programs.

Comments? Georgewilliamherbert 20:56, 24 November 2006 (UTC)

I believe that MaxDB should be included. This is one of the few SAP compliant database and one of the very few open source enterprise database. True, that the current version of the article sounds like advertisement. Maybe that's why it is not included? I can try to find the time to improve this article. Mbaudier 14:42, 28 February 2007 (UTC)

Suggested redesign[edit]

I've redesigned a copy of this template on my scratch page, in a way I believe improves the way the products are listed. Discussion is welcome, of course, this page is on my watchlist. Feel free to use it, or improve upon it further. :) -- Limn 17:56, 12 January 2007 (UTC)

         Looks good to me Will henderson 18:06, 30 January 2007 (UTC)
Looks good to me - I suggest we use it! Alex Jackl 08:21, 31 January 2007 (UTC)
I agree, the new one does look better. I suggest we use it as well! Cgsguy2 16:05, 31 January 2007 (UTC)
I do agree as well. It's much clearer. I would also add MaxDb (see my previous comment above) Mbaudier 12:57, 12 March 2007 (UTC)
I suggest we go ahead with implementing Limns redesign of the template. Unless anyone objects, I think this should be done asap. As far as I can see, there are some DBs missing from the list at Limns page, which are in the list in the current template, namely: Paradox | eXtremeDB | Filemaker Pro | Greenplum | H2 | Microsoft Access | Mimer SQL | MonetDB | Oracle Rdb | Perst | UniVerse. Also, the Candidate key article is not in Limns table. These should maybe be added in? Jerazol 08:41, 13 March 2007 (UTC)

List of Database products[edit]

It seems this list is continually getting added to, with more or less notable products. Should there maybe be some sort of criteria for what is included in this list? For instance the link to Alpha Five, which is basically a sales pitch for that particular product. Jerazol 07:36, 8 April 2007 (UTC)

What kind of criteria do you have in mind? What you see so far on Alpha Five was not intended to be a sales pitch, it's what I dug up for two editors that were repeatedly marking the article for deletion due to lack of nobility (which they no longer dispute). PeetMoss 15:19, 8 April 2007 (UTC)

PeetMoss: Nobody was "marking the article for deletion" — the {{notability}} tag is not a deletion tag, but a cleanup notice. It would help if you actually read what it says before reverting repeatedly.
As for a criteria for the list, I agree that we need one; the navigation box is too big already as it is. But I can't see any obvious places where to draw the line. -- intgr 17:09, 8 April 2007 (UTC)
I have a suggestion that can help make Wikipedia a more inviting community Intgr: don't bother editing articles unless you're willing to go a little further then a game of tit-for-tat when confronted or when your actions are questioned. The obvious is not always so obvious to others. You may also want to note: "if you actually read it" is a statement that suggests a high degree of arrogance on the part of the party making the remark. PeetMoss 17:45, 8 April 2007 (UTC)
Considering the massive influx of vanispamcruftisement articles, sometimes even drafted by marketing departments (I've been directly contacted by a few, Wikimedia Foundation is dealing with them daily[1]), I personally think Wikipedians aren't being ruthless enough when it comes to articles about companies and products. I seriously cannot see why would an encyclopedia need to cover hundreds of different database products, as opposed to for example, generic academic articles about databases; but let's drop that argument, it's not leading us anywhere. -- intgr 18:10, 8 April 2007 (UTC)
I find the company/product articles extremely useful when it comes to learning about the development and evolution of databases, whether they are notable or not. PeetMoss 19:21, 8 April 2007 (UTC)
Sounds like you would be looking for something in the lines of "history of database managements systems", rather than "Alpha Five" then. It is fine to mention non-notable products in such an article if they are indeed relevant to the history of databases. -- intgr 19:58, 8 April 2007 (UTC)
I think a good criteria would be to only include links to articles that aren't stubs. I'll work on this today ... -- Mikeblas 18:04, 8 April 2007 (UTC)
Yeah, that's a good start. -- intgr 18:10, 8 April 2007 (UTC)
I agree. In addition, the article to be included, should maybe have reached a certain maturity with contributions by a number of editors. EXtremeDB is an example of an article which I think should maybe be excluded on the basis that it's so far only been edited by a sole editor. Jerazol 19:18, 8 April 2007 (UTC)
Thanks for chipping in, Jerazol. I've gone through and removed stubs, including tagging a few. I think I've also got to the point where the template is included on all the articles that are actually in the template, which wasn't the case before. -- Mikeblas 21:14, 8 April 2007 (UTC)
I disagree. Rather then pick and choose which one's are in this template I think it would be better to simply include a link to an article listing all of them, stubs or not so people will continue to expand on the articles. PeetMoss 19:21, 8 April 2007 (UTC)
Unfortunately, such an approach doesn't scale. There are lots of DBMSes, and listing them all would make the template intrusive. I don't see inclusion or exclusion from this template as a significant motivator for the improvement of articles. Perhaps a better approach would be to remove the list altogether, and let the template focus on pure database management topics without vendor or product involvement. After all, there is already Category:Database management systems and List of relational database management systems to collect products; perhaps the template should link to those, instead. -- Mikeblas 20:01, 8 April 2007 (UTC)
You misunderstand. I was suggesting the same thing. I don't understand the point of providing links to particular DBMSes in this template. Because it doesn't say otherwise I assumed I was looking at the complete list of database products with articles or stubs that I could contribute to. The following link is all that is needed in this template: List of relational database management systems. PeetMoss 14:58, 12 April 2007 (UTC)
So, how about something like this: User:Jerazol/tmpldb. This way, we reduce the size of the template as well as solving the problem with the ever expanding productlist. Jerazol 16:16, 12 April 2007 (UTC)
Yes, just like that but add: "List of relational database management systems". PeetMoss 20:19, 12 April 2007 (UTC)
That is the "Relational" link at the bottom, right next to the Object-oriented link.Jerazol 20:25, 12 April 2007 (UTC)
I'm for getting rid of the product lists. The proposed template is still very large, though, because it's full of whitespace. -- Mikeblas 18:10, 12 April 2007 (UTC)
Well, let's see how this works out, and if necessary see what we can do to further reduce the size.Jerazol 19:46, 18 April 2007 (UTC)
The criteria recently used to purge certain database products from the template list seems quite arbitrary to me. So, you want to cut an article if only one editor has contributed to it? If another editor chimes in on one of these pages, they're back on the template?
How about criteria based on the DBMS's level of innovation (that is, it offers something new rather than being an also-ran in an over-populated category)? Or the size of its user community, or its growth/adoption rate? You may respond that these criteria would be difficult to apply. That points to the fact that this list should strive to be inclusive -- it calls itself a list of databases not a list of the biggest and oldest databases or databases whose Wikipedia articles are authored by more than one editor.
Really, I think if such a list is going to exist, the Wikipedia user community is served better by a list that illustrates the breadth and depth of this technology category.
And, by the way, I am in favor of the list continuing to exist. It is a way for interested parties to discover "what else is out there" when they are investigating a technology. That is a major benefit such a template can provide. What's the harm of a broader list, that it will consume a few more square centimeters on the PC screen? It's at the very bottom of the page, after all -- that's hardly intrusive.Ted nw 21:32, 17 April 2007 (UTC)
The problem is, such a list will eventually become a beast. A template like this shouldn't form the dominant part of a page. Besides, the list is still easily accessible, through the links to "list of (rdbms|ordbms)" in the same section.Jerazol 07:50, 18 April 2007 (UTC)
You write, "the list is still easily accessible" -- that assumes the list will be moved off the template, correct? Is this going to happen?24.19.89.253 Ted nw 18:14, 18 April 2007 (UTC)
Yes, through the links to list of OO/Relational databases. Jerazol 19:46, 18 April 2007 (UTC)

ACID[edit]

Isn't it odd that this template links to "ACID" but not to "Transaction"? I think there might be an issue about inclusion of concepts, too, aside from the product list. -- Mikeblas 18:18, 8 April 2007 (UTC)

It does link to Transactions, but in the Objects list. I'm not sure who decided this distinction, and I'm not sure why a transaction is considered an object. Jerazol 13:01, 9 April 2007 (UTC)
Oh! There it is! -- Mikeblas 13:08, 9 April 2007 (UTC)

Smaller Template[edit]

I think there's already some concensus that the existing template is too large. I've put together a smaller version at User:Mikeblas/databases which doesn't use quite as much space. Is there any objection to switching to that layout? -- Mikeblas 12:26, 9 April 2007 (UTC)

I'm sorry, but I don't see it justified to switch to a layout which is, imo, a lot less readable than what we have now, in order to save 140px of screen space. I think what we've got now is fine, as long as we're able to prevent the database products section from expanding beyond control. Jerazol 12:50, 9 April 2007 (UTC)
140px is a linear measure; shouldn't you should be thinking about area? -- Mikeblas 12:51, 9 April 2007 (UTC)
Let's not get to hung up on my highly inaccurate attempt at illustrating why I don't think we should use your new layout. The point is, I don't agree that there is a problem that needs fixing. Jerazol 12:57, 9 April 2007 (UTC)
Hung up? I'm just trying to understand your opinion. -- Mikeblas 13:08, 9 April 2007 (UTC)
Okey, my opinion is that we should not switch the layout, ie. I object to switching the layout.Jerazol 13:16, 9 April 2007 (UTC)
Er, I guess it's more accurate to say that I'm trying to understand the reasoning for your opinion. But maybe it's not worth the effort. -- Mikeblas 13:23, 9 April 2007 (UTC)
The reasoning was explained in my first reply. I think your proposed layout is a lot less readable than what we already have, and I don't think the modest savings in space makes up for that.Jerazol 13:28, 9 April 2007 (UTC)
I certainly agree that the current template is huge, but your draft indeed reduces clarity present in the current template. Perhaps we can break it into smaller boxes where only relevant sub-boxes are shown on articles — the only example I know is {{crypto navbox}}, which transcludes sub-boxes like {{crypto hash}}, {{crypto block}} (wow, that list is worse than the database products list ever was). This would increase relevance as well as reduce the size and number of links. Live examples can be seen at Wikipedia:WikiProject Cryptography#The main crypto navigation box. -- intgr 13:25, 9 April 2007 (UTC)
I'm not sure I see the loss of readability -- the text is the same size and the same font. Project Metalworking does different boxes, similar to what I think you're demonstrating with the crypto stuff. That can work, but it takes a bit of work to avoid confusion over which project subset template gets placed in which articles. Do ya'll want to go for that approach instead? -- Mikeblas 00:26, 10 April 2007 (UTC)
Yeah, the metalworking navboxes work similarly in concept. -- intgr 19:55, 18 April 2007 (UTC)

Maybe now that the "Database products" section is so much smaller it would make sense to make it a sub-section of "Implementations of database management systems" like "Types of implementations" and "Components" are. Also, the three sub-sections might be re-arranged, with the larger group "Types of implementations" could be moved to the bottom. Just something to make it take up a little bit less space, and maybe make it a little easier to read... Sample here: User:SqlPac/databases SqlPac 03:22, 19 April 2007 (UTC)

Slight adjustment to sample at User:SqlPac/databases to make bottom half look more like the top half of the template.SqlPac 14:21, 19 April 2007 (UTC)
Sure, go for it. I have no objections at least. Jerazol 15:27, 19 April 2007 (UTC)
I'll give it a couple of days and if no one objects I'll change it on Monday.SqlPac 16:35, 19 April 2007 (UTC)
I guess it's a step in the right direction, but it's not much smaller. It's still mostly whitespace. -- Mikeblas 16:37, 19 April 2007 (UTC)
Looks good to me, just be bold IMO. -- intgr 17:45, 19 April 2007 (UTC)
Don't be boldSqlPac 22:27, 19 April 2007 (UTC)
I stand corrected. :) -- intgr 22:44, 19 April 2007 (UTC)
I'm with you :) I subscribed to be bold until I was corrected around 17 April 2007 myself:)SqlPac 01:38, 20 April 2007 (UTC)

Added "Null" to Topics in SQL[edit]

With no objections, I've added Null to Topics in SQL.SqlPac 14:27, 17 April 2007 (UTC)

Well, since this template is quite widely used, it's a good idea to suggest changes here, so that others have the opportunity to object before you make the change.
As for including NULL, in the template, I have no problem with that, but wouldn't it fit better under the "Concepts" heading? Had the article been purely about the SQL NULL keyword, fine, but the article also includes the concept of null "values" in a database. Jerazol 17:28, 17 April 2007 (UTC)
Adding it under "Concepts" heading sounds fine to me. I didn't know how widely this template was used, but now that I do know I think I need to decline making the change you recommended above and defer to anyone else who wants to take that particular risk. Thanks. SqlPac 19:24, 17 April 2007 (UTC)

Slight adjustment to template[edit]

Since there were no objections to my post on Friday, I went ahead and modified the template slightly. Basically eliminated the third section at the bottom of the template, combining it with the second section. Also rearranged the bottom section to reflect the same format as the top section. SqlPac 18:21, 23 April 2007 (UTC)

BTW, if space is at a premium, has anyone considered doing something like this? Sample Template. I'm no graphic artist, so this is not the best that could be done, just a random idea for maybe saving a little space on the template. SqlPac 19:34, 23 April 2007 (UTC)

Nah, having to tilt your head to read them is an obvious loss in usability IMO. -- intgr 20:13, 23 April 2007 (UTC)
That's the problem with most things you could do to save space on it. Short of actually eliminating something from it completely, just about anything else you can do will affect readability... Ah well... SqlPac 00:38, 24 April 2007 (UTC)

Add Codd's 12 rules to Template[edit]

Suggest adding Codd's 12 rules to the template - possibly under the Concepts sub-heading. SqlPac 15:20, 29 April 2007 (UTC)

Object-oriented DB vs Object-relational DB[edit]

Comparison of object-oriented database management systems is confused with comparison of object-relational DMS. Perhaps both categories need to be added

Rethinking this template[edit]

I'm going to try completely reorganising this using template:navbox. The current version is fiddly and very markup-heavy. Chris Cunningham 11:17, 18 October 2007 (UTC)

Done. What we may lose in diagrammatical clarity I think we make up for in brevity and standardisation. Comments? Chris Cunningham 11:27, 18 October 2007 (UTC)
I think that's a big improvement, with negligible loss if any in clarity. This template still needs more work, though, on its logical organization of topics. Also I think it would be best simply to sort the entries within each section. - JCLately 16:28, 18 October 2007 (UTC)
It's more about Databases, not Database Management Systems now. Title should be changed, and it should be included on pages such as Relational database. 129.255.123.137 (talk) 18:01, 26 June 2008 (UTC)

Category:Database management systems[edit]

I recommend that the template NOT blindly put articles that use it into Category:Database management systems as that category is currently very large and some work has been done on subcategories to break it up. For example, I don't think that the articles on SQL keywords and clauses should be put into this category. So, I propose that a new parameter be added to the template such as add category to indicate whether the category should be added. If not specified, the default would be to add it to the category to retain backwards compatibility with existing behaviour. Also, why is it sorting every article it adds to the category by "*"? RedWolf (talk) 19:21, 22 August 2008 (UTC)

Second that, or is there a standard for which articles upon which this gets placed? A number of vendor-specific DBMS's, notable or not, include the template, perhaps the only article including it should be the ones explicitly linked from the template? Cander0000 (talk) 06:24, 28 May 2009 (UTC)

Recent edits.[edit]

UML and other[edit]

Why exactly is it that the "Implementations" should not be part of this template? I don't really care one way or the other about the "Data"-section you added, but please don't remove parts of the template without giving a reason for this. Jerazol (talk) 10:45, 22 September 2008 (UTC)

I got no clue what are you talking about, in regards to the "Implementations" you are referring to. Initially I remove ORM thinking that they are not part of the database at all. But after reading it, I found out that is totally not true, most didn't people don't recognize the object aspect of database because of misunderstanding. So I will try to explain in brief.

Database are most divided into 3 parts. The DBMS (Relational, Distributed and Object). I think you guys are pretty familar with relational and distributed DBMS. Where Distributed (is the big Parallel Computing part, which has various topics such as: Grid Computing, Cloud Computing, Cluster Computing...etc.) Relational is commonly use with server side scripting like php, cfml, jsp...etc.


Objects are a bit complex (they are used in

  • Scripts: ECMAScript, JavaScript, VBScript, C#
  • Components: BHO, Plug-ins, Widgets, Applets
  • Exchange: ActiveX, ADO (ActiveX Data Objects), OLE


  • Models: COM, UMLcontroversial
  • Database: DLL, RDO (VBScript), SDO (Semantic SOA database), Microsft JET Database Engine (Microsoft Access)
    • Virtual Machine (a higher studies of Object notation based database that utilize the engine communication. Similar Hypervisor work similarly.) (e.g. JVM)
Classification (Higher Studies of Databaes), lower studies of Object Notation
Higher Studies of Database, Higher studies of Object Notation is known as Virtualization, which emphasize on the effiency of Workflow management and Traffic Engineering of Hypervisor interface, this is significantly important in the study of database, since modern database is not design to such integrity that the technical workflow of each protocol is concurrently unbalanced. Even Parallel Computing-subvariant Database like Cloud Computing, Distributed Computing have a limitation (well for now at least.)

ECMAScript 4.0 Standardization Crisis

Not sure about: J#, J++ and FastCGI uses the object database too?

Simply put it they stored the information in an actual file (.bin) and access it using interfaces. Which is the same method as flat file database. However, since this mechanism is not easily hosted, as these type of database are more commonly use in shopping (iTunes downloading), accounting...etc. Some companies like Microsoft develop runtimes and other extensions on top of SQL for easier access. Other companies like Adobe, choose to download the object as a swf. Firefox on the other hand implant it as a more user friendly method. (aka Ubiquity) <ref>.

I don't why this scope of the database should be removed or eliminated, if Distributed Computing Database is allowed. There is a lot more to it about the different types of database that we haven't discovered, so if you guys only want relational and distributed then this template is totally biased, and it doesn't serve well as a template purpose of simple navigation.

Note: that the describing retrieve as in access is a incorrect nomenclature, but I couldn't really figure out a good computer science term to describe it. --Ramu50 (talk) 21:13, 22 September 2008 (UTC)

I just undid Ramu50's last change - I have specific concerns that Knowledge base, Information architecture, and Semantic web aren't really DB terms, and that HSQLDB and SQLJ are completely non-notable and out of place where he put them. Along with other concerns about the rest of the changes, I thought best to just undo and ask for more discussion here...
Ramu - Those first three topics really aren't database topics per se. They are conceptually related to a lesser degree, but that's not the same as being DB concepts or topics. The last two... There are at least hundreds of database programs, and thousands of access methods / scripting tools out there for working with databases, and there's no reason that those particular two are notable enough to fit under the high level topic in the template. Those two are really not that notable or well known in the field.
Please discuss here before making any more changes. Thanks. Georgewilliamherbert (talk) 21:55, 15 October 2008 (UTC)
Ramu - to follow up, you just reverted back to your version, claiming "WP:Notability, doesn't that encompass the skeptism on minorities topics." ...as your edit summary.
Please don't do that - please discuss here first, revert the template second.
Regarding the notability policy... This is a template, not an article about those individual topics. Notability of the particulars used in a summary template like this has to be established more strongly and with stronger consensus than for just having articles about them. All of them clearly deserve articles. But there has to be more support, more connection with the topic to justify them being in the template. So far, you have not addressed those concerns with any of those topics. I am not sure what your work or educational background is with these topics, but I'm familiar with the research and work with them professionally, and I believe that the Knowledge base, Information architecture, and Semantic web topics are simply not Database topics, and that HSQLDB and SQLJ aren't sufficiently notable within the overall topic to deserve space on the template.
Again: There are hundreds of database engines and thousands of database access / scripting tool sets out there. NONE of them currently is listed in the section you added SQLJ and HSQLDB to. They're simply not that important / notable enough, and it's the wrong section anyways.
Please stop changing the template and discuss here in more detail. Thank you. Georgewilliamherbert (talk) 22:11, 15 October 2008 (UTC)

Give evidence that The Semantic Web isn't considered a form of database. You need to go over with your old ideas on database, database is simply a medium of storage.

Database is simply a collection of information. The range of the flexibility, functions or exprsesions is not limited, in fact you could argue that Database itself might not be a design flaw at this stage of computing. Commonly people think database are just a storage software, but that is true, if it were why design the tree structure database of a Registry keys, why would MIT spend time on classifying the a database informations, its obvious there the design of the current database is probably posing a scalability problem similar to what happen back in the Intel Pentium stage design, whos to say when you implant an interface, a classification or perhaps a middleware in the database for more flexible and cross compatibility (across multiple database) and the structure isn't a database anymore. Who are you to judge what the infrastructure of a daabase can be or can not be.

KnoweldgeBase Database is just a simple way of putting it as an artificially intellgient database, so you can probably process things faster such as (instead of having the need to use Google Crawler to search things through a broad range of algorithm, database should probably consider classifiying their information like a library so things can be easier to find and the range would be smaller, therefore when expressing things are more cleary, concise, coherent, structural..etc. (XML Schema). Also who says database can't be in lightweight form like Cloud Computing.

Main Source: http://en.mimi.hu/computing/database.html


Registry are database. DLLs are database. Main Source: http://en.mimi.hu/computing/database.html

A database can be as simple as a shopping list or as complex as a collection of thousands of sounds, graphics, and related text files. Database software is designed to help users organize such information. Even Sun Microsystems have make the capability of making StarOffice spreadsheet into a flexibile table that can imported and exported, that is not a database? Just because you don't think that is one of the growing technologies, doesn't mean your definition of notability is correct.

http://www.maran.com/dictionary/d/databas/index.html Microsoft Article also states that Access is RDBMS.

A program that helps you manage large collections of information. You can use a database to 
store, sort, and easily find information. Two popular database programs are Lotus Approach and 
Microsoft Access. See also FLAT FILE DATABASE and RELATIONAL DATABASE. 
The registry database consists of "keys" and "values". A key is a "place" to store a value. A 
key may store many values, or it may have subkeys, each which store values. The Windows 2000 
registry has five top level keys as listed below.

http://www.gtscompanies.com/glosscomp.html

A shared collection of logically related data, designed to meet the information needs of 
multiple users in an organization. The term database is often erroneously referred to as a 
synonym for a “database management system (DBMS)”. They are not equivalent. A database is a 
store of data that describe entities and the relationships between the entities. A database 
management system is the software mechanism for managing that data. 

DBMS definition

DataBase Management System: A software application system that is used to create, maintain, 
and provide controlled access to user databases. The DBMS masks the physical details of the 
database storage, so that the application only has to know about the logical characteristics 
of the data, not how it is stored. Because of this, unlike a flat file, a database makes 
applications easily portable across (hardware and operating system) platforms. See also: flat file. 

Masking is an abstract term for the basic theory of lightweigh database (aka Cloud Computing). --Ramu50 (talk) 22:20, 15 October 2008 (UTC)

Again - this is a template. Even more so than for overview articles about a topic, our policy on undue weight for minority or unsupported viewpoints is very important for Templates like this.
I don't know of any researcher or writer who's described the Semantic Web as a Database system. You have to provide some sort of source that connects them. You can't just assert that to be true - there has to be some sort of source for that.
The rest of what you're saying above does not challenge what I've said, and does not support the changes that you made. Again:
  • Knowledge base, Information architecture, and Semantic web topics are simply not Database topics
  • HSQLDB and SQLJ aren't sufficiently notable within the overall topic to deserve space on the template.
If you have sources to the contrary please provide them.
I am also concerned about the rest of your changes but haven't reviewed them in enough depth to have specific comments at this time. Georgewilliamherbert (talk) 22:31, 15 October 2008 (UTC)
Following up: you added to your comment:
Database is simply a collection of information. The range of the flexibility, functions or exprsesions is not limited, in fact you could argue that Database itself might not be a design flaw at this stage of computing. Commonly people think database are just a storage software, but that is true, if it were why design the tree structure database of a Registry keys, why would MIT spend time on classifying the a database informations, its obvious there the design of the current database is probably posing a scalability problem similar to what happen back in the Intel Pentium stage design, whos to say when you implant an interface, a classification or perhaps a middleware in the database for more flexible and cross compatibility (across multiple database) and the structure isn't a database anymore. Who are you to judge what the infrastructure of a daabase can be or can not be.
This is nonsense. It's random topics jumbled together and not connected. Please focus on the issues I have raised. Georgewilliamherbert (talk) 22:39, 15 October 2008 (UTC)

A random topics, you got a long way to learn. The issue is you are the not able to learn to accept, while the industry already can accept these flexible theories. Its quite obvious your level of understanding doesn't show your knowledge on computing science. Middleware interface implementations is your so-called random organizations. Do you even understand anything about computer science.

I do wonder what you put the Wikipedia:NPOV policy, when the inforamtion I talk about is just a bunch of possible metholodgies that the industry is rearching on I don't know what is your problem of not willing to accept since the information are not synthesis since and I am not the only one saying it,

I've seen a lot of template that deals with topics that are not easy to define, therefore people put principle topics on there for guidance, so how is my topics irrelevant.


e.g. Template:Parallel_computing


I have the right to put what I think is correct and provide guidance for others. That being said, the guidance does the serve the purpose of navigation which is what template are meant to be.

By the way the Database Middleware, I though is a possible theories, but I didn't know there was already something out there. Read this article.

Database Middleware.

Note that SearchSOA own by TechTarget has 6 million worldwide subscribers, so don't even try to pull off for notability and other irrelevant policies. --Ramu50 (talk) 22:47, 15 October 2008 (UTC)

Ramu, I have studied databases and database theory, and computer architecture, and worked in the field. Please don't assume that I don't know what I am talking about.
You still have not cited references for the specific claims you're making and I have challenged.
Please stay focused. Please provide citations for these claims I have raised questions about. Georgewilliamherbert (talk) 00:34, 16 October 2008 (UTC)
Following up regarding the techtarget.com article there - What is your point? The article isn't describing anything relevant to what you added to the template or what I'm challenging about the template. It's not a relevant reference to what you added. Why did you try and cite it? Georgewilliamherbert (talk) 00:37, 16 October 2008 (UTC)

semantic web[edit]

See article by Tim Berners-Lee [2] "The semantic web data model is very directly connected with the model of relational databases (...) Indeed, one of the main driving forces for the Semantic web, has always been the expression, on the Web, of the vast amount of relational database information in a way that can be processsed by machines.". This appears to clearly belong on the template. --Enric Naval (talk) 01:50, 16 October 2008 (UTC)

Knowledge base[edit]

A knowledge base is usually implemented on a database, but so does every other system that has lots of text that needs to be queried, like inventory management applications. I couldn't find sources stating if knowledge bases are actual databases per se. Not sure if it belongs on the template, Ramu50, can you present some source explaining the link between knowledge bases and databases? --Enric Naval (talk) 01:50, 16 October 2008 (UTC)

Information architecture[edit]

This just looks a concept related to databases, even if it's very loosely related... It probably belongs on the template. --Enric Naval (talk) 01:50, 16 October 2008 (UTC)

Example of KnowledgeBase Database, DNA database, documentation database. http://n1.cs.fiu.edu/SemanticWrapper.ppt#317,4,Features of Semantic Wrapper

Can function as a stand-alone application and/or be plugged into a hetergenous mult-database system

http://n1.cs.fiu.edu/SemanticWrapper.ppt#261,7,Semantic Data Model Benefits (Cont.) --Ramu50 (talk) 02:42, 16 October 2008 (UTC)

Windows registry, DLLs, and shopping lists[edit]

DLLs are absolutely not databases. DLLs are executable images, almost identical to EXEs except for a flag bit that says "this is a DLL." Does Ramu50 also claim that EXEs are databases?

Ramu50 quotes an article (itself unreferenced) that says "A database can be as simple as a shopping list..." Well, yes, I could store a shopping list in a simple database consisting of one table with one text field per row. Or in a text file created with ed, for that matter. However I think you would have a tough time convincing your "Data bases 201" instructor that this was a database whose implementation should earn you credit for your class project.

The Windows registry is something of a corner case. It can be viewed as a database; it certainly is a hierarchical data store providing a naming mechanism for each level of the hierarchy. However one hallmark of most databases is that someone defines a set of tables, and the fields or columns in each table are well defined. The Windows registry is nothing like that; its low-level structure is left completely up to each user or client program. It is really much more like "a file system for very small files" - this is a quote from Brian Willman, one of its architects.

Oh, of course you are going to say that you can consider file systems as databases too. And really, this has much more credibility than the claim of "shopping lists." But if you start listing file systems like FAT, NTFS, ext3, ODS-2, etc., in this template you are going to be overruled. Similarly for flat text files.

"Database is simply a collection of information." So any simple data structure in memory like the Windows KPROCESS or IRP is a "database" to you? How about a stack? A FIFO doubly-linked list? Nonsense!

It is true that when complex subjects like databases are taught, simple examples like "shopping lists" are used in the first lectures. But not because anyone considers them "databases." They are used simply as a way of gently introducing the student to what may seem to be an intimidating subject. These simple examples are quickly left behind once their usefulness is outgrown.

If you insist that "database is simply a collection of information", this really does include things like flat files, file systems, and everything in between... and you have just diluted the term "database" so much that it no longer has any useful meaning.

In any case it is not the purpose of a template to be all-inclusive, to list everything that one editor thinks might conceivably be considered related to the topic. Jeh (talk) 08:36, 16 October 2008 (UTC)

Well for the me I think of dll ability to store icons as a database. If it wasn't a databse, then the concept of In-memory database meaning loading a mini-database into memory wouldn't even exists thus the idea of cloud computing (which does focus on the concept of lightweight database wouldn't even exists today). I don't know what you even brought things like Windows KPROCESS, IRP and FIFO doubly-linked list, they have no mechanism of a database.

Semantic Web is considered a form of database, because the artifical intelligence of classification can build the database into a tree-structure system, this in turns can speed up the SEO.

KnowledgeBase is database, because of a capability of analytical. If says the current database is a database, then let me ask you how is a database different a spreadsheet, a garbage can. Because they are just pile of information that classify things unwisely. Those mechanism of things like group by, order by in SQL work well in Buisness Intelligence, but for things like biochemistry it is totally useless, say a research want to find a listing a protein that can heal the skin very fast, he search for find a list of protein with healing capabilities, then you are telling me that the database response by giving a mixture of biohazard protein, and pig protein is ok? Yea very smart of you to invent something that can do its job, but because the lack of classifications or methodologies you use in a database it kills people instead of healing people.

On a higher-level of study if the database only has those simple function, then when you impor those information into a CAD simulation programs, it would quite obviously lack (the footnotes ability of letting users know the info is biohazard, acidic, dangerously radioactive...etc.) Putting it on a more realistic scale, if the database contains all types of protein including life from other planets, then you assign virus is null, when virus' protein can give you the best anwer. Thus you will NEVER find the best answer, and the effiency will always be limited and if a newer species is found under Kingdom virus, the database ability remains to be rather bias because it considered all other information to be null when analytical information can possibly map out the simple information as a Venn diagram, a flow chart, an algorithm...etc. --Ramu50 (talk) 18:47, 16 October 2008 (UTC)

Ramu, what is your computer science background and education? Georgewilliamherbert (talk) 19:35, 16 October 2008 (UTC)
"Much my knowledge come from CompTIA A+ and reading thousand of article on the internet." Ramu50 in talk:AT Attachment (The A+ certification esentially says you are qualified to assemble PC hardware.) A few months later he claims knowledge of a very wide range of IT mechanisms. Jeh (talk) 20:54, 16 October 2008 (UTC)
Ramu50, the ability of a DLL (or an EXE for that matter) to store icons, fonts, sound clips, etc., as "resources" is rather that of an embedded file system, not a database. In the "resources" mechanism there are no concepts of rows, columns, relations, etc. Again, if you consider this as a "database" then you must certainly consider a traditional file system as a database as well.
I brought up things like KPROCESS, IRP, and FIFO list because you (Ramu50) claimed, in defense of your very inclusive philosophy, "database is simply a collection of information". Well, a KPROCESS is indeed a "collection of information" - it's the root data structure that defines a process to the Windows kernel. An IRP, same concept: It's a "collection of information" describing an I/O request. A FIFO list is likewise a "collection of information", specifically the collection of the elements in the list.
But by saying "they have no mechanism of database" you are agreeing with me that the claim "database is simply a collection of information" is specious. It admits far too many examples (like simple in-memory data structures) which are clearly not databases.
I believe that to most practitioners the term "database" implies both a collection of information and storage of metadata that provides at least some degree of semantic description of the information. In this sense, a traditional sequential file or even an ISAM file consisting of records containing fixed-length fields is not a database: any program that consumes such files has to have knowledge coded into the program that tells it which field occupies which character positions, which fields are to be interpreted as binary integers vs. character strings, etc. Whereas with a database, at least some semantic information is in the database. For example, consumers can simply refer to fields (columns) by name instead of by byte offsets within a record (row). This distinction is completely missing if you believe "database is simply a collection of information". I therefore maintain that any "databases" admitted to this template under the "simply a collection of information" principle don't belong here.
I would disagree though with the notion that only examples of relational databases should be included. Although not much in favor these days, hierarchical databases were historically very important and a few of the most significant should be mentioned. Jeh (talk) 20:06, 16 October 2008 (UTC)


And why would I want to reveal my full education to others that I don't know, questioning things on Wikpeida:Adminastrators' Board and on irrelevant talk page, I should of report both of guys to other adnminastrators. --Ramu50 (talk) 21:29, 16 October 2008 (UTC)

Everything you have written on WP (aside from deleted pages) is part of the public record and may be quoted at any time. On the other hand, your threats to "report to adnminastrators" (sic) seem to me to go against the guidelines at WP:Talk_page_guidelines#Behavior_that_is_unacceptable. If you think you have a legitimate complaint don't threaten, just follow the procedures at WP:DR or perhaps WP:WQA. As for this page, please stick to the subject at hand. Do you have any substantive, on-topic responses? Jeh (talk) 21:41, 16 October 2008 (UTC)


Are you asking for the conflict of ATAPI to be brought again or what. Any yet despite that you seem to have hatred against what other has to say about things when you alone already violated the Wikipedia:Talk_page_guidelines#Behavior_that_is_unacceptable. Prejudice against someone opinion. Reporting Georgewilliamherbert is totally acceptable to Wikipedia and legally if I want to for racial discrimination

He's claiming deeper knowledge of computer architecture but not sourcing claims. This is an 
area which I've studied and worked in professionally and he seems to just be spouting nonsense 
so far. He could be a non-native-english-speaking research grad student who's just not 
communicating effectively in english or something, but he's not listening to me, and I can't 
protect the template when I'm involved in a content dispute on it etc

Wikipedia:Administrators'_noticeboard/Incidents#Uninvolved_admin_on_Template:Databases

--Ramu50 (talk) 21:57, 16 October 2008 (UTC)

If you start "the conflict of ATAPI" up again I will once again use e.g. WP:RSN (where the consensus was ~"Ramu50 is wrong", and one ~"Ramu50 is simply being stubborn") and WP:RFC. I believe that WP:POINT will also be applicable.
I see no "racial discrimination" in anything GWH wrote. (Plenty of people of all races have trouble expressing themselves clearly in English; that comment therefore has no racial component.) I too have noted that you do not express yourself well.
Please be aware that threatening legal action is also specifically in violation of WP policy: No legal threats. In fact it appears from WP:NLT that you can be blocked from editing for doing so.
I really think you should bring complaints against both of us at WP:WQA: "an early step in the Wikipedia Dispute Resolution Process... a non-binding noticeboard where users can report impolite, uncivil or other difficult communications with editors, to seek perspective, advice, informal mediation, or a referral to a more appropriate forum." Sounds about right. I suggest this because it is clear to me that nothing will be resolved here; bringing your complaints to a wider forum will be beneficial to all. i.e. stop threatening to report people and just go do it, or don't say anything more about it. Your comments here can then be kept on topic.
In spirit of which, I ask again: do you have any substantive, on-topic responses? Jeh (talk) 22:26, 16 October 2008 (UTC)

Focus of this template?[edit]

It seems to me the locus of the dispute is that some editors want to extend the template beyond relational databases, and others oppose. Probably the best thing to do is to rename it to "relational databases", as not to mix too many things in the same template. VG 08:59, 16 October 2008 (UTC)

Agree. To implement this, the "implementations" group should be moved to Template:Database implementations, and the links to the different database types (like Network model and Network model) should be moved to Template:Database models. --Enric Naval (talk) 00:18, 17 October 2008 (UTC)
So, in a few days I'll go ahead and make the changes if there are no issues raised. --Enric Naval (talk) 21:20, 20 October 2008 (UTC)
Well, at the end I created Template:Database implementations and models. Please review it. --Enric Naval (talk) 20:16, 23 October 2008 (UTC)

In nobody opposes, on a few days I will create Template:Relational databases. When I have splitted everything relational-related, I'll cleanup this template so that it mostly links to lists of stuff and to the other templates, as a general template should do. --Enric Naval (talk) 23:24, 24 October 2008 (UTC)

I don't oppose this - I think it is a right thing to do, but I am curious, what will be left in this template after everything that is specific to relational databases is moved to another one ? Most of the links in this template currently are relational database specific. Wikiolap (talk) 15:33, 25 October 2008 (UTC)
Well... not much really.... mainly stuff that applies to all databases types and not just relational ones, like Transaction processing, CRUD, Data dictionary, Partition (database) etc. There is stuff that should be on both templates, like Database transaction and Query optimizer, and there is stuff that I still have doubts about, I will need quite a lot of time to sort it out.
Damn, maybe we should just divide this template into relational-specific stuff and general stuff, I'll have to look at it... --Enric Naval (talk) 18:02, 25 October 2008 (UTC)

Template talk:Database implementations and models#new --Ramu50 (talk) 21:04, 18 November 2008 (UTC)

Object vs object-relational[edit]

The bottom bar here says "Database products: Object-oriented (comparison) · Relational (comparison) · Document-oriented".

"Object-oriented" links to "List of object database management systems" -- systems like db4o and ZODB. "comparison" links to "Comparison of object-relational database management systems" -- systems like Oracle and PostgreSQL.

These are completely different, and I'm not sure why they're presented in a way to make OODB and ORDBMS sound similar. How can we fix this?

Oh, I see there is a "Comparison of object database management systems" page. Could we simply add a fourth section to the bottom bar, so Object and Object-relational are separate?

Recent activity by The Fortunate Unhappy (talk)[edit]

This relatively new editor has, just today, taken it upon himself to

  • remove the SQL section from this template
  • replaced the references to this template with references to the SQL template on about 20 SQL-related pages

This may or may not be a reasonable thing to do - I'm not an SME here. But somebody who is an SME needs to take a look at this activity and see if this obviously highly motivated person needs to have a bit of brake applied. Particularly since his contributions show utterly no participation in Talk pages for any of the SQL-related pages. Jeh (talk) 23:45, 21 February 2010 (UTC)

It appears the Template:SQL is a relatively new development, but it has been incorporated into all of the SQL commands articles. It looks like a good thing because it provides some add'l references for SQL as a language. The question is to continue to list popular SQL commands on this template are not? Since not all databases are relational and/or use SQL, perhaps its not as good a fit here. Cander0000 (talk) 13:27, 24 February 2010 (UTC)