Semantic and Mediawiki: free and little by little approach

As an early Mediawiki user and evangelist I will only write about Mediawiki as a CMS.

Nowadays Content Management System refers mainly to Database Management System and Mediawiki stores the content of the wiki in a database. This storage is as easy as writing freely in “edit” mode and save. You can use or not some mediawiki markups to improve presentation of the text (bold, italic…) and link one page with another or refer to an external website.
So MediaWiki is the most powerful way to quickly store and retrieve any data
But at the end of the day you can SEARCH in the wiki every single word you want BUT you will NOT be able to ASK a basic question (query) to this wiki database.

Example:
– search : “Berlin” > you will get an amazing page of results with all the pages which titles contain the word “berlin” and all the pages which contain the word “berlin” in the core text content.

– ask :”What is the capital of Germany” > no way.

The Semantic Mediawiki Extension adds a “dissipated structured database layer” to MediaWiki. Any data, anywhere inside the text of any MediaWiki page, can be selected to be structured as having some values of some property. Properties are similar to the traditional fields of a structured database. So adding property and values to some data in your Mediawiki allow two tremendous features:

1- Ask (query) : for example in your international company you decide to run one unique wiki. This wiki has to be multilingual. A nice way to know where are the pages written in French is to do a simple query to the database. You decide then to create a Property:Language. To do so:

1. you open a new page with title “Semantic Properties” : everybody will be invited to check it before “editing” a page in the wiki.
2. this new page contains a very short list of mandatory properties according to your wiki target use.
3. each property is written in the page with the following mediawiki markup : [[Property:Language]]
4. when you save the “Semantic Properties” page you click on the link ‘Property:Language’
5. the page with title ‘Property:Language’ explains the different value of Language: English, French, German, Italian, Portuguese, Spanish, … This is not an exhaustive list, any user can add a new value (ex: Chinese)
6. new page has to contain this tag [[Language::English| ]] or [[Language::French| ]] to be reported as an English or a French page.

2- Report (Printout) : every query could be reported or print out in a customized format. This is a hugh and valuable improvement to compare with the result SEARCH page of standard Mediawiki.

3- Forms: on my point of view it is the less operating feature as it reduces the “free editing” feature which gives more power to the wiki content compared to blog content. Enrichment of content is open in a wiki and this is what enterprise 2.0 are looking for.

To wrap it up, Semantic tool must not kill the free access to knowledge that Mediawiki gives, offers to organisation.
Let a free folksonomy prevail and do not impose a presupposed taxonomy. Life and time will build it as it should be…

See also : http://iphylo.blogspot.com/2009/08/wikispecies-is-not-database.html

Related articles by Zemanta

• Social Media Is Changing The Rules For CRM (networksboise.wordpress.com)