Semantic media Wiki

Semantic media Wiki

MediaWiki is an open source software that can be used to create the Wikipedia that everyone has used in the past. As a demonstration of DX, it digitizes non-electronic information (such as paper documents and drawings) using OCR and other methods, and organizes them in the form of Wikipedia so that they can be searched, shared, and edited. It will be easy to create demonstrations that allow users to concretely imagine the scenes of use when they are digitized.

For information on setting up and starting up MediaWiki, please refer to “Setting up MAMP and media wiki and how to use them easily“.

Semantic MediaWiki is a knowledge management system that can be seamlessly connected to other web systems by embedding a framework for describing the metadata of resources on the web called RDF (Resource Description Framework) in MediaWIki. It is a system that can be turned into a knowledge management system that can be seamlessly connected to other web systems.

It is similar in appearance to the general MediaWiki, but implements extensions such as “inline queries” to bring in content from outside, various forms of “semantic forms”, “semantic drilldown” to organize content, and “mediawiki widget,” which is a function to embed videos, etc.

Reference books include “Working with MediaWiki, 2nd edition” and “MediaWiki: Wikipedia and Beyond”. The former is a good reference book for those who want to use mediawiki.

Chapter 16 of Working with MediaWiki 2nd edition is the chapter on Semantic Mediawiki. Here is an excerpt from the introduction to chapter 16.

“Semantic MediaWiki

Semantic MediaWiki is a very important extension to MediaWiki that defines a framework for storing data in a wiki and querying it – this has the effect of turning a wiki, often considered a mere storage system for text and images, into something more database-like effect. smw is an abbreviation, which is great on its own, but when used in conjunction with spin-off extensions, it becomes (dare I say it) magical. smw has over 50 spin-off extensions, ranging from data entry to browsing and searching, visualization, extended storage, and every other aspect. While some of the extensions are more popular than others, this document will focus on the most essential extensions.

SMW, when used in conjunction with its extensions, can transform an ordinary wiki into a kind of collaborative database. This is because the version history stored in a wiki means that all data can be open to editing by any number of people – something that is almost impossible with an application supported by a standard database.

Why is it called a “Semantic MediaWiki?” The word “semantic,” in its most general form, denotes meaning. It denotes the underlying meaning, not the presentation or precise wording (i.e., “syntax”) of the text. In the contemporary context, the term “semantic web” became a hot topic in the early to mid-2000s. Ironically, the meaning of the term “Semantic Web” itself is ambiguous (see here). But the main idea behind it is to pin down the underlying meaning of the text that readers see online, a meaning that can be reused and processed by humans and machines. And that is also the main idea behind Semantic MediaWiki.

Semantic MediaWiki may not be the most important MediaWiki extension (ParserFunctions would have that title), but it is clearly one that has the greatest life of its own. As of 2012, SMW is used by 500 to 1,000 wikis, and the number is constantly growing. SMW has its own website (semantic-mediawiki.org), mailing list, IRC channel, and its own conferences (SMWCon twice a year). No other MediaWiki extension comes close.

Semantic MediaWiki was created in 2005 by Markus Krötzsch and Denny Vrandečić. Originally conceived as a feature for Wikipedia, it was conceived as a data storage method to eliminate the need for hundreds of thousands of manually generated lists and categories on Wikipedia. Its use in regular MediaWiki was of only secondary importance to its creators, at least initially. While the Wikimedia Foundation took a “wait-and-see” attitude toward SMWs, that soon changed as regular wiki users discovered their benefits and began to embrace the technology. However, with the Wikidata project (headed by Vrandečić), which began in mid-2012, the initial dream may come true. Wikidata is a very exciting project that aims to create a single data repository for all Wikipedias in different languages, allowing structured data (infoboxes, links to other languages, etc.) to be populated automatically. It is planned that the piece of code that handles the backend storage of semantic media wikis will be separated into a separate library, possibly named “DataValues,” and that both SMW and Wikidata will use this new library for data storage.

The structure of Wikidata as currently planned is quite different from the way SMW was originally proposed for use on Wikipedia. This is largely due to the fact that Wikidata is intended to support hundreds of languages simultaneously. And the syntax for storing and querying data with Wikidata will most likely be quite different from the standard SMW syntax. So it is possible to overestimate the fact that Wikidata’s storage component will be SMW-derived code. Still, if Wikidata is successful, it could raise the profile of semantic media wikis considerably. This would be a nice side effect of Wikidata’s main goal: to create the largest structured database of general knowledge information in the history of the world.

But enough about Wikidata. The rest of this chapter will focus on the semantic media wikis used in regular wikis and the many advantages they can offer. If you got this book just to read about core MediaWiki, then hopefully you will read the following chapters as well.

How SMW works:an example

Let’s say you have a wiki about wine. Now, you want to be able to see a list of all Chardonnay wines grown in the south of France. A typical wiki, be it Wikipedia or whatever (even a wiki that is not MediaWiki for the most part), has basically two options: either compile that list manually on some wiki pages, or compile it on all such pages (assuming there are pages about all Assuming there is a page about wines) can be tagged with a category such as “Chardonnay wines from the south of France”.

These two types of actions are done all the time on Wikipedia, and many other wikis as well. The first option is to create the list manually, which requires a lot of effort and must be corrected each time a new wine page requiring that list is added, or whenever some error is discovered; in the second case, the list (the category page) is generated automatically, but but information must be painstakingly added to each page. And if we expect our users to do that, we need to give them precise instructions on how to add categories, what to name the categories (“the South of” or “Southern”), and in general, what the ideal data structure should be. Should there be a “Chardonnay” category for every country listed on the wiki, even if the country only has one or two wines? Conversely, should countries or regions that produce many wines be further divided by year? Or should years be tagged in a separate category?

Semantic MediaWiki provides a solution to this problem. Instead of compiling a list or having an overload of categories, one can define a single infobox template intended to be placed on wine pages. Thus, instead of managing a large and perhaps disorganized set of categories, the data structure can be kept simple and the complexity moved to the query that displays the data.

As for infoboxes, it may still be difficult for users to learn how to add and populate infoboxes. In that regard, the Semantic Forms extension, which will be discussed in the next chapter, provides forms such that users do not need to look at the underlying wikitext syntax to create or modify data.

Finally, combining SMW with other SMW-based extensions allows you to do more than simply display information in lists or categories. Information can be displayed in tables, wines can be displayed in amap, wines can be aggregated and broken down by country, year, etc., and faceted searches can be performed on all of these fields to find wines of interest to the user.”