Visualization of the Franco-Prussion War

MediaWiki – Software moving towards the future (Part 3 of 3): Communication, Wikidata, Translation

Communication and notification – the end of the classical discussion pages

Another MediaWiki construction site is delivering good news: The MediaWiki Communication System. This concerns the discussion pages. These will soon disappear, in the form we know them now. More precisely: The discussions pages will fundamentally change and merge with the notification system.

The reconstruction of the MediaWiki communications system will take place through two new extensions:

  • Echo allows the individual following of changes, gives an overview of the whole system and is a framework for a variety of communication services. Echo is already in Wikipedia as a new notification system.
  • Flow makes discussion easier. One can more easily follow discussion processes; answers are shown via Echo. And much more. The aim of these developments is to build up a modern discussion and collaboration system for all Wikimedia projects. An interactive prototype is already online.
The Flow prototype
The Flow prototype

In the end, MediaWiki will have access to a new communication layer, which has long been overdue. It is important to note that the combination of articles and discussions – a characteristic and principle for the success of MediaWiki remains. The articles will not be separated from the discussions which relate to them. The linking of articles with discussions is a basic principle which makes MediaWiki unique, giving it something special compared to other solutions. Flow and Echo can really bring this basic principle of MediaWiki to its full potential.
However, many in the Wikimedia community are sceptical. For them, the current systems is absolutely adequate. Behind this, there is an interesting debate.

Erik Möller writes: “Do we want discussions to take place in a document mode, or do we want a structured commentary mode?”

This is the difference between saving discussions on a page or in a database with other possibilities for sorting and filtering.

Even though saving onto a page has a certain charm – particularly with respect to archiving, it is unlikely that this principle can be maintained for long. The advantages of following different discussions in a central place are convincing and overriding.

Collaborative data collection – Wikidata & Friends

A further signpost to the future is the development of collaboratively managed databases. The project OpenStreetMap is already well known, and creates map material using the wiki principle.

Wikidata follows a similar path, a collaboratively managed database. The different language versions of Wikipedia need not only pictures and videos, which are currently collected centrally in Wikimedia Commons. Providing metadata centrally is just as important.

Before, each language version needed to be changed by hand.

 

Excerpt of the Wikidata page of New York City (Original: https://www.wikidata.org/wiki/Q60)
Excerpt of the Wikidata page of New York City (Original: https://www.wikidata.org/wiki/Q60)

 

To provision all Wikipedias and for all use cases, is a very ambitious aim. After all, the data is not included in a homogeneous data schema. The data samples differ enormously by topic and cannot be centrally predetermined. This can only be done by authors. The authors can freely specify the data fields and composition themselves. They can add to new data fields or delete them. This also means that authors need to agree on uniform nomenclature and consistent data management. Further, the sources of the data must be shown so that one knows where a number, name or value has come from.

The effort is, however, worthwhile: One of the next steps will be to display the data better, in order to, for example, show personal connections in politics and history better. This will be significantly easier with Wikidata. For example:

 

Visualization of the Franco-Prussion War
Visualization of the Franco-Prussion War

This will develop, in a similar manner to the project Wikimedia Commons, a publicly accessible store of data, under a free licence, meaning a part of the web’s memory will become open to all.

This development is, therefore, pioneering for the free web. Initially, data will be collected for the Wikimedia projects, but very soon other projects will be able to consider using Wikidata software. For example, when data is to be collected about a specific specialised area which is too in-depth for Wikipedia; or when it is necessary to keep control over the data sets. I am thinking of Lobbypedia or more extremely Wikileaks. This requires protected Wikidata infrastructure, which is kept by a circle of trustworthy people.

The end of the category pages and SemanticMediaWiki?

Alongside the build up of this database infrastructure, another MediaWiki feature will soon belong to the past. We are talking about category pages. Up until now, the category information (indexing) was inserted in the wiki source text. This concept has, like discussion pages, its charm, but it leads to very laborious management in the long term. Due to this, it is likely that Wikidata will eventually replace the category system. The category pages are, however, often also used as practical portals. This function must be maintained. But these are the topics of interest for the years to come, and the discussion has only just started.
This is also true for the future of SemanticMediaWiki. Some see Wikidata as the successor technology on the horizon. For this, though, Wikidata needs to develop significantly as regards implementation and adaptability. And as Semantic needs absolutely no extra technology, and is being driven on by a very lively community, there is no end in sight. It is actually more likely that the Semantic community will anticipate many developments which will only later be taken on in Wikidata.

Automatic translation

Lastly, I would like to highlight another very promising development: Great strides have been made in the automatic translation of content. The extension Translate, which has been in use for a while now, is developing to translate system texts from open source software projects. Henceforth, this function will be extended by a technical translation aid which will already be able to suggest whole text passages as translations and improve their quality significantly because one can use Wikidata. The first versions were introduced at Wikimania 2014 and the end of the second step has recently been announced. Languages are being added one by one and the functionality is being continually extended.

Final thoughts

All these developments are focused on the needs of Wikipedia and possibly on its sister projects. It is, however, necessary to make these developments useful and available in the long term for other use cases and non-Wikimedia projects. For this reason service providers will have to provide the necessary services – from development and customisation up to maintenance and support at a high level.

 

Links

 

You Might Also Like

Leave a Reply