The Betawiki statistics for October are in again. With the request for Western Panjabi we now support 321 linguistic entities. Of these only 102 provide the most basic support, 61 languages cover 90% of the core messages and 24 provide a 90% coverage for the extensions used in Wikimedia Foundation projects. The good news is that compared with last month, the numbers are essentially up except for the WMF used extensions.
When you compare this with a year ago, you will find that we have come a long way. I was told at the Wikimedia Conferentie Nederland, that with the move to Betawiki by the German localisation effort more people had become involved and that the quality had gone up. At the same conference i learned that many people are just not aware of its existence and significance; I hope that we may expect some Frisians at Betawiki.
Because of the continuous development of MediaWiki it is really hard almost impossible to maintain the localisation for a language. We have an increasing number of developers resulting in daily updates to the MediaWiki messaging and for the majority of our languages, it is impossible to keep up.
After such a realisation, the question is how to support these languages better. It is arguably best for them NOT to run on the bleeding edge of MediaWiki software. Running on bleeding edge should be reserved to those languages who are able to maintain their localisation.
With this discrimination in the support of languages, we give added weight to our stable software. This will improve the usability of MediaWiki for our end users.
When we make such a division, the only argument should be the quality of the localisation effort. This would provide a powerful incentive for language communities to maintain their localisation in order to have the latest and greatest of what MediaWiki has to offer. It is crucial to keep in mind what this is to achieve; to provide our readers, our editors with the best support in our projects and localisation is essential.