The Cross Lingual Wiki Engine as it is today is only a first step in the direction of collaborative translation. By providing basic tools to support translators, the project allows better collaboration in a multilingual context. However, there is still very little we know about how translators want to collaborate. The tools described previously collect a significant amount of information about the translation processes. In order to understand the information, advanced data visualization and aggregation tools will be required.
Combined with user feedback, the obtained information will allow to identify the kind of modification required to the existing set of tools to enable collaborative translation and help increase the adoption rate. By identifying usage patterns, we will be able to isolate the impediments and improve the solution to remove them.
On a shorter time scale, issues and potential improvements have already been identified and should be addressed. At the time of writing this paper, the key issues are:
- Content difference may highlight changes already incorporated in some cases. This situation can easily be replicated, but more research is required to eliminate the problem.
- Page relationship broken if linking to a new page. A solution to this problem is already designed, but the changes required are incompatible with the TikiWiki release cycle at this time.
- No support for error correction. Wikis are typically very flexible towards human errors. However, the current implementation does not provide such flexibility.
Also, the objective of the CLWE is to make collaborative translation widely available. At this time, the base implementation is made in TikiWiki. However, the same concepts and the lessons learned could be applied in multiple other wiki engines.