There’s a proper discussion on automatic translation in the comments on the Hungarian LibWorld article. Especially the translation of so-called agglutinative languages seems to be difficult. Gamoia (see her article on Spanish library blogs) states:
I see there’s still much work to do. But I really think it is worth to do it.
And there is an interesting project dealing with this problem, called EuroMatrix:
EUROMATRIX will apply the most advanced machine translation (MT) technologies systematically to all language pairs to achieve this goal. For most translation directions that involve languages of the new and near-term prospective member states, the project will provide baseline MT functionality for the first time. For language pairs where MT systems already exist, EUROMATRIX will design and investigate novel combinations of statistical techniques and linguistic knowledge sources as well as hybrid MT architectures in order to combine the accuracy of rule-based techniques with the adaptivity of data-driven approaches.
The project is led by Hans Uszkoreit. One of the methods to find the best translatoin tools will be organizing competitions like the one last week in Edinburgh.
Hopefully the result will be freely available as open source software.
Sources:
Heise: Das Europa der 23 Sprachen – Computer sollen häufiger übersetzen
News4Linguists: The new European project EuroMatrix develops automatic translation systems for all official languages of the EU