I was startled to find this gem in EU directive 2019/1024 Art.9 ¶1:
Where possible, Member States shall facilitate the cross-linguistic search for documents, in particular by enabling metadata aggregation at Union level.
Even if you neglect the “cross-linguistic” specification, merely making public documents searchable is a huge leap of progress in the EU. And I think all member states are currently breaking that law for the most part, as we are generally forced to rely on private sector ad surveillance garbage from Google and Bing to find most public sector docs.
Sure there are a few scattered search tools for some very specific collections of documents. But most public documents are not at all indexed in any publicly administered search tool.
Of course the “cross-linguistic” specification is quite interesting because document translations are sometimes performed but the result is often not shared and even more often not searchable. E.g., for some reason a university or institution in Belgium (possibly public sector) went to the effort of creating a good English version of a big piece of the Belgian Economic Code. I was lucky to stumble into it out in the wild. Per the directive (which is hopefully transposed into national law), someone who searches for that section of Belgian economic code should get a reference to the unofficial English version along with the French and Dutch versions. But they certainly do not because the national legal statutes search site is hard-coded for just French and Dutch.
This touches on a recent question I asked. If the EU were to obtain an English version of transposed directives, in principle they should be furnishing that to the public. There’s one snag here though: the open data directive seems to exclude the EU itself from Art.9.