is there any anglo-saxon source about information retrieval out there, that does not stand in relation with a defense department?? so far everything i was looking at was funded by either darpa, cia, nsa, naval department, defense evaluation uk, blablabla research??

just an example out of others:
The purpose of the project was to develop automated systems and tools enabling multilingual and multimedia information capture, search, retrieval, summarization and reuse. The system, built on the underlying Informedia system ...
(europeans involved. at mIi project they worked together with university of karlsruhe)

... concepts, technology and infrastructure, is designed to access textual, audio (radio) and video (TV) information, to index, categorize, retrieve, summarize and analyze it, in one or multiple languages. We focused primarily on the Serbo-Croatian language to demonstrate viability and practicality of proposed concepts ...

no na. the project started 1997 and ended last year, so to say. and some people say serbo croation language doesn´t exist. i once had an endless debate about that topic)

... other target languages were german, french, italien, spanish, japanese and korean).

what they did was building a prototype for a ...

"multilingual browser of text, video and radio material that accepts English queries and returns the most relevant Serbo-Croatian, German and English language reports or segments in their original language, in full or summary form. For example, this would enable an analyst to compare divergent American and foreign reporting of the same event or topic.

however, it seems they put their emphasis on german, sk, and english.
:

"We built and delivered functional broadcast news-focused systems to multiple, network-connected, offsite locations including DARPA and NSA. Network delivery issues were being addressed and system architecture was being redesigned to improve performance when anticipated project funding was curtailed.

they used and (open source) for speech recognition and translation, lycos for information retrieval – as far as i remember brewster kahle mentioned something at scope that alexa will get included in lycos around 1998 -), for machine translation.
results?
topic detection: "allowed the user at least some (sic!) judgment about the returned stories".
multidocument summarizer: "Beyond single-document summarization, a synthesized summary of a set of documents -- such as those output by the retrieval engine with respect to an analyst's query -- often proves more desirable.
dynamic language modeling: they used web text corresponding to cnn, ap and reuters news, reducing speech recognition errors to 19% on news stories. (well i guess if you are analysing the same source/people all the time, the program should get it after a while. so it doesn´t sound soo great.)
video OCR: "The overall recognition results are good enough for use in news indexing."