I've also started adding related papers and information about XML query processing. We should probably put together a online bibliography of papers, but this is it for now.
This is Bolo's stash of papers to look at. They are not niagara specific, but things of DB interest that I have stumbled across when looking for info.
The Timbre project is directed by a UW Alum, Jignesh Patel at the University of Michigan, Ann Arbor.
[Jianjun] Two major differences between their work and ours, I think, are: first, they focus more on updates since they adopt a data warehouse approach, while we assume data still reside on the Internet; second, they pay more attention than us on supporting modules, such as crowler, event detector etc.
[Jianjun] They claim they are building a XML data warehouse for the XML data on the Internet. They also want to keep the content in the data warehouse to be consistent with data on the Internet. Thus they have to consider updates and use crawler and event detector for loading and monitor changes on the Internet.
"goxml search is a scalable, context-based xml search engine that delivers precise results."
[Qi Su] The parent company, xml global also offers GoXML DB, "a native XML database with a high performance query engine."
Xindice is a small scale XML database donated to Apache by the XML:DB project
The benchmark has one document and 20 queries modelling an auction site. The CWI group has obtained results for the Monet database system. They have also registered the domain name http://www.xml-benchmark.com!
This is XMark.
Some/many of these are full-scaled XML projects, but I didn't have a writeup to go with them. Hence the short-form listing...
Niagara currently uses the Xerces-C parser. However Niagara has its own XML framework which allows it to use any number of XML parsers... and they can even be used simultaneously.
These benhmarks are often purveyors of XML datasets as well.