OpenSearch is a bad standard. It was written at Amazon and published as a “standard” with no consultation or committee. It was just “hey folks, we’re Amazon, so give us results the way we want them."
One example of badness is the “language” parameter, which says the search client desires search results in the desired language. Pretty fuzzy, but appears to mean filtering the documents to only those in that language. What about the language used to tokenize and lemmatize the query? What about the locale for that? What about other filters? Every search engine separates filtering and ranking queries. What about specifying the syntax used for the query? So, just junk. It is substantially inferior to the federated search we built at Verity (using SOAP, yuk). If you want federated search, use GData. That is a pretty good design, even though Google has mostly moved on from it for their proprietary APIs. https://developers.google.com/gdata/ wunder Walter Underwood wun...@wunderwood.org http://observer.wunderwood.org/ (my blog) > On Apr 26, 2019, at 6:20 PM, Erick Erickson <erickerick...@gmail.com> wrote: > > None. See: https://issues.apache.org/jira/browse/SOLR-2143 and > https://issues.apache.org/jira/browse/SOLR-2419 for the last times I know of > that any consideration was given for supporting it and the reason why not. > > Best, > Erick > >> On Apr 25, 2019, at 1:02 PM, DESDCS <daniel.s...@hrsdc-rhdcc.gc.ca> wrote: >> >> What version of OpenSearch specification does Solr currently uses? >> >> >> >> -- >> Sent from: http://lucene.472066.n3.nabble.com/Solr-User-f472068.html >
