Search (SRU) API

• How do we configure an API query to return the same (or very similar) results as a manual search via <http://www.isni.org/search>?
• Are there parameters to filter queries by identity type: person or organization?
  • Can we use the sort function (http://www.loc.gov/standards/sru/sru-1-1.html#sort) for this? To use XPath of "personalName" or "organisationName" for parameter "path" and "omit" for parameter "missingValue"? (but what should we fill in for "schema"? Is it pica, isni-b, or something else?) (LM)
    
• Where are creation class codes defined?
• What’s the default result sort based on?
• How does sort by dc: properties relate to entities with multiple titles/publications?
• page 2 "Example search by name keyword (pica.nw):" Does not explain what "pica:nw" means. (ICQ)
  • "pica.nw" is the index code used in the URL for searching by "name keyword"; "pica.na" is for "name" and "pica.isn" is for "ISNI number" (LM)
    
• page 5 "decade of publication" Doesn't say what standard it's using by searching "197" for 1970-1979. for the country it instructs us to use and ISO standard, but for date it doesn't (ICQ)
• page 6 same for "dates of identity" do they want the dates in any particular format? (ICQ)
• page 8 states "Freely available ..." and gives some examples (e.g. freely available titles include those from VIAF) ; it doesn't explain what would be the NOT "freely available" information for each type (ICQ)
• Is there a cap on number of queries/records retrieved per day? I got returns with blank <srw:recordData> using one computer but with full records using another computer. When I use the manual search interface (http://www.isni.org/search), it says "Your data limit has been reached". (LM)
  

Submission (AtomPub) API

• How are we using local identifiers in our requests?
  
  • Do we need to avoid duplication between institutions within PCC?

• Lacking inverted names for co-authors, what’s the best way to incorporate this data?
• Is there some Chrome equivalent to the recommended testing app, HttpRequester?
• Is it best to start with a manual load of a tab-delimited file, or start with AtomPub XML directly?
  • Answer: Janifer Gatenby recommends no more than a "glance" at tab-delimited; starting with a manual upload of XML, instead, and then moving on to direct use of the API
• Page 16: the section on replacing metadata confused me a bit, especially this part "There is no difference between this request and a request for new assignment. The ISNI metadata given (identityInformation – identity) must be complete as it is used to replace existing data in the case of a match on local id (requestorIdentifierOfIdentity)." (ICQ)
• Page 24-25 don't understand the requesting status of "legal deposit" and then removing it (ICQ)

External documents

Harvard google doc: https://docs.google.com/document/d/1-qEnZiZ8O88N3trwiktLOShXDOlfQqcIg5FyGjhXXrY/edit