Sync across multiple InSurge repositories
FYI that there was a breakout session at last week's Access conference on sharing of social content across bibliographic systems. The InSurge model may work well for this.
I started to take a look at the existing Alpha1 InSurge code but uncovered several questions. I'm not a coder at all (I can talk basic pseudocode concepts) so I apologize if the answer are in the code itself.
I took a look through the insurge-client.php and insurge-server.php but don't see anything in there that looks like a "sync InSurge repository" function, though I did see room in ../config/insurge.ini to enter an ID and key for a parent InSurge server.
1. Is the piece that syncs InSurge data across repositories built yet?
Also, from my review of the code, I'm not clear:
2. How to refer to the same edition of an book across systems (seems to refer to a bnum which looks like a local systemID
3. Whether a uid (userID) is or isn't shared across InSurge repositories.
--
Steve Toub
Product Manager, BiblioCommons
Steve, great questions. I chose not to include the code I've written for the actual Insurge sync in the beta because I feel that there are a lot of questions I still need to answer, and I wanted to get some real-world data to work with while I build out that functionality.
The specifics you're bringing up are good examples of issues that I'm trying to resolve. The table you're looking at is used to track local social data only. I'm planning on using a hierarchical approach to key "works", starting with ISBN, then stdnum, then LCCN. I have a feeling I will need to incorporate a service like xISBN in the future and as Insurge really grows, I'll need to work out the fiscal issues with OCLC. I would also like to work with Tim at Librarything--they've got a very substantial database of 'works' that Insurge might take advantage of.
My thought on user IDs is that once data leave an organization, its userID is scrubbed. The only originating identifier is the site ID so that you could subscribe only to content from the institutions you wanted content from.
Getting feedback like this is very helpful. There are a myriad of small little details like this and I'm sure I haven't thought of them all. I would love to talk with the folks at BiblioCommons about this in more depth because I think we're in the same ballpark.
Thanks, John. I'll follow up with you over email.