I think my question is general enough to not be version-specific, but if not, we're running Foundation 2010 with minimal development (SPD, Vis Studio, etc.) capability.
I have a 3500-document repository that recently got converted to a SharePoint library. Our metadata is almost non-existent and we're relying on full text document searches to find content. Since keywords were not part of the initial repository, they were created from scratch in a spreadsheet a few years ago. Col 1 is the document name, Col 2 is a column delimited list of keywords. This is the data I'm trying to populate into SharePoint's metadata columns.
Here's where I'm having a hard time wrapping my head around things.
Let's say that an Excel record is this:
document_foo.doc; keyword_foo1,keyword_foo2,...,keyword_foo8. I plan to cut-and-paste the records from Excel to a Datasheet view of my library.
My question is this: if I paste a list (
keyword_foo1,keyword_foo2,...,keyword_foo8) into an empty metadata column does that whole string of 8 keywords become the sort value? or would Sharepoint recognize
keyword_foo1 as a seaparate and distinct keyword from
keyword_foo2? That's out intention, but I suspect it won't.
But if I create ten distinct keyword columns, and reformat the single Excel keyword field into up to ten separate keywords in ten separate columns, I end up potentially missing something if I search in
keyword_foo2 and the value lives in
keyword_foo4. With that, I run into the potential of having a list of thousands of keywords from which users get to choose when creating a new document--that seems needlessly unweildy to me.
In a perfect world, I'd like to see multiple choice check boxes created from the keyword list; but then I'd have to add a carriage return instead of the spaces in the filename, right? And I'd have to figure out a way to prevent a list of a few thousand keywords from accumulating over time.
I'm a little too close to the project too see it clearly. Does anyone have any suggestions or strategies for me?