1) Warn on broken dragndrop 2) complete metadata
1) That drag'n'drop for PDFs is broken in linux is known, so it would be nice if recognizing my useragent as linux, zotero wouldn't do whatever it does when I tried to drag and drop (freak out, make it look like my library is deleted). A little popup "this won't work for you!" is much preferable to exploding.
2) On some cites (example: The Lancet) I'll be able to save a citation, but the metadata will be incomplete (missing page numbers). It would be nice to have the same function which takes retrieved PDF metadata and finishes it with scholar be able to pull these missing bits in.
2) On some cites (example: The Lancet) I'll be able to save a citation, but the metadata will be incomplete (missing page numbers). It would be nice to have the same function which takes retrieved PDF metadata and finishes it with scholar be able to pull these missing bits in.
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736%2809%2960250-6/fulltext
I think you can take as given though that some articles will initially have e-pub ahead of print and then acquire page numbers later. If not Scholar, pubmed or another index.
Looking through my library from before when I started checking to make sure I had the metadata, I see a few other embedded RDF which got some of the data (title, date, author), which is enough to find the article.