You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi Ross, thanks for bringing the incorrect publisher to our attention.
The publisher given by the CrossRef API for the "Open access and altmetrics" paper is "Wiley-Blackwell". It appears that "Wiley-Blackwell" is sometimes being listed as "International Studies Perspectives 2015. DOI: 10.1111/Insp.12104" in SHARE.
We are looking into the root of the problem and will keep you updated.
@rossmounce Thank you again for report this issue.
We have put a temporary fix in place for your article, the publisher will now show up as originally reported by CrossRef.
We are going to remove the incorrect metadata from the other articles until we have a permanent solution in place.
I'll close this issue once we've finished removing the incorrect metadata.
Activity
rossmounce commentedon Jul 11, 2017
Not just my article either. Lots of articles seem to be incorrectly tagged with the metadata from that particular ISP publication!
https://share.osf.io/discover?publishers=International%20Studies%20Perspectives%202015.%20DOI%3A%2010.1111%2FInsp.12104&q=open%20access%20and%20altmetrics

laurenbarker commentedon Jul 11, 2017
Hi Ross, thanks for bringing the incorrect publisher to our attention.
The publisher given by the CrossRef API for the "Open access and altmetrics" paper is "Wiley-Blackwell". It appears that "Wiley-Blackwell" is sometimes being listed as "International Studies Perspectives 2015. DOI: 10.1111/Insp.12104" in SHARE.
We are looking into the root of the problem and will keep you updated.
chrisseto commentedon Jul 12, 2017
@rossmounce Thank you again for report this issue.
We have put a temporary fix in place for your article, the publisher will now show up as originally reported by CrossRef.
We are going to remove the incorrect metadata from the other articles until we have a permanent solution in place.
I'll close this issue once we've finished removing the incorrect metadata.