I'm like others -- I'm Googling every few weeks to find a way to do this. I'm amazed that a solution hasn't been implemented yet, given that a large fraction of articles has incorrect metadata...
This is really difficult to accomplish without introducing errors -- something that, in my opinion, is worse than incomplete metadata.
Zotero developers are to be commended for their careful approach.
A few of the many examples of the problems faced when trying to update metadata:
Elsevier: Most articles in journals are not assigned page range numbers but an article-number. However, quite a few journals are fully paginated. Early ahead-of-print articles (before being assigned to a volume) will have a page range that is really the number one followed by a "-" followed by the number of pages. Once the article is assigned to an "in-process" volume the article may have true page numbers assigned but may (for a short time) retain false pagination as the page range. How does anyone know when an update to metadata is complete and when you must wait for another metadata update to what appears that it might be complete? Elsevier pushes "updates" that both contain new metadata for some articles but that also contains the same old incomplete metadata (but flagged as new). Further, for several years now, most journal volumes aren't separated into issues. However, some journal titles are separated into issue numbers. Without a way to tell if the journal is divided into volumes and issues or simply into volumes how do you know when the metadata is complete. Sometimes articles are in a preliminary status and assigned to a volume (of a journal that actually has issues) but not yet assigned to an issue.
Emerald Group: Similar problems to what is mentioned above for ahead-of-print items with the further problem that within some (maybe most) journals; when the article is assigned to an issue it might receive a second valid DOI. For many weeks the old DOI will take you to the ahead-of-print version while the new DOI will take you to the version with complete metadata. Only later will both DOIs connect to the same version of the article. Using the "old" DOI to fetch new metadata will not work until the old DOI points to the newest version.
There are similar but not so alarming complications with Springer, Sage, and Taylor and Francis journals. There are probably similar but different problems that don't come to mind immediately.
The metadata that is pushed to databases may seem (judged by the date of the push) to include updated metadata but sometimes the blank metadata field will be replaced with other useless junk such as "ePub", "PP", "pending", etc. Emerald Group has several terms that they use for ahead-of-print status and their update may be a useless change to a different term for a-o-p status.
Upgrade Storage
This is really difficult to accomplish without introducing errors -- something that, in my opinion, is worse than incomplete metadata.
Zotero developers are to be commended for their careful approach.
A few of the many examples of the problems faced when trying to update metadata:
Elsevier: Most articles in journals are not assigned page range numbers but an article-number. However, quite a few journals are fully paginated. Early ahead-of-print articles (before being assigned to a volume) will have a page range that is really the number one followed by a "-" followed by the number of pages. Once the article is assigned to an "in-process" volume the article may have true page numbers assigned but may (for a short time) retain false pagination as the page range. How does anyone know when an update to metadata is complete and when you must wait for another metadata update to what appears that it might be complete? Elsevier pushes "updates" that both contain new metadata for some articles but that also contains the same old incomplete metadata (but flagged as new). Further, for several years now, most journal volumes aren't separated into issues. However, some journal titles are separated into issue numbers. Without a way to tell if the journal is divided into volumes and issues or simply into volumes how do you know when the metadata is complete. Sometimes articles are in a preliminary status and assigned to a volume (of a journal that actually has issues) but not yet assigned to an issue.
Emerald Group:
Similar problems to what is mentioned above for ahead-of-print items with the further problem that within some (maybe most) journals; when the article is assigned to an issue it might receive a second valid DOI. For many weeks the old DOI will take you to the ahead-of-print version while the new DOI will take you to the version with complete metadata. Only later will both DOIs connect to the same version of the article. Using the "old" DOI to fetch new metadata will not work until the old DOI points to the newest version.
There are similar but not so alarming complications with Springer, Sage, and Taylor and Francis journals. There are probably similar but different problems that don't come to mind immediately.
The metadata that is pushed to databases may seem (judged by the date of the push) to include updated metadata but sometimes the blank metadata field will be replaced with other useless junk such as "ePub", "PP", "pending", etc. Emerald Group has several terms that they use for ahead-of-print status and their update may be a useless change to a different term for a-o-p status.