Failure to Import PDFs

kdonovan11

I am a fan of the web import feature (the Chrome plugin) but it seems to automatically import very few PDFs these days. I have access to all the PDFs through a web proxy, but it seems the plugin cannot liaise with the systems of major publishers (including Project Muse, Taylor & Francis, JStor and others). I think this is because they require an extra click when opening a PDF from the article's webpage. For example, JStor requires clicking that you approve their Terms & Services. T&F requires you to choose between a standard or enhanced PDF.

Is there anyway to fix this?

adamsmith

could we get a couple of URLs of specific articles (ideally exactly as you see them, including your proxy)

kdonovan11

Sure:

https://muse-jhu-edu.proxy.lib.umich.edu/article/627377#.V7BmdSpSpCw.twitter

http://www-tandfonline-com.proxy.lib.umich.edu/doi/full/10.1080/03085147.2012.718625

The standard URLs would be:

https://muse.jhu.edu/article/627377#.V7BmdSpSpCw.twitter

http://www.tandfonline.com/doi/abs/10.1080/03085147.2012.718625?journalCode=reso20

adamsmith

if you hover over the URL bar icon, what does it say for both of these? I suspect that Zotero can't handle the hyphenated URLs used by U Mich's proxy. (Most proxies don't do this; e.g. I have http://www.tandfonline.com.libezproxy2.syr.edu/doi/abs/10.1080/03085147.2012.718625?journalCode=reso20 , which works fine).

Project Muse is broken generally, though. I can look into that, though it may take a little and won't help as long as you're going through a proxy.

For you, using VPN from off-campus instead of the we proxie should help: http://its.umich.edu/computing/wifi-network/vpn

kdonovan11

Okay, you're right that using the VPN solves the problem for T&F (though not for Muse, since it's broken generally). I can aim to use that in the future.

I'm not sure what you mean w/r/t "hover over the URL bar icon", though.

kdonovan11

For what it is worth, the PDF did not import for either of these URLs, either (lacking dashes):

http://www.jstor.org.proxy.lib.umich.edu/stable/646233?seq=1#page_scan_tab_contents

https://culanth.org/articles/206-making-democracy-count-opinion-polls-and-market

adamsmith

When you hover over the "Save to Zotero icon" (that used to be in the URL bar, though isn't anymore, i.e. the icon that changes on the type of content), Zotero tells you which translator, i.e. script, it's using on the page you're looking at.
E.g. for https://culanth.org/articles/206-making-democracy-count-opinion-polls-and-market, it just says "Embedded Metadata", i.e. Zotero just uses generic data on the page, which is why it misses the PDF.

For JSTOR, you have to accept the download conditions once per session. I don't believe we can work around that, sorry.

kdonovan11

I see.

For the T&F with proxy, it says: Atypon Journals.

For T&F without proxy, it says: Taylor and Francis+NEJM.

For the Project Muse w/ proxy, it says: Embedded Metadata.

For JSTOR w/ proxy, it says: JSTOR.

adamsmith

yeah, so that's all as expected -- as you see, Zotero doesn't recognize T&F when you go through the proxy and as such and so you don't get the PDF. The same will be true for Project Muse even after it's fixed. For JSTOR, see above.

adamsmith

(Project Muse is now fixed and should work with VPN)