Renaming & changing item type in collection(s)

richard p

When the search engines & others fail to recognise metadata on a pdf, I still import the pdf into Zot & then 1) generate an item type (eg Journal Article or Report), & then 2) Add relevant data (title, authors, yr, etc) in RH column. 3) I then attach pdf as child. (& 4) I delete (move to trash) the original pdf)

Assume pdf is in "My library" & also in "Collection A". If I rename etc as above in "My Library", these changes seem to be copied to relevant item in "collection A", but NOT vice versa. this is based on a few test cases - can you please confirm (& if correct, could there be an option for a generic change in one collection impacting all collections, altho accepting there may be occasions when not desirable (??) - hence an option button)

Cheers

R

adamsmith

No, that's not correct. If you have a single item, all changes you make to that item will affect that item globally.
The only way a change to an item in a collection would _not_ affect it in "My Library" was if it was a duplicate rather than the same item.

I also don't understand why you have to delete the original PDF. If you move it to a parent item that's all you'll need to do. It doesn't copy.

aurimas

I also don't understand why you have to delete the original PDF. If you move it to a parent item that's all you'll need to do. It doesn't copy.

I think richard is referring to the downloaded PDF file on his Desktop/Downloads folder. You should be able to use Shift+drag to move the file into Zotero instead of copying it in (it's a little finicky though, at least on Windows)

richard p

adamsmith & aurimas

Hi - thanks for replies. Yes - aurimas is correct the pdfs are physically imported when the FF/Zot link doesn't automatically download the pdf. 2 things I will check:
1) aurimas - that a shift+drag will allow me to add pdf as a child to "parent report/journal/whatever" (various L & R clicking whilst dragging does not work); &
2) that a change in "Collection A" is replicated globally - certainly i had a couple of exceptions to this rule when i was checking/testing ...

Cheers

R

adamsmith

exceptions to 2) are pretty much technically impossible. If you saw what looked like one, you were almost certainly working on a duplicate. This is literally the same item, regardless of whether you look at it in a collection or in My Library.

aurimas

aurimas - that a shift+drag will allow me to add pdf as a child to "parent report/journal/whatever" (various L & R clicking whilst dragging does not work);

If you're dragging from the Desktop, refresh the Desktop afterwards (F5). Windows 7 (I think XP used to have similar issues too) sometimes shows you items on the desktop that are no longer there (at least that's the issue I'm seeing on my end)

richard p

Thanks guys - I will confirm tomorrow (UK time)

R

richard p

aurimus - hi -

I have accessed a research gate page with opportunity to download paper (https://www.researchgate.net/publication/259126318_Integrated_Landscape_Initiatives_for_African_Agriculture_Development_and_Conservation_A_Region-Wide_Assessment)

In FF with Zot open, I click on rectangular page icon in url bar - I get title etc of parent & what apperas to be child snapshot. So I manually download "To Zotero" & get pdf listed as parent item (journal item)

I then "Shift +drag" & get the "no go" black circle with diagonal line.

Interestingly, I was doing similar earlier (not on research gate, but existing pdfs in my middle column of Zot) & I couldn't get the shift+drag to work, but kept trying with progressive pdfs, & it suddenly "clicked" & worked - ie seems v sensitive to mouse position (dreaded Win 8.1). But have tried many times on above RG paper & no joy

Cheers

R

adamsmith

you have "retrieve metadata" checked when downloading the PDF (otherwise the PDF wouldn't appear as a journal article item type but as an attachment item type). Since you already have the metadata, that doesn't make sense. So download the PDF with that box unchecked, then drag it to the Zotero item.

You don't need to hold shift for that. Shift is only relevant to
a) move items into Zotero from your file system
b) move items between Zotero collections

It won't have any effect when attaching a PDF to a Zotero item.

richard p

adamsmith - yes - I do have the box ticked - will try again tomorrow

Cheers

R

adamsmith

that box makes perfect sense if you don't already have an item with the metadata. So it depends on the specific situation.

richard p

Agreed - unfortunately predicting which files will "download seemlessly" & which won't is the question ;-)

Thanks

R

aurimas

But, presumably, at the time you're downloading the PDF, you already failed to import the attachment via URL bar, so you already know what you want.

On a side note, it's a shame that ResearchGate doesn't offer any good metadata export (they offer COinS, which, for some reason, is not working on Chrome). We could write a dedicated translator for their site, including PDFs.

richard p

"But ..." yes - altho that requires I think!

As a biologist I do occasionally use RG (all my papers are on there for free download), but it is slightly quirky. In a similar vein, I am suprised that google is still limiting searches from Zotero ...

Ultimately tho, you guys are doing an amazing job

Thanks

R

aurimas

I am suprised that google is still limiting searches from Zotero ...

Can you elaborate? We fixed some issues with retrieve metadata, so if you're encountering problems there, we would want to know.

richard p

Because of my work patterns, when I have access to decent download speeds I will frequently downloads 100s of pdfs from WoS etc, & then seek to collate & file when in the field or when time allows. Last year, google was pretty rigorous about limiting searches to ?100 per 24 h period. I have noticed that it now requires "captcha" responses (annoying when setting a large job in motion only to return later & finding such a response required) - after a few of these 'responses' google gets pissed off & tells me taht is enough for now! But I have now at least got past the stage when I started out using zot of havong a good few 1000 pdfs to have their metadata extracted - so (with a bit of patience) not too much of an issue for me now.

aurimas

Last year, google was pretty rigorous about limiting searches to ?100 per 24 h period. I have noticed that it now requires "captcha" responses (annoying when setting a large job in motion only to return later & finding such a response required) - after a few of these 'responses' google gets pissed off & tells me taht is enough for now!

Right, so that's the improvement (instead of failing, Zotero now relays the captcha request from Google), except that I have not been able to hit a hard block myself even after 10 captcha requests. I haven't tried this for some time though, so perhaps this has changed. There's not much Zotero can do beyond what we already try.

It's a bit odd that you would have that many Google queries though, since if there is a DOI in the PDF, Zotero won't even talk to Google. Is it common for these PDFs to not contain a DOI? I suppose you may accumulate all of the non-DOI PDFs, since Zotero now proceeds with metadata retrieval even if you hit a block with GS. So, on a first pass, it will retrieve metadata for all PDFs containing a DOI and then you're left only with the ones that need to go through GS.

Anyway, just curious. We probably can't improve this for the time being.

richard p

As I said earlier - you guys are already doing a brilliant job, & I find it pretty amazing that you can "assimilate & accomodate" such very diverse sources (& presumably formats) of data. My thesis was written pre-word-processor days, so the advent of MS Word (after WordStar) & being able to cut and paste the 50 key references was an amazing liberation. Problem now is 5000 key refs & another 5000 I haven't read!!

But in answer to your Q, quite a bit of the "grey litt" lack doi's (even world bank & other "international organisations" do not provide relevant metadata - which is so stupid (& time-wasting)....), and also the scientific litt from earlier yrs where papers are simply scanned in. Roughly 1/3rd of my papers are still n pdf format with no metadata.

Regards

R