Sync does not sync record numbers

I’ve been helping my wife with referencing her thesis.

(She uses Mac, I’m on Windows 10. We use Endnote X7.7.1.)

At some point - over a year ago - we synched her library to a new library on my laptop.

Secure in the knowledge that the libraries were synching, she’d do some work on a chapter of her Thesis, then I’d sort out the references - she’d often put them in quickly and I’d tidy them up with the correct bits in the correct fields etc.

She’s getting close to hand in time now, and we’ve discovered that there are quite a few references with completely different record numbers in the two, originally identical, synched libraries. This means that when I open a chapter on my laptop, I’ll find a citation with an author, year, and record number, but the library will have a different record number; and if I make changes because it was entered incorrectly, they aren’t synced on her laptop… The idea was that we’d both be able to work on the same documents, and use the same library. 

We need to get the libraries identical again. I tried copying one of the references on her mac, and syncing again. The reference was duly duplicated on both laptops - but the record number was different on the two computers.

What we’d really like is a Dropbox like system, where we can make changes on either computer, and the identical file will be synced, including all the correct record numbers. 

Any suggestions as to how best to get to that point? 

I suspect we might have to cancel syncing on both libraries, create a new version on one of the computers, sync it to the other - so we have two identical libraries - and then use this new version. I guess to stop them getting out of step we’ll have to be very careful to not update one library while the other is open - should this work?

I’ve spent about 4 hours trying to work out how to do this today, and we want to hand in at the end of the week. We thought Endnote would handle all the references without problems; we’re very disappointed.

Remember when syncing a library that yes, the record numbers will change.

Usually you would make sure to update citations and bibliography then move the document to the other machine. Then open the synced library and first update citations and bibliography in the document. This will match the citations, record numbers to the local library.

If you unformat the citations and then move, yes the citations will miss-match. Go back to the original machine, update citations, move the document and update once again. If that is no longer pssible, you will have to  match the citations.
You will have to go through them all, save and backup the document. Always have back-ups of your library and Word document as you work.

This can work, but you can’t unformat the document.  It will recognized the same references in the synced libraries or at worst, will use the integrated “travel library” versions of the citations.  Another trick is to routinely copy the the record numbers to the label field, and use the label field in your temporary citations.  This is described in the “little book”  pg 107


 To bookmark or send this article to a colleague, please use the following URL: http://endnote.com/kb/129310

To download, please click this link.

Hi,

I am having the same problem with unmatching record numbers across two different desktops.

Was the solution: ‘cancel syncing on both libraries, create a new version on one of the computers, sync it to the other - so we have two identical libraries - and then use this new version’ and then being careful to ensure syncing occurs correctly at the end of each Endnote session? Or did you find an alternative solution?

Thanks in advance.

As I said below, I copy the record numbers (using the copy/move/change tool) to replace the label field and then append a prefix and suffix to them (to make them unique as I find labels are often treated as “contains” rather than as “is” fields by the temporary citation matching - although they may have fixed that).

then change temp citation in preferences to use label rather than the record number.   I have also used the pages field instead, which doesn’t require regular copying.    

1 Like

I’m brand new to EndNote - have used RefMan for more than 20 years.

Isn’t the correct solution that the so-called sync actually sync properly? !

curiously, i set up a new library on one computer and synched it to  my Endnote account.  Then i created a ‘new’ library on another of my computers and synched (sic) this

none of the record numbers is the same as in the original ‘new’ library (which had not been touched in the interim)

I’m loathe to go further with EndNote and not be able to do create proper reference lists on different computers, which was a major reason why I even contemplated switching away from RefMan.

Hello,

Leanne is correct. Record numbers are not syncronized with the records. EndNote online does not retain record numbers, and record numbers in each individual library will be unique, and not syncronized through the online version. The method for transfering documents seamlessly between computers is to have them in the “formatted” state with a bibliography at the end, which can be reformatted on the new computer and the record numbers saved in CWYW will be updated to the library on the new compter. We do not recommend using the record numbers as unique identifiers in EndNote, unlike our previous programs like Reference Manager or ProCite where these numbers were more flexible. This is a limitation in the EndNote program.

Thanks,

Colin G.

Customer Care - Technical Support Representative
Clarivate Analytics
Phone: +1 800 336 4474
clarivate.com

The recommended method of keeping references properly synced seems FAR to complicated and subject to problems.

I’m thinking that I would be MUCH better off to turn off syncing in Endnote and keep my EndNote library in a folder that is synced to my cloud storage service. In that way, the record numbers will always match and the libraries will be nicely synced between all my machines.  

If my logic is flawed here, please enlighten me. 

Thanks very much.

Leanne’s suggestion makes great sense as a workaround. Any suggestions for a library that has 2,500 records in it? 

Thanks.

You can copy from record numbers to the label field on 2500 records, no problem, using the change/move/copy tool   

Actually, these days I rely on the Accession numbers (in the temporary citations in preferences) that come from pubmed for the articles.  It seems to find other reference types even if that field is empty, or you can sort on accession number and copy from record number to that field.  You can select and show any record where the field is empty (sort library on the accession field). If my colleagues also do this, we have no problem co-authoring papers!  

Thanks Leanne,

So rather than the “label” field, you’re using Accession Number in your temporary citations. And for references not including an Accession Number, you do a change/move/copy procedure to copy the record number into that field?

You mentioned earlier that when using the label field, you had to insert characters before and/or after the record number to ensure the citation linked to the correct reference in your library. Is this no longer needed when using Accession Number?  

Finally, to be sure I understand - I see how this would work like a charm for doing the copy procedure on all references in my library w/o accession numbers. It doesn’t matter that these references all have different record numbers on my different PCs. Once I establish the new Accession Numbers, those numbers get synced to the matching references on all my other machines, as well. Is that correct? 

Say I import some new references without accession numbers and don’t get around to copying over the record numbers for a few days. Meanwhile, through online sync, these references all exist, but have different record numbers on my office PC, home PC and laptop. When I get around to copying over the record numbers to the Accession Number field, should I always be sure to use the same computer to do that? Otherwise, I think there’d be a chance of getting duplicate numbers in that field for different references. Does that make sense?

Thanks!

to be honest, I haven’t updated the empty accession numbers, except for a few old records that I still cite, where I had to refind the records in pubmed and copy the accession or doi, so it could update record in the original library – I still wish endnote could combine duplicates, but then I would like the option to let me define WHICH record number would be retained!   

I personally haven’t had any issues and acc numbers appear to be unique, however,  99% of the time, I use my own computer or a computer with access to my library on the network drive (and yes, I know the developers recommend against a networked drive used for the library, but after 19+ years – I am absolutely an “old dog”. ).  I only use the cloud version of the library when I am traveling with my laptop and am usually not doing that much heavy editing of manuscripts.  Mostly retrieving PDFs for review, etc.  Even then, I try to remember to copy my library to the laptop before leaving, so that I can work offline.  

Thanks Leanne,

Your last comment about the cloud drive leads me to my last (maybe) question.

I have a single, large Endnote library that I do not share with anyone. However, I use it with 5 different PCs, depending on where I am. I like to have it on the HD of each computer. 

Basically, I see two options moving forward:

  1. I turn OFF Endnote’s syncing feature and put my library in a folder that IS synced with the same folder on my other PCs through a cloud drive service. That would keep the library synced across all devices, with matching record numbers, etc.

  2. I turn ON Endnote’s syncing feature and put my library in a folder NOT synced through my cloud drive service. This would also keep my library synced across all devices, but without matching record numbers, so I’d have to use your accession number strategy. 

#1 would be a bit less work. #2 would ensure I’ve got an up-to-date copy on Endnote’s server (which I rarely use, however). I’m leaning toward #1, but wondered if you’ve got any thoughts. 

Thanks!

Syncing folders thru cloud drive service absolutely requires that you NEVER sync while a library is open on any one of the machines.  So if you leave it open and forget, and it syncs,  and database files are open, that will absolutely corrupt the library.  I think that is just too dangerous.

I manually sync thru endnote on my main machine - occassionally, especially after adding records, so I have the backup.  

Leanne,

Thanks very much for walking me through this. I did notice a few hangups using the cloud drive strategy, as you mentioned. I ended up going with your Copy record number to accession number strategy, and it worked like a charm. 

You’ve been a great help. Much appreciated!

I have to wonder WHY this utterly insane decision was made not to include record numbers in syncing. It makes the entire sync/share feature… virtually useless. Essentially, it means you have a synced library which is not synced.

Working with formatted citations as suggested above is NOT a useful option in a program like Word, for several reasons:

  • It’s horribly slow. Having to wait 30 seconds for the entire bibliography to be recreated every time you need to add or change a citation in a document is completely unworkable
  • It’s prone to errors that will either cause mistakes in output or corrupted documents
  • It requires that EndNote is installed on all contributors’ machines

There is a reason unformatted citations are there – they are so much faster, safer and more practical to work with when several people collaborate on the same document.

But they also require record numbers to remain accurate across computers – which is not, it seems, possible. Why on earth would unformatted citations be made to rely on precisely the one piece of information that is not synced in a synced library? The sheer insanity of this boggles the mind!


Anyway, now that I’ve ranted and raved for a bit, perhaps someone can help me figure out a solution for my current problem:

I’m currently editing a book together with another editor. The book comprises ten chapters written by eight different authors. The final product must have one joint bibliography for all chapters at the end of the book. Most of the authors do not have access to EndNote at all and don’t know how it works. Myself and the other editor maintain a synced and shared EndNote library containing all references from all ten chapters. The library currently has about 6,000 sources, and it is not uncommon to have more than one source where author name and year of publication are identical (i.e., what will show up as Smith 2018a and Smith 2018b).

Several of the chapters are very long (some over 150 pages) and on average have perhaps eight or ten citations per page – so probably over a thousand in a single file. Working with formatted citations in these documents is completely impossible because it’s so slow. And of course, if an author, when working through a round of proofs, fixes something in a formatted citation by simply replacing the text, that change is obliterated the next time anyone makes any kind of actual EndNote change – something we cannot risk. So formatted citations are right out, and we’ve been working with unformatted citations as much as possible.

The chapters are currently in separate files, each file going back and forth between editors and authors for proofs; but in the end they will have to be combined into a single file in order to create correct formatting for citations and the joint bibliography across the entire book, before being imported into InDesign for proper typesetting. This will be my job, and it’s what has me really worried.

Since EndNote references have been added by two different people, using a synced library whose record numbers apparently don’t match as we had just assumed they would, it seems likely that many, many citations will end up having an incorrect record number in the unformatted citations in the document. This will prompt EndNote to ask me to manually choose the appropriate source in each case that cannot be unambiguously detected by author + year + record number.

In cases where there is only one possible match, this will mean that the final formatting will take a long time (manually confirming thousands of citations), but it shouldn’t cause any serious structural issues. If I could be sure there are no ambiguities, I could just delete all the record numbers, since EndNote will at least properly match an unambiguous citation by author + year alone.

In ambiguous cases, on the other hand, I will have to somehow figure out all over again which source is being referenced – something we’ve already done when adding the EndNote references, and exactly what the record numbers were supposed to keep track of to begin with. This is something that requires being able to see the context in which a given citation appears and checking which work by the person cited makes more sense in the context – but of course you can’t see the text in the document when EndNote/CWYW is working on updating the bibliography and asking you to select the correct source for a given citation.

Can anyone think of some way to achieve the goal of combining the ten files into one, hitting “Update Citations and Bibliography”, and then actually getting all the citations to match with the correct sources in the EndNote library without having to manually select them all – particularly the ambiguous ones?

Edit: I just noticed Leanne’s link to the ‘little book’ which explains how to use the Label (or in my case, Accession Number) field as a permanent record number. This gets me some of the way. We started out using a single library stored in a Dropbox folder, which caused us a fair amount of problems but at least meant that the record numbers were consistent; we only recently switched to syncing/sharing a library. Luckily my copy of the library is the one shared, so _most _of the record numbers in the document (hopefully) match mine. The only ones that won’t work should then be the ones that were added by the other editor after we switched over to a shared library