I have an evergrowing pile of unread but interesting paper in my download folder. With the size growing from several hundred MB to more than 2 GB, it is becoming hard to find a pdf file when I really need it.
Back in 2015, I was annoyed and decided to look seriously for some tools that would manage the mess. I was looking for something which could retrieve metadata when importing pdf and searches fast in the database when I provide only a couple of keywords and then name of corresponding author (“I need the Jessell paper on *digits*”). Mendeley seemed fine for these purposes then, so I spent some time and organized my files under it.
One month later, I noticed the mobile app of Mendeley stopped working after an update. Specifically, it kept logging itself out. I did not worry too much because I spent most of my time in front of my laptop, but I was surprised it took a couple of days for the support team to reply “Please be patient, we are working on that” in the official forum and then another 2 weeks to release a fix.
Having said so, Mendeley served me quite well after that, so I started to forget the mobile app event and to recommend it to my colleagues.
Then, it broke badly again in early June 2018. After an update of its desktop app, it refused to open any pdf and wrongly complained those files were corrupted. I checked Twitter and was surprised again that the bug was reported and seemed to be epidemic in late May, but the buggy update was still online. People were complaining missing highlight and messed-up references in their Word file, and in the end, it took them almost one month to fix this one.
With two events of slow response to bugs that affect the majority of their users, the technical support for something backed by a publishing tycoon seems to be too weak, and because of its instability, synchronization of files and highlights started to look more like a disadvantage than an edge. After all, synchronized files and highlights would be gone when service is down.
So, I am back on my search of tools recently, and here’s my brief opinion:
- Pdf rendering in ReadCube is still slow on my laptop. Though other functions are fine, a pdf reader preventing browsing is a deal breaker to me.
- Metadata retrieval becomes much better in Zotero that it was in 2014, and it uses default pdf reader which usually works fine. Since my need for reading on mobile device is not high, synchronization with Dropbox might suffice.
- F1000Workspace has an amazing web clipper and highlighter, but the display of highlight on their online pdf reader is far from acceptable, with highlighting overlapping with each other in a sentence. Its storage is completely online, so it might alleviate my shortage in disk space.
- The interface of Papers 3 is still as slow as I tried it 3 years ago, but the pdf reader works just fine.
- Mendeley is generally fine if it does not break. When it breaks, it’s serious and takes long to be fixed.
- The built-in search in MacOS does a good job in searching the text inside pdf, so perhaps no reference manager would also be an option.