I have been a Papers user since v. 1. It is my research inbox, because it remains best of class for:
1. automatically (more or less) matching stray PDFs to their metadata--i.e., giving me a full cite without typing.
2. renaming the PDF according to my favored convention and putting it in a folder with the right name--this way I can find it easily: across platforms, using Alfred etc.
Plus it has good library search. But personally, I think Papers has always been bad at most other reference-manager functions, so I use Endnote when I write.
The Papers 1-->2 transition was a disaster. The feedback was so bad that the developers took their support forum private to stop the negative word-of-mouth--and it's still private, btw. So I wasn't surprised when the 2-->3 transition was also a fail. Maybe even worse, since meanwhile the developers had sold out to one of the big publishing conglomerates, and were focused more on a PC version than on making the Mac app work.
Now (1/16), more than a year after Papers 3 release, bugs in Papers 2 have finally forced me to switch. On the plus side, match still works. The minus side outweighs this gain, however:
1. The grey-on-grey interface with limited customizability is not readable by my middle aged eyes. That's the main problem with the interface, although as the top review also commented, there are numerous annoying lapses. For example:
The match button is sometimes there, sometimes not. Tabs are not tabs. There are even more clicks than before to do basic tasks--e.g., you can't just enter an author's name--you have to enter it as a search term, then move the mouse to select it from the list. The uncustomizable top bar gives quick access to some functions I use all the time (e.g., go to Library), but is cluttered with other functions I never use (e.g., author list).
I could go on, but you get the idea: the Papers 3 interface adds like 10% to the frustration of whatever you're trying to do.
2. If I stick with the file-renaming function, Papers will not sync across platforms. When sync is enabled, Papers 3 gives all PDFs unintelligible names. Thus I can either have sync or file renaming, not both. So now my research is stuck on one machine.
There is a new "reading list" function which allegedly allows a limited number of papers to sync across platforms. That would be a minimally adequate substitute for sync. Alas, it doesn't work--only about 25% of attached PDFs get uploaded.
3. Large numbers of minor additional annoyances: "Citations"--a Papers add-on--is like whackamole: no matter how many ways I turn it off, it keeps on re-appearing. There doesn't seem to be an automatic warning about duplicate copies any more. I'm getting random hangs. Occasionally, a imported PDF is disappearing. Occasionally an archive of an article's webpage is getting imported--wtf? Database "checkpoints" are constantly being created, for what reason I know not; I have 250 MB of them just from today.
Software is usually just a product. I guess I feel close enough to my research to feel personally aggrieved by the thoughtless, lazy, shoddy design of Papers 3. Increasing the amount of frustration and wasted time in the world is unethical. Eventually I'll carve out some time, do some research, and move my 6K item library to software that respects the user.