Dear all,
I have regretted upgrading/updating to EN V20 (currently on 20.6). I am getting the nag screen to “upgrade”/update to V21. I am quite skeptical about upgrading as I have not been impressed with V20 and fear it would only get “more broken”. There are many flaws, cosmetic issues, and functional quirks, and the absence of responsiveness by the EN team does not engender confidence. I’d be interested in advice on if there is any value to upgrading to V21. Can someone make the case?
Cheers,
Chris
I down (up?) graded to ver 9 when v20 came out, and i gave v21 a try. it fixed some but not all of the problems, so i’m sticking w ver9 as long as i can–or until they restore functionality that was lost from ver9 (and some from v8 as well, esp. the ability to search & replace in the field attachment field)
Hi @marcbecker,
Are you using macOS by chance, and if so, which version? I am trying to see whether EN X9 works past Monterey. I’ve been sticking to EN X9 because of all the issues of V20&21 reported. I would like to upgrade my macOS, but don’t want to be forced into EN 21 because of an OS upgrade. I posted about this long ago, but received no responses (Endnote X9 and Ventura on M1 or M2 silicon, macOS 13+?).
Thanks,
(another) Chris
Ver9 on macOS Sonoma 14.1.1 with a MacBook Air M1 (2020) chip.
I presume that at some point ver9 will quit working w/ mac upgrades, and I bought ver21 naively hoping that it would correct more of the problems w 20. I’m dreading the day when i can no longer use legacy versions and they haven’t fixed these problems. very disheartening.
1 Like
Hi Greeley. I can only share my personal experience. I can second marcbecker. To date, X9 still remains the best release ever as far as responsiveness, features and bugs are involved. I used it on Windows and Intel macOS. The performance was especially good on Windows and occasionally laggy on macOS with massive databases.
However, PDF import filters work better in 21 vs. X9. Now, take this with a grain of salt, as it may simply be the result of improved online databases.
I more or less skipped version 20 (pre-ordered it, read about its flaws, never installed it, asked for a refund and never got a response, …) and to cut a long story short, used it briefly and upgraded to 21. With my relatively brief version 20 experience, I prefer version 21 as it is better supported. I no longer use it on Windows, however. The Sync behavior is better with version 21 compared with 20. Previous update cycles considered, we may be close to the next iteration of EndNote, so it may be worth waiting.
UPDATE:
P.S.: Since my post, I took another objective look at Bookends. I exported a large database to Bookends. Bookends is slower to load the library when starting the program. From there, however, things are more or less even, but with significantly fewer annoying pauses in Bookends than in EndNote. EndNote Sync has become very slow and further deteriorates performance. Considering EndNote’s performance deterioration since its peak with X9, Bookends is clearly the better solution for me as a macOS user, not to mention its integrated browser and ability to perform very specific searches using SQL and Regular Expressions. EndNote’s PDF annotation tools may be better than Bookends’, but I use an external viewer to annotate anyway.
Having gone from 18 to 20 just in time to get the free update to 21 given that there have been no major improvements for over the last decade or so (just a shuffling of features and bugs it seems) I won’t move to 22 unless the whole of this forum sings it praises. As you can usually upgrade for 2 (or sometimes 3) versions, I will wait until V23 or V24 before upgrading again (if at all)
I sometimes think Clarivate just put some part-time developers on for a few weeks to change things so they can justify a new version so they can charge money.
You are correct. The changes are mostly platform specific rather than functional, e.g., supporting 64-bit architecture, supporting Apple Silicone, etc.
For the next iteration, I would expect integration of AI tools that are now all the rage across all industries that, if correctly implemented, could provide a compelling reason to upgrade.
AI is a marketing word with no real basis in reality. It is just that the data sets are getting refined (and a lot bigger) and the processors more powerful with a lot more memory (my PC runs 128GB RAM) so they can do in seconds what would have taken a month before.
I have a friend who programmed weather analysis SW for the UK Met Office many decades ago. He told me that they have refined the mathematical models and the datasets have increased by orders of magnitude, at the same time the speed of operation has gone up at a faster rate. No magic or intelligence, just they can do a lot more, a lot faster.
Sort of like (made up numbers) Instead of running a 12 scenarios in 12 hours, they can run 100 scenarios in 12 minutes. It also means they can, in terms of weather forecasting, run in real time. IE 20 minutes, not 20 milliseconds. It is the same with 99.999% of other “AI” out there. Not to say there are not 0.001% of research projects that are not looking at real artificial intelligence, but that is something else.
Sorry, RANT over 
I am not sure where EndNote would benefit from, where is commonly called AI? Where would you see it being applied in Endnote? This is a serious question.
One use case would be better heuristics for finding related citations that would enhance say metaanalyses. These do not necessarily need to run on your local machine, but on servers. So you can imagine there will be a subscription service for their use.
That is a non-starter for a LOT of users. It is one thing going on-line to search a database, but another to use an online system for that sort of search. For a start, it is likely to be run on a server in a hostile country
Also you are not talking artificial intelligence but more along the lines of what I mentioned above and that is used in many applications now. In particular, photo and video editing software.
So I think they could do what you suggest on the local computer. It might be helpful for quite a few users.
What might be better is that sort of system built in to the Catalogues of the Libraries when you put in a request So you can do the search with “exact match” to get the obvious ones as we do now and then again with “AI Match” to get the fuzzy matches.
I’m not posting to argue. PubMed is hosted in a country considered hostile by other countries. Open AI runs on Microsoft Azure, Google’s, Amazon’s, on their own. Generative AI in all of its iterations is the current frontier in information technology. Yours and my arguments are irrelevant.
I take your point, but, I only use them when I go to them to use it. I don’t have them as part of an app installed on my computer. Clarivate can put the AI into endnote to work locally. It does not need to talk to any server. The more useful place to put it, is in the external Library catalogues, like the British Library or Library of Congress. Again that would be in their database system.