Sorry if this has been answered elsewhere, but is there any idea on zotero compatibility with the new apple M1 chips as the Mac transitions away from intel?
A question was asked whether Zotero is compatible with Apple's new M1 chip. The answer was:
"Apple Silicon Macs should run all existing Mac software (except for kernel extensions and virtual machine apps) from day one using Rosetta 2.We intend to support Apple Silicon natively in a future version, but that won't happen before sometime next year.
However, I do not fully understand this answer. I urgently need a new MacBook, but I want to make sure Zotero is compatible for M1.
As I understand it, Zotero is not yet compatible with M1, but we are working on it. However, we do not yet know when Zotero is compatible. Is this correct? If so, do you have any idea when Zotero will be compatible for M1?
As I understand it, Zotero is not yet compatible with M1, but we are working on it.
No, that's not correct.
As I say, Apple Silicon Macs should run all existing Mac software, and that includes Zotero. Many people are already happily using Zotero on M1 Macs.
The one thing to be aware of is that, if you use Word, there's a minor workaround required for Word integration, but once you do that it works fine. It's not a reason not to get an M1 Mac now if you want one.
A future version of Zotero will be updated to run natively at full speed on M1 Macs, but that's irrelevant for current usage. Most software that hasn't been updated to run natively on M1 Macs will still actually run faster than on Intel Macs (after a ~10-second delay the first time you start it) simply due to how much faster the M1 Macs are.
I do not see how comparing Intel and M1 Macs here is relevant. Word works perfectly well with Silicon and I want it to run on Silicon. Switching it back to Rosetta does not really sound like a great solution, and it's actually a step back for M1 users. And even if it runs faster than on Intel, it does not mean that the solution is more stable than having a natively supported app.
@brendan2: I think you're misunderstanding something. The posts above are months old. Current versions of Zotero work perfectly fine with Word running natively on Apple Silicon. Zotero still runs under Rosetta and will for a while for technical reasons, but it still runs as fast or faster than it did before — that's all I'm saying above.
@brendan2, I use Zotero on my daily driver (M1 MacBook Pro) and have no issues using it within Rosetta and native Silicon Word. They work as well as they ever have on my old Intel Mac or on my Windows desktop.
If anything, it is surprising how well everything works given that Zotero runs through an emulation layer.
@dstillman sorry for digging up this old thread, but you said Zotero will add native M1 support “sometime next year” which is 2021. Is there any timeline on this?
While performance under Rosetta is fine, Zotero has been the only Intel app remaining on my Mac for quite a while now so I wonder when it will become a universal binary as well.
Did you try restarting your Mac as recommended in the support link you provided? This happened to me when I first used my new MacMini I found the support page and Rosetta was available after a restart.
@Cryss: You should ask Apple about that. That string exists for some reason (potential patent disputes?), but as far as I know Rosetta isn't actually disabled anywhere, so that's almost certainly a bug.
Hey guys! Just a quick reminder, LibreOffice 7.2.0 for Apple Silicon was finally released… which means Zotero is now the last non-native link in many academics' toolset chain.
I finally have access to M1 machines, and will do some tests to see how it gets along with Zotero 5.0.96.3 ASAP and let you know how it went. In the meantime, can we get an ETA on native AS support?
"which means Zotero is now the last non-native link in many academics' toolset chain."
I agree – all other software I use runs natively on M1 (Firefox, Office, Libre Office, Latex). May I ask which reason hinders compiling Zotero for M1 (Apple Silicon)?
May I ask which reason hinders compiling Zotero for M1 (Apple Silicon)?
Zotero is currently based on an older version of Firefox from before they made massive changes to the platform, and Zotero requires extensive technical changes to work on current versions. Moreover, even current versions of the Mozilla platform on Apple Silicon don't yet provide everything we need, since we use functionality that Firefox itself doesn't use on macOS.
We've been working on rewriting all the parts of Zotero that need to be rewritten for later Firefox versions, but we have to balance that work with other development that benefits everyone, such as the new PDF reader and note editor. Those of us who use Macs are as excited about Apple Silicon as anyone, but the fact remains that Zotero will run faster on M1 Macs under Rosetta 2 than on most or all Intel Macs, so this just isn't a huge priority.
(That said, once Zotero 6 is out, getting Zotero running on a current Firefox version, and ultimately Apple Silicon, will be a major focus. There are at least a couple 14" MBPs on order between us, so we have somewhat of a vested interest in this.)
@dstillman any consideration of simply porting the iPad app to MacOS via Catalyst? (Excellent app btw, even though it is still technically in testing! They recently released TestFlight for Mac as well...)
@dstillman, sure it is not a replacement. If it is easy enough though (as Apple makes it sound to be like an on/off switch), you can call it Zotero-Lite or Zotero-Mobile or something like that and release it. I think many users use Zotero as a default reading app. That's what I mostly use Zotero for, personally, to read academic papers and related items in a centralized and intuitive place --- and for that, the iPad app would definitely cover 90--95% of my needs on the Mac. Sure, when I need to export a collection as bibliography, then I can use the other app or do that from the web version, etc.
Just an idea, and especially if it is easy enough, it won't harm anyone, especially through TestFlight. This way you may find yourself dealing less and less with all the keen and eager Apple users ;)
Thanks for all good and hard work on this. It does make a difference!
@ngmahfouz while I would normally agree with the two-tiered approach which you're proposing, when it comes to Zotero in particular, it doesn't seem like a great idea, and I'll play “devil's advocate” here for a bit now.
As was already stated, Zotero 5.x does indeed run fine on M1 Macs under Rosetta 2, and Apple won't rid macOS of that compatibility layer for the foreseeable future; aggressive as they usually are when it comes to those, something tells me they will keep it for much longer than Blue Box/Classic, Rosetta 1 and x86-32 support, seeing just how huge their installed base and software are at this point compared to what it was during those earlier transitions and considering the rumours that they may keep an Intel-based Mac Pro box (or even introduce an updated one) around for a few more years.
Also, it was already established that this rewrite is taking, or will take, a considerable toll on feature-rich updates. Honestly, I'd rather see Zotero devs just focus on getting a one-size-fits-all, Apple Silicon-native version of Zotero 6.x out there and save some resources for other user-facing features (dark mode support, I'm looking at you!), than wasting time on a Catalyst app which will, for all intents and purposes, become redundant once that inevitable update becomes live.
Finally, the very idea of having a “Zotero lite” option available for macOS (and only macOS, unless there's some other kind of “lite” Windows apps that I'm missing and which might make sense in that platform) seems to be a bit misguided, as it would induce user confusion, and perhaps even discourage users from making use of “full” Zotero's advanced features and plugins or even force them to switch from one Zotero version to another once they decided to do so. The whole Mendeley Desktop vs. Mendeley Cite conundrum is already proof of just how bad of an idea that is in practice, simple as its implementation might seem to be at the outset.
As an aside, I'm happy to finally see a bit more info on this transition and a renewed commitment to it, late as it might seem at this point. There is, effectively, still time to finish it without any disruption to our workflows, so I'm not too worried about being stranded without a citation manager or having to delay some otherwise desirable hardware purchase because of this.
@joaofrgomes thanks for your thoughts on this and I appreciate you taking the time to add more context.
Here's the thing though: Zotero already provides iPad and iPhone apps (I think they share the same basic code?) and I personally really like the iPad one! All I am saying is: If it is really easy to enable the iPad app on Macs, I think that's something worth pushing out (through TestFlight --- both iPad and iPhone apps are through TestFlight anyway, and Macs recently got TestFlight). I would NOT want the developers to spend any additional time on this though, so I would only advocate for it if the effort is minimal; or I would just tick the box that allows the iOS/iPadOS app to be downloaded on MacOS without further support (i.e. "use at your own risk").
As for the naming and other OSes: I will leave that up to the developers. But the iPad/iPhone app is called "Zotero" and it should be renamed if we are going to be strict about naming and features and all that stuff.
Also, btw, yes Zotero works flawlessly via Rosetta but I think it could run much better if native; whenever I use it, it ends up one of the highest consuming apps in terms of memory and energy... (but that's not a big deal for me personally as I tend to use Zotero minimally)
Lol don’t be sorry because you should NOT be. And yes, that’s fine… as long as you’re happy with shipping a half-performant app… but that still works totally fine for all practical purposes imo :)
As I said, I mainly use the iPad app and that’s been great, so thanks again for maintaining and releasing the iPad app!!
(Update: editing my poorly worded and defensive comment; sorry!!)
@ngmahfouz: If you're going to post here, please don't be obnoxious. You asked for something. We explained the reasons we're not going to do it — because the current version runs fine, because the iOS app isn't a replacement for the desktop app, because it would create a massive amount of user confusion, and because it would waste developer resources that are focused on improving our actual products. (And don't believe what Apple says: you can't create a high-quality Mac app from an iOS app by checking a checkbox. You only need to look at their own still mostly atrocious Catalyst apps to see that.)
The macOS app isn't "outdated" — it needs an architecture update and a visual refresh, but it receives far more development hours and has vastly more functionality than the iOS app (including tons of new functionality in the Zotero 6 beta), and that will always be the case. When people talk about "Zotero", they're generally talking about the desktop app. I'm glad to hear that you like the iOS app, but if you think it's in any way comparable to the desktop app, you're simply unfamiliar with the scope of Zotero functionality.
Anyway, as I've said, once Zotero 6 is out, we'll focus on updating Zotero for Apple Silicon. There's nothing more to say on this.
Upgrade Storage
We intend to support Apple Silicon natively in a future version, but that won't happen before sometime next year.
"Apple Silicon Macs should run all existing Mac software (except for kernel extensions and virtual machine apps) from day one using Rosetta 2.We intend to support Apple Silicon natively in a future version, but that won't happen before sometime next year.
However, I do not fully understand this answer. I urgently need a new MacBook, but I want to make sure Zotero is compatible for M1.
As I understand it, Zotero is not yet compatible with M1, but we are working on it. However, we do not yet know when Zotero is compatible. Is this correct? If so, do you have any idea when Zotero will be compatible for M1?
As I say, Apple Silicon Macs should run all existing Mac software, and that includes Zotero. Many people are already happily using Zotero on M1 Macs.
The one thing to be aware of is that, if you use Word, there's a minor workaround required for Word integration, but once you do that it works fine. It's not a reason not to get an M1 Mac now if you want one.A future version of Zotero will be updated to run natively at full speed on M1 Macs, but that's irrelevant for current usage. Most software that hasn't been updated to run natively on M1 Macs will still actually run faster than on Intel Macs (after a ~10-second delay the first time you start it) simply due to how much faster the M1 Macs are.
If anything, it is surprising how well everything works given that Zotero runs through an emulation layer.
While performance under Rosetta is fine, Zotero has been the only Intel app remaining on my Mac for quite a while now so I wonder when it will become a universal binary as well.
https://discussions.apple.com/thread/252157153
I also tried the terminal approach. Same message.
I finally have access to M1 machines, and will do some tests to see how it gets along with Zotero 5.0.96.3 ASAP and let you know how it went. In the meantime, can we get an ETA on native AS support?
An Apple Silicon–native version of Zotero is still a ways out.
I agree – all other software I use runs natively on M1 (Firefox, Office, Libre Office, Latex). May I ask which reason hinders compiling Zotero for M1 (Apple Silicon)?
We've been working on rewriting all the parts of Zotero that need to be rewritten for later Firefox versions, but we have to balance that work with other development that benefits everyone, such as the new PDF reader and note editor. Those of us who use Macs are as excited about Apple Silicon as anyone, but the fact remains that Zotero will run faster on M1 Macs under Rosetta 2 than on most or all Intel Macs, so this just isn't a huge priority.
Just an idea, and especially if it is easy enough, it won't harm anyone, especially through TestFlight. This way you may find yourself dealing less and less with all the keen and eager Apple users ;)
Thanks for all good and hard work on this. It does make a difference!
As was already stated, Zotero 5.x does indeed run fine on M1 Macs under Rosetta 2, and Apple won't rid macOS of that compatibility layer for the foreseeable future; aggressive as they usually are when it comes to those, something tells me they will keep it for much longer than Blue Box/Classic, Rosetta 1 and x86-32 support, seeing just how huge their installed base and software are at this point compared to what it was during those earlier transitions and considering the rumours that they may keep an Intel-based Mac Pro box (or even introduce an updated one) around for a few more years.
Also, it was already established that this rewrite is taking, or will take, a considerable toll on feature-rich updates. Honestly, I'd rather see Zotero devs just focus on getting a one-size-fits-all, Apple Silicon-native version of Zotero 6.x out there and save some resources for other user-facing features (dark mode support, I'm looking at you!), than wasting time on a Catalyst app which will, for all intents and purposes, become redundant once that inevitable update becomes live.
Finally, the very idea of having a “Zotero lite” option available for macOS (and only macOS, unless there's some other kind of “lite” Windows apps that I'm missing and which might make sense in that platform) seems to be a bit misguided, as it would induce user confusion, and perhaps even discourage users from making use of “full” Zotero's advanced features and plugins or even force them to switch from one Zotero version to another once they decided to do so. The whole Mendeley Desktop vs. Mendeley Cite conundrum is already proof of just how bad of an idea that is in practice, simple as its implementation might seem to be at the outset.
As an aside, I'm happy to finally see a bit more info on this transition and a renewed commitment to it, late as it might seem at this point. There is, effectively, still time to finish it without any disruption to our workflows, so I'm not too worried about being stranded without a citation manager or having to delay some otherwise desirable hardware purchase because of this.
Here's the thing though: Zotero already provides iPad and iPhone apps (I think they share the same basic code?) and I personally really like the iPad one! All I am saying is: If it is really easy to enable the iPad app on Macs, I think that's something worth pushing out (through TestFlight --- both iPad and iPhone apps are through TestFlight anyway, and Macs recently got TestFlight). I would NOT want the developers to spend any additional time on this though, so I would only advocate for it if the effort is minimal; or I would just tick the box that allows the iOS/iPadOS app to be downloaded on MacOS without further support (i.e. "use at your own risk").
As for the naming and other OSes: I will leave that up to the developers. But the iPad/iPhone app is called "Zotero" and it should be renamed if we are going to be strict about naming and features and all that stuff.
Also, btw, yes Zotero works flawlessly via Rosetta but I think it could run much better if native; whenever I use it, it ends up one of the highest consuming apps in terms of memory and energy... (but that's not a big deal for me personally as I tend to use Zotero minimally)
As I said, I mainly use the iPad app and that’s been great, so thanks again for maintaining and releasing the iPad app!!
(Update: editing my poorly worded and defensive comment; sorry!!)
The macOS app isn't "outdated" — it needs an architecture update and a visual refresh, but it receives far more development hours and has vastly more functionality than the iOS app (including tons of new functionality in the Zotero 6 beta), and that will always be the case. When people talk about "Zotero", they're generally talking about the desktop app. I'm glad to hear that you like the iOS app, but if you think it's in any way comparable to the desktop app, you're simply unfamiliar with the scope of Zotero functionality.
Anyway, as I've said, once Zotero 6 is out, we'll focus on updating Zotero for Apple Silicon. There's nothing more to say on this.