I've been Synergy user years ago but then I got fed up with amount of issues that simply couldn't be resolved (third mouse button issues, wrong key mappings, special keys pressed during crossing the boundary could mess things up on the new computer). So in the end I just put two keyboard and mice for two computers on the table.

Synergy left the impression as something opposite of "it just works".

And it doesn't even support basic gestures. Useless with a trackpad.

Synergy/Barrier straight up don't forward multitouch, which really sucks because part of the reason I want to use a Mac is because of their phenomenal support for gestures baked in throughout the OS.

Actually, I think Barrier does, at least between my Mac and my PC. But it's the weekend now, so I'm not going to fire up my work machine for anything...

It does not work well for multi touch, even from Mac to Mac in my experience.

You're right. My use case is also Mac to Mac and Barrier/Synergy does not provide the UC experience. No multi gestures support - and the thing that annoys me the most - trackpad scrolling events aren't mapped one to one, you end up scrolling lines on the target, not the precision you're expecting.

To be honest I'm not sure if Barrier developers could actually mimic that without resorting to some sort of undocumented APIs and/or obscure techniques.

The APIs are all there on Quartz events, it’s just that nobody seems to use them :(

In my case, my always-on workstation is a windows machine, and I also need a 3-button mouse for a decent CAD experience. So, I just give up on gestures and multitouch anyway (I prefer this consistency over having slightly different behavior across machines). Probably for similar reasons, it would be difficult to make barrier work as we wished on MacOS and still be multi-platform.

I agree. I'm a Mac user, still I've been using Synergy for years to do the very same thing with a Linux Desktop. I remember using it in like 2006. Now, Synergy doesn't run on iPads, and Universal Control is certainly "better" than synergy (in technical terms) because the Apple Behemoth invested a lot of money into this, but I can't help but wonder.. why: I don't see the use case, there's not a single app on the iPad that I'd like to control from the Mac.

i use an ios chinese dictionary called pleco that, for some reason, the developers don’t allow to be installed on macos. i primarly use my macbook to study, and before this update, i just checked definitions using the app on my phone. now, i have my ipad directly next to my macbook on a stand, and i can easily drag the mouse across and and look up entries without needing to unlock my phone and peck out or copy/paste characters. these kind of small optimizations to workflow are really helpful.

I think Pleco's developers are more focused on improving the dictionary than porting it to computers (they have an Android version too), but it would be very nice to have a computer version that can look up words if you right click them, for example.

yeah, i can understand that. i should have mentioned that i’m using an m1 mac, and as far as i understand it, there’s no extra overhead on the developer side for enabling an ios app to be downloaded from the app store and used by someone with an m1 + macos. it’s actually something the developer has to explicitly disallow (someone should correct me if i’m mistaken).

If you were using it as a second screen to watch something or do some basic browsing for reference material, they are probably hoping this will address the desire to cast the screen contents from the Macbook. If you seamlessly are passing your mouse and keyboard inputs, you'll likely just have your iPad do the video streaming or browsing directly.

But you can already cast the screen contents as well. It's called Sidecar!

At the expense of the systems resources. With Universal control your leveraging the iPad’s instead of your Mac’s.

Could be nice for mobile devs though, you don't really have to switch devices to test apps anymore. Just have a nice iPad stand next to your desktop and you'd be good to go.

That is indeed a nice setup I hadn't thought about.

My “why” is because it allows me to use the native iPad versions of apps which only have resource hungry and sometimes more quirky web/electron versions on desktop. Currently the iPad is responsible for handling Slack and Discord, but that list may grow.

Can't you install iPad apps on Macs now? Seems like that would be an easier solution to your needs.

There are ways but last I knew, if the dev has disabled installing the iOS version on macOS (as Slack and Discord regrettably have), you need to turn off SIP to be able to decrypt the iOS app bundles, then re-sign them and re-enable SIP.

My only M1 mac is a company machine and I don’t like the idea of disabling SIP on it, even temporarily.

>there's not a single app on the iPad that I'd like to control from the Mac.

Any software that has a native iOS version, but only has an Electron version on the Mac comes instantly to mind.

The Electron version of Slack's resource usage can be insane.

But do Synergy and Barrier Just Work? It has to Just Work, otherwise it's a defective product, or a rough prototype for some Apple product.

See, that's the thing. My guess is you had to configure Synergy, and as a techie you're blind to this effort. But it is still effort, hence a time sink and a cognitive burden. The Apple version is none of these and a joy to use because Apple puts the emphasis on making it Just Work with zero configuration if possible.

Hundreds of millions more people will experience multi-device control for the first time with Apple Universal Control, than will have ever heard of Synergy.

> Apple puts the emphasis on making it Just Work

They really, really don't. I have constant Bluetooth issues with all of my Apple devices, and since all of their flashy stuff relies on Bluetooth most of it breaks every couple of days.

Universal clipboard failures, AirDrop failures, AirPod pairing issues when trying to switch between devices, and total Bluetooth stack crashes.

They get things working to 80% and then just stop maintaining them.

> But do Synergy and Barrier Just Work? It has to Just Work, otherwise it's a defective product.

I don't own any apple products so correct me if I'm wrong, but if there was a way to make a program to "just work" will Apple allow it? From what I've heard devs complain apple's walled garden is pretty restrictive when it comes to things you can and cannot do especially when mobile devices are involved.

I’m an Apple developer, and own lots of Apple products.

Apple gear and software always has a low-level advantage, as they own the SDK, but I haven’t seen them actually use this to deliberately disadvantage competitors (unlike some other companies that I won’t mention).

That said, I have seen them “Sherlock” developers, though; where they implement integrated versions of products made by independent devs. This has been fairly controversial, as they have not always paid/given credit to the developers they plowed under. That sucks, because it would, literally, cost them nothing to simply mention the product that inspired their work. I suspect that’s because lawyers.

I tend to like Apple for the same reason that other developers hate them. That “walled garden” is a big fat PItA, but it is also the reason for an enormous, lucrative, market. We need to learn to take the bad with the good, if we want to be happy.

> Apple gear and software always has a low-level advantage, as they own the SDK, but I haven’t seen them actually use this to deliberately disadvantage competitors (unlike some other companies that I won’t mention).

Try to write code for macOS that can stream data from disk as fast as Logic Pro. Maybe there's really no magic, and I'm just an idiot, but I certainly couldn't do it.

I’m not exactly sure what that has to do with disputing the quoted sentence. It’s actually what I was talking about. Everybody knows that Apple has "internal APIs," giving them access that non-Apple software does not have. This does make it difficult for independent developers to offer competing functionality (this has happened with me), but I have not felt that Apple has deliberately targeted competition. It's just that they have the ability to deliver something that serves the user, and their internal access gives them an advantage.

As an independent developer, I can certainly understand why I would not be allowed to access some of these private APIs, and I don't feel that it is because Apple is afraid of any competition from li'l ol' me. I have always assumed that it's because access at that level may bypass some security methods.

You have just described "deliberately disadvantaging" the competition, which was what you said you did not believe Apple had done.

The approach he described is consistent with non-malicious failure to turn a new API into something robust, reliable, external, and well documented. If you’ve got a hot new way to move bytes the long term stability of the API is a real concern. You break things if you change it.

No. "Deliberately disadvantaging" is doing something like adding code to slow down disk write operations used by a direct competitor.

Listen, I'm not gonna play this game. I said my piece, you said yours, we're done.

Do you really trust the average individual to install apps that with zero configuration can read your screen from any of your devices?

This walled garden is 100% a feature.

I’ve used barrier for years and love it. It’s great for separating personal and work devices but having access to both at once. But it requires some effort to run if the server is a static always on workstation, and constant effort if the server is a laptop which goes to sleep and/changes its IP address. I would never recommend it to someone that might need support.

> Hundreds of millions more people

It's a pretty niche feature. I have used both Barrier and Synergy in the past but as a workflow, visually manipulating multiple computers like this sucks. Few people need to do it and life is better if you can avoid needing to do it at all. Multiplatform development is the greatest win but still, the more you can minimise on-target testing, the better. A fast test harness on your dev machine is much nicer.

IMHO remoting into a machine is much less hassle than having a dedicated screen. Most use cases typically involve poking it now and again and having an occasional window on your main display is nicer than giving up desk space and properly aligning another display. If it's not part of your permanent setup then this extra display is often an ergonomic pest. You may be able to share peripherals but if you have to contort your neck to see it then ugh, it's not fun. It can be ergonomically nicer to use their own dedicated peripherals on a side table rather than sharing them.

If it's about sharing data then obviously network accessible storage (cloud, NAS or mount a remote machine's drive) is better.

Does anything that relies on “Linux on the desktop” just work? Of course not. I say this as a user.

How do I universal control my iPad from my Linux computer? Just Works within your own ecosystem, Never Works outside of it.

Yeah, but nothing sucks quite like Synergy.

On Windows if you accidentally move your mouse off of the login screen you are _completely_ unable to user your client computer. You literally need to use the power button to recover your machine. This issue persisted for years before I gave up on Synergy and all of its suckiness.

This sounds a lot like the famous comment from yore that Dropbox “was just rsync between a computer and a mounted FTP server” or something.

Your comment makes it sound like not needing to setup apps, not having limitations and having all complexity abstracted away on an OS level are “polishing” or nice to have.

For most people, those thing are the feature.

Chance my mom will use Synergy? 0%. Chance she’ll use Universal Control at some point? I’d say pretty high.