Hi,
I just heard about the upcoming macOS Sonoma.
Now I am wondering about what problems with Arduino will it cause.
Will it fix some of Ventura’s USB problems?
What will it change?
Will it break stuff?
Just curious.
I’m looking forward to it though.
1 Like
There is a beta version for developers and a public beta over the summer if you want to test it out… you’ll get your chance to signal issues if any !
2 Likes
Thanks.
I will take a look at that. 
To be true I am a bit lazy to do that.:
What if it breaks all my Arduino stuff? 
That’s life on the bleeding edge
2 Likes
I am not looking forward...
An old engineers' saying:
If it ain't broke, don't fix it...
3 Likes
Yep,
But still……. The new features are tempting.
I hate upgrades as they take a lonnnngggg time to finish.
That’s why I am still using macOS Monterey for one of my computers.
And Ventura might mess up my stuff.
I read the reports.
Might as well just wait for Sonoma.
What features? Screen savers? xd
Funny how even Apple makes them first on the list of the new stuff. This update is useless. It could be a minor update for the apps and that's it – that's not OS–RELATED stuff.
I will not upgrade. Actually, I can't upgrade. It's not available for my MBP2015. Missing nothing though xd
That explains it. You are just Jealous not to be able to enjoy the bleeding edge 
Joke aside, it’s tons of small improvements over the years here and there that enrich the user experience and make you more productive (or make a big mess from time to time). I love continuity for example.
(Some enhancements in security do require OS level stuff that you don’t see (like the work of moving drivers out of kernel extensions ) or the work on passwords)
2 Likes
That's why I said: "It could be a minor update for the apps and that's it"
I grep macOS Sonoma - Apple for kernel, driver, extension – see nothing :>
Yea, I know, I know, not sexy enough to advertise
Anyway, really, I don't see how this update would improve my life. If the upgrade was paid, most of the people would probably skip it. OK, I don't know most people. I never paid for a screen saver. Some people do 
Haven't read through the preview.
But this happened for Ventura:
A new app called FreeForm, It's great!
A Weather app.
A redesigned System Settings.
Personally, I think those small updates are great.
Only con of Ventura that I know:
Breaks some USB and serial stuff.
And I haven't even encountered that.
Yeah - that’s the consequence of moving stuff out of kernel extensions and third party writing the drivers waiting until after the facts to start thinking about embracing the new APIs (they knew for a couple years it was coming).
When you think about it, all the traffic with external devices, printers as well, etc was going through some cryptic third parties drivers that were installed at kernel level so with quite augmented rights. As an attacker if you Manage to convince someone you have a legit driver and have them install it then you had let the fox in the henhouse… (I am always worried of people downloading random drivers from anywhere on the internet to get their arduino working…).
3 Likes
When you start having thousands of small new features scattered throughout and working across the ecosystem then I don’t see that as minor but sure you could say it’s just apps update. Reality under the hood might be different as apps had to embrace new APIs that come with the new OS to offer some of the features or services
Then of course With your old hardware (nothing pejorative there - I do own and still run a few ones that are older than yours) you don’t see The work on supporting iOS apps on Mac with Apple Silicon that opened up the possibility to leverage a brand new ecosystem nor the low level work to make the whole system more energy efficient (hardware and OS working closely together).
Whether you want to stay anchored in the past because what you have is enough for your needs or prioritizing budgets ( it does not come cheap to always update the macs) or whatever reason - that’s OK. The fact that a 10 or 15 year old Mac still works great is good News.
I would just be more sensitive with those machines which no longer can get security updates and might have attack vectors. Don’t have those in your DMZ and don’t install untrusted stuff in your home network…
or
Yep, great tradeoff xd That only proves my point: waste of time.
Want a good weather app? THAT'S AN APP. U should be able to find one. Or Apple could release it any time. And not destroy BASIC OS stuff, like… USB communication xd
there was a valid reason for that change (moving more stuff in user space) as explained above.
At the 2019 Apple Worldwide Developers Conference (WWDC), Apple announced a plan to begin deprecating the usage of kernel extensions (KEXTs) as a part of an ongoing effort "to modernize macOS, improve security, and create reliability with third-party software and security providers to ensure compatibility with operating systems upgrades."
KEXT deprecation began with the release of macOS Big Sur a year later (see. Deprecated Kernel Extensions and System Extension Alternatives - Support - Apple Developer and the likes) but deprecated does not mean no longer functional right away
there was one support engineer comment that sums it up:
The basic idea is that, as we introduce new user space functionality, we will deprecate the corresponding type of KEXT. For example, this year we’re introducing a wealth of new NetworkExtension features, and we’re deprecating NKEs. Similarly, for various I/O Kit families, kauth, and so on. However, there are still KEXT types that have no user space equivalent — the ones that immediately spring to mind are the remaining I/O Kit families and VFS plug-ins — and in that case the equivalent KEXT functionality is not deprecated.
I should stress that deprecated does not mean removed. To continue the example above, NKEs still function on 10.15. However, I would not expect that to continue for long. If you’re creating a KEXT for one of the subsystems which have user space replacements, the time to act is now
Tightened security means that some of the stuff you could do in the past no longer works indeed... And there are bad guys out there happy to leverage those possibilities... So I'd say this is a welcome move even though it created some issues with some drivers.
1 Like
Well, I'm getting paid for writing software, so I kinda know that what APIs are, etc. Of course, when they introduce some new API, they might want to demonstrate it on some new shiny app. But if they really wanna show the usefulness of the API, they should open source the app -> because then I could really see how "new" is "better" Otherwise, who's to tell if the new API really did make any difference, maybe it could be done with the old API, just as good?
Anyway, like I stated before, macOS Sonoma - Apple – first thing they promote is "screen savers" – that's really not a reason to upgrade a working, stable system. That could be a tip of the icing. That could be "one more thing" to sweeten the deal. Not a reason to upgrade.
And most of the features I see are not a reason to "risk upgrade". I know, for a person that uses a computer to check out facebook on safari and message with friends, there are different priorities.
But hey, again: messages is an app, safari is an app. How can Google make innovations with Chrome without changing the whole OS, while Apple suddenly must change the OS to improve an app? It doesn't. It's just marketing.
Yep, I certainly like that. And I'm still getting security updates. I would argue that my MBP generation has the best keyboard+touchpad Macs have ever had. When I broke my 2014 model 3 years ago, I just bought 2015, because I didn't want anything newer. Especially I will not buy into the bullshit of "8 GB of RAM is enough" with the new models, when clearly not even 16 GB is just for a minimal setup of 50 Brave Browser tabs xd Where's the space for Vagrant virtual machines? xd
I'm hoping this machine never dies. Although, it's already feeling slow. I'm tempted to upgrade. Now, with the new M1/M2 models, I'm really getting tempted. Until the moment I hear about all those issues for professional development, while the focus is on… "widgets" xd
well if you are in that business you know that offering an API is a long term commitment (decades often) to the users of that API, so this is something you handle carefully.
Developing and shipping some apps using those internal frameworks at first is an option to test it out in the real world too.
If you are a large company, you usually would iterate internally on the API a few times and test it out in real life, tweaking it as internal developers provide suggestions or hit bugs. Once you feel the API is solid, useful, proven and you've ironed out the quirks, then you can provide it to the external world to leverage and you provide sample code and documentation.
For example look at what they provided with the DriverKit when moving away from kext - and documentation such as
you clearly are not the typical end user like my mom either. They have great options for those with advanced needs with the new Macs introduced at WWDC this year. costs $$$ but I'm sure the outcome is great (being very tempted)
BTW, offtopic, but I've just watched part of it yesterday, and highly recommend: Chris Lattner: Future of Programming and AI | Lex Fridman Podcast #381 - YouTube
The interview is with the guy behind LLVM and Swift -> now doing an AI platform. Really cool guy. He got me interested in the Mojo even though I despise Python
yeah Chris Lattner is quite an engineer deserving the "geni" part of the title !