This is one of those tough cases where software cuts both ways.
Some people are smart, informed developers that install a trusted tool to monitor their traffic and have legitimate reasons to want to inspect Apple traffic. They're dismayed.
Most people are the opposite and this move protects the most sensitive data from being easily scooped up or muddled in easily installed apps, or at least easily installed apps that don't use zero days.
Is the world better or worse due to this change? I'd say a touch better, but I don't like the fact that this change was needed in the first place. I trust Apple, but I don't like trusting trust.
I'd argue this opens up a giant attack surface where malicious software will try to route its command and control communication through a protected service. Do we really want to trust that Apple will keep all 50+ of these privileged services fully protected?
I think it makes the "world" slightly worse in that it will be harder to discover malware. Little snitch has a small user base, but it's been used to identify many forms of malware and protect many more people once the threat is identified.
Yes I agree with your first part. There are real drawbacks.
But it's like installing a custom HTTPS cert in your OS to inspect potential traffic that malware may use through, say, a Google Doc or Sheet. It's helpful to true professionals dealing with highly sensitive information, but it's ultimately a bigger source of compromise for the vast majority of software users.
I don't think there is an easy answer here. That's why I said I thought it made the world a "touch better" and I can see from your response that you understand the tradeoffs roughly as well as I do based on the wording of your response. The fact is that contemplating these hard tradeoffs belie the underlying truth: Securing computers is hard and getting harder and the stakes keep going up. I can't say if this move by Apple will ultimately be worth it, but I certainly understand the predicament they are in. This is no easy work.
> Do we really want to trust that Apple will keep all 50+ of these privileged services fully protected?
No.[1] That's what people need to start understanding.
Even if you decide to trust that someone will attempt to act in your best interests (you really shouldn't, see Google's extinct "do no evil" mantra), you can't trust anyone to do so perfectly.
All this aspirational goodwill that fans express on behalf of their favorite FAANGMUULA is the tech equivalent of flat earthing. The facts are simple: no software is perfect, you can't trust any software.
If you can get into apple’s system processes, you are already on the other side of the airtight hatchway. You can make sufficient changes to the system at that point that you can certainly mess with any user-installed firewall monitoring.
I think this is the case where you can have traffic monitoring set-up on your home router or any other network gateway available. It will be slightly more troublesome, but not impossible.
Even if you believe all the MPs / representatives are trustworthy and intend to act in your best interests, their competence is going to be limited, so we need to checks and balances and a limit on their power.
The decision is questionable, but you can always inspect traffic from the machine outside it, I would even say that's preferable in context of malware.
If I install Little Snitch, it's because I trust Little Snitch to be responsible for my computer's network traffic, over and above anyone else.
I recognize that this won't necessarily apply to all users or all apps, but there needs to be a way for the user to designate trust. Apple services and traffic should not get special treatment.
5 years ago I found LS was unable detect any traffic out of a VMWare virtual machine running on the same Mac. Sure the VM is running through some installed virtual network adapter, but if that's all it takes an attacker can set up one of her own. Cool Hollywood interface but I gave up on LS as a serious security tool right there.
If you don’t trust Apple then you need something more than little snitch. Apple is responsible for both hardware and OS. What delta in security or trust is little snitch going to offer over Apple?
Trust relies on faith or evidence, the overwhelming circumstantial evidence is that Apple can not be trusted with anything other than their commercial interests.
You can not trust Apple with anything else, therefore you must have faith.
Why doesn't each individual user have the final say over whether she wants to accept the change or not? There is no option presented to the user:
[ ] Do not trust Apple, trust only me
You say "Some people are smart, informed developers" but in this case, it appears Apple is treating every user as the same.
I am not a "developer" (nor am I particularly "smart") and yet I monitor traffic to/from computers I own. Maybe some incorrect assumptions are being made about so-called "users". I find it perplexing that any company should be able to prevent me from monitoring traffic to/from computers I own. I own the computers, I pay for the bandwidth. I do not buy Apple computers for the Apple software.
Actually, I don't think this is about trust. I mean, when I use an Apple OS, I (should) trust them, as their software has access to all my most sensitive digital information.
However, making it impossible to route the traffic of the system apps through a VPN of my choice (whatever the reason), is just broken functionality.
Absolutely. For example, I think that the lockdown of the bios was a move that hobbled developers like myself that installed custom bios extensions. I used to be able to run raw linux on real hardware. Now I need to use a commercial virtual machine just to get the dev environment I want.
The difference between the two is subtle, but true. I want true masters that understand what the tradeoffs are to make those hard choices for themselves. I want the rest of the world to have a blanket of privacy and security that protects everyone.
Especially the elderly that are too trusting with what they believe.
Have you used little snitch? It very clearly allows all apple traffic by default, and if you modify something that would affect it, you get a huge popup explaining what will happen and have to click on a red button to confirm.
> Is the world better or worse due to this change?
This is the false shortcut behind any attempt to weaken security. Security makes access harder, therefore let's weaken security to improve access.
The fact is that weakening security also makes malicious behavior easier and/or more likely. Changes like this are bad particularly because Apple users pay for a protected walled garden.
> Some people are smart, informed developers that install a trusted tool to monitor their traffic and have legitimate reasons to want to inspect Apple traffic. They're dismayed.
Wouldn't say I'm that smart. Wouldn't call myself a developer either. But I'm still kind of dismayed. I used to love macOS (or OS X to be precise), but the clock has been ticking for years now. Near every decision made about macOS future goes in the wrong direction (for me). Right now I'm looking at Manjaro. But still, I need the Adobe CC suite to get my work done, so I will have to use two machines. I hate running two computers. But that's probably where I'll end up.
Either Apple doesn't trust Little Snitch and shouldn't let it interfere with any apps, or Apple does trust Little Snitch and shouldn't block it at all. There's no reason to implement this halfway.
If the data is so poorly protected in transit that a firewall app on the system is a concern, something has gone very wrong indeed. It's just going to see that your Apple services on your Apple device are speaking to Apple servers.
There's an availability consideration here, but that's about it.
Tech savvy users are not just the minority. They're also cheap. They've been conditioned by the FOSS movement to think all software should be free as-in-beer. (The people who started FOSS didn't say that, but that's what it's become.) They say they want free as-in-freedom, but since they are not willing to pay for it they don't exist. Those who pay set the agenda for everything.
Developing a truly polished operating system with a whole ecosystem of services is far, far beyond what volunteers and hobbyists can achieve. It's just too much work. It also requires focus and coordination and someone who is able and willing to say no. Without that the FOSS community rewrites everything over and over again instead of doing the not-fun parts of programming like fixing bugs and edge cases.
TL;DR: we get what we pay for. We don't pay for freedom so we don't get it.
Where are these weird anti-FOSS statements being bred from?
> Those who pay set the agenda for everything.
And this different from non FOSS software how?
> Developing a truly polished operating system with a whole ecosystem of services is far, far beyond what volunteers and hobbyists can achieve.
As someone who uses Linux as my primary workstation I disagree. My coworkers that use Mac or Windows seem to have about the same number of issues overall. I mean- look at the article this is about. I’m pretty confident that would be much harder to get away with in the Linux community. Gnome shell is more polished than windows or macOS were at the same age.
> It also requires focus and coordination and someone who is able and willing to say no.
Clearly you haven’t dealt with the Gnome folks who are perfectly willing to say no to features some users scream for. Or read any of Linus’s rants about nvidia.
> They say they want free as-in-freedom, but since they are not willing to pay for it they don't exist. Only paying users matter.
Citation needed. If you look at app store pricing models the opposite seems true. If I were going to take a random guess I would say that tech savvy users use open source software to avoid anti-consumer bullshit more than anything else.
I'm happy to pay for good FOSS and open hardware and I'm paying. Also I'm trying to avoid any proprietary and especially cloud-connected things. You are generalizing too much, there are enough people who are happy to pay for trustworthy software and hardware. Just noone cares.
> Tech savvy users are not just the minority. They're also cheap.
Bologna. I spent $4,000 for this MBP, and I've spent many hundreds on accessories, and thousands of dollars on software to run on it. I do everything on it. It is the center of my digital life.
That being said, the day I go to do something on this machine and find that I can't is the day I go buy a sub-$1,000 PC laptop, and go back to Linux (which I ran on the desktop for 19 years). Apple should be very careful how hard they squeeze here.
Apple seems to do all kinds of weird networking _stuff_. For instance, during wakeup, your T2 equipped Macbook will wait for a DNS response and then use said DNS response to synchronize time via NTP before letting the user use the keyboard. Probably checking timestamps on signatures for the keyboard firmware, or something stupid like that. This only happens if it happens to have a default route.
Similarly, all macOS machines will test a DHCP supplied default route before applying it by trying to reach something on the internet. So if you happen to have some firewall rules that block internet access, no default route will be applied until the internet check times out.
I won't share the other sentiments about the above, but is it really that hard to document these behaviors?
Apple touted the T2 chip as the bee's knees in security. Now, we have a vulnerability that cannot be defended against. However, Apple went all in on the security of this T2 chip so that you cannot replace the SSD (besides the method to manufacture). I appreciate the desire at making a device difficult for a bad actor to get to your data, but they epicly failed and ultimately only made an user-hostile device. Oh, and the laptops with these chips also had the world's worst keyboard. Absolute trash.
> I appreciate the desire at making a device difficult for a bad actor to get to your data
That's what FileVault is for. I don't understand what's the problem T2 is trying to solve by its existence. Being able to use something else to read the data from a drive you pulled out of your computer, after decrypting it with your password, is a feature, not a bug. T2 is a regression, not an improvement in security. You can't call it a security product if you keep the master key, which Apple does.
The new keyboard is no longer horrible beyond index. Unfortunately, it's merely adequate, which at least in my book is unacceptable for any $1k+ laptop, let alone $3k+.
> Apple went all in on the security of this T2 chip so that you cannot replace the SSD
That's not a security thing, really. It's easy enough to layer encryption on a normal SSD. It's their desire to make it some kind of do-everything auxiliary chip, which has the end result of weakening security.
Plus don't talk about display. Its has serious flaw. Like most macbook 2017 have lines on bottom due to apple placing controller in tcon board. What a trash .
Oh wow! This probably explains why every now and then when I wake my MacBook Pro from sleep it says no keyboard is connected! I thought I had some hardware problem on a basically brand new machine. Glad to hear it's only a stupid software problem!
> but is it really that hard to document these behaviors?
I imagine it is, given the bureaucracy of a big company. Apple's documentation has long been really dreadful, mostly nonexistent and where it does exist, usually incomplete and even wrong. I've assumed it was because the code itself is developed by isolated groups while the documentation presumably has to touch all sorts of people (publishing, translation, language checks, ...) in a kind of Conway's law.
However, hard or not, writing comprehensive documentation is quite doable. I have never been a fan of the Windows programming model but I have long admired not just MS's documentation but the amount of effort and commitment they obviously put in.
Apple cares about some things but in this regard it appears they simply don't give a shit.
> Apple seems to do all kinds of weird networking _stuff_. For instance, during wakeup, your T2 equipped Macbook will wait for a DNS response and then use said DNS response to synchronize time via NTP before letting the user use the keyboard. Probably checking timestamps on signatures for the keyboard firmware, or something stupid like that. This only happens if it happens to have a default route.
When did they start doing this? I'm still using High Sierra on my 2018 MBP work laptop, because the keyboard and trackpad was freezing for anywhere up to 5 minutes or more with Mojave after a wakeup (usually after a long sleep). Downgrading to High Sierra fixed it, but fighting with the machine was such a pain I haven't dared touch it since.
I'm wondering if you're describing the problem I was having, but could never figure out.
Unrelated but has anyone often had Chrome going on cpu usage rampage and unresponsive fairly frequency on 'wakeup from sleep'? It's almost certain to happen if the chrome has been updated and waiting to be restarted.
That's how typical Apple "magical/just works" features are implemented, i.e. very ugly behind the curtain.
Documenting means revealing the edge cases and the limitations, which engineering knows is the best kind of documentation. But marketing people are invested in the "magic".
The keyboard thing is new to me, wow that sucks. The other one sounds like a workaround for captive portals. I think there is some documentation on that wrt Safari and the built in networking, but it was mostly a workaround needed to deal with wifi hotspots that intercept dns until you pay/subscribe, and it causes safari to look hung - so they had to make it clear it wasn’t their browser hanging since it couldn’t make SSL connections.
OS is a weird design. It lets the machine belong to Apple/MS/Google not we, so they could update whatever they want or query to their website secretly. You can't even stop them because once you installed you agreed for all. You don't have choices to partially agree. It makes me feel like when you have a cecal surgery, the doctor also took out your foreskin for auto-updating.
You actually just helped me diagnose a really annoying bug I've been having lately. When I wake up my Mac from sleep mode the keyboard and mouse are unresponsive for a up to a few minutes in some extreme cases, sometimes I even have to hard reboot. I found online that it was related to VPNs trying to restore their connection but I could never find the link between the keyboard and the VPN.
It was also compounded by the VPN setting I use to disable all traffic until it successfully reconnects. Meaning whether my computer works or not is dependent on my VPN providers reliability.
Now that I know Apple thinks I need an internet connection to wake up my laptop securely I'm quite pissed by this. Brand new $4k laptop is a paperweight if my VPN can't connect.
I'm working from home now, and in my company we use Tunnelblick for vpn into corp network. VPN has time-based OTP so it never gets saved.
Sometimes when my MBP goes to sleep it loses wifi connection and VPN disconnects. When it wakes up, Tunnelblick asks for password, but it doesn't restore routes (I guess?). Basically no internet until I either enter password or click disconnect. At that moment I'm typing in my OS password and pressing Enter.
What then happens is that it waits for ≈30 seconds and then logs me in, as if it made a network request and waited until it timed out.
Could it be related to the issue you're describing?
I was trying to figure out how my routing table was set up on my iPad and I found out that iOS doesn't expose any interface to routing tables, at any level of privilege. Very frustrating.
I think this is probably wrong. I don’t know what the interface is, but on my iPad running 14.0.1 this app shows a Routing Table that looks okay to me. https://networktools.he.net/
I should've clarified - it only does this if there is a default route. Funnily enough, whilst the firewalls in the original twitter post would possibly fail to catch this traffic, PF will block it just fine.
When I had the authenticate with watch option enabled, and for some reason the watch lagged, the Mac didn't allow me to log in with my password or finger.
> For instance, during wakeup, your T2 equipped Macbook will wait for a DNS response and then use said DNS response to synchronize time via NTP before letting the user use the keyboard.
Aha so this is why I need to put my MacBook back to sleep after waking on a spotty WiFi connection or when it was previously connected to vpn which timed out during sleep!
Would certain go a long way to explain why waking my MBP up after going AFK involves an affair that requires me to undock it from my vertical stand, entering password, and awkwardly trying to place it back into the stand, reconnecting peripherals while slapping the BT keyboard endlessly so it doesn't go back to sleep after login.
> your T2 equipped Macbook will wait for a DNS response and then use said DNS response to synchronize time via NTP before letting the user use the keyboard.
Holy shit, this is why my macbook sometimes won't let me log in for like 15 seconds on my shitty cellular hotspot connection? Absurd. Apple software has fallen so far from just 10 years ago.
"You have to trust Apple", it's said. But I suspect that if you actually knew how much your Apple devices were phoning home to Cupertino, you wouldn't trust Apple anymore. Using Little Snitch (the kernel extension) was a real eye opener for me. Especially when I allowed Little Snitch to block all Apple processes (by disabling the built-in iCloud Services and macOS Services rule groups).
This may be a good time to remind folks of my blog post where I explain how Catalina phones home when you run unsigned executables, including shell scripts! In the article I mentioned that you can prevent this with Little Snitch. But that was the LS kext. Is it even possible anymore?
https://lapcatsoftware.com/articles/catalina-executables.htm...
Let me just quote one comment from the HN discussion of that article: https://news.ycombinator.com/item?id=23278253
"Making this about speed is burying the lede. From a privacy and user-freedom perspective, it's horrifying. Don't think so? Apple now theoretically has a centralized database of every Mac user who's ever used youtube-dl. Or Tor. Or TrueCrypt."
It's all too easy to dismiss the privacy violations that we're not aware of. Out of sight, out of mind.
Why would the most successful company in history—a success gained in no small part through protecting users, selling hardware and services instead of their data, and promoting and enhancing privacy as a first-class feature—do that sort of thing? What possible benefit could such a centralized database serve? How's that gonna make them more money?
I'd love to apply Occam's Razor to Apple's network connections. Those entitles should not be multiplied without necessity. That's why I use Little Snitch!
Seriously though, Tim Cook has been absolutely trashing Apple's hard won reputation by relentlessly pushing (via push notifications no less) TV shows and other garbage "subscriptions" on computer buyers. It's not what I signed up for when I became a Mac user many years ago.
That quote—“Apple now theoretically has a centralized database of every Mac user who's ever used youtube-dl.”—is somewhat misleading.
Apple doesn’t get script contents, it only gets a hash. Of course, if Apple really wanted, they could maintain a DB of hashed contents of every possible version of youtube-dl script, and do their best to match it up with what users execute. However, even that far-fetched scenario falls apart the moment you wrap youtube-dl invocation in a convenience script—as only the hashed content of the script you invoke is submitted for notarization check, not every binary or script further launched by it.
That totally breaks my use case for Little Snitch: working tethered. When I tether my laptop it thinks it has free reign with the bandwidth and all of the little background processes can kill my data in a few minutes. With a firewall, I can grant access to only the processes that I need to get my work done.
Now, I guess I have to run some external firewall between my laptop and my phone. ... or better yet, abandon Apple.
For what it's worth, my hacky solution to this is this script which kills all the background processes that use significant bandwidth. If you're interested in how I came up with the list of processes, I can share the BitBar [1] script I wrote for monitoring per-process network usage (I wrote a small wrapper around nettop that logs to a db, which is read periodically by my BitBar script to show me the per-process usage:
I use Trip Mode for that (https://tripmode.ch/). Though, it's not unlikely it'll have the same issues described in the OP, it does seem to block Apple stuff on Mojave.
I had replied to parent as well, but then saw your note. So I deleted, and pasting my agreement here:
> free reign with the bandwidth and all of the little background processes can kill my data in a few minutes
New: TripMode 3, made for macOS 11 Big Sur. Easily control your Mac's data usage on slow or expensive networks.
Drastically optimize your Mac’s data usage by automatically blocking unwanted background updates. Keep control with the new live monitor and data usage reports. Reveal domains where your apps send your data to. Now with a redesigned, easier than ever UI.
Last year Apple introduced 2 flags on the network: “constrained” (the Low Data Mode toggle) and “expensive” (most cellular and personal hotspots). These are intended to let the app make intelligent decisions about what network requests to do. For example, “expensive” networks should disable background or speculative fetches and only fetch what the user asked for.
Presumably Apple apps that bypass the network filter are making use of these flags already, to avoid unnecessary network traffic.
Is there no chance for little snitch to block app store? I just have a demo ver of little snitch and will buy it for blocking all apple service. I always connect the internet through my phone outdoors. The bandwidth is limiting...
If Microsoft did this in windows, or Google did this in chrome, would we see so much defense of this strategy? Or could it be those rose coloured glasses that HN tends to view Apple through.
Or more like "users are literally brain dead and cannot be trusted to change the channels on their TV" coloured glasses. If you only trust your users to watch TV, then get into TVs instead of computers.
We don't fault the maker of a drill when a careless user drills a hole in their hand. We fault the user for being careless. At what point do we start doing the same for computers? The advantage of physical power-tools is that their mechanism of operation is readily apparent, open, understandable, predictable. If Apple really cares about their users, they should start investing in making software open, understandable, predictable. This is a much harder problem, and probably less profitable, than just building another TV, but I'd rather live in that world than this one. I don't need another TV.
Btw, when I've been testing a "kill switch" on Windows (firewall configuration that doesn't allow internet access without a VPN running) using the built-in firewall, I discovered that
- Chrome adds a Firewall rule on installation that grants it access to all networks, bypassing kill switch configurations.
- Microsoft has an "Allow app through Firewall" [1] dialog that manages all of the rules for its apps and services along with some third-party apps. These rules again tend to allow everything, and at least on earlier builds from like 2018 they would reset to allow everything on _every_ update.
I wonder if it would make sense for Little Snitch to continue supporting their kext-based solution in parallel to the new one, possibly only for users who are willing to disable SIP.
You might argue that disabling SIP for a security product defeats the point, but I'm not sure if that's necessarily true. SIP effectively delegates trust away from the user and towards Apple, which is fine as a default—but the calculus may be different for experienced users, like the ones who use Little Snitch.
Kexts are used by Apple internally, so I'd be shocked if they were removed from the OS completely. Third party kexts may be deprecated, but as long as SIP can be disabled it will always be possible to load your own.
Background: I've written my own kernel extension that works in similar manner to Little Snitch, but does a lot more, including SSL MITM and on-demand packet capture, that I've been using for more than 10 years now.
It's a fact that Apple has continuously moved to lock down macOS in ways that are antithetical to folks that want full control over their operating system. To many of us that moved on from Linux on the desktop, the combination of a stable/uniform/attractive desktop environment with a Unix core that had great developer documentation -no longer the case!- and nicely-designed APIs was too much to resist. Unfortunately, the push towards consumers and Apple's increasingly one-sided my-way-or-the-highway approach (fueled by security concerns that to me are completely irrelevant, if not a huge annoyance and waste of time) means that a lot of us oldschool Unix hackers were left out in the cold.
I don't plan to upgrade past Mojave and at some point in the future I will move back to Linux.
Linux on the desktop and Linux on the laptop (heh) has definitely improved. It _sometimes_ needs a little tweaking to get it right, but KDE/Plasma also happens to offer that level of "tweakability" that should satisfy almost all semi-mainstream users (at least anyone coming from Windows or Mac).
Compared to my first Linux laptop (a Sony Vaio circa 2000), my current XPS 13 works as well as any Mac laptop I have ever owned, and all the hardware that you would "expect" to work (but probably didn't work as smoothly 10 or 20 years ago) Just Works (WiFi, external displays, excellent battery life/sleep, etc...)
Based on the complaints I have heard about Apple hardware and MacOS over the past few years, I'd even argue that Linux-on-the-desktop isn't any less stable or harder to get working than a Mac.
Just wanted to add another compliment for KDE (specifically Plasma). I've been using KDE Neon as my daily driver for a few months now and it's amazing. Connects to my android device to share notifications and clipboard content, is heavily customizable and themeable, the whole OS feels very snappy and uniform in terms of UI/UX, and installing alongside Win10 and macOS in a hackintosh setup with full LUKS disk encryption was a snap through the installer GUI. Absolute 10 out of 10.
Same! I purchased a Razer Blade Stealth 13 and put Linux Mint on it end of 2019. I have been really pleased with the entire thing. I don't do anything crazy (web browsing, simple budget spreadsheets, watching videos, viewing family photos) and it works perfectly. I was an avid mac user for many years because of bash/BSD but the march toward locked-down hardware and software really pushed me away. The only thing I miss are the glass trackpads and the fantastic gesture support.
How is desktop search? Spotlight (mac desktop search) is a killer feature for me -- fast, reliable, smooth, all straight out of the box. Meanwhile, I've wasted many hours trying to get desktop search up to the same standard on Windows and Linux. That was years ago (for linux, at least), hopefully things have improved. How is linux desktop search doing today?
I've become a huge fan of Linux Mint. It looks amazing and unlike before now there are no driver related issues (the thing that kept me from using it all this time).
The only thing I miss is Photoshop but I really can't think of a single reason besides that to not use Linux anymore.
Require the user to authenticate, then provide full control? Yes this provides a vulnerability pathway, but it's not like Apple software updates don't already provide this type of access.
> I've written my own kernel extension that works in similar manner to Little Snitch, but does a lot more, including SSL MITM and on-demand packet capture, that I've been using for more than 10 years now.
I'd be interested to read more about this, and maybe even use your kext. I'm currently MITM'ing all of my SSL traffic[1] for a different, esoteric reason: I insist on using a 7-year-old version of macOS, and it doesn't natively support modern SSL ciphers, so I have to add it in with an mitm proxy.
I've run into a handful of issues with various software that I've had to work through as they arise, but if you've been doing this for ten years you've probably seen it all already.
hahaha. I also don’t plan to upgrade past Mojave. To me Catalina was a trainwreck and at this point I think I’m loosing a lot of trust I used to put in Apple.
this is compounded by the fact that I love Little Snitch and it has basically exponentially improved my life when it comes not only to browsing the web but when using any app on mac.
I tried catalina and... why? why did they dumb down mail? This is like the beige apple box era all over again. lame decision after lame decision and everything turns to mud.
The linux desktop experience is still quite in a state. I will likely do the same and suffer Linux, but I think many will go back to windows as WSL continues to improve.
I switched back to linux two years ago for exactly the same reason.
It was painful at first, but it's worth it. The only things I still miss are the visual feedback in the UI (lots of little stuff) and the feel of the trackpad.
But the customizability has more than made up for that in productivity. Like being able to edit the source code for the window manager.
A great example of why you need defense in depth. Ideally you'd be running the local firewall on your box, as well as an external firewall.
That being said, this is not ok behavior on Apple's part. There shouldn't be a way for traffic to go around the firewall like this, even if it is just Apple apps.
Because as Apple well knows, once you make a backdoor, someone will figure out a way to exploit it.
> Because as Apple well knows, once you make a backdoor, someone will figure out a way to exploit it.
I can't help but see this as the real reasoning behind the change. With EARN-IT on on the table and antitrust cases looming, they've got every reason to bend over and give governments whatever access they can.
Of course not, which is why this still isn't very cool. If you're super paranoid you can always carry around a small router or a pi to attach to the wifi and be your external router though.
Readit News
Some people are smart, informed developers that install a trusted tool to monitor their traffic and have legitimate reasons to want to inspect Apple traffic. They're dismayed.
Most people are the opposite and this move protects the most sensitive data from being easily scooped up or muddled in easily installed apps, or at least easily installed apps that don't use zero days.
Is the world better or worse due to this change? I'd say a touch better, but I don't like the fact that this change was needed in the first place. I trust Apple, but I don't like trusting trust.
I think it makes the "world" slightly worse in that it will be harder to discover malware. Little snitch has a small user base, but it's been used to identify many forms of malware and protect many more people once the threat is identified.
But it's like installing a custom HTTPS cert in your OS to inspect potential traffic that malware may use through, say, a Google Doc or Sheet. It's helpful to true professionals dealing with highly sensitive information, but it's ultimately a bigger source of compromise for the vast majority of software users.
I don't think there is an easy answer here. That's why I said I thought it made the world a "touch better" and I can see from your response that you understand the tradeoffs roughly as well as I do based on the wording of your response. The fact is that contemplating these hard tradeoffs belie the underlying truth: Securing computers is hard and getting harder and the stakes keep going up. I can't say if this move by Apple will ultimately be worth it, but I certainly understand the predicament they are in. This is no easy work.
No.[1] That's what people need to start understanding.
Even if you decide to trust that someone will attempt to act in your best interests (you really shouldn't, see Google's extinct "do no evil" mantra), you can't trust anyone to do so perfectly.
All this aspirational goodwill that fans express on behalf of their favorite FAANGMUULA is the tech equivalent of flat earthing. The facts are simple: no software is perfect, you can't trust any software.
1: https://www.cvedetails.com/vendor/49/Apple.html
Even if you believe all the MPs / representatives are trustworthy and intend to act in your best interests, their competence is going to be limited, so we need to checks and balances and a limit on their power.
I recognize that this won't necessarily apply to all users or all apps, but there needs to be a way for the user to designate trust. Apple services and traffic should not get special treatment.
Trust relies on faith or evidence, the overwhelming circumstantial evidence is that Apple can not be trusted with anything other than their commercial interests.
You can not trust Apple with anything else, therefore you must have faith.
I am not a "developer" (nor am I particularly "smart") and yet I monitor traffic to/from computers I own. Maybe some incorrect assumptions are being made about so-called "users". I find it perplexing that any company should be able to prevent me from monitoring traffic to/from computers I own. I own the computers, I pay for the bandwidth. I do not buy Apple computers for the Apple software.
However, making it impossible to route the traffic of the system apps through a VPN of my choice (whatever the reason), is just broken functionality.
The difference between the two is subtle, but true. I want true masters that understand what the tradeoffs are to make those hard choices for themselves. I want the rest of the world to have a blanket of privacy and security that protects everyone.
Especially the elderly that are too trusting with what they believe.
This is the false shortcut behind any attempt to weaken security. Security makes access harder, therefore let's weaken security to improve access.
The fact is that weakening security also makes malicious behavior easier and/or more likely. Changes like this are bad particularly because Apple users pay for a protected walled garden.
It's pretty clear what they think - they allow basically any app to access the network on ios.
Wouldn't say I'm that smart. Wouldn't call myself a developer either. But I'm still kind of dismayed. I used to love macOS (or OS X to be precise), but the clock has been ticking for years now. Near every decision made about macOS future goes in the wrong direction (for me). Right now I'm looking at Manjaro. But still, I need the Adobe CC suite to get my work done, so I will have to use two machines. I hate running two computers. But that's probably where I'll end up.
Either Apple doesn't trust Little Snitch and shouldn't let it interfere with any apps, or Apple does trust Little Snitch and shouldn't block it at all. There's no reason to implement this halfway.
There's an availability consideration here, but that's about it.
Deleted Comment
Developing a truly polished operating system with a whole ecosystem of services is far, far beyond what volunteers and hobbyists can achieve. It's just too much work. It also requires focus and coordination and someone who is able and willing to say no. Without that the FOSS community rewrites everything over and over again instead of doing the not-fun parts of programming like fixing bugs and edge cases.
TL;DR: we get what we pay for. We don't pay for freedom so we don't get it.
> Those who pay set the agenda for everything. And this different from non FOSS software how?
> Developing a truly polished operating system with a whole ecosystem of services is far, far beyond what volunteers and hobbyists can achieve.
As someone who uses Linux as my primary workstation I disagree. My coworkers that use Mac or Windows seem to have about the same number of issues overall. I mean- look at the article this is about. I’m pretty confident that would be much harder to get away with in the Linux community. Gnome shell is more polished than windows or macOS were at the same age.
> It also requires focus and coordination and someone who is able and willing to say no.
Clearly you haven’t dealt with the Gnome folks who are perfectly willing to say no to features some users scream for. Or read any of Linus’s rants about nvidia.
Edit: formatting
Citation needed. If you look at app store pricing models the opposite seems true. If I were going to take a random guess I would say that tech savvy users use open source software to avoid anti-consumer bullshit more than anything else.
Bologna. I spent $4,000 for this MBP, and I've spent many hundreds on accessories, and thousands of dollars on software to run on it. I do everything on it. It is the center of my digital life.
That being said, the day I go to do something on this machine and find that I can't is the day I go buy a sub-$1,000 PC laptop, and go back to Linux (which I ran on the desktop for 19 years). Apple should be very careful how hard they squeeze here.
With trust you get trade. Trade is commerce and the more trust you have the more money changes hands.
If I could firewall my phone I would upgrade every year no question.
Similarly, all macOS machines will test a DHCP supplied default route before applying it by trying to reach something on the internet. So if you happen to have some firewall rules that block internet access, no default route will be applied until the internet check times out.
I won't share the other sentiments about the above, but is it really that hard to document these behaviors?
That's what FileVault is for. I don't understand what's the problem T2 is trying to solve by its existence. Being able to use something else to read the data from a drive you pulled out of your computer, after decrypting it with your password, is a feature, not a bug. T2 is a regression, not an improvement in security. You can't call it a security product if you keep the master key, which Apple does.
Deleted Comment
That's not a security thing, really. It's easy enough to layer encryption on a normal SSD. It's their desire to make it some kind of do-everything auxiliary chip, which has the end result of weakening security.
I imagine it is, given the bureaucracy of a big company. Apple's documentation has long been really dreadful, mostly nonexistent and where it does exist, usually incomplete and even wrong. I've assumed it was because the code itself is developed by isolated groups while the documentation presumably has to touch all sorts of people (publishing, translation, language checks, ...) in a kind of Conway's law.
However, hard or not, writing comprehensive documentation is quite doable. I have never been a fan of the Windows programming model but I have long admired not just MS's documentation but the amount of effort and commitment they obviously put in.
Apple cares about some things but in this regard it appears they simply don't give a shit.
Developer docs for most of their libraries are usually just the method name in a large font and the parameter types and that's it.
Anyone want to tell him about Microsoft's Azure or .NET documentation?...
It's the same all over.
When did they start doing this? I'm still using High Sierra on my 2018 MBP work laptop, because the keyboard and trackpad was freezing for anywhere up to 5 minutes or more with Mojave after a wakeup (usually after a long sleep). Downgrading to High Sierra fixed it, but fighting with the machine was such a pain I haven't dared touch it since.
I'm wondering if you're describing the problem I was having, but could never figure out.
Documenting means revealing the edge cases and the limitations, which engineering knows is the best kind of documentation. But marketing people are invested in the "magic".
Marketing people have too much sway at Apple.
It feels rather heavy handed as there are ways other systems have worked around this that don't hijack routes.
It was also compounded by the VPN setting I use to disable all traffic until it successfully reconnects. Meaning whether my computer works or not is dependent on my VPN providers reliability.
Now that I know Apple thinks I need an internet connection to wake up my laptop securely I'm quite pissed by this. Brand new $4k laptop is a paperweight if my VPN can't connect.
Sometimes when my MBP goes to sleep it loses wifi connection and VPN disconnects. When it wakes up, Tunnelblick asks for password, but it doesn't restore routes (I guess?). Basically no internet until I either enter password or click disconnect. At that moment I'm typing in my OS password and pressing Enter.
What then happens is that it waits for ≈30 seconds and then logs me in, as if it made a network request and waited until it timed out.
Could it be related to the issue you're describing?
I just wish the font rendering situation on Linux was better though. Text (in browsers) just looks so bad on Linux compared to both Windows and mac.
... and what if your network is down? You can't even use your keyboard?
The machine is basically frozen at login until some timeout hits.
Deleted Comment
Aha so this is why I need to put my MacBook back to sleep after waking on a spotty WiFi connection or when it was previously connected to vpn which timed out during sleep!
Quite annoying.
Holy shit, this is why my macbook sometimes won't let me log in for like 15 seconds on my shitty cellular hotspot connection? Absurd. Apple software has fallen so far from just 10 years ago.
I really wish Apple executives were forced to use their computers on crap wifi. Who am I kidding, I would imagine Tim Cook hasn't used a Mac in years.
This may be a good time to remind folks of my blog post where I explain how Catalina phones home when you run unsigned executables, including shell scripts! In the article I mentioned that you can prevent this with Little Snitch. But that was the LS kext. Is it even possible anymore? https://lapcatsoftware.com/articles/catalina-executables.htm...
Let me just quote one comment from the HN discussion of that article: https://news.ycombinator.com/item?id=23278253 "Making this about speed is burying the lede. From a privacy and user-freedom perspective, it's horrifying. Don't think so? Apple now theoretically has a centralized database of every Mac user who's ever used youtube-dl. Or Tor. Or TrueCrypt."
It's all too easy to dismiss the privacy violations that we're not aware of. Out of sight, out of mind.
Why would the most successful company in history—a success gained in no small part through protecting users, selling hardware and services instead of their data, and promoting and enhancing privacy as a first-class feature—do that sort of thing? What possible benefit could such a centralized database serve? How's that gonna make them more money?
Seriously though, Tim Cook has been absolutely trashing Apple's hard won reputation by relentlessly pushing (via push notifications no less) TV shows and other garbage "subscriptions" on computer buyers. It's not what I signed up for when I became a Mac user many years ago.
Apple doesn’t get script contents, it only gets a hash. Of course, if Apple really wanted, they could maintain a DB of hashed contents of every possible version of youtube-dl script, and do their best to match it up with what users execute. However, even that far-fetched scenario falls apart the moment you wrap youtube-dl invocation in a convenience script—as only the hashed content of the script you invoke is submitted for notarization check, not every binary or script further launched by it.
Now, I guess I have to run some external firewall between my laptop and my phone. ... or better yet, abandon Apple.
And really, put a sleep in there of at least a second or so or this'll be a huge resource hog.
> free reign with the bandwidth and all of the little background processes can kill my data in a few minutes
New: TripMode 3, made for macOS 11 Big Sur. Easily control your Mac's data usage on slow or expensive networks.
Drastically optimize your Mac’s data usage by automatically blocking unwanted background updates. Keep control with the new live monitor and data usage reports. Reveal domains where your apps send your data to. Now with a redesigned, easier than ever UI.
https://tripmode.ch
Glad to know stopping shit like that is no longer an option.
Presumably Apple apps that bypass the network filter are making use of these flags already, to avoid unnecessary network traffic.
Deleted Comment
Or more like "users are literally brain dead and cannot be trusted to change the channels on their TV" coloured glasses. If you only trust your users to watch TV, then get into TVs instead of computers.
We don't fault the maker of a drill when a careless user drills a hole in their hand. We fault the user for being careless. At what point do we start doing the same for computers? The advantage of physical power-tools is that their mechanism of operation is readily apparent, open, understandable, predictable. If Apple really cares about their users, they should start investing in making software open, understandable, predictable. This is a much harder problem, and probably less profitable, than just building another TV, but I'd rather live in that world than this one. I don't need another TV.
- Chrome adds a Firewall rule on installation that grants it access to all networks, bypassing kill switch configurations.
- Microsoft has an "Allow app through Firewall" [1] dialog that manages all of the rules for its apps and services along with some third-party apps. These rules again tend to allow everything, and at least on earlier builds from like 2018 they would reset to allow everything on _every_ update.
This was such a pain to deal with.
[1] https://az767233.vo.msecnd.net/images/Security/win8_winfirew...
You might argue that disabling SIP for a security product defeats the point, but I'm not sure if that's necessarily true. SIP effectively delegates trust away from the user and towards Apple, which is fine as a default—but the calculus may be different for experienced users, like the ones who use Little Snitch.
It's a fact that Apple has continuously moved to lock down macOS in ways that are antithetical to folks that want full control over their operating system. To many of us that moved on from Linux on the desktop, the combination of a stable/uniform/attractive desktop environment with a Unix core that had great developer documentation -no longer the case!- and nicely-designed APIs was too much to resist. Unfortunately, the push towards consumers and Apple's increasingly one-sided my-way-or-the-highway approach (fueled by security concerns that to me are completely irrelevant, if not a huge annoyance and waste of time) means that a lot of us oldschool Unix hackers were left out in the cold.
I don't plan to upgrade past Mojave and at some point in the future I will move back to Linux.
Compared to my first Linux laptop (a Sony Vaio circa 2000), my current XPS 13 works as well as any Mac laptop I have ever owned, and all the hardware that you would "expect" to work (but probably didn't work as smoothly 10 or 20 years ago) Just Works (WiFi, external displays, excellent battery life/sleep, etc...)
Based on the complaints I have heard about Apple hardware and MacOS over the past few years, I'd even argue that Linux-on-the-desktop isn't any less stable or harder to get working than a Mac.
The only thing I miss is Photoshop but I really can't think of a single reason besides that to not use Linux anymore.
I think everyone would appreciate ideas for solutions.
I'd be interested to read more about this, and maybe even use your kext. I'm currently MITM'ing all of my SSL traffic[1] for a different, esoteric reason: I insist on using a 7-year-old version of macOS, and it doesn't natively support modern SSL ciphers, so I have to add it in with an mitm proxy.
I've run into a handful of issues with various software that I've had to work through as they arise, but if you've been doing this for ten years you've probably seen it all already.
1: https://forums.macrumors.com/threads/fixing-maverickss-outda...
Yeah, me too!
this is compounded by the fact that I love Little Snitch and it has basically exponentially improved my life when it comes not only to browsing the web but when using any app on mac.
Apple is going to lose developers.
How do we go about replicating this sort of per-process network visibility/permission on Linux?
It was painful at first, but it's worth it. The only things I still miss are the visual feedback in the UI (lots of little stuff) and the feel of the trackpad.
But the customizability has more than made up for that in productivity. Like being able to edit the source code for the window manager.
That being said, this is not ok behavior on Apple's part. There shouldn't be a way for traffic to go around the firewall like this, even if it is just Apple apps.
Because as Apple well knows, once you make a backdoor, someone will figure out a way to exploit it.
I can't help but see this as the real reasoning behind the change. With EARN-IT on on the table and antitrust cases looming, they've got every reason to bend over and give governments whatever access they can.