The author complains "For some damn reason, it matters which port your external disk is plugged into when you install or update macOS".
The reason is simple and perfectly understandable.
DFU is very low level. It happens very early in during Boot ROM and before the Mac has even entered Low-Level Bootloader. Which is why its also USB-C with no Thunderbolt support.
Boot ROM code is, by necessity and for robust security, kept to a bare minimum.
Bus 0 Receptacle 1 is designated the DFU port in the Boot ROM.
Hence the limitation to one port.
Widening support to >1 port would mean you would have to introduce extra logic into the Boot ROM code (port iteration, conflict resolution etc.).
Apple also makes it easier to achieve that privacy:
- They put all the privacy controls in one place in Settings so you can audit
- App developers are mandated to publish what they collect when publishing apps to the App Store.
> - They put all the privacy controls in one place in Settings so you can audit
That’s true. On Pixel Android, there’s several unrelated places in the various settings for the device and for the Google account to take care of and see that they do not collide. And for every function there’s always some sort of small print like “it’s all private to you unless you choose to share” - but to use any of the features/services you have to “share” like with Google Photos and Calendar and Tasks, you lose track of what you share with whom in the end. So essentially not only the metadata is collected but also the content and nothing’s private as a result, at least that’s what I got to understand. And even if you ask Google to delete your personal information, it will retain it for a while for compliance purposes.
As for
> - App developers are mandated to publish what they collect when publishing apps to the App Store.
I believe that’s still moot and rather a voluntary disclosure that no one vets. I’ve seen apps with no collection stated on App Store but deviating privacy policies, or app functions that contradicted their own privacy policy.
From what I heard and read, I understood that as a well-meant idea but still a misconception on the consumer part due to lack of enforcement by Apple.
> From what I heard and read, I understood that as a well-meant idea but still a misconception on the consumer part due to lack of enforcement by Apple.
I'm not familiar with the detail so I cannot comment directly on what you are saying. I don't have the time to go read up on it right now.
But what I would say is that many aspects will be indirectly enforced by Apple (and can be audited/enforced by the user) through the privacy controls (location services, microphone, camera etc.). Clearly that does not cover everything, but it covers a large chunk.
Apple have also made it impossible to for example get a device-level ID, you can only get an app-level pseudo-device-id. So there are various code-level enforcements too.
> As a paying customer I'm concerned about the company's focus being blurred when there are 3.8k open issues on their Github repo and my company has been tracking some particular issues for years without progress.
I feel exactly the same way.
So many open issues, the majority thoroughly deserving of a resolution.
I would rather they get their house in order on the core product first before rushing out shiny new things .... because the shiny new alpha/beta things will only exponentially increase the number of open issues.
To be fair, I entirely understand the absolute necessity for this.
The reason for its introduction is before hand the PHVs (Uber etc.) of this world would, instead of using the car parks, go up to the drop-off area and wait there.
Because there was no charge and no penalty, what they would do is drop off a passenger and then sit there waiting for their next job to ping on their screen.
This became a particular problem at Heathrow T5 where the drop off area is relatively tiny.
The result would be that at busy hours, private individuals attempting to drop off their friends and family would be unable to find space and end-up double-parking and causing safety hazards.
For a while they tried to use airport Police to enforce it, but the volume of PHVs was just far too great. Hence the cameras, charges and penalties were introduced.
It should also be noted that at Heathrow, if you do not want to pay the £7, you can instead drop people off for free at the Long Term Car park and they can get the shuttle bus back to the terminal.
Rather than charge everyone £7 or more for a drop off, wouldn't it make more sense to charge the people abusing it an absurd amount? I'd much rather see a £25 fee after 90 seconds and an additional £125 fee after 5 minutes than £7 for 30 seconds.
It seems less about making things more efficient and more about just squeezing a little bit out of money out of everyone.
In San Francisco we have toll tags called FasTrak. You can pay for parking at the airport with it. Of course, there, it's just the normal, pretty high airport parking rates, but there's no reason you couldn't use such a tag for enforcing quick free drop offs and pickups with exactly that much precision. Enter the drop/pickup area with your toll tag, if you're out in 3 minutes, no charge. 5 minutes, $4, and if longer than that, $20/hour or whatever. It's not like computers mind doing that math.
> I wonder how many actual terrorists they pick up for saying "I'm here for terrorism"
Its like those stupid questions on US immigration forms, e.g.
"Do you intend to engage in the United States in Espionage ?"
or
"Did you ever order, incite or otherwise participate in the persecution of any person ?"
It's like, really ? Do they seriously think someone who should answer yes will really answer yes ?
Might as well just turn up at the immigration desk, slap your wrists down on the counter and invite them to handcuff you .... why bother with the form !
> the purpose of the form is to generate convictions for lying on the form.
Yeah but if the immigration officer has reason to question you about those sections of the form then surely they have more than enough evidence of the underlying crime anyway ?
It’s often an easier case to prove that you lied on the form when you said you came to the US with no intent to commit espionage than it is to prove that someone committed espionage.
It basically unlocks a second set of potential facts that they can use to bring a criminal case (or revoke a visa, etc).
Intent to commit espionage is not a crime (but committing or attempting to commit it is) Lying on the form is. It is probably easier to demonstrate intent to commit espionage than to catch them in the act.
Wouldn't it be easier to make those things illegal and then prosecute them instead of the lie? For prosecuting a lie you need to prove 2 things, the thing lied about and the lie itself, so it seems like a more difficult prosecution for no reason. Also how does every other country in the world manage to not have these questions?
> Also how does every other country in the world manage to not have these questions?
You sure about that? Many other countries have what would be considered odd questions on their forms.
Also, saying "every other country" is a mighty wide brush. There are a whole lot of countries where the rule of law doesn't come first and they can simply do what they want if they suspect you of anything regardless if they have a law or not.
This is what happens when a legal system and a political system is taken over by specialists with little to no other skills.
Instead of politics being about setting policy to work toward desire outcomes, politics becomes about ensuring the viability of future political processes. Instead of the legal system being about defining crime, establishing punishment and carrying out said punishments it becomes about ensnaring others in legal "gotcha" moments like lying on a form. Society is not safer because of the outlawed nature of lying on a form. Society is not better off because someone is convicted of lying on a form. The individuals who participate in the prosecution are better off because it gives them an opportunity to advance their career.
Making false statements to federal officials is itself a crime. The intent of having those sections is to be able to have legal recourse against people that lie on them, which hopefully deters people that would lie on them from attempting to immigrate in the first place.
> Now more than ever, trusting a US jurisdiction VPN provider ? No thanks !
The whole point of Obscura is you aren't trusting any single company. A Swedish company and an American company would need to collude to cause a problem. Unless you know something I don't?
> The whole point of Obscura is you aren't trusting any single company.
First, Mullvad's infrastructure has been independently audited.
Mullvad integrity has also tested as proven by a legal case where they were subject to a search warrant when someone was trying to claim copyright infringement.
As far as I can tell, Obscura has not had anywhere near the same scrutiny.
Second, obscura is the first hop is it not ?
Therefore it may well "only" relay the traffic to the exit node but it is still a relay and hence open to SIGINT analysis by the US.
I would have thought therefore using Mullvad's built-in multi-hop mode on their audited platform would be the wiser decision ?
Hence why Mullvad is being used as the exit point.
You have full e2ee between yourself and Mullvad but crucially Mullvad don't know who your IP. Five eyes are already doing SIGINT on behalf of both the US and the UK government before my connection even reaches Obscura so I lose nothing but potentially gain privacy.
How is it you think a single company (Mullvad) having access to my IP and what I am browsing is less secure than splitting it up amongst multiple providers one of which being Mullvad with that audited platform you talk about?
If I wanted Tor on top I'd layer it on top too but that would still be a single point of failure.
It's open source which means I can trust having the app installed if I build from source (or I can just use Wireguard directly). I then know I'm directly connected to a Mullvad Wireguard node by checking the public key here: https://mullvad.net/en/servers
Other than Wireguard protocol being broken there is no way for Obscura to snoop presuming I check the public key. I'm not saying I trust Obscura, I'm saying with their model I don't need to trust them which is vastly superior. Nor do I need to trust Mullvad.
You keep hand waving around that Obscura are somehow untrustworthy but you have steadfastly refused to address the fact that their model does not require trust. If you trust Mullvad (which you are claiming to) please show an attack that would work to breach this model. You can't.
Yeah. It was formally announced in the November 2025 budget and launched today.
reply