jonesv

joined 2 years ago
[–] jonesv@lemmy.ml 2 points 2 years ago (1 children)

The whole point of end-to-end encryption is that you don't have to trust their server: it cannot read your messages. Then for metadata, the question is about what metadata they are receiving at all (if they don't receive it, then you don't care if the server is proprietary) and what they do with it (e.g. for the private contact discovery, the idea is that you can verify that the code receiving your contact list is doing what it should (it's open source), and you can verify that this code is the one running in the secure enclave.

You don't need the whole server to be open source: only the important parts.

[–] jonesv@lemmy.ml 1 points 2 years ago

Let me ask: why does it matter to you?

[–] jonesv@lemmy.ml 4 points 2 years ago (3 children)

What's that closed-source server side? What does it do? Source?

[–] jonesv@lemmy.ml 3 points 2 years ago

As far as I am concerned, I use Matrix as a replacement for Slack/Discord/IRC, where I absolutely don't count on e2ee. And Matrix does that really well. For e2ee direct messaging, I use alternatives like Signal or Threema.

[–] jonesv@lemmy.ml 5 points 2 years ago

First, that's completely unrelated to the very concept of MITM. Second, it also shows that you have no clue about how such protocols work (in Signal, in SimpleX, or anywhere else). I really don't understand why people who are really into secure messengers often don't really care about how they actually work... I mean it is damn interesting!

[–] jonesv@lemmy.ml 3 points 2 years ago* (last edited 2 years ago) (2 children)

Not my point. Their webpage says that others (Signal included) are not protected against MITM (in the case of Signal, there is a note saying "if the server is compromised"). Which is plain wrong.

[–] jonesv@lemmy.ml 1 points 2 years ago

But probably those who made this attack were hackers, right? So "hit by a hacker attack" does not say that hackers are malicious, it's just a way of saying that it was an attack made with computers (and not with, say, fighter jets).

I don't think it's inaccurate or generalizing (hackers are not necessarily cybercriminals, and cybercriminals are not necessarily hackers, but cybercriminal who attack a computer system with a hack are indeed hackers). It's just a shortcut for "hit by an attack by cybercriminals who happen to be hackers, and used a skillset commonly attributed to hackers to execute their attack".

If that makes sense :)

[–] jonesv@lemmy.ml 3 points 2 years ago (4 children)

Oh, yet another messenger that pretends that it's the only one that can prevent MITM. Abusive marketing, or plain misunderstanding of what e2ee means? Anyway I wouldn't trust them just for that.

 

A super interesting essay about energy efficiency policies and rebound effects, and why it may be counter-intuitive.

"To focus on energy efficiency is to make present ways of life non-negotiable."

[–] jonesv@lemmy.ml 2 points 2 years ago

Signal introduced closed-source server side code last November.

What? I'm not aware of that. Source?

[–] jonesv@lemmy.ml 9 points 2 years ago

“Secure messenger” and “requires a telephone number” are not compatible concepts.

Following that logic, could we say that "secure messenger" and "requires a computer" are not compatible concepts, because the computer could be compromised? I mean, in the Twilio situation above, users got informed that the conversation key had changed (suggesting that they should verify the keys again if it matters to them). Now if your phone is compromised, you're screwed, whether or not your secure messenger requires a telephone number.

[–] jonesv@lemmy.ml 2 points 2 years ago

So instead of writing “Signal hacked by hackers”

Pretty sure it was "Signal attacked by hackers", but I get your point about "Signal hacked by hackers", though I don't think this would be worth an entire blog post :-).

trying to put in my mouth something I did not say.

On the contrary, I am trying to reformulate what I understood, so that you can confirm (or not) that I got your point. Don't assume that people who disagree with you are in bad faith, and you'll generally have a better experience communicating.

Anyway, that's not constructive, let's stop here.

[–] jonesv@lemmy.ml 4 points 2 years ago (4 children)

This seems completely off-topic to me. I never said I have nothing to hide. The Signal client app (i.e. the part that you can audit, compile and run, not the server) provides a lot of privacy already: e2e encryption via the excellent Signal protocol, private profile, private groups, sealed sender. So in terms of metadata, the Signal server never knows what you write, who is in which group, and to whom you are writing. Again, from the client code that you can audit yourself before you run it.

On top of that, leveraging the secure enclaves, the Signal server (tries to) guarantee(s) the private contact discovery (based on the hashes of your contact list). Which means that if you trust the SGX enclave, all that the Signal server knows is your phone number. If you don't trust the enclave, then you can assume that the server got access to your contacts when you did the discovery (i.e. when you installed the app).

That's very, very, very far from saying I have nothing to hide.

 

I believe that getting underrepresented groups into software development is a good thing. This is not a controversial opinion until you start talking about felons.

Today’s guest is Rick Wolter. He’s an iOS developer who served 18 years in prison for second degree murder. Rick killed somebody and for some that’s all they need to know about Rick. But today’s episode is about Rick’s path to redemption him, teaching himself to code in prison, smuggling in a Python interpreter, and then getting out and trying to get a job as a dev when you’re a felon.

25
submitted 2 years ago* (last edited 2 years ago) by jonesv@lemmy.ml to c/security@lemmy.ml
 

Users of the Signal messaging app got hit by a hacker attack. We analyze what happened and why the attack demonstrates that Signal is reliable.

 

Dependencies in CMake should be handled with find_package, not add_subdirectory. This posts tries to explain why and how.

view more: next ›