this post was submitted on 08 Dec 2024
64 points (89.0% liked)

Privacy

4388 readers
71 users here now

A community for Lemmy users interested in privacy

Rules:

  1. Be civil
  2. No spam posting
  3. Keep posts on-topic
  4. No trolling

founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
[–] ctag@lemmy.sdf.org 0 points 2 weeks ago* (last edited 2 weeks ago) (2 children)

I don't trust Signal. Haven't used it since it went down when people and capitol rioters fled WhatsApp and signed up. My understanding is it's a brittle centralized system just like WhatsApp.

AND back when I did use it, the app had dark patterns that included spamming all your contacts when you set up the app.

Matrix still needs work, but it is the future in this space.

[–] Kalcifer@sh.itjust.works 2 points 2 weeks ago* (last edited 2 weeks ago) (1 children)

Matrix still needs work, but it is the future in this space.

Matrix can send encrypted events ^[1.1]^, but, imo, the Matrix protocol is a firehose of metadata ^[1.3][1.2]^. I'd argue that metadata leakage doesn't lend itself well to anonymity; if one seeks anonymity, then I think they should seek to reduce their metadata footprint, as, logically, any information is better than no information when trying to identify someone.

References

  1. "Client-Server API". "Matrix Specification". Matrix Foundation. Version: 1.12. Accessed: 2024-12-09T02:21Z. https://spec.matrix.org/v1.12/client-server-api
    1. §"Sending encrypted attachments" (§10.12.1.7). https://spec.matrix.org/v1.12/client-server-api/#sending-encrypted-attachments.
    2. §"Extensions to m.room.message msgtypes" (§10.12.1.7.1). https://spec.matrix.org/v1.12/client-server-api/#extensions-to-mroommessage-msgtypes.
    3. §"Events" (§10.2.1). https://spec.matrix.org/v1.12/client-server-api/#events-2.
[–] ctag@lemmy.sdf.org 1 points 2 weeks ago (1 children)

I think that's fair, maybe I should have said efforts like Matrix.

But I'd also view a singular commercial company's no-cost product as not being a long term bet on privacy/anonymity.

[–] Kalcifer@sh.itjust.works -1 points 2 weeks ago* (last edited 2 weeks ago)

But I’d also view a singular commercial company’s no-cost product as not being a long term bet on privacy/anonymity.

Oh I completely agree with the idea that a federated service, like Matrix ^[1]^, is far preferable to a centralized one, eg Discord ^[2]^; I have no issue with using Matrix (I personally use it extensively) I just think one should be careful with the idea of using it with the intent of being anonymous.

References

  1. "Matrix Specification". Matrix Foundation. Version: 1.12. Accessed: 2024-12-09T06:41Z. https://spec.matrix.org/latest/.

    Matrix defines a set of open APIs for decentralised communication, suitable for securely publishing, persisting and subscribing to data over a global open federation of servers with no single point of control. […]

  2. "Discord". Wikipedia. Published: 2024-12-07T05:34Z. Accessed: 2024-12-09T07:03Z. https://en.wikipedia.org/wiki/Discord#Features.
      • The license is listed as "proprietary". This was interpreted to mean that Discord is the owner of all of their infrastructure.
[–] Kalcifer@sh.itjust.works 1 points 2 weeks ago (1 children)

I don’t trust Signal. Haven’t used it since it went down when people and capitol rioters fled WhatsApp and signed up. My understanding is it’s a brittle centralized system just like WhatsApp.

Imo, there are more components to trust than service reliability (iiuc) — eg: trust in the underlying protocol, trust in the governing body etc.

[–] ctag@lemmy.sdf.org 1 points 2 weeks ago (1 children)
[–] Kalcifer@sh.itjust.works 1 points 2 weeks ago (1 children)

Is that an "agree" as in you hadn't considered what I said, or that you agree to that in addition? If it was the latter, I should clarify that I wasn't adding supplementary information — I was outlining what I thought was a flaw in your rationale (eg argument from ignorance) for distrust in Signal.

[–] ctag@lemmy.sdf.org 2 points 2 weeks ago (5 children)

It's an agree as in I don't really feel like arguing with another user here. I don't buy the point about metadata when Signal, a centralized service like Discord (why are we talking about Discord?), may be able to scrape it too. Or the point about anonymity when Signal is far from the right tool for that purpose too, see above "spams your contact list."

For reliability, I'm not concerned with how much RAM Signal's servers have. What I should have highlighted is that Signal can nuke your communications on accident / on purpose / under coercion. And it's proven because they've already done it before. Mitigate that by having a backup system set up? That necessarily doubles your surface area for breaks in privacy or whatever a given user is worried about. So starting with Signal in the first place doesn't make sense to me.

[–] Kalcifer@sh.itjust.works 1 points 2 weeks ago* (last edited 2 weeks ago)

[…] And it’s proven because they’ve already ["nuked" communications] before. […]

Would you mind providing a source of this? This sounds interesting, and good to know.

[–] Kalcifer@sh.itjust.works 1 points 2 weeks ago

For reliability, I’m not concerned with how much RAM Signal’s servers have. […]

I'm not sure that I understand this statement. What does RAM have to do with with Signal's infrastructure reliability?

[–] Kalcifer@sh.itjust.works 1 points 2 weeks ago

[…] What I should have highlighted is that Signal can nuke your communications on accident / on purpose / under coercion. […]

Are you referring to the possibility that they may be able to block communications, as they are a centralized service?

[–] Kalcifer@sh.itjust.works 1 points 2 weeks ago

I don’t buy the point about metadata when Signal […] may be able to scrape it […]

I agree that it it within the realm of possibility, but, imo, this is independently verifiable, as the Signal apps are open-source ^[1][2][3]^ and offer reproducible builds (except iOS ^[2.1]^) ^[1.1][3.1]^. See this section on Signal's metadata for some more concrete information ^[4]^.

References

  1. "signalapp/Signal-Android". Github. Published: 2024-12-06T21:08:08.000Z. Accessed: 2024-12-11T05:38Z. https://github.com/signalapp/Signal-Android.
    1. Filepath: "reproducible-builds". Published: 2024-10-29T18:36:34.000Z. Accessed: 2024-12-11T05:40Z. https://github.com/signalapp/Signal-Android/tree/main/reproducible-builds.
  2. "signalapp/Signal-iOS". Github. Published: 2024-12-04T21:19:41.000Z. Accessed; 2024-12-11T05:41Z. https://github.com/signalapp/Signal-iOS.
    1. "Reproducible builds" (#641). Author: "Jeroen Massar" (massar). Issues. Published: 2015-03-03T09:16:05Z. Accessed: 2024-12-11T06:11Z. https://github.com/signalapp/Signal-iOS/issues/641.
  3. "signalapp/Signal-Desktop". Github. Published: 2024-12-04T22:57:07.000Z. Accessed: 2024-12-11T06:12Z. https://github.com/signalapp/Signal-Desktop.
    1. Filepath: "reproducible-builds". Published: 2024-11-21T03:14:21.000Z. Accessed: 2024-12-11T06:15Z. https://github.com/signalapp/Signal-Desktop/tree/main/reproducible-builds.
  4. "Signal Protocol". Wikipedia. Published: 2024-11-30T04:32. Accessed: 2024-12-11T06:18Z. https://en.wikipedia.org/wiki/Signal_Protocol#Metadata.
    • §"Properties". §"Metadata".
[–] Kalcifer@sh.itjust.works 1 points 2 weeks ago

[…] starting with Signal in the first place doesn’t make sense to me.

If you have the means to choose something more secure/trustworthy/robust than Signal, then I think it would be in your best interest to do so! I personally would recommend SimpleX, if possible.