Molly (Signal fork) has not been updated since 3 months and is vulnerable

People use Signal because they don't want to be compromised when they just leave their phone / laptop unattended for a couple minutes on a bus or a train. They don't want to be in a situation if one day their laptop gets computer virus and they have to assume all these accounts are compromised and have to restart their digital identity all over again.

https://support.google.com/android/answer/15341885?hl=en

There are some tools available to make management even more straightforward, like Insular, a fork of Island: https://f-droid.org/packages/com.oasisfeng.island.fdroid/

They intentionally store very little data about their users but their "mailbox" model explicitly means that the server (or an attacker) can figure out who is talking to whom. And that's fine! Signal isn't trying to be an anonymity network, so while stuff like sealed sender are nice features it's important to remember that anonymity breaks are explicitly not part of the Signal threat model.

"Don't stand out" only makes sense as advice for anonymity tools to avoid bifurcating the anonymity set -- it doesn't make much sense when you're taking about a service which links your account to your phone number and thus every server request is explicitly identifying you.

You're also hedging against mandated government backdoors.

A single organisation like signal can easily compelled to add a backdoor (or leave the territory) like the EU is trying to do. Going after thousands of servers individually is a whole different problem.

The only valid reason to be against an optional feature is the maintenance burden, I don't think it's too much here but I also can't speak for the Signal team.

If you don't know how computers work, yeah it sounds great. But those who do know how computers work see it for what it really is: a bucket of shit.

I only have 2 people in my contacts that use it now. Every now and then somebody will install it for a couple weeks and ditch it again.

> Easily? There's no easy in maintaining and releasing two different versions of your app, especially for a small non-profit.

It's literally just another job on your CI. They got a 50 million seed donation in 2017 and reported almost 30 million in revenue in 2024. They are not exactly small. If they were VC founded and not a non profit they would be a unicorn.

Putting them on Signal requires me to convince them to download an app from the play store, with thr same name. The app is very similar to messenger or whatsapp and they're used to the functionality already. So many of my friends made an account so we can talk.

Anything more involved and they stay on meta's garden. Any successful messaging app needs to prioritize "normies" to get users.

This is not something exclusive to iOS; Android does the same thing.

Best of luck!

Copy the backup off your phone and use https://github.com/bepaald/signalbackup-tools to extract everything.

It seems to me like the app itself should be able to perform this operation and restore data? Why is this not just fixed in the app?

Further, I have been frustrated with similar design decisions with Signal - but how can they ensure you can export your messages without allowing letter agencies to do the same? Falling back on platform security guarantees is a rather good choice in terms of the economics - Signal need not gatekeep access to their private data - it is android and iOS that do the heavy lifting there. I’d love to move my messages between devices and sims - but I understand why that is not feasible for now.

As it stands it's a proprietary app with a lot o marketing to convince me otherwise.

They can give you the right to reuse their source code for building your own service using your own network (an aspire to give you this right), but still forbid you to connect to their service on their network with third party clients.

In practice, they don't seem to enforce that a lot (anymore).

But clearly connecting to the Signal servers with an outdated, modified fork is against Signal's goal.

> I think this is exactly what he wanted

You think, users should be required to carry a cell phone or make use of a debit card or credit card, in order to message another user, is exactly, what Matthew wanted?

A self-described anarchist who described the original cypherpunks as his teenage heroes?

When XMPP+OTR had already been, among competent people, the prevalent norm for over half a decade when he gave that DEF CON 18 talk?

That is just crazy, man.

In that DEF CON 18 talk, Moxie talked about scope of choice and how the scope of choice changed to "do I want to carry a cell phone, or not?". A notion that Matthew's dream was to impose on the world a demand to carry a cell phone or a debit/credit card transaction on others, such a tax, to enjoy what he and his friends and colleagues already on top of pure Internet Protocol, is on its face absurd.

_

> Spam is not a plague because registration remains expensive

The issue of spam is fully mitigated with an invite token, a 16 byte string ; "is", not "can be", because other systems in production have already been doing this.

The way it works is simple. If Alice wants to receive a message from Bob, Alice can share a connection string username+token to Bob. Then Bob can message Alice and for that time not other users of the hosted centralized namespace.

Very simple concept.

Simple enough for a mediocre dev to implement without an LLM.

The spectre of spam does not justify the demand for "checking in" with your local PSTN tower, nor the friction involved with an uninvolved and unwanted third party named Google (love Google overall but this is such a bastardly and disgusting concept), nor does it justify demanding a debit/credit card transaction become involved.

The spectre of spam does not justify any of this horse shit.

"Hey Bob, here is my username to message me, provided you subject yourself to this horse shit", where such horse shit antithetical in a fundamental way to the values espoused founder of Signal.

The new, grossly degenerated Scope of Choice.

Perhaps some who call themselves "hackers" or software developers may have been actually fooled by this false spectre of spam. I was never among them.

Even worse, the phone numbers, Google Play accounts, debit cards and credit cards, these are cheap. Among the scammer underworld these are a plenty and fungible.

An invite token personally sent by Alice or read aloud by Alice to Bob, is not.

But if somebody wants to harass a specific user on Signal, that's relatively cheap to do.

_

> Not to mention that sending ongoing traffic to your own server tells everyone where and when you're posting

Uncle Jeff, AWS, who doesn't tell exactly everyone but a lot of people, not excluding ICE at any given time, knows when you are posting, from where, and to what other users. The last of which is specifically obviated by a separate namespace hosted on a separate server.

On the occasions I have run Signal's server software, I and other users did mitigate these concerns. We do appreciate that others also have these means, or not.

Another messaging system, you may have already read its name actually, has already solved these problems in a way that they are successfully mitigated in a wholly transparent way for casual users.

_

> Alternative servers are not sapping user metrics.

This imperative is not the users' problem and runs counter to the best interests of the users.

_

> Its tradeoffs all the way down.

Right. In Matthew's view, the horse shit is a tradeoff.

I believe there are other tradeoffs available that are ethical, not more ethical, but ethical, and carry much less technical complexity.

I'll just say that the main issue with SimpleX is that it lies in its marketing material that it has no identifiers. What they really mean is that they're not adding new identifiers in the app, it doesn't mean they're actively removing existing identifiers, that is, your IP address. The client currently routes messages through two hosting providers which reduces incidental metadata collection. But it does fuck all if the hosting providers collaborate unbeknownst to the users.

That being said, SimpleX has really good support for Tor, in that if you just install tor package to Linux over apt or whatever, and configure like two items in the app's proxy settings, the client will automatically switch to SimpleX Onion Service servers instead of routing to clearnet servers through exit nodes. I can respect that.

What I can't respect is the threat model being treated like a fine print that's buried beneath menus: When you claim to be the most private messenger out there you need to act the part.