Hacker News Reader

Comment by leksak | Hacker News Reader

It's a struggle getting non-privacy minded people to change to Signal.

That's because they refuse to make it seamless in a way iMessage is. Or Telegram. Both of those will work on several devices, have dedicated clients, synchronize messages and iMessage will transparently use SMS (on all platforms).

Signal does nothing of the sort - it's impossible to backup conversations, they won't sync to a new device, it won't work on a tablet, it's SMS support is buggy and incomplete and their desktop client is based on a dead technology (and limited to a single machine without SMS support). It also doesn't offer anything over builtin SMS/call support like Telegram does. To top it off, it hijacks SMS store meaning you can't even use OS-based software to backup at least SMS conversations.

I can't give Signal to non-tech users and just say "hey, use this instead of SMS", because it won't work and it'll lose their conversations. I can do so with iMessage.

> their desktop client is based on a dead technology (and limited to a single machine without SMS support)

They're working on an Electron version (Chrome apps aren't dead yet, so they still have some time), and I'm using Signal Desktop on two machines.

Are you perhaps also using it on two Android/iOS devices?

No, one Android phone, two computers.

>Electron

Absolutely disgusting.

I'll double my RAM every year if that's what's necessary for cross-platform apps that are truly easy to develop.

Regardless of your opinion of it, it's not "dead technology" like Chrome apps are, i.e. apps written using it are not guaranteed to stop working at some point.

I wouldn't be so sure. Electron (libchromiumcontent) is a patchset on top of Chromium and it can be easily broken by upstream changes.

But its deprecation hasn't been planned, and a binary you provide won't suddenly be upgraded and stop working. That is not the case for Chromium apps.

Yeah, I'm actually more worried about Electron devs manpower/ability to catch regressions on the very big and simultaneously fast moving codebase where the upstream project doesn't really care about their use cases. This must be total PITA. Otherwise it's an ok platform, I guess.

iMessage doesn't sync to new devices either. Only new conversations going forward are available since each device has its own encryption key.

My guess is it's because it still feels a little rough around the edges as a chat application, which shows its open source origin (most open source projects lack sufficient funding, etc)

Every 6 months or so I try pushing it onto "normal" people, and then I realize something is not working quite right. A while ago it was the Signal desktop app that wasn't importing contacts. I also have an issue now where I can't delete/hide a contact from the contacts list in there, even though it doesn't appear on the mobile app.

Speaking of which, I wonder what's their plan for the desktop app, as I believe Google has deprecated those types of apps in Chrome and I don't think it will work anymore in about a year. I've also used WhatsApp more because the native desktop app seems much more reliable.

Then it was messages arriving too slow (waiting many minutes to be "pushed" as a notification, which is pretty bad for a chat application). I think it still has this issue.

The video call still doesn't seem to work as well as the WhatsApp video call. It seems buggy and the connection is not of the same quality. WhatsApp video calls are more resilient than Google's Duo, too, which switches to audio often.

Even the ringtone the video call uses by default is worse than the one WhatsApp uses. I know it seems like a little thing to complain about, and if it was just this one little thing, I probably wouldn't complain about it. But it's all of these together that still stop me from pressing others too much to use it.

I would also like to complain about OWS/Moxie's inability to "build a community". Yes, many cryptographers and security people like the app for the good tech it uses and such, but the organization is not actively trying to build a community or care much for it from what I've noticed. I think Durov and Telegram work on that much more, and it's important. In fact, it may be one of the primary reasons why Telegram is so much more popular.

OWS/Moxie barely even post on Twitter anymore. They release a blog post every few months. It's hard to "get excited" about the app or want to follow them closely too much when they do this.

I think this is also the reason why they're losing/have already lost an opportunity to become the default secure messenger in the Democratic party to Wickr, which is much more active on outreach and stuff like that, even though Clinton's team used Signal in the campaign and they got a lot of good press for it. But they're now blowing all of that away because of their "silence" and non-marketing focus.

Tracking the changes to the Signal Desktop Application on the WhisperSystems github, it looks like they're moving to an Electron app (like Slack). Thumbs down from me for code bloat, but it'll be nice to have a separate app and not be required to keep Chrome open all the time.

Video calling on Signal works a little differently than WhatsApp to my understanding; it's decentralized and so the quality is lower, but the Signal server manages to keep from collecting any data, and backdoors can't be added (it just sets up the call and then bows out).

> Thumbs down from me for code bloat, but it'll be nice to have a separate app and not be required to keep Chrome open all the time.

In the sense of having a separate instance of Chrome, dedicated to Signal, open all the time:-)

It's probably still the best, quickest cross-platform solution — which is just sad. Although I wonder how bad the current Java compatibility story is: if they used Java maybe they could reuse some of their Android code.

> My guess is it's because it still feels a little rough around the edges as a chat application

When I tried it in 2015, it frequently broke, lost chats, lost history, etc. Security means nothing if the secure app itself is unusable.

The infuriating thing is that Telegram makes it harder. Some of these have been convinced to give Telegram a try sometime in the past, some actually still use it and like it, but in any case: they believe that Telegram's privacy was good enough, and why would they give yet another app a try now?

You're right. Privacy is good enough and apps do not suck.

Telegram privacy is not "good enough" for a commonly used communication platform. Most messages are actually stored in plain-text on middleware servers.

I'd be surprised if some state actors don't have access most Telegram messages.

If you're switching from WhatsApp to Telegram, you're actually going backwards in terms of privacy, so if that's what you were led to believe, you've been misled.

Not sure if it changed but one of the most frustrating aspects of it has been finding out who actually still uses signal. The few people I knew used it stopped using it and that eventually sealed the deal.

My whole family use it. It's really quite nice.

It's a struggle getting privacy minded people to switch to signal, since it requires giving out your phone number. This is especially an issue for privacy minded women. The Signal team really did not think that through.

Reply on news.ycombinator.com