Here is their answer:
Hi,
Thank you for your thoughts on the announced SMS removal. The blog post describes all of the biggest factors in making this decision, but I know this is a change that is difficult to adjust to, so I wanted to chime in with some additional info that might give some more context.
1. RCS (Rich Communications Services) is coming, and it doesn’t play well with Signal. I once had a situation when I was sending SMS to one of my friends via Signal, but I wasn’t seeing any of their responses – this was because their app was automatically responding via RCS, which wasn’t delivered to Signal. This is going to continue to get worse, and Signal cannot add RCS support because there’s no RCS API on Android. Honestly, the days of any third-party SMS app are numbered.
2. Proper SMS/MMS support is hard. Signal has to support thousands of devices running dozens of versions of Android. Now multiply that by the hundreds of cell carriers running an inherently bad/buggy protocol, and you’ll start to understand the weird MMS bugs we can run into. And any time spent trying to fix them is time invested in an insecure protocol.
3. SMS/MMS has plenty of its own bugs. Remember that incident a few years ago in which everyone got old Valentine’s SMS messages delivered 9 months later? It was an SMS protocol bug for which some users blamed Signal. Other weird bugs like temporarily-split MMS groups, bad image quality, and the general inability to leave MMS groups are flaws in MMS that also get attributed to us.
4. Spam. My goodness, SMS spam is a real thing, and many people who use Signal cannot tell the difference between SMS spam and Signal messages if both come through Signal. They think we’re responsible for the spam.
5. Finally, Signal having SMS support gives a lot of people the wrong impression of SMS. They think that because SMS is being sent through Signal, it’s actually secure or as secure as an encrypted Signal-to-Signal message, and that’s just not true. We can add unlocked padlock icons to each SMS message, and we can label the message compose box as “insecure”, but the misunderstanding would continue. The only thing we can do is store the SMS messages encrypted on the device, but in my opinion that matters very little when anyone who wants your SMS messages can just get them all from your cell carrier.
In short, SMS is on its way out in general, and in a world where Signal supports SMS, all of SMS shortcomings are often attributed to Signal itself, all while confusing people into thinking their SMS messages are secure.
In my opinion, a secure SMS app does not exist. Just choose the one with the best layout or usability, and preferably one that supports RCS (which I believe at this point are Google and Samsung Messages), because at least then there’s some chance that they might end up being encrypted in the future.
I hope that helps give some more context. And please know that I understand this is difficult to adjust to. I can relate. I’ve used Signal as my SMS app for over 6 years, but I truly think it’s for the best.
They tried to make a bazillion messenger apps, all of which failed, and now they try to piggyback on an existing standardized protocol but don't expose any APIs for other apps.
While at the same time pushing this "Apple is bad because green bubble" narrative because Apple doesn't support it. It's somewhat amusing in some ways. Companies (Google or Apple) are never on the consumers side, and yet we fall for it all the time.
So today it's either Google or Samsung messages and Jibe, otherwise RCS is essentially useless. From what I know Facebook wanted to get on board at some point, I'm not sure why this didn't work out.
The fact Google publicly whining about the iMessage lock-in is pretty rich.
Saying "this is a better SMS app" got people on-board and let them "upgrade" to secure messaging. That's why I started using it in the "TextSecure" days.
But, sadly, I agree with Signal's reasoning here. Mixing the two protocols was annoyingly complex. If someone stopped using Signal, messages you sent to them would never arrive - with no notification. And there's no obvious way to "downgrade" to SMS.
I was working on RCS a decade ago. I'm glad to see it is finally getting somewhere - but I'm sad it is at the expense of better and more secure protocols.
I certainly wont mourn for this. Sms gets super expensive to other countries (outside the EU), it also gets super expensive when you don't pre-pay for a bundle (it's 2 eur for 1000 messages, then 25ct per message after that, ridiculous). Imho dutch carriers did everything they could to kill sms, and now I'm happy for it. The only bad thing is that my society at large seems to have chosen Whatsapp as it's replacement...
Really? Do "we"? Everyone keeps warning of these issues while getting laughed out of the room. At some point I have to believe that most of society actually actively wants to be screwed over.
I wanted a no-code solution to push my bank-sent SMSs into a real-time transaction database. Apparently this is impossible on iOS, and on Android it's incredibly unreliable (IFTTT & Tasker keep getting shut down by the OS). Fortunately, before writing my own solution, I found messages.google.com/web. I have to scrape this, but that's simpler than maintaining an Android app.
Somehow, it's 2022 and if I want real-time banking information I have to write code to consume SMSs OR hand my banking credentials over to a third party to scrape my bank statements.
But there are threads in their issue tracker where Signal people would disagree with basically functionality of responding to text messages on dual sim devices. "We'll always use the Signal registered number as default".
One part of this is a terrible(but working) protocol, the other is really weird product management where basic needs of the consumer are brushed aside with a "I know best approach" that doesn't really work when you're not in the shoes of apple and have the ability to just replace SMS altogether.
If you need to actually send files etc use another one of a multitude of messaging apps out there including the latest Google TalkMsgChatThing (but it works)
I can imagine that's less of an issue in the US; do you pay extra for text messages and calls that go across state lines?
Took a lot of grumpy users before they backtracked.
I use and like Signal. I'm not smart enough to disagree with some of their technical decisions, but I wish they could have found a way to make SMS/RCS work.
The only large group of people who still primarily use SMS to communicate person-to-person is Android users in the USA.
Every other country has settled on either Telegram, WeChat, WhatsApp or FB Messenger, or other niche apps. These apps work on both iOS and Android and often also Windows. I haven't sent an SMS in probably 12 years. I don't know anyone who has.
It's only in the US that iMessage is so prevalent that Android users have to use SMS, the only other way of messaging iOS devices. And the US is quickly becoming a de-facto iOS only country. It already has more than 50% market share, even 80% among young people.
With the US going (almost) full iMessage and the rest of the world having already settled on another app there simply no point to supporting SMS.
- Signal -> Close friends & family I've convinced to use it.
- WhatsApp -> Most of my friends.
- SMS -> School notifications, 2FA, shipping updates, etc.
- Facebook Messenger -> Elderly relatives
- Telegram -> That one relative who wants to use this instead of Signal.
Is there a consolidated messaging app that the HN community recommends?
I can't be the only one suffering from messenger bloat.
And their president also commented on it a bit at https://www.theverge.com/23409716/signal-encryption-messagin...
This was Signal for me. I understand the reasons to get rid of SMS support but I strongly disagree. Unfortunately I don't have enough people in my life on Signal to justify keeping it around as yet another messaging app.
The specifications are public, but yeah it requires a lot of work to get into.
That being said, I think neither of those limitations apply to Google's "RCS" (which bypasses carriers, so they can't get their own IPv6 connection and I don't think they can use SIM challenges for auth), so I think it should be possible in a 3rd party app.
Almost all phone plans have unlimited (domestic) texting and calling and differentiate themselves with data, reliability and ”free” perks like subsidized Netflix. You really have to go out of your way to find a plan with limited texts/calls.
But somehow people started using it because it was "more secure" than whatsapp.
Do you have any data to back it up? I have trouble believing that, but I am admittedly biased against Apple devices.
That's obviously not true. I live in a European country where a lot of people are still using SMS.
But that's pointless to have inside the Signal app, which is for person-to-person communication. I wouldn't even want those messages in WhatsApp even if it could do it.
I do still occasionally get work conversation initiated via SMS rather than WhatsApp especially if that comes from a phone which is associated with a task or job. Like the out of hours mobile phone which is moved between people.
By definition, Signal is not going to have bells and whistles and niche features, to keep the codebase lean and easier to avoid security weaknesses.
If you want an app with many features, use Telegram.
(They did however ship emoji reactions in Signal because most people actually like it and the other apps do have it; curiously missing in Telegram).
Has there been any new press on this? The last nugget I heard was https://www.android.com/get-the-message/, which didn't really announce any progress.
I think it would be more accurate to say that ongoing communication via SMS messages isn't common at all any more. They're like a protocol negotiation handshake.
1. Requires data connection, either WiFi or mobile data. SMS does not.
2. On Android, it goes through Google Messages. I don't trust Google.
So, I replaced Messages with another SMS app that doesn't spy on me and disabled RCS. If I wanted to do rich messaging through data connection, I would use Signal or similar.
So, RCS is not a replacement for SMS. It's just another IM protocol. And we have enough of those.
As for others replying through RCS, that's a failure of the messaging app, should recognize that the counterpart is SMS and reply in SMS. It's an attempt to force RCS...
SMS is for fast short messaging between mobile users, and that is its killer feature, it's universal. Every mobile phone supports it.
Signal dropping SMS support makes no difference for me, never used it for SMS, I used Signal for the secure conversations. Two different methods of contact.
I don't understand why people on this board keep forgetting that central Europe and the US is the minority of the world population.
Everyone keeps bringing up WhatsApp. But it seems that everyone has all but forgotten that WhatsApp became so popular not because they only focused on the US market, but because they went around the globe and specifically targeted feature phones as well. I.e. they understood that their own home turf isn't enough to make a dominant chat application.
In the UK, where I am, most contracts have unlimited SMS. I think even the cheapest PAYG plans include massive bundles.
Lots of companies send out reminders by SMS because it is universally accepted. Not every customer has WhatsApp.
I now have everyone I talk to regularly on Signal (30-40 people), but it took years.
For me personally, Signal will turn into a messaging app among dozens of similar apps, of which I have enough already. SMS was why I stuck with Signal over the years I used it and now I have no real reason to stay on it.
Most of society is brainwashed by commercials, don't even understand half of the words techies use and would call you pretentious if you are not using iOS or Android
Signal's rationale is just Signal's own reluctance to build an umbrella messenger. And given they do drop SMS, still won't introduce usernames it's very hard to actually sell it as a WhatsApp replacement.
And now, with WhatsApp supporting password protected cloud backups and up to 2Gb attachments, I'd say Signal will loose the userbase it acquired during the hype and Musk tweet.
In fact, during 2020 Belarus protests, Signal did nothing to support it's own operations during internet semi-blackout in the country, while Telegram tweaked their server side to provide at least some possibility to know what was happening in big cities. So what are the values of Signal — I don't even know. But they sure did support pillagers and rioters in the USA.
To be even more brazen, Signal is not Apple. They stopped innovating. And they don't have enough political power to convince people do things the new way. Even their zero knowledge server is worthless. Check out the story on FBI cracking down on the leader of some right wing proud boys type of armed group. They tracked him and then compelled to give up access to Signal.
Their innovation stopped at providing solid cryptography that was adopted by most decent messengers already. And they aren't visionaries with cancelling SMS.
UPD: the funniest part is that the service that drops the SMS support still relies on SMS to provide account registration.
This is just an unprecedented level of sarcasm.
I was hoping to read here someone had forked it here. I won't be continuing with Signal.
Some months my data volume doesn't last till the end of the month. I use SMS instead of the Signal protocol then.
Yes I might be a minority, but if you're not the market leader, cutting out minority groups of users will not make you more successful.
What's your mission: Giving secure communication to everyone or become the next WhatsApp?
You can see various threads where people complained about the forced Apple emoji e.g. https://github.com/signalapp/Signal-Android/issues/3712 and https://github.com/signalapp/Signal-Android/issues/3675
Signal's own FAQ says
> Signal Android includes built-in emoji functionality for consistency between platforms
https://support.signal.org/hc/en-us/articles/360017561992-Wh...
My initial response on it being announced in the Signal app was "Oh no, that's terrible!". Followed by "Meh, all software goes to shit eventually. Now it's Signal's turn". Now I already don't really care anymore.
I have been using Silence [0] as my sms app for a day and don't really miss the Signal sms integration anymore.
What bugs me more is that the text message export from Signal seems incomplete. Oh well, I will get over that as well.
(Funnily enough Silence is a fork of Signal for SMS only. I thought it looked quite familiar but just realised it is the case)
The developers were adamant that they know better than me what I want from my SMS tool. And then proceeded to work on crypto and stories.
Any proof? If you're calling MTProto 2.0 'insecure' then you should know it's already been audited multiple times in the last 2 years. If insecure means not using E2EE, then I guess the whole infrastructure of the internet is insecure.
> It's not even anonymous
It's more anonymous than Signal is. It requires phone number to register but you don't need to share a phone number or any personal detail to communicate with people.
Source https://www.ofcom.org.uk/__data/assets/pdf_file/0011/222401/...
SMS decline is probably inevitable though.
Signal team is also actively hostile to any 3rd party client usage of their service.
And that's always cheered on this website - just remember the RCS topics where people were making fun of attempts to add some basic standardisation to this mess.
I can't remember the last time I got such a response. It's a challenge to get them to respond to complaints
BS. I visited city I grew up recently, met with a few (9+) people (25-32 years old) and only one of them had WA, most haven't heard about Signal, everyone simply uses SMS. It's simply multiplatform, works with their gradmas and no one wants to install __another__ app to send messages to people. No one cases about RCS that will be used to push QR codes and ads, people will use SMS for its simplicity and reliability. I'll be dropping Signal and moving to WA once Signal drops SMS support.
WhatsApp is a closed protocol owned by Facebook. It has its uses but relying on it is a mistake.
https://www.statista.com/statistics/271561/number-of-sent-sm...
E.g. EU countries are split between plain SMS and WhatsApp as the preferred messaging platform for automated interactions with services. Google not offering an RCS API would be like pushing everyone to use WhatsApp for automated interactions. Would Google risk that? Probably not.
By pulling the plug from SMS support now, Signal just makes it certain that they will lose user base, no matter what Google does.
And it is also kind of disrespectful to anyone who spent their personal time advocating the use of Signal to friends and family.
The stats show a significant drop as mobile data became cheaper and richer services became available, but still quite a lot of traffic.
I suspect that the people I see using Nokia and Samsung dumb phones will continue to use SMS, so traffic will fall to a sustained tail.
https://www.statista.com/statistics/271561/number-of-sent-sm...
Telegram has amazing user experience. It's available for any platform, the messages are always backed up, the apps are high quality and responsive and they have great features for group messaging and group organization. They even give you a library you can build your own Telegram client with.
It's *great to use* - something that Signal people never prioritized and always rather pushed their sometimes horrible preferences down peoples throats.
Alternatively if you are in Europe/UK the openbanking API is quite widely used now
From the answer and the wikipedia page it sounds like this might be a decision that the other side's client (SIM? carrier?) makes, to reply via RCS.
And my device would be able to receive this only via data connection? Or would my carrier detect that I'm not currently in 4G covered area (middle of Germany for example...) and send the text via regular SMS?
I'm a bit worried by the first point, really. What does that mean for e.g. pinephones or google free android? Do I risk silently missing text messsages?
Basically, SMS used to be a big revenue driver for operators. That business has dried up almost completely. The notion of paying per message is just completely gone. So, operators stopped caring about SMS a long time ago. In the same way, call minutes are increasingly less relevant. It's all about 4G and internet now.
I think you just described the core of the problem: Signal has a very US-centric view of the market, and has no clue that SMS is actually still relevant elsewhere, and a low-hanging fruit for capturing user-base.
It genuinely surprises me how often the HN crowd falls for it.
I would have expected them to be more aware than the average consumer of how tech companies use lock-in to trap their users.
Perhaps there's a blind spot because lots of HN folks work on projects or for companies that (hope to) use the same tactic?
The hidden gem is that basically Google is taking over control of RCS, and by proxy "text messages". There will only be one Android implementation, and it's Android Messages. It (finally) comes with E2EE messages with something resembling the Signal protocol, but you lose the possibility of writing a custom UX for that, as was possible since Android 1.0. It's particularly ironic during Google's campaign pushing Apple to support RCS.
> The average mobile connection sent 51 messages per month in 2020, 17 fewer than in 2019.
I'd love to know the median, I assume there's a number of power users that drives up the average. Or bots that are sending out thousands of messages a day.
My wife's boss communicates with all her employees by SMS (mass SMS - works like group communication, both ways).
AddEdit: Airlines send their notifications and links to boarding passes etc. via SMS. Dentist and doctor appointments, other public office appointments (e.g. my upcoming passport renewal), document notifications (from pension fund insurance companies for example), public warnings ("Toxic fire nearby - close your windows"), and more, are via SMS where I live.
Or (bluntly): why not Matrix (and/or XMPP)? What makes WA so much better that you're willing to go all-in with this company?
Wow, it's really dead.
However most of my instant messaging is done on a computer using discord, so I might not be in the prime user base of these apps
> I don't get it how having a dozen of messaging apps and remembering who uses what is better than a simple SMS
Indeed, especially now that Telegram is taking off by times in Europe (in Belgium / Spain / France at least Telegram is getting used by a lot of people) and that some people now refuse to use WhatsApp.
In addition to appointment reminders from doctor/dentist/notary/whatever and delivery tracking numbers I still exchange SMS with quite some people.
It's not as if it was exactly hard to open and reply to a SMS you just received from someone: takes exactly the same time as answering using WhatsApp or Telegram.
> If you're calling MTProto 2.0 'insecure' then you should know [...]
If you're calling "secret chats" the default, then you should ask around or try to use telegram on desktop or just open telegram and see how much stuff is actually encrypted.
Jokes aside, I see SMS as a useless protocol; because it cannot be used for identification, and neither can anything be encrypted nor verified without another communication channel.
It's also not in the power of the end user to decide whether or not their number gets reassigned, blocked, or does work at all. Most US people seem to think that it's normal to have "one" number for years on end. For the rest of the world, it's not true.
For example: If I don't use my SIM card to make phone calls (which get billed) for 6 months, it's gone and reallocated to a different person.
SMS is used by companies to send notifications and asking for confirmation, even (ouch!) banks. I haven't sent one in more than a decade.
Which is not to say that I can't see the rationale, but I didn't know and it surprised me.
The next thing I wondered is how much work it is to maintain two entirely separate messaging systems in one app. How much did this drag them down over the years? I agree with you and hope that it can finally get Signal moving forward, there are so many missing features compared to Wire and, especially, Telegram that it is currently a pretty tough sell to move anyone away from Telegram.
My UK-based employer seems nonchalant about expecting me to agree to be subject to the laws and courts of California in order to receive internal company newsletters delivered via a 3rd party.
While I agree this is harmful to the user (or unpatriotic, if you prefer), it's extremely common thanks to the state of the global economy since the 1980s.
I'd wager that 99% of people in the UK would now be unable to contact their friends and family without relying on at least 1 large U.S. company.
Interesting, here in Germany, almost everyone I know has Signal and WhatsApp with some people using only one of them. Telegram I encountered from one US American living here, and from people into conspiracy theories.
Uh, I’m from Germany and had the same mobile number for over 20 years.
If it doesn't support SMS I have zero reasons to use it I may as well use anything else that had less friction.
I liked your application enough to pay a reoccurring subscription, removing this functionality means I've cancelled my subscription.
SMS is big in Europe (yes, Europe is not a country. I just mean "dozens of countries in Europe"). All courriers have plans with SMS focus.
Currently, it looks like they are focusing on social networking features: stories, emoji stuff, better link previews. Basically everything that competition already did. The roadmap is not public, so I wouldn't take guesses as what may come next. But...
..."dropping support of X as a feature" is some kind of new transcendent approach to product development incomprehensible to common earthlings.
I'm not taking anyone with me, they already have Signal and WA and obviously SMS. I'm uninstalling Signal and not recommending it again. I can already message Signal contacts using WA or SMS. I don't need Signal for that, and I'm not keeping 3rd messaging app.
It's a ̶̶u̶s̶e̶ pay it or lose it thing. AFAIK, typically applies to prepaid/pay-as-you-go SIMs, not on contracts.
For my case, it is that I have to make a 12EUR top-up every 3 months. The top-up credit will expire if I don't make another top-up on time. After a few months on zero credit, you get you incoming calls blocked. And after a couple more months, your SIM is de-registered.
Transferring your number is always possible, yes. As long as you're still the registered owner of the SIM number.
"It doesn't matter for the use cases I don't care about" - what a selfish look at the world.
Besides paying for parking by SMS and other services in Europe there's also M-Pesa and similar services[1]
It happens in iMessage and RCS threads as well.
I don't think the reasoning for their implementation is as well thought out as you claim here.
RCS - and VoLTE for example - is in this way no different. It's just a data message that gets sent to a configured endpoint (either on Google or your carriers' servers, depending on your carrier preference).
Of course, the carriers still remain the dumbass link in this chain as usual. I've just had to deal with a carrier that charges VoLTE call configuration setup connection as normal foreign data roaming with minimum pricing. The phone ate through 30EUR of costs despite having mobile data disabled.
"Reduced" is a very kind characterization, even if it's literally true - in my experience (with Vodafone and Three) this means ~64kbps which makes even messaging apps functionally unusable.
Since I started traveling more, I use my eSim slot for a ~7€ data-only plan from one of the discount MVNOs just to avoid being caught in this situation.
As for Telegram, people mostly use it to consume news. It basically replaced RSS readers for common people. Although its install base is relatively high, I have yet to receive a single private message over Telegram.
Telegram has had emoji reactions since 2021 -- https://economictimes.indiatimes.com/magazines/panache/now-y...
Fine by me. If they want a better experience, I have almost every messaging app on the Dutch market bridged to my Matrix server so I don't care, they'll just have to live with the lack of features if they want to chat with me (or install something like Signal).
I don't care about RCS and other ISP standards that exist to squeeze more money out of texting. I'll use Telegram/Signal/WhatsApp calling before I'll use my phone app because my subscription doesn't include free minutes (and I barely call anyway) so I've gotten the benefits of tech like WiFi calling through VoLTE for years before ISPs bothered sending their VoLTE profiles to my phone.
Typically, if I try to reach someone but they don't answer I'll follow up with an SMS to explain what I tried to call for. If it's a co-worker, or an acquaintance, I'll send an SMS to make sure they receive my message. SMS is more reliable, doesn't require you to know which apps the other party runs, and it comes through to the recipient as SMS is a basic service in the telecom networks.
When I get to know a new person there's a transition to "oh, you're on $APP too", and I might starting moving non-urgent messages to $APP. But if you don't know the other party well or there's a question of reliability, what is an alternative backbone for messaging if not SMS?
Even two years ago, I had to actually change a phone number because I couldn't transfer my number from one provider to their own reseller. O2 Scheißladen.
MTProto is the name of the:
1. Cloud Encryption
2. E2E encryption
algorithm at Telegram. MTProto 2.0 is not just secret chats, a different implementation is used for cloud: https://core.telegram.org/mtproto/AJiEAwIYFoAsBGJBjZwYoQIwFM...
Both cloud and e2ee consist of what's called the MTProto 2.0 algorithm.
Huh, I guess I got lucky staying with Viag Interkom and then O2 (which was an automatic switch when O2 bought them) for so long, I only really switched providers in 2020 which was long after the EU regulation was in effect.
SMS is:
1. not controlled by a single company
2. a different network than the internet
3. a fail safe for people who don't use apps or are unable to at a given time for some reason (inc 2fa)
4. a fail safe for a "small group of people" who are suffering the consequences of a natural disaster.
Though perhaps not economically feasible for certain companies, supporting redunancy is as much an honorable goal as privacy.
Here in the UK, whatsapp is the default first try, at least with anyone I've interacted with. The dog groomer even messaged me on whatsapp to tell me dog is ready, completely unprompted.
In whatsapp, messages have always been on-device / in-memory, where they belong, doing a p2p sync/transfer
> 1. Is There a Secret Chat On Telegram Desktop?
> No. Due to Telegram secret chat's end-to-end encryption and the requirement for permanent storage on the device (and not using the Cloud to store data), Telegram does not have the secret chat feature on Desktop or Web Telegram. They may add this feature on their desktop version in the future, but for now, it is not safe enough to have it.
Whatsapp has caught internationally but it's Facebook and its desktop app is a crashing dumpster fire.
Viber is another popular app, but has too many ads and visual noise.
Telegram has caught on as a good alternative for all, because it does everything good. Apps are functional, fast and stable. Interface is clean. Also Telegram channels were genius idea to increasing market penetration. Nowadays all social networks are heavily abused by bot abusing abuse feature (hehe). Basically any post containing "politics" let alone "war" content can be taken down by abuse spam. Be it facebook, twitter or reddit, all the same. So political and social "influencers" are rapidly creating backup or new main channels in the Telegram to post "controversial" information, and people reading news and blogs in Telegram will also message there too.
Apparently Germany has ~8bn SMS for 160m contracts (don't ask me why there's an average of two contracts per person), which is like 50 a year. Edit: that number seems to include automated messages.
Not in Hungary, you still have to pay per message here unless you choose the most expensive plan.
Some chat apps bootstrapped by riding emerging markets with exorbitant SMS costs. Others bootstrapped by cross-promotion from existing popular communication networks.
Signal's strategy was to replace your current SMS app as-is and then incentivize you over to their network one contact at a time. It didn't work, for the stated reasons, which honestly sound like failures of product design and engineering more than anything else. Apple's Message app has the same strategy and it's working.
We haven't actually seen what Signal's next strategy will be, which is maybe a timing mistake as announcing a viable alternative could have taken some heat off the SMS retirement announcement. Or maybe there isn't a next strategy?
'The super secure chat network' is not a bootstrap strategy, it's a build it and they won't come strategy; so I think it's fair to remain bearish on Signal's future even after reading a rational take on why writing SMS apps are difficult.
It became common to get atleast 1000 sms in your plan, combined with maybe 100mb of data back them, then unlimited sms with 1gb of data, and then slowly data went upwards while sms can't go up from "unlimited". 100MB is not enough to leave "the internet" running 24/7 on your phone, so internet-based chat services were unusable for general reachability back then (we're talking about early symbian and stuff like msn messenger era), and you just sent an SMS (becase 1000 is enough for everyone... except teenage girls back then(.
Users can switch to another OS if they really cared that Apple refuses to use industry standards, hamstrings their own mobile browser to bolster app sales, and violates antitrust laws with their ban on third party browser rendering engines.
Maybe coming preinstalled on all i-devices helps?
------------------
This will relegate Signal to niche use.
I don't even know if I will keep it and I've been a booster for years. I'll have to remember which ~5 of my contacts use Signal, and mentally update that list when/if any join?
Can't take on that cognitive load for everyday comms.
-------------------
"actively" is a big word, there are several 3rd party clients and no big push to make them stop. They don't want widespread 3rd party clients though.
2. That's not a requirement even though Google does their best to kill all alternative implementations. (Samsung had one, but Google paid them a lot of money to get rid of it. I have to admit I don't know who is left with their own RCS client)
> So, RCS is not a replacement for SMS.
It's 100% a replacement to SMS, it's literally defined in the same spec as SMS over LTE, as an upgrade to SMS. Though yeah Google made it, this... uh... thing. RCS is technically a federated standard, but Google killed every part of the federation and some parts of the standard.
Apple phones are simply good to use and work well. Me getting a generic phone and installing Arch Linux in it or something will have little effect on its own. But it will make my life very inconvenient!
Also see [1], they have every bridge's features well documented.
[0] https://news.ycombinator.com/item?id=33181636
[1] https://news.ycombinator.com/item?id=33179047
[2] https://community.signalusers.org/t/signal-blog-removing-sms...
Of course, it's not really "anonymous" if a nation-state wants to come after you, but that's not the threat model for most people.
https://twitter.com/matrixdotorg/status/841424770025545730/p...
This is precisely why RCS and MMS support aren't important. I just need Signal to deliver SMS authentication codes and notifications. For person-to-person communications I'll use Signal protocol.
So, I replaced Messages with another SMS app that doesn't spy on me and disabled RCS. If I wanted to do rich messaging through data connection, I would use Signal or similar."
3rd party SMS apps use a Google provided API to send and receive SMS messages on an operating system provided by Google. I'm not sure how you think using a 3rd party SMS app would be protecting you if Google cared about spying on your messages.
RCS through Google Messages is end to end encrypted while SMS is not, so you haven't eliminated Google in theory spying on your messages but you have assured that your carrier, the recipients carrier and whomever controls the OS and app on the destination device can potentially spy on your messages.
I receive dozens of SMSes from banks with one time passwords for 2FA and payments notifications, from delivery companies to notify me about progresses in my orders plus some spam. It's easier for them to use SMS than anything else because every phone receives SMS right out of the box.
Group of people using it to communicate might be small, but group of people using SMS to get verification codes, messages from delivery company about scheduled delivery, doctor appointment reminder etc. will be for sure 90% or more phone users at least in Europe
So I find it very stupid to say nobody use SMS, because pretty much EVERYONE use them, just not to communicate with other, but to receive messages and you still need some app to receive these messages. So while I can communicate with almost everyone through Whatsapp I still need to use SMS app almost every day and I'd rather have all my communication consolidated in one app than must switch between apps to check the codes or other stuff, man I wish Whatsapp supported at least receiving SMS, actually I'm pretty sure that would be enough for majority of people which don't really need to compose message.
Bridging chats of different technologies doesn't work well/at all (i.e. Signal bridge + WhatsApp bridge users in a single room) but bridging external chats (DM or group) into Matrix works very well. Some services need a daemon running on a phone (i.e. WhatsApp) and that's very annoying, but where possible these bridges all run in the cloud.
If you trust third parties, you can also go the easy route by getting a subscription from EMS (https://element.io/matrix-services/ems-pricing) or Beeper (https://www.beeper.com/). I personally prefer to keep my messages and encryption keys on devices I control, but others prefer to let someone else take care of it all and I respect that.
It's relatively straight-forward to set up a bridging server if you're comfortable with Docker and YAML files. You can read how to set up a Matrix server here: https://matrix.org/docs/guides/free-small-matrix-server and here: https://github.com/spantaleev/matrix-docker-ansible-deploy/b...
If you use the Ansible playbook, all you should really need to do is run through the setup, fire up a Matrix client, start chats with bot accounts, and follow the instructions on the guide (usually sending /login to a bot and authenticating your account with whatever service you're bridging).
Your Matrix account doesn't have to be on the same server as your bridges, which is a setup some seem to prefer. You can set up a Matrix server just for bridging so that you don't need to set up all the VoIP features and performance tricks while keeping your own server dedicated to just bridging stuff. This does break some nice features (i.e. double puppeting, a bridge feature) but it also makes your own server less of a single point of failure if you ever do get talking on Matrix.
Users cannot switch to another OS because if they do they won't be able to communicate with social circle. This is also completely artificial because the networks have otherwise pretty much the same feature sets and are only distinguished by their accreted userbase.
Concurrent cannot compete because they can't gain enough users to get a critical mass
I wager that no single entity should have so many captive users.
Regulation is clearly in order
But the nature of these things is X gains traction. Y wants a piece of X's pie so cuts X off. X realises that it's dependent on the Ys and so launches its own service.
We've seen the this with Netflix. We've seen the former with Twitter.
Ideally the stars would align so that it's in everyone's interests to support an open protocol and we kind of have those in SMS and email. Except these have their own issues.
I am not HN community, but there is Beeper or Texts.com possibly others, there are also (other) Matrix bridges, but it's PITA to set them on your own.
Personally I'd just move family and (old) relatives to Whatsapp and you will have everything consolidated in one app used by everyone anyway. That one relative would have easy choice, either Whatsapp or SMS if they wanna talk to me, not keeping extra app for one special snowflake.
Plus you need to keep SMS app to receive all those codes, shipping updates since Whatsapp sadly doesn't support SMS.
That's also my setup - Whatsapp+SMS, used to have also Signal years ago with family before we ditched it en masse after PINgate for Whatsapp, my mother has also Facebook, I think father only Whatsapp, sister I don't talk to has also Whatsapp, wife has Whatsapp and (Google) Messages which she use just for receiving SMS.
If I would be moving somewhere my family (parents, wife, kids) I'd go for Element (Matrix) - decentralized network, various apps to choose from, no phone required.
Another alternative but without (video) calls would be using some email app like Delta Chat or Mailtime for instant messaging, that would require no signing up for new service, I like the idea, though I guess messages would be quite slow.
Btw. Messenger and Skype (Lite) supports SMS, so since you use Messenger anyway you could ditch SMS app and Signal after they remove SMS, if you wanna keep more IM apps than having everyone on Whatsapp.
Read before replying. I literally said "for person-to-person communication".
100% agree here, we've collectively picked convenience and shiny objects over everything else so often that we're left with a handful of companies with way too much power and reach
> Regulation is clearly in order
I disagree, or at least hope,this isn't our best or only option left. If it is the only thing that would work though, at least its something
I don't have your phone number. I was talking about my 4 Signal contacts, they are on WA too.
I love that "there are two big players in the market with a literally 50%/50% split" means that iOS dominates. iPhones have always been popular among teenagers, if Millennials are any indication it levels out over time. It's just one of many weird demographic splits, not some grand trend. People need to not fall into the fallacy that "something is popular with young people" means that thing will remain popular as they get older.
iOS is more popular among women, young people, liberals, professionals, upper-middle and high-income, people with post-secondary education, and urbanites.
As well as the inverse of the above Android is more popular with people who work in IT, and people who follow tech news.
There's a lot of people here claiming that their personal use is representative of their country, or of Europe as a whole. I get SMS from a lot of people. You don't, probably because a lot of the people you know are on Facebook/Whatsapp and it's more convenient for them to stay with that platform. That doesn't mean that they don't use SMS for anyone else. It just means that you are bubbled.
Apple's iMessage/SMS handling is a perfect model for how Signal should have approached SMS support: a firm hand of features and UI reinforcing the fact that SMS sucks, but reasonable enough support that you can send and receive SMS/MMS just fine.
This is the signal president's stated reasoning on why they couldn't have a blue-message (secure) / green-message (insecure) dynamic, akin to iMessage. I think she is being disingenuous, and that the development effort of MMS is the driving motivation, but I am cynical.
- we don't have the dev resources to keep SMS/MMS working
- with RCS becoming more common and no Google API for RCS, we don't have the ability to fully replace Google Messages any more
- since we can't support SMS/MMS on iOS, we decided to kill the feature for Android users for the sake of maintenance
They shouldn't pull punches and pretend that this feature removal is for the "protection" of users who accidentally send SMS instead of Signal messages. That's a strawman, anyway -- if a user manages to send SMS in the Signal app, it's because the person they're trying to communicate with doesn't have Signal installed... so there inherently isn't a secure communication path. Users who pay per SMS should disable it in the app settings, and you can easily add a popup the first time you send an SMS/the first time you open the app to make that clear.
This is 50% Signal trying to streamline development, 50% Google's push for RCS (and their lack of APIs to build alternative RCS apps on Android). They should be honest about that instead of making up nonsense about misguided users getting confused.
I think you know this based on the rest of your post but there is no “another SMS app that doesn’t spy on me”.
> The death of SMS is hardly specific to central Europe and the US.
I hate to be a grammar Nazi, but since you're specifically attacking my wording I have to correct you. This statement would only be correct if it was: "The death of SMS for person-to-person communication is hardly specific to central Europe and the US."
So yes, your statement is incorrect. It's not dead, far from. It might get there eventually, but definitely not yet.
maybe now that signal is switching off SMS it can implement user handles that people can share instead of their number. once they do i'll give it a try
Yes, but data tariffs were also expensive, while you can send SMS with regular (no-data) tariff.
Personally, I use signal as a replacement for Hangouts.
on this paper: https://nebuchadnezzar-megolm.github.io/
I don't agree with your take or arguments, and you seem to keep branching off pejorative comments on their organization and product instead of actually discussing the points. I think the conversation would be more productive if we focus on the same point, i.e.:
* Focusing on what matters most is a good idea, as nobody serious about secure messaging uses SMS
* Your argument: irony-covered "dropping features is not a good product development approach".
* My counter-argument: it **is** a valid approach, why support a feature that was useful in the past, but it is now dying/not aligned with your core value proposition?
* Your other argument: their focus is on social networking, and some disdainful comment that "the competition already did it".
* My argument: how is catching up with well-established user behaviors across other messaging platforms a bad thing?
(edited for formatting)
now give me Matrix client with basic SMS support (I don't need even MMS) and I'm installing it immediately to replace my dedicated SMS app
I did same love with family as poster, from Signal with family, WhatsApp and SMS just to WhatsApp/SMS combo though already years ago after PIN nagging
Did you mean to reply to a different comment? Mine was a reply to the GP, to answer the very narrow questions he asked about the US.
Which is insane. An SMS replacement where Verizon and AT&T users cannot message each other is insane.
Worth reading the response from Matrix as well (https://matrix.org/blog/category/security).
My first reactions are to wonder how many of these issues are associated with federated (as opposed to fundamentally decentralized) group chat in general. Matrix seems to be taking the position that some of these issues ultimately relate to trust vs lack thereof in the homeserver as a bottleneck.
I also wondered if there was a good security model for federated or decentralized group chat at all at the moment. I can't remember offhand if Briar was adding groups or not, but that's not federated.
Is this the comment you would give when someone says they intend to take a notebook with them when travelling in addition to their phone?
The only practical issue raised by https://nebuchadnezzar-megolm.github.io/ which we didn’t already fix is the question over whether servers or clients should control group membership. Our position is that it’s okay for the server to control it as long as clients are warned if malicious users/devices are added. Fixing it properly is Hard: for instance, if you are chatting in a room and it turns out that a remote user kicked another remote user, but the kick was delayed in reaching you, you could keep chatting away encrypting messages for a user who is no longer in the room and theoretically should not be receiving them. Is this a security flaw? Or is this just how causality works? So we’re dealing with problems similar to that; hopefully we will be able to switch to client controlled membership by end of year.
tptacek’s derision is not very constructive.
No one is branching off, but a pretext that your personal take on things must comply with some sort of argumentation protocol that is the only valid blueprint for discussion isn't convincing. Moreover you've managed to somehow unwrap you single comment into a fully fledged dialog while ignoring that their roadmap (the big picture) is not exposed to the public. Given we can only judge isolated decisions, they seem what they are — rather not aligned with the expectations of the userbase.
Personally, I see a pattern of Signal making news in rather negative connotation rather than positive lately.
When it got traction, I felt like it's a new day and the future is bright. But since then, they went with a series of rather ambiguous decisions that sidetracked from previous claims.
EOS for SMS is again one of controversial decisions, I mean, we're in a thread started by a person that went above to clarify reasoning behind the press release. And before it was a year of server side repos without any commits, and then the public got a feature no one asked for — MobileCoin integration. And echoes of intent about it are still heard across the table.
https://matrix.org/blog/2022/09/28/upgrade-now-to-address-en...
Matrix can't even load 100 old messages properly with E2EE enabled in a room. Signal can't even handle scale when it comes to chat groups and communities. There's no anonymity in both either as Signal doesn't even allow you to hide your phone number and Matrix leaks your metadata to all involved participants like crazy.
Telegram doesn't use E2EE but the privacy and security are in no way compromised.
The whole fuss about "They can read your messages" holds a very negative assumption in the first place about them reading it and then also assumes everybody's threat model involves inferior UX of managing chat backups like WhatsApp just to keep messages away from cloud.
Just take a look at your threat model and decide what you want, not everybody wants an E2E encrypted chat app because we know the compromises that we have to make with E2EE and I'd rather have my chats on cloud encryption than my local device, considering how many features Telegram allows me to have with cloud sync.
In fact, all the ios users I communicate with do it over whatsapp or telegram.
Today I had an ios user sms me asking if he could send me a voice message. I honestly have no idea if that would have worked but I directed him to whatsapp (that he already had installed) and it was frictionless thereafter. I expect I'll never see another sms from him again.
So if I'm not traveling I stay with the cheapest data option for my occasionally otg stuff and on heavy travel months I choose larger packages, because I found myself using more often, for example as access point to notebook.
I also have an Kaufland (Telekom) Prepaid card, that said it would provide basic, very slow internet, for free so that text messages over chat works, bur I don't know if I got the wrong APN settings, it has problems in my second slot or it only works if you top it up regularly, but that internet and rest never really worked, even the account management over the website doesn't really work for me.
Even cheapest monthly subscriptions have at least 500 messages per month bundled, with more typical monthly plans (~$15/mo) having unlimited calls, SMS/MMS messages and around 30 GB of 4G/5G data.
Oh, no, please! What I want is the other way around: turning Signal into Telegram, i.e. keep bolting features onto Signal until it has feature parity with Telegram, or even what Telegram did five years ago. That would be a dream.
I disagree that e2ee can fundamentally not deliver Telegram's experience, at least not far off. It may need more local processing and indexing (storage), but generally it's all possible. It's just a ton of work that Telegram has sunk many millions into and will cost even more to do securely.
What you are absolutely wrong about, however, is claiming that it's all the same.
> Telegram doesn't use E2EE but the privacy and security are in no way compromised.
There are various scenarios in which your data on Signal is safe in ways that it is not on Telegram, and more actors can see your data on Telegram than on Signal. Thus, both security and privacy are impacted. That much is plain as day. Whether that is worth the trade-off, is up to you.
It's fine to have opinions and a conversation about whether the whole e2ee concept is silly, but please don't give your friends and family false senses of what the practical impact is for privacy and security when choosing these trade-offs by saying it's all just as safe and identical.
That seems reasonable: Company-issued phones, LTE-Routers, some undercounting of M2M, and gerneral churn (I changed provider so I had 2 SIM-cards this year).
This is a big call. I live in a country where SMS is still standard. Most communication with friends and family is done by it.
Can you please provide a source?