Default disappearing messages
signal.org
signal.org
Fine with whatever the default is, but I feel like the subtext here is that Signal thinks people's phones are going to be increasingly compromised so this will be a more necessary, and the "remember the passing acoustic conversation" is just a way to put a less depressing spin on that.
For some people (e.g. journalists) this is exactly what they need, because forgetting to set disappearing before sending a message could be problematic, and I'm glad they added this feature for that use case.
It is also still missing a key point we've been asking for for awhile, message archival. That's the big thing keeping me from using disappearing messages, because I don't want to take a screenshot anytime anything important is said. That's a terrible way to search for messages. With an arbitrary delete function (only limit I see is Signal's servers don't accept delete commands after 24hrs) you can have consent from the other party about archiving messages while disappearing messages is on. This provides a tremendous amount of utility.
I used to defend Signal a lot here. But with a lot of decisions they've made in the last year they seem to be just another company that "knows better" than the user. We don't need another tech company that's disconnected from users and makes changes just to make them[0]. (Beyond that, Signal isn't large enough to make people stick around when dumb decisions are made) I think this is a fine stance to take when it comes to the security portion --there are fairly objective "best practices" here-- but not for UI/UX and privacy functions. But with the latter there's a lot of subjective preferences and needs by different people and groups. Signal, you do not need to be and look like WhatsApp. Just be Signal. The progress has been very slow, new features come with footguns, and there is a lack of communication. They only seem to take communication from their community forums which has an extreme echo chamber and is the textbook example of "we know better than you" mentality. It's very frustrating and it's caused me to stop recommending Signal to people (my friends did the same for similar reasons). I just will not understand why when given the chance to continue to grow Signal decided to shoot themselves in the foot.
Boy does this hit close to home, whenever I think of Firefox and its accelerating decline.
Yeah, that's my read too.
The article's title is confusing and should be changed. I was worried they were going to change the defaults on me, and I was going to have to rush in and reset things.
But that is exactly what the default should be and what's needed to thwart large-scale blanket surveillance of the whole population.
If the mere fact that ephemeral messages are enabled raises a flag, that kind of defeats (part of) the purpose.
Yes, the messages are only on your phone; Signal doesn't have them.
While I haven't had my devices seized, two acquaintances had their device seized for an investigation. They were just on the wrong place at the wrong time.
This meant that months and months of our chats were potentially read by a third party.
Phones are compromised by default. E2EE doesn't matter if someone else owns both ends.
The issue is a matter of degree. Everything and everyone are compromised to some degree; perfection is not the standard.
They will, just now it's the supposed "good actors" that are doing it. See Apple's recent photo scanning announcement. I have no doubt governments will start "encouraging" even more phone monitoring software.
Police and authorities can have logs (either of everyone or high value targets) and those people won't have access to their own messages to defend themselves.
It even means chat conversations can be entirely fabricated by one side, and the other has no way to refute it.
Someone gets hit by a hack like Pegasas and it is only able to scrape a small history of messages.
At the end of the day this is going to be a per user decision. Security and privacy are different things. What's the correct privacy options for you is going to be different from another person, while we are likely to share the same optimal security strategy.
> Police and authorities can have logs (either of everyone or high value targets) and those people won't have access to their own messages to defend themselves.
> It even means chat conversations can be entirely fabricated by one side, and the other has no way to refute it.
I don't think that's a realistic threat, and if it was, the defense you propose wouldn't work.
Given that Signal is E2E encrypted, I don't see the mechanism where the police and authorities could even have logs without compromising the endpoints. Furthermore, even if they did fabricate a message, it's not like your phone is some inviolable log of what you actually sent. Even with Signal today, you can delete messages on an ah-hoc basis, so their rebuttal (that you sent it but deleted it from your phone) would be exactly the same.
In some systems, e.g. PGP, messages are digitally signed. This means when I send you a PGP mail, I'm actually providing proof - which anybody can verify - that I really sent that email.
If what you wanted the message platform for was to agree legal contracts maybe that feels like a reasonable structure. Alice sends Bob a PGP mail "I agree to X" and Bob now has a signed message proving this agreement. Bob couldn't have faked that, only Alice could have signed it.
Signal deliberately doesn't do that. Signal messages have cryptographic integrity protection (so Charlie can't send Bob a message pretending to be from Alice) but not signatures, so Bob can fake messages from Alice to Bob because Bob is a participant and so has the keys.
As a result Bob knows if the messages are fake or real, but can't prove that to anybody else. That feels much more like what you want from a platform you use to chat with people, not for making legal contracts.
PGP does the identity verification out of band. So you can simply not sign your message and completely eliminate the problem in the first place.
So it seems odd to suggest that PGP is inferior to what Signal does.
>This means when I send you a PGP mail, I'm actually providing proof - which anybody can verify - that I really sent that email.
PGP prevents access to signature information with the encryption. So you would need access to the private key of the person I sent the message to before you can verify my signature.
What is the word "effectively" doing in your sentence? I suggest that the word you needed is in fact just not. The messages can be authenticated but they are not signed.
> Signal releases data to allow your correspondent to forge that signature after they have received your message.
Nope. Bob always has the ability to produce forged messages apparently from Alice (or Charlie) to Bob.
What may have confused you is that Signal can eventually send Bob's expired authentication keys in cleartext. At this point any eavesdropper could make forged messages from somebody to Bob, that seem authentic to anybody except Bob (and presumably whoever didn't actually send them) but are rather old news.
> PGP does the identity verification out of band. So you can simply not sign your message and completely eliminate the problem in the first place.
In this case your message can't be authenticated, a very different scenario. Now Bob can't tell if this PGP email is really even from Alice because it lacks a signature. Perhaps useful for an anonymous tips line or something, but rarely what we want for messaging.
> PGP prevents access to signature information with the encryption. So you would need access to the private key of the person I sent the message to before you can verify my signature.
Nope. Just need the signed message. If the message was also encrypted then its recipient can decrypt it and show the decrypted but still signed message to other people as proof of what was sent, without revealing their private key.
This is very often undesirable, and yet it's the only option provided in PGP.
>What may have confused you is that Signal can eventually send Bob's expired authentication keys in cleartext.
Is this a new feature? Originally the decision was made to only allow correspondents to forge messages, not everyone as in the case of Off The Record.
If you signed a PGP message and did not encrypt it then you specifically wanted it to be possible for everyone to verify the signature. A trusted correspondent would not decrypt and release your signed message so the only time that would happen would be if your correspondent was compromised. In the Signal case you would have to claim that your trusted correspondent forged a message from you. That wouldn't really work.
In the case of an untrusted correspondent you would not want to sign your message at all or would sign it with a non-public identity. PGP allows you do do either.
In what sense do you believe X3DH is "signed" ? Is this going to be more handwaving where you treat authenticated as "signed" just because that's how you'd do it in PGP ?
The recipient of a first X3DH message does know for sure it is from the sender, because they're the only one that could have performed the steps needed without the recipient's private keys.
But if (like the recipient) you know their private keys, you can just forge messages that are entirely plausible.
Consider an initial message purportedly from Alice to Bob but let's suppose we're Bob and we're forging it. How do we go about this?
We need to do three DH calculations. First one, is Alice's long term Identity crossed with Bob's prekey. Alice would do this by knowing her Identity private key, but we know Bob's private prekey, so we use that with Alice's public long term Identity.
Second one, Alice's ephemeral crossed with Bob's long term Identity. Alice would do that using her ephemeral private key, but we know Bob's private Identity key, so we use that with any key X.
Third one, Alice's ephemeral crossed with Bob's prekey. Alice would use her ephemeral private key again, but we know Bob's private prekey, so we use that with key X again.
We now have a secret key we can use to encrypt our forged message "from Alice" and we write key X into the headers where Alice would have written her public ephemeral. Our message is indistinguishable from genuine. Only Bob (who forged it) and Alice (who knows she didn't send it) know this is a forgery and neither of them can prove it.
> In the Signal case you would have to claim that your trusted correspondent forged a message from you. That wouldn't really work.
If the Secret Police are content that you are a Blasphemer or you plotted against the Emperor or whatever without proof, nothing would "really work", regardless of the technology used or not used. That's not interesting here because we can't do anything about it. If the Star Chamber will have you executed based on hearsay, that's just as true for literal whispers as for a Signal message.
In contrast PGP provides them with unshakeable proof even though it didn't need to.
* https://signal.org/docs/specifications/x3dh/
... I understand that the prekeys are signed. I will confess that I have no deep understanding of how everything works together. Signal protocol is somewhat baroque.
A loose equivalent in the OpenPGP world would be a user's temporary encryption keys, which may change sometimes while your signing key stays unchanged. I don't know how common this practice is today, I know it was common years ago. Clearly knowing Alice's most recent temporary encryption public key, signed by Alice's signing key, would not prove a message you showed me was from Alice.
> Signal protocol is somewhat baroque.
I agree it seems more complicated in principle than OpenPGP, although I think it's hard to make that argument for the practical deployment of OpenPGP because of the complexity of delivering compatibility.
But Signal's complexity has a specific purpose. For example those three DH pairs are combined together with a KDF so that 1. Bob learns whether the message is really from Alice (assuming Bob is not the protagonist of the movie Memento and so remember if he forged the message); 2. Alice knows the message can only be received by Bob (to the extent she verified Bob's identity, which is an out-of-band problem), and 3. A fresh random Forward Secret key is used for the conversation. Three desired properties, run DH three times.
It is definitely possible that cryptographers will come along and say "Oh, we can do this in a single function" and write like fifty pages of difficult maths and one day we get a replacement for Signal with "simpler" (although like Elliptic Curve maybe not simpler to explain) cryptography than this. Definitely possible, but these moving parts all have a purpose AFAIU today.
The default default (i.e. for new users) will still be set to keep messages.
I've also found it somewhat therapudic to let go of the past in other places as well. This included throwing away tshirts I got at a successful startup 10 years ago and even cleaning out my "digital closet" by delete my 10 years of location history on Google and old emails.
Playing legacy board games, which require destroyibg pieces of the game as it progresses so you can not back out or do it again a different way, introduced me to this feeling and I enjoy it.
Funnily enough, around the same time I made this mental change I also started keeping a journal.
I write it down in the same place. I have a pinned conversation with myself that I drop notes (typed/audio). Also useful for recording audio messages which you can play back to yourself before forwarding to a Signal contact (I hate that there’s no preview before sending audio).
I know HN has a higher proportion of the reasonably-paranoid EFF-loving population, and I appreciate their push for options like Signal, but do you folks really prefer acknowledging that a secret number has changed every time someone screws up new device setup? I don't know how many times I've been saved by searchable chat in gMail or got a key piece of info from my Location History, or answered someone's question by scrolling back through https://myactivity.google.com/ - am I the exception?
And, hypothetically, if we were scheming terrorists or dissidents in a totalitarian state, who here would trust their conversations to _any_ mainstream communication tool, no matter its privacy reputation?
Messages in my phone are now the only remaining words from someone I once knew. I now and then scroll through them like they were an old shoe-box of letters.
I don't know (and will never be able to ask) what emotional connection they had of their own messages. I only know that I am glad to still have those...
So yeah, I agree... having things disappear after a reasonable amount of time is a great feature. Sadly, most networks (eg. twitter, facebook,...) don't have this.
That's why I use a utility called Shreddit and some custom scripts to delete all my reddit/Twitter/whatever posts except the latest N.
I had to search for it, and assume that you are referring to this utility: http://apps.palmtronix.com/shreddit/android/
It makes claims of zeroing out the local storage, but does not seem to do anything with regard to erasing your social media footprint. I would like your recommendation on scripts which comprehensively delete posts/history on Reddit, Twitter et al. and make them unrecoverable.
edit: I just noticed a post by @randomuser87178 which probably answers my query
There are several third-party search engines which use the Pushshift API underneath, most prominently Camas Reddit Search [2]. Ironically, it is way faster and more thorough than Reddit's actual search feature. You can do a full-text search of a specific user's history and filter by subreddit, score, and post date in a couple of seconds.
Since Pushshift contains comments and posts in the state they were made when they were originally posted, editing or deleting them from Reddit proper, or deleting your Reddit account, does nothing to prevent them from being searchable in Pushshift.
Pushshift does support opt-outs, but they have historically been slow to process them. Apparently a new opt-out system will be in place as of this weekend [3].
That said, Pushshift isn't the only Reddit archiving game in town, just the most visible, so even if you opt-out of Pushshift your post history is still probably in somebody else's archive. Also, I'm pretty sure the opt-out just removes your posts from the public API, I think the historical dumps still have your data in them if somebody were motivated enough to download all 80+ TB of them.
Your best bet for preserving privacy on Reddit is to (1) never post identifying details, (2) change accounts periodically, and (3) keep a few alts for commenting on different subreddits, especially if posting on subreddits that somebody can infer identifying details from (eg: the subreddit for your city).
[1] https://pushshift.io/ [2] https://camas.github.io/reddit-search/ [3] https://www.reddit.com/r/pushshift/comments/p2fm0n/deletion_...
I don't know why so few people consider this point. People change, social norms change, so it's very easy for an older conversation to be used in such a way as to make someone look horrible, even if they were not, and are not, a horrible person.
I'd like something like 3 months / 13 weeks, but 3, 6 and 12 months seem like good durations.
[1] https://community.signalusers.org/t/longer-timeframe-more-op...
[2] https://github.com/signalapp/Signal-Android/pull/6814#issuec...
That's sad to know. But we need to face the fact that Signal is not a community driven project. It has always been a moxie driven project. So it's not surprising that he would do (or not do) things that the rest of us don't understand or agree with.
Short enough to hopefully nudge people into immediately writing the whole thing off as a terrible idea and then choosing not to enable it.
Apparently dumb shit like this is what Moxie had in mind when he wrote the infamous blog post about why disallowing personal (patched) builds and third party clients gives them the advantage of being able to control the user experience in a way that they wouldn't be able to if people were using an unsanctioned client.
> We’ve also added the ability to set custom timer durations on your conversations, so that some content can be gone in 60 seconds and others can exist for 18 minutes or 4 weeks.
> As far as longer durations are concerned, it's tricky, because backups do not include disappearing messages. That can cause unexpected behavior for a user that set their timer to 18 months.
The variable duration of the message adds one additional layer of complexity to the already messy game of social communication.
It's not like a session at the park or a phone call that once you close it, it's over.
I already had trouble understanding the concept of ephemeral message, like you say something to me but expect me to not remember it in one hour, so why say it in the first place.
It gets even more confusing when people don't remember what they have told you in the past and change their mind.
I am also confused by the fact they acknowledge that this is pointless in an adversarial scenario. It's like submitting to the rules of social norms instead of playing reality ; But people who are fluent in social communication will instinctively use it for soft lies while others suffer trying not to infringe stupid rules.
It's like a social norm that restrict usage of external memory (and by extension internal memory). Why not make alcoholic beverage consumption or taking memory-suppressant-drugs mandatory then ?
This change of rule distort the game and not for the greater good as only cheaters will take advantage. Promises now only hold for the duration of the timer.
What value of the timer, am I supposed to put so that people are comfortable. I don't necessarily want to pressure them in answering quickly, but also I don't want to always be sending weak signals by always configuring a different default, and I don't want to handle the various other people timers preferences.
Semantically here it's like the self destruct message they send in Impossible Mission but with a very long timer, it's a new primitive I'm not used to handled yet.
A garbage-collector for your memories like I am inviting the Langoliers into my phone.
I would find it odd if someone were writing down everything I said. I already find it quite strange when I write something (on Signal or similar), and the person responds with a quote-reply of something I said years ago. "But I thought you wanted to visit Moscow?" or even "I'm sure you said you wanted to visit Moscow" are quite different from "here's the message you sent on 2019-02-05 at 14:32 telling me you wanted to visit Moscow".
It's not a super alien concept, but it requires getting used to a new social norm. It is difficult, but I don't think one can really get it without actually embracing the concept without questioning it.
If people are not adversarial then all extra information that you can handle properly is good.
It will also help people perceive a lot more depth in other people. And that will help good relationships grow and help avoid bad relationships.
It will help alleviate miscommunication issues and give more weight to intent.
This is not going to change why I don't use Signal as much though. I try to partition my conversation modes if possible (since not many people I know use Signal). If there's anything very sensitive or isn't needed after a while, I'd use Signal. The fact that it doesn't even have chat backup and restore for iOS (I'm not talking about transferring from one phone to another when they're in close proximity) forces me to avoid using it for important stuff that I'd prefer to have around. I use Telegram for the conversations that matter, and the fast search feature in Telegram has come to be of great use for more times than I can count or remember. Not to mention Telegram is far too feature rich than Signal can hope to be in the next few years.
Of course, if the other party would prefer history to be deleted, it's generally nice to go along with that.
Can I do that without the other party knowing or being able to override it? I would check it myself but signal requires a phone number so I can't.
We already have email for things we need permanent record of.
If I could be assured of privacy in the recordings (which I think isn't achievable today) I'd jump at the chance. I don't think I have a particularly bad memory, but it's not exceptional and won't be getting better as I get older. The idea of enhanced recall is very exciting to me.
I wish the technology was something available to everybody should they choose to partake. Society would have to adapt. What is human history, though, if not a narrative about society adapting to technological change?
People who turn on transient messaging, it's their right, and I get it, but it's also hobbling me from doing what I'm quite good at and basically exposing what's for me an upsetting functional disability, which bums me out a bit.
(If I really cared I could take notes elsewhere, yes, but jeeze that's a lot of work for daily chitchat)
Like that?
That's not to say that having a persistent log of communication might not expose someone's pattern of lying. If it did I'd definitely incorporate that knowledge into how I interact with that person. It's not something I'd ever state explicitly, nor that I'd necessarily tell anybody else, but it would be factored-in to how I deal with that person.
If you only do it to prove you were right, stop doing that, it's not a good look.
I'm not using my persistent records to "correct" anybody, per se. For very specific "offensive" scenarios, like taking a vendor to task for failing to meet requirements, for example, I'm glad of having the details in black-and-white. Good business dealings are done in writing, in my experience. It keeps all parties accountable and.
Aside from that "offensive" scenario my old communication records are there to augment my memory. They helps me assure I'm living up to expectations and promises, and they give me super-human ability to recall minute details.
If referring back to old communication suggested a pattern of inaccuracy or out-right lying on somebody's part I'd certainly use that to inform my future dealings with them. I think it'd be gauche to speak about it overtly to anybody (except maybe immediate family) but it would color my interactions.
It seems fine to me if someone makes an error in recollection to correct them (if it's about something material). I mean there are more and less nice ways to do it, and you don't necessarily want to do it about everything, but it seems like a very odd rule of thumb to me to say "it's not a good look". Maybe you have a particular personality archetype/pathological behaviour in mind that I'm not familiar with?
Is it about the awkward relational-imbalance that might be created between people who check records and people who can't be bothered? (the former able to use their spare time to reduce the status of the other party within the relationship, who lack the time/stamina/desire to reciprocate).
I know some super-pedantic people, who might engage in such correction pathologically. With those friends I actively resist ever making any distinction/categorisation that might be used against me in the future. It's a real arms race!
What sort of error? The GP said "where other parties recall a verbal conversation differently than me", and what's the benefit there? You have two differing views of a conversation, you can say "oh well I meant X" and move on. There's no benefit in figuring out who got it wrong just to apportion blame. Just leave it ambiguous and clarify the intention when you realize there was a misunderstanding.
> Maybe you have a particular personality archetype/pathological behaviour in mind that I'm not familiar with?
It's nothing pathological, some people just tend to want to go back to prove they were right, but proving the other person wrong doesn't tend to be productive.
I recognize that two people may interpret a conversation differently. The ground truth of what was said has value, though.
If I know what was said I can work to tailor future my interactions with that party in an effort to prevent misunderstandings. I will certainly also try to suss-out if the other party has a pattern of apparently choosing to recollect conversations differently for their own ends (or maybe just has memory issues). It's going to color my interaction with that person. I don't "blame" them but I'm also not going to ignore information that could help me in future interactions. I'm not going to speak about it (because I think that would be in poor taste) but I'm certainly going to use that information to my benefit too.
If I find out that, though reviewing old communication, that my recollection is faulty I'm going to make things right, make amends as necessary. I'm also going to try to prevent it from happening in the future, too. I genuinely don't want to be a person who choose to recollect conversations differently for my own ends. That's impolite and not something I would ever want to do.
I see your point, though I still think that deleting them is a better default, as I prefer my everyday chats to be ephemeral. Certainly I wouldn't want them lasting for years.
I relish knowing about those, too. That's a learning moment for me.
> It's not that common (at least in my group) that this is done out of malice ...
I don't see that in my personal interactions either. I've seen it in business interactions enough to look out for it. I've worked for and along side ethically-challenged individuals who consider agreements subject to interpretation, to their advantage, indefinitely.
(An aside - Thanks for the continued interaction here. I am enjoying it, though I should get back to work. >smile<)
Interrupting conversations with “lemme check that” followed by fondling a phone is a nasty habit anyway.
re: interrupting conversations - Yeah. That sounds like a dick move, and not something I'd do. The whole "I'll go look this up on xxx reference site" that some people inject into conversations can be maddening. Rarely do I hold a phone or look at a screen while I'm talking to others. I wish more people did that.
I do think there's good value in referencing old communication if it's something the parties agree has value. If I'm working with somebody to make a decision that could be better-informed by way of looking up old communication I'm going to suggest it, but I'm not going to actually do it unsolicited.
I'd think a better idea would simply be to accept this as reality, and have a config entry or utility that lets you batch delete messages marked "ephemeral" (or better in this context, "unimportant").
The feature has two functions. The first is to decrease the severity of the consequences if a conversation partner's device falls into the wrong hands, say a government who wants to use chat messages as evidence, a corporate competitor who stole the phone, or just a friend who is rifling through your unlocked phone.
The other purpose is to make it easier to participate in a social contract where your chat history won't be kept around forever for all participants to dig up in the future. Yes, someone could modify the client code, or take screenshots, or do any number of things to preserve data. You still have to take care about what you send. But most people won't, because it is additional effort and generally people want to adhere to social contracts.
Security is never all or nothing. It is always about taking as many reasonable steps as you can to limit your risks.
Came here to say this but someone beat me to it. I imagine forging a screenshot to 'put words in someone's mouth' is hard to do though. You'd have to find a way of injecting words into the Signal app by messing with the code. You'd also have to use the correct font, make sure the metadata of the file is consistent, etc
Since there's no permalinks for a Signal message, you could forge anything you wanted, and you just have to get some dummy to believe the screenshot is an actual screenshot of a conversation. Treat screenshots of messenger apps with plenty of skepticism.
> This is not for situations where your contact is your adversary — after all, if someone who receives a disappearing message really wants a record of it, they can always use another camera to take a photo of the screen before the message disappears. However, this is a nice way to automatically save storage space on your devices and limit the amount of conversation history that remains on your device if you should find yourself physically separated from it.
> This is not for situations where your contact is your adversary — after all, if someone who receives a disappearing message really wants a record of it, they can always use another camera to take a photo of the screen before the message disappears.
/s
Hence why Lord whatever is in trouble.
It seems to be extremely common in govt now. To the point where actively requiring surveillance of personal devices of government officials seems pretty reasonable.
nobody asked for this sort of thing, just like no one asked for signals bizarre crypto currency or the security passphrase stuff that comes with a weekly reminder. what we did ask for was a full post-mortem of the outage that crippled the service six months ago, and a better more consistent (and buildable) source code released to the git repo regularly but we still didnt get those.
we've pushed (albeit for a while now) for a decentralized signal...one that would arguably be more resilient and reliable, but still havent seen it. It feels like the projects just doing whatever Moxie wants.
While you make some valid criticisms this functionality seems firmly related to Signal's core goal of truly private messaging.
It's a simple swipe on the message, and either party can unsave it. Of course my ex-girlfriend just saved everything.
Snapchat's UX is (was?) really incredible for privacy, and did a great job of mimicking the IRL experience, they just didn't bother with E2EE unfortunately.
Chat is all about what happened recently and not about what happened weeks/months ago. Ephemeral messages are a good default.
While I agree in principle that email should be used for things that are not ephemeral, chat is used for a lot of things where history becomes important like
- A command a coworker pinged you which has some special flags for a build
- A doc that was shared with you
I'm curious what the legality of that is under the GDPR and other preceding data-protection regs...
If you're using Slack in a personal capacity, then don't you have the legal right to demand an archive of your messages without paying them anything?
Thats all.
I'm not really interested in the disappearing message either if I wanted that I could just use Slack Free edition :troll:.
In all seriousness though, my messages are part of my memory. They're shared experience with my partner and a record of decisions.
I understand they're useful for some people, but they're an anti-feature for me.
Thanks for saying this, I moved to Telegram[1] before I had a chance to learn about/suffer from Whatsapp take on backups, but it would have been devastating to lose years of messages with people over a bricked phone.
[1] I know Telegram has a bad rep due to their home cooked crypto and maybe other stuff, but I like the app, I vaguely trust the author, and it works fine for me.
Is that because you assume the NSA aren't reading all your messages or you don't mind if they are? Because the whole problem with dodgy crypto is that it still seems to "work fine for me" even when it isn't.
Disappearing messages are to prevent my kids reading the messages between me and their mother ;-)
If NSA actually read my messages that would be kind of funny.
If they have such a hack I'm sure they won't spend it on me or any other Joe Sixpack.
Maybe ФСБ reads it but I have a hard time believing that too.
My threat model consists of (not in any particular order):
- my kids,
- random internet strangers,
- Facebook (who'll sell you to whoever pays),
- Google or a Google-like entity with runaway "AI" and arrogant customer support[1] (who can lock out from my accounts for no reason)
Besides, despite all the talk about E2E-encryption being so important, Signal (which I cheer for) IIRC has had a nasty XSS-exploit in their desktop allowing remote control of the pc, their phone client has been sending pictures to others than the recipient and probably a couple more.
Ad I and others have pointed out repeatedly:
There is more to security than encryption. It doesn't matter if it is E2E-encrypted if it is delivered to the wrong contact. And it doesn't matter if it is E2E-encrypted if whoever knows the secret can send you a specially crafted message and take control over your machine.
[1]: I originally wrote employees but I don't think the average Google employee is more arrogant than others, they just come across that way because of policies created far above them.
Not caring about your messages being read, or being willing to trust the operator of the service, is fine, but in that case I struggle to see why you'd pick Telegram, given that their whole selling point was about privacy and encryption when their encryption didn't actually work.
It is secure for my purposes.
Throwing away threat models and the meaning of secure and encrypted just to be able to define something that competes for attention as "not secure" and "not encrypted" (from other threads) isn't very high level, it is basic framing.
edit:
> Not caring about your messages being read, or being willing to trust the operator of the service, is fine, but in that case I struggle to see why you'd pick Telegram, given that their whole selling point was about privacy and encryption when their encryption didn't actually work.
Remember, when Telegram came around WhatsApp was un-encrypted. Not point-to-point encrypted, just encoded. You didn't need to know or crack a key, you just needed toknow the protocol.
Also, the main selling point of Telegram as I remember it wasn't security, it was dumb IMO, it was "we will be free (no cost) forever while WhatsApp is going to charge you money".
Finally it isn't that I don't care if my messages are read, K just cannot care if my messages are read by NSA or FSB. If they go after me nothing will stop them.
It's basic, but it's valid. I was responding to all your "It doesn't matter if it is E2E-encrypted if xyz" points, which are faulty logic; if whether it's E2E-encrypted matters in your threat model, then whether it's E2E-encrypted still matters in your threat model even if other aspects of the system are secure.
> Remember, when Telegram came around WhatsApp was un-encrypted. Not point-to-point encrypted, just encoded. You didn't need to know or crack a key, you just needed toknow the protocol.
Citation? I don't remember it happening that way at all; Telegram marketed themselves by attacking WhatsApp on a couple of minor vulnerabilities in side features like video handling (which was legitimate in some ways, but extremely hypocritical given the bigger weaknesses in Telegram's own implementation). If message text was readable that would have been a much bigger deal.
I wrote basic framing.
It is a valid and effective marketing technique but not something you should rely on in a technical discussion.
> if whether it's E2E-encrypted matters in your threat model, then whether it's E2E-encrypted still matters in your threat model even if other aspects of the system are secure.
Then we agree :-)
I've been communicating over email, irc and lots of other channels that were not E2E-encrypted for years and I still have to.
I'll have to live with the fact that my bank, my government and a number of others have unencrypted, realtime access to far more sensitive data than Telegram.
The fact that someone can read my Telegram messages consisting of perfectly innocent photos is way down on my list as long as they who do it has a lot to lose by doing it ;-)
(I.e. if you had access it would be a problem. If any serious agency has access they will never use it against me because then they give away that they have the capability. In the same way Telegram has everything to lose by letting employees snoop. Same as Google: I do not fear them snooping my mail, I fear their AI will kill my account and no one will tell me why.)
> Citation? I don't remember it happening that way at all; Telegram marketed themselves by attacking WhatsApp on a couple of minor vulnerabilities in side features like video handling (which was legitimate in some ways, but extremely hypocritical given the bigger weaknesses in Telegram's own implementation). If message text was readable that would have been a much bigger deal.
I've followed Telegram quite closely so I am fairly certain about their marketing message early on. As you can see it also put them in a bad light so thats something too.
I've been unable to find references for the point about just being encoded, not encrypted so take that with a grain of salt for now. Some evidence can probably be found in an old unofficial client repo or someone who wrote one of those can confirm.
I might be wrong but remember this was back in 2009. It was a whole different world back then.
I used and liked WhatsApp despite all this because just like with Telegram perfect security didn't matter for the kind of communication I have on messengers back then either.
> WhatsApp was initially criticized for its lack of encryption, sending information as plaintext.[198] Encryption was first added in May 2012.[199][200][201] End-to-end encryption was only fully implemented in April 2016 after a two-year process.
from the "Security and Privacy" section of :https://en.m.wikipedia.org/wiki/WhatsApp
Happy now?
Again: this wasn't thought to be a big deal back in 2009 and I don't want to beat them up over it.
The person I wrote to wrote:
> If message text was readable that would have been a much bigger deal.
For anyone who has read so far, keep in mind such rules only apply to Telegram ;-)
Now: can you admit that WhatsApp was unencrypted and you were wrong?
But I think my original point still stands, because at no point has Telegram been a more secure alternative to mainstream messenger options (such as WhatsApp), despite their marketing and branding being security/privacy-oriented and their "breakout moment" being a targeted, hypocritical call-out of WhatsApp security. Yes, I'm holding them to a higher standard than WhatsApp. But I think that's completely fair given that security and privacy is Telegram's supposed selling point, whereas WhatsApp never positioned itself as anything more than a way to send messages to each other.
If Telegram had come first then I could understand why people would keep using it. But I don't get why anyone would switch to it (except for the owned-by-Facebook thing I guess, which sure, is much more of a question about what your threat model is).
Why don't you look at discussions here and see why people use Telegram?
Many, maybe most of us don't think of it as specially secure.
If you ask why people use it they'll tell you it is fast, stable, almost bug free, well designed and everyone else in their group uses it.
I've yet to see anyone using cryptographic security as a selling point.
Telegrams biggest claim to fame security wise is their censorship resistance and that they get away saying they've never give authorities one bit of user data, year after year. If they had a secret deal with anyone that should soon become obvious.
Edit:
> and FWIW that does reduce my trust in WhatsApp (who I've never been the biggest fan of in any case).
I'm consistent here: I liked them better back then when they were a scrappy startup trying to do good in a world of ads and tracking.
Technical issues can be fixed, as they have done.
The incentives formed by being bought for $17bn (or 19 or whatever?) by Facebook, they cannot easily be fixed.
Yes, messages are now E2E-encrypted but we now have thousands of people working trying to exploit metadata etc instead of 50 guys trying to avoid that.
example of heavily encrypted: multiple rounds of properly configured AES with multiple keys split on multiple servers.
example of non heavily encrypted: zip archive with PaSsWoRd as password.
Sure, it's not actually saying a lot. My point is that's the very first thing in their marketing message, so they're very much advertising themselves as secure/private. By comparison if I pick an arbitrary cloud storage service - Dropbox - their first line is "Bring your files and cloud content together with the tools your team wants to use".
what are referring to here? I vaguely know about delivery bugs and protocol patches
I don't defend Telegrams history when it comes to crypto,just pointing out that there are massive double standards here:
- WhatsApp is the good one (sending data unencrypted until 2012, and to add insult to injury they sent it on port 443)
- Telegram is the bad guy (sent data encrypted but used a dodgy algorithm early on)
Absolutely not universally true:
Encryption that protect against everyone except NSA is still infinitely much better than sending plaintext.
Internet banking can't work with plaintext.
It can work perfectly fine even if everyone knows NSA could break it.
An example is E2EE, it is a nice property but today it has significant UX implications that makes it bad default in my opinion for a popular personal messagin app; in other places it is essential, just not there.
Moreover the main threat model is not NSA hacking my device or the app's servers, it is app developer selling my data and governments making "lawful" requests to service providers. In both these metrics signal ranks better as far as I understand, but it is more of an individual choice.
If we are worried about the NSA the best thing to do is to move out of SMS/phone calls as fast as possible on mediums that do not allow passive monitoring and are likely to resist government pressure.
But Telegram's whole selling point is privacy and security. If security is not your priority (which is totally reasonable!) there are plenty of other messaging apps that are much more established than Telegram, and much more honest about not being particularly privacy-oriented.
> An example is E2EE, it is a nice property but today it has significant UX implications that makes it bad default in my opinion for a popular personal messagin app
Doesn't seem to have been a problem for e.g. WhatsApp.
> Moreover the main threat model is not NSA hacking my device or the app's servers, it is app developer selling my data and governments making "lawful" requests to service providers. In both these metrics signal ranks better as far as I understand, but it is more of an individual choice.
> If we are worried about the NSA the best thing to do is to move out of SMS/phone calls as fast as possible on mediums that do not allow passive monitoring and are likely to resist government pressure.
Both these points seem like complete non sequiturs. If you want to avoid the app developer selling your data or providing it to governments, you need E2EE, there is no other way. Passive monitoring hasn't been a thing for about a decade, any non-joke messenger is at least using TLS, and it's not like people were using SMS/phone calls until Telegram came along with the revolutionary new idea of an internet messaging app.
> there are plenty of other messaging apps that are much more established than Telegram.
Not where I live, the only comparable app is whatsapp and after that various social media platforms probably.
> [E2EE] Doesn't seem to have been a problem for e.g. WhatsApp.
It is, porting backups between phones (last time I tried) requires both phones to be active.
> and it's not like people were using SMS/phone calls until Telegram came along with the revolutionary new idea of an internet messaging app.
They were using whatsapp, which in my opinion is worse than telegram and uses E2EE as a PR shield. I personally do not trust facebook to deliver a trustworthy app (they would have to be OSS with reproducible builds at least) I do not care that my messages are encrypted on their servers, they can steal them from my phone storage directly. It is a matter of lack of trust towards facebook.
I keep that in mind when discussing things.
If I want to start doing something shady, I can pick up something else than Signal, since I don't want my phone number ending up on random people's phones.
Telegram's initial marketing heavily emphasised privacy. If I search for Telegram then the very first search result blurb starts "Telegram messages are heavily encrypted". Maybe they don't explicitly claim to be E2E, but they absolutely are selling themselves as an encrypted, privacy-friendly messenger.
> If I want to start doing something shady, I can pick up something else than Signal, since I don't want my phone number ending up on random people's phones.
Completely agreed, I dislike the Signal hype as much as anyone. But I trust Telegram even less.
https://telegram.org/privacy#3-3-1-cloud-chats
> All data is stored heavily encrypted and the encryption keys in each case are stored in several other data centers in different jurisdictions. This way local engineers or physical intruders cannot get access to user data.
I want to find a better cell plan with a company I am not ethically opposed to but Fi is so good that it makes it really really hard to beat.
I avoid phone based 2FA where possible, but the few that I have are sent there.
But it's not really for non-US, and they've really started cracking down on people that use it exclusively internationally.
Which is a real shame. It's absolutely the service I want/need as digital nomad.
Do you have to constantly cycle through numbers, do you have a VOIP number?
I'm note vertis, but I wont give out my cell phone number, so: Companies that require 2FA via phone just don't get my business and about all banks in germany offer chipTAN. The chipTAN devices are all made by shady companies and are less than open but I still trust them more than any smartphone or the mobile network.
Or, at the very least, support people who have a phone that is not Android or iOS by allowing them to register from a desktop application. Does not seem misaligned with Signal's aim for privacy.
Pros
Free
Very good encryption
Almost no metadata kept
Protocol independently audited
Seamless to use on Android
Disappearing messages
E2EE text, voice, and video group chat
Cons Requires a valid phone number to register
Hosted on Amazon Web Services (AWS)
[1] https://protonmail.com/blog/whatsapp-alternatives/You can manage to use Signal with most features without the need to run proprietary blobs by registering your phone number using signal-cli or axolotl (and maybe alternatives, but I haven't tried). They can run on a regular desktop or on a GNU/Linux phone, and then pairing Signal-Desktop with it. This is not straightforward, and forget convenience on the phone for now (I haven't managed to make groups work on UIs based on signal-cli though they theoretically should, Axolotl does not fully support new groups and by that I mean it's missing essential features like accepting an invitation).
Then, the blobs you are running are "only" in your phone's modem, unless you also managed to use some service providing SMS from the internet.
This is totally unsupported because it depends on alternative Signal clients that are not allowed on the Signal's network. For some reason Signal-cli and Axolotl don't seem to have been asked to stop from the Signal team, but it happened to a Signal fork, Free-Signal, which only goal was to de-blob Signal. And they are in their own right to do so, I'm not questioning this.
This is painful, really. Signal is so close to be great. Matrix / Element works well enough. It has rough edges but I (re)-discovered that Signal has some too and I'm not speaking about this whole situation.
Edit to clarify: Signal sets up a background connection instead, which still results in real-time notifications (the same way that Wire works on a phone without Google Play services)
A year ago or two, I wanted to build Signal for Android without the proprietary library that talks with google play services. It was possible by modifying the source code a bit, because the code checks for the presence of the handles the absence of the Google play Services and enables some sort of fallback.
I would not know if it is still the case today, I don't currently own an Android device.
Signal is an analogue of the tor browser, most people get diminishing returns after an adblocker/incognito mode.
they did iPad. presuambly the secure region coding for iDevices is better?