Is this a private and secure messenger app or is it all a lie?

>privacy
>internet
pick one

>>57522677
Privacy implies encryption, which immediatly draws attention to your communications unless everyone encrypts.

>>57522677
Not as private and secure as they want you to believe.
A lit more private and secure than most competitors and with nice features.

>>57522677
Join:
https://telegram.me/joinchat/AWZzfQe_IZ2LKUiAusLXGw

>>57522982
What is it?

File: 1422978890106_2.jpg (23KB, 337x367px) Image search: [Google]

23KB, 337x367px

>>57522982
What is this group?

>>57522982
NASA is knocking on my door,.
Thanks, /g/

>>57523386
okay spaceman

>>57522677
It's both secure and insecure, depending whether you use "secret chats" or not.

Alright, getting out the tripcode for this one, as suggested in a previous thread about this app. (Why does it get so many? It's almost like it has an army of paid shills or something...)

Sorry anon, Telegram's shit.

Cleartext all over the place by default, yet it's promoted as a secure messenger; MTProto for secret chats, which are not on by default, is bizarre custom garbage using IGE mode (as Matthew Green and I discussed on Twitter, like trying to bake a cake if you've never heard of cake ingredients) and Pavel Durov won't fix it.

To quote the grugq, "I can't possibly think of a worse combination for a secure messenger".

If you want a phone messenger, your best bet right now (despite a few imperfections) is Open Whisper Systems' Signal.

File: kerroppa_joku_toinen_vitsi.jpg (59KB, 640x427px) Image search: [Google]

59KB, 640x427px

>>57523450
Hello Moxie, still trying to smear the competition that created a protocol that's both easily secure and convenient at the same time?

File: 3a3c48bad836b853ed.png (45KB, 780x724px) Image search: [Google]

45KB, 780x724px

>>57523450
>Alright, getting the tricode out for this one
Are you some kind of big shot around here or...?

Not OP but new to this but looking to go into security and cryptography.

So what exactly makes it so shit? Say, for example, anon were to simply use secret chat all the time. What's the issue? Surely if the encryption process is custom then it's under less scrutiny, as in, people have been trying to break well-known protocols forever and then this new custom one hits the scene.

Telegram claims that secret chants are protected by end-to-end encryption and that nothing is stored in the Telegram cloud.

Like I said, not an expert or anything, but what's the issue?

File: cf861ae5ea85912769.png (56KB, 780x822px) Image search: [Google]

56KB, 780x822px

>>57523545
Meant to post part II

>>57523545
The "issue" is that it's a new approach to the thing. It's untested, unproven.

File: eatadick6969.jpg (133KB, 508x396px) Image search: [Google]

133KB, 508x396px

>>57523579
Good security goes unnoticed. Has there ever been any major leaks of conversions, specifically secret chats, from Telegram?

>>57523492
If I were Moxie I wouldn't have Signal require phone numbers, would have tried private set intersection for contact discovery if I had any at all, and would open federation now the protocol's been documented. Signal is imperfect, but it's still pretty good and passed its review, unlike Telegram.

>>57523545
Look up Matthew Green's thread about it for more info. Data integrity, lack of AEADs, MITM issues, and clearly enough metadata is stored for several countries to attack extremists using the group features, which are all unencrypted.

>>57523619
There were some leaks, but they were caused by things not exactly related to Telegram itself, such as stolen phones. There also was some SMS and phone number intercepts, but that comes with the phone # identification, which I personally hate.

>>57523627
What kind of reviews? Server code review? Or what?

Oh, and Telegram has supported encrypted groups for a while. Your information is outdated.

>>57522677
Something that needs your phone number for authentication cannot be private.

>>57523704
>There also was some SMS and phone number intercepts, but that comes with the phone # identification, which I personally hate.

They also fixed that by sending the auth code to already logged-in devices.

>>57522982
>https://telegram.me/joinchat/AWZzfQe_IZ2LKUiAusLXGw

Thanks. Now you fuckers got my number.

>>57522677

>2016
>Privacy

>>57524629
What did you expect?

No messenger will ever be private and secure.

>>57522881
ok but all telegram traffic is encrypted soyou dont draw more attention than anybody else thats using it and its not like youre trying to hide from the NSA who doesnt care anyway, just any lesser 3rd party who might want to data mine you.

>>57524629
since when does telegram share phone number with strangers?

File: hardchoices.jpg (52KB, 427x575px) Image search: [Google]

52KB, 427x575px

>>57522677
>something that offers some level of privacy.
>something that offers a lower level of privacy.
Hard choices are hard to make

>>57524777
It only shares your public handle if you have one

>>57522677
PGP Encryption is the only way to go

>>57524832
>decent instant messenger
>"copy paste a gpg key here for le encryption"
>just works
>every other app spontaneously combusts from the awesomeness

Maybe someday lad.

>>57524832
>no perfect forward secrecy
I'll pass

>>57524777
>since when does telegram share phone number with strangers?

Since always

>>57522677
I'm using it as a stop-gap solution until Matrix has a proper desktop client and audited OTR support

>>57525067
>until Matrix has a proper desktop client

lol

>>57525079
It's gonna be a long wait, isn't it

>>57523704
Of the published MTProto protocol. I haven't looked at it since then, so I might give it another look at some point if you say it's changed? I could use a good laugh.

MTProto's inexplicably bizarre design choices are difficult to justify. It's certainly not the kind of unsurprising, clear, solid, defensible design that you see in strong protocols.

>>57525067
Oh yes, matrix.org mentioned their olm end-to-end proposal when discussing secure federated messengers, and I do plan to take a look at that. I think at this stage that much of the analysis surrounding the Signal protocol can probably be reused for olm, but we'll see in good time.

If the UK wants to ban it, it's good.

>>57522677
LibreSignal>Telegram

>>57522935
This. It's the best option right now, and multiplatform as fuck. Hopefully it will be replaced with something more secure but as feature-rich in the next few years.

>>57527045
moxie marlinkike threatened to sue LibreSignal maintainers so they dropped the project

>>57522881
Telegram is NOT encrypted by default. You have to create a "secret chat" for it to be encrypted and, even then, it uses its own ecryption. Furthermore, the fact that terrorists groups love it makes law enforcements organizations keep an eye on it. That said, feature-wise, it's still one of the best messenger apps.
If you don't care about that and you're explicitly looking for security, I recommend you check out Riot.
https://riot.im/
Full disclosure: I haven't really used it but it's open source, descentralised and uses standard encryption (I can't find it right now, but I'm pretty sure I read somewhere it was PGP). It also allows you to create your own server, if you're that paranoid.

>>57522677
Has anyone been getting messages in Arabic on telegram? I got one today and I googled it and the first thing that showed up was YouTube video : https://www.youtube.com/watch?v=Tn6MJw_BovE

File: IMG_20161115_124456.jpg (13KB, 695x597px) Image search: [Google]

13KB, 695x597px

>>57528815
Here is the screenshot

>>57528815
>>57528844
Also, why was there automatic reply