Geekly Articles each Day
📜 ⬆️ ⬇️

Black PR Telegram. Who to believe?

image Recently, on Geektimes, a fuss was raised with the article “Bad Telegram” or How I did not take money for the black PR of Telegram in Habrahabr . As a result, they found out that the acquaintance of Burumych reads his daughter's correspondence and that the greeting “Good afternoon” is better than “Good day”.

In order to throw in useful information to the fan, we and Edison specialists made a selection of publications about Telegam and secure messengers so that the inquisitive reader could independently conclude (and not get a “paid” expertise) which we should trust and use for our purposes. Pro level of confidence / yellowness of the media I propose to decide the reader yourself.

The flow of adequate and inadequate information about the security of communication (including messengers) is growing and will grow, and I really want the competent, independent and understandable analytics to be found on Habré.
')
What criteria to use for evaluating the safety of instant messengers can be seen from fighters for digital immunity - Electronic Frontier Foundation (EFF). By the way, the question is, are these criteria exhaustive or additional (for example, about masking metadata) needed?

To increase the degree of objectivity and independence, please comment in the comments of those who understand the issue about the security of instant messengers.

Based on what data can you draw conclusions?


Strategic Intelligence Information Work Washington Platt

Electronic Frontier Foundation
www.eff.org/secure-messaging-scorecard

The Electronic Frontier Foundation (EFF) has published a serious analytical material (last updated 2015-11-03), assessing the degree of security and privacy of mobile and online messengers. EFF awarded points to the participants, evaluating the applications by seven parameters.

  1. Is the data encrypted during transmission?
  2. Is the data protected from reading by service providers?
  3. Can the user verify the identity of his interlocutor?
  4. Is the correspondence history protected from decryption when intercepting the current key (this question implies that the encryption key must change constantly, and the keys used are safely deleted along with the random data on which they were built)?
  5. Is the solution code open?
  6. Are the encryption methods described in detail?
  7. Has an independent security audit been conducted in the last 12 months?

More about the criteria in English
METHODOLOGY

Here are the criteria for various communication tools.

1. Is your communication encrypted in transit?

This criterion requires that all user communications are encrypted. If you’re not interested in a network connection, it’s not recommended. We do not require that metadata (such as user names or addresses) is encrypted.

2. Is your communication encrypted with a key the provider doesn’t have access to?

This criterion requires that all user communications are end-to-end encrypted. It means that it will be generated and stored at the endpoints (ie by users, not by servers). Endpoints but it is not necessary to end the keys or synchronize the keys between two devices. It is fine if users have public keys are exchanged using a centralized server.

3. Can you verify your identity with identity?

It is a criterion that has been compromised. Two acceptable solutions are:

There are no translations available.

A key exchange protocol with a short-authentication-string comparison, such as the Socialist Millionaire's protocol.

Other solutions are possible between the users and the cryptographic channel. For the scorecard, we are simply not evaluating the security of that mechanism.

4. Are your communications secure if your keys are stolen?

This is a criterion that requires it to be encrypted with the ephemeral keys. It is imperative that these keys can be reconstructed. Note that this criterion requires criterion 2, end-to-end encryption.

Note: For example: TLS with a Diffie-Hellman cipher suite) and non-forward-secret end-to-end encryption plus an explicit guarantee that you are not logged by the provider. It is not a cessation of secrecy at all.

5. Is the code open to independent review?

This criterion requires that it can be independently compiled. Although it is preferable to be a license. This is a case in point: backlash, back doors, structural problems. Note: When the tools are used, it is not a complete OS. This is a compromise, but it’s beyond the scope of this project.

6. Is the crypto design well-documented?

This criterion requires clear application. This is a white paper written by the audience of professional cryptographers. This must provide answers to the following questions:

What are the algorithms and parameters (such as key sizes or elliptic curve groups)?

How keys are generated, stored, and exchanged between users

Their key

This is a clear statement about what the model aims to provide. This should also be a clear statement.

7. Has there been an independent security audit?

This criterion requires an independent review. This is a review of the instrument of the development team. Audits by an independent security team within a large organization are sufficient. It is not a problem.
[ source ]


Top list of the most protected instant messengers according to EFF:



Caution long image
Summary table




Log] [aker


Habr


Kaspersky


Security lab


Tj


Roem.ru




RBC
Officials decided to limit the communication of Russians in messengers (December 9, 2015)
The authors (of the restriction draft law) propose to introduce such a notion as “information and communication services” into the laws “On Information ...” and “On Communications”. Their activity means the transfer of “text, voice and image messages that are technologically inextricably linked to communication services provided by third parties on data transmission networks of telecom operators”. Representatives of Internet companies and operators, having familiarized themselves with the document, came to the conclusion that we are talking primarily about instant messengers, although theoretically, social networks can also be affected by the law.


PS
According to a study conducted by Kaspersky Lab together with B2B International, 62% of respondents do not consider online messengers safe, 61% do not trust IP-telephony, 60% are wary of video chats. Nevertheless, 37% of those who took part in the survey most often use online messengers for communication, 25% - social media messengers and 15% - VoIP services.

The main findings of the study



One of the critical vulnerabilities of modern messengers (including Telegram) is unprotected metadata (attackers can get a list of contacts to determine the circle of communication, network access time, etc.), and very often the fact of communication is much more important than the content of communication. What tools to solve it (exclude from the system centralized servers or something else) I propose to discuss in the comments.

UPD
ru_crypt :
“The latest results from the analysis of the MTProto protocol: eprint.iacr.org/2015/1177.pdf . This paper shows that this protocol is not persistent in the IND-CCA model, i.e. It is possible to convert any ciphertext to some other ciphertext that can be decrypted into the plaintext source. Despite the fact that the attack is theoretical, the authors recommend not to use self-made solutions, but to implement well-tested designs, and they suggest changing the protocol so that the attack does not take place. ”


ValdikSS :
“I don’t like Telegram, but, in general, it is definitely better than other popular messengers, at least because there is an opportunity to make an encrypted chat with a more or less normal matching of the key fingerprint, but for normal chat, only the transport is protected, as well as , in fact, the majority of other popular messengers, and the media, for the most part, describe Telegram as a super secure solution that even terrorists use, so you, like, will definitely do it. I consider such an exaggeration on the part of the media to be quite a big problem, no kidding, because Such news attracts many technically illiterate new users to Telegram. If in the same Tor security is built architecturally, then the security of the correspondence of illiterate Telegram users rests only on the principles of developers about data disclosure and physical security of servers, and in the case of compromise, not only the users themselves suffer, but you too, if you corresponded with the user, the data which, for example, was given to government organizations. ”


zhovner :
“This topic is much bigger than all of us, because such important technology for humanity as IM and audio-video calls cannot be managed by one company. There are all sorts of work groups supporting industry-critical libraries and programs, such as openssl, apache. I think some IETF should do this.
It’s just that at some point they squander this and open standards didn’t manage to do commercial work. The fact that IM is now monopolized by companies is terrible. Ideally, all instant messengers should maintain a unified minimum standard sufficient to communicate with anyone, like email, and all of their branded pieces to be implemented within the network for their users.
Imagine that emails could only be sent between outlook, but you can’t go to gmail anymore. Or you could only call from Nokia to Nokia, but not to Samsung. This is exactly what the messengers look like right now. ”


J_o_k_e_R :
“The most obvious and obvious file telegram: centralized servers. Take the ass those who manage them - and everything. No telegraph. Even if no one read my correspondence in the past.

This file follows from a slightly less obvious file - the closeness of the source code of the servers. ”

Source: https://habr.com/ru/post/273001/

More articles:


All Articles