0

In a recent discussion re Tory Minister and British Home Secretary Amber Rudd's insistence that WhatsApp install a backdoor for government monitoring of encrypted data, I brought up Enigma as an early example of a system that was broken due to an exploit that revealed a backdoor, and cited it as something Rudd (a history graduate) should be aware of.

Many people disagreed stating that a backdoor is something that is intentionally designed into the system. My understanding was a backdoor was something that is intentionally designed or something which is unintentional and discovered as an exploit.

It is perhaps a question of semantics, but would one consider it to be a backdoor as a result of an exploit, or even an example of cracking a system due to an exploit, and if not, why?

  • 1
    The Allies distributing Enigmas they knew to be weak to other countries after WW2 with the intention of intercepting their communications could be considered a backdoor. -- "After the end of World War II, the Allies sold captured Enigma machines, still widely considered secure, to developing countries." – CodesInChaos Oct 6 '17 at 11:25
  • 4
    unintentional weakness != backdoor - backdoors are designed – schroeder Oct 6 '17 at 11:45
3

I'd say that for something to qualify as a backdoor it must be intentionally designed to allow access. Otherwise every exploit becomes a backdoor, and the term thus meaningless. The cryptographic weakness of the Enigma was a design flaw, and not something intended, and it would therefore not qualify as a backdoor.

Do note that no malicious intent is required - it is e.g. not uncommon that ISPs include backdoors in the routers they ship to their customers to simplify support. Their intention is not that the backdoor should be exploited, but of course if often is anyway.

Attackers often install backdoors after they have penetrated a system with the help of some exploit. The backdoor is malware designed to allow access to the system (i.e. there is intention) without having to use the original exploit again. This is practical, since the vulnerability that made the exploit possible might be patched.

Note the difference between the original exploit and the backdoor here. The exploit itself is not a backdoor, but it could allow the attacker to install one.

In the case of the enigma machine, there was only an exploit and not a backdoor.

  • Thanks for the great answer. This seems to be the pervasive view in the argument and the reason people disagreed, but (and I think this really is a question of semantics) pretty much all the definitions I've found describing backdoors talk about it being intentional or unintentional eg searchsecurity.techtarget.com/definition/back-door "attackers often use backdoors that ... install themselves as part of an exploit", although the attacker here does design the backdoor as part of the exploit and it doess not neccessirally describe a flaw in the design. Is this where I'm misunderstanding? – Munki Fisht Oct 6 '17 at 13:26
  • @Anders that really explains it - yes I understand now. A backdoor is always by design, but that could be legitimate (as in intentionally by the programer) or illegitimate (designed by a hacker who has already gained access to the system). For Enigma, even though they learnt more and more flags to help compute that day's decryption over time which reduced the computing power, they still had to go through that process. Thanks. – Munki Fisht Oct 6 '17 at 15:10
5

There are two parts to this question.

(1) Is a backdoor something that is put there intentionally?

This has already been answered by Anders in their answer. To summarize, yes, because otherwise every vulnerability might as well be called a backdoor.

(2) Was the Enigma cracked because of a backdoor?

No. The Enigma was cracked because it was vulnerable to known-plaintext and guessed-plaintext attacks. However, all cryptosystems at the time were. The cipher implemented by the Enigma was still (one of?) the most advanced of its day.
The operators, however, were the ones using predictable sequences. So the cracking of the messages was largely due to operator mistakes - human failure, rather than a failure of the cryptosystem. In this sense, the breaking of the Enigma was not due to a "backdoor".

Further reading, and source used: Cryptanalysis of the Enigma on Wikipedia.

  • @MunkiFisht If you wanna know more about how Enigma was cracked, I recommend to you the movie 'The Imitation Game' in the movie it is explained how Alan Turing cracked Enigma. I hope you haven't already seen it ;) – WasteD Oct 6 '17 at 13:16
  • I have, but it's a great movie - although as far as I know, pretty historically inaccurate :( – Munki Fisht Oct 6 '17 at 13:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.