Bug Research ENS* ( Turkish alphabet )

Using the Turkish alphabet, you can register similar domains. For example, I have a personal ENS - obchakevich.eth, but if I turn on the Turkish alphabet, I can register ENS - obchakevıch.eth. The only difference will be in the Turkish “ı”, which does not have a dot. It is very difficult for an ordinary user to notice the difference, and this can lead to fraud. Let’s also take the lido.eth domain as an example. We register the same domain with the Turkish alphabet and get lıdo.eth which will be available for registration (example below) The solution to this issue may be to prohibit the creation of similar domains in which 1 or more letters will be changed using the Turkish or other alphabet. Entering a domain in a single Latin alphabet ( English ). Video - [bug ENS maın]


The problem here is that the domains are written in Turkish (Latin alphabet), but in Turkish there is a peculiarity in them “i” looks like this “ı”. The service misses this spelling.

That is, I do not mix languages, English and one letter of Turkish, I write everything in Turkish. If you look at my screenshots and videos, you can see that those domains are available and no one has registered them, and they are very identical to the original (already registered domains) such as 1inch. This means that no one knew about this before me, and while there is no publicity, we can solve this so that people do not start registering new domains in this way. Thereby increasing the no ways of fraud with ENS domains.

Solution to this problem

Exclude the Turkish “ı”, and when writing in Turkish, the service replaces it with “i” or indexes it as “i” and does not allow registration.

Please see the ongoing discussion in the normalization thread.

The ı character will already be disallowed with the new library. https://adraffy.github.io/ens-normalize.js/test/resolver.html#obchakev%C4%B1ch.eth

1 Like