-
Notifications
You must be signed in to change notification settings - Fork 18
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[enhancement] avoid collisions with real contacts #2
Comments
Thanks for the ideas, but it sounds a bit complicated. And I'd rather have all fake names in one bunch (alphabetically), so they clearly stand out (for the phone's user) from real contacts. Yes, the current scheme is a bit easy for someone to recognize as fakes. OTOH, if everyone just uses the default settings, the given fake names will appear in lots of Contact lists, and thus look more genuine. I'll think about it. Thanks. |
I have programmed this method (quick and dirty in a script) and it generates a lot of fake emails like these ones: Miah Moran [email protected] If you are interested, see here: https://codeberg.org/kollo/X11-Basic_examples/src/branch/master/All/Internet/fake-email-names It is a sort of BASIC dialect, sorry, but maybe you can read it anyways... However, of course one need still to find a way to get these fakes out of the way of the user on the phone. Another thought: What about adding fake profile pictures from deep fakes..... |
Yes, good point. |
I have an idea for enhancement of the contact generation principle. It is hard for the app to identify fake contacts it has created, e.g. for later removal. Using a certain name pattern (all names starting with the letter "z") is simple, but the poisoned database also can be easily cleaned. So I would suggest the user to enter a 4-digit PIN at the startup of the app. All contacts will be coded with that PIN so that later fake contacts can be identified only if you know that PIN.
How to do it: Maintain two lists of 1024 names, 1024 surnames and 1024 last names. Now randomly combine them, (gives 20 bits code) and also combine them with the 12 bits generated by the PIN. Together we now have a 32 bit value for each combination. Now calculate a hash from these 32 bit (lookup table or 8bit checksum). And discrard all combinations which have a hash other than one speicific, say 0. Then all fake contacts can be identified and the chance of a false positive (to be removed from the contact list) is 1/256 (in this example with a 8 bit hash) which probably is good enough. One can use a 6-digit pin to further bring it down.
Lists with 1024 Names can easily be obtained/generated. One even can use customizes list according to the nationallity or so.
The text was updated successfully, but these errors were encountered: