Disclaimer (sort of): here I am talking from a Software developer point of view AND about open source projects, which is true in most of the cases where I saw the behavior I am about to describe.
I read a blog post yesterday that is only the last example of a bad practice I saw often: disabling authentication/security in applications only because it is annoying to type the password “every time”. My point of view in those cases is do not do it. Not just because there is a reason to have such security layer - which is true BTW -, but because there are at least two better solutions (and I will give you three):
It might be a bug. In that case, the right thing to do would be to open a ticket in the bugtrace and the brilliant thing to do would be to fix that bug.
The application already provides an automatic authentication method. OpenPGP keys (GPG, PGP, and the likes), encrypted password files and a lot of other things (really fashinating some proximity sensor based mechanisms for example) might be available and the worst thing here is that I am not learning how to use those important technologies if I just “hack” the software security.
If the application does not provide anything like above. This is similar to point 1, I have a great opportunity to contribute to the project in a really valuable way!
Hacking, when legal, is good to learn new things, but sometimes it is not the right solution.