How do open source companies avoid harmful contributions and find useful ones?

Question

Since wikipedia is open source, I can change anything I want. But what happens if someone removes, adds bad content, changes useful content? What does wikipedia do to prevent this?

The last question is available for any other open source software: linux, php, etc. How do they know what contribution is useful and what is harmful? Thanks.

Answer 1

Open source development is not exactly the same as Wikipedia. First, developers can still control what code they accept. For another, you have nothing to stop creating your own version with custom changes. This is done either to test them before serving them upstream, or if there is some disagreement as to what should or should not be included.

Open source developers typically review, run, and test all of the resulting code, and also rely on reputation to determine how much attention the code requires.

The Linux kernel is a great example. I run the kernel with Con Kolivas palette sets. These patches will probably never be accepted into the main core, but he will save his patches on his website if someone wants to use his changes anyway. It doesn't seem like he created a completely new kernel, he just needs to periodically update the fixes so that they still work.

Linus Torvalds highlighted some of these ideas in his Google Tech Talk about his Git VCS:
http://www.youtube.com/watch?v=4XpnKHJAok8
Of particular interest is his concept of a "network of trust" and his strategy for appointing people responsible for certain parts of the kernel (for example, Alan Cox used to be responsible for the TTY code and personally approved all corrections for this part of the code.)

Eric S. Raymond also wrote a lot of great material about open source development in general. Its relevant documents are listed here:
http://catb.org/esr/writings/homesteading/

Answer 2

Browsing process

In the case of Wikipedia, this is done by a special peer method with some exceptions. Highly contested pages are moderated, and changes must be approved.

Similarly, when a patch is submitted to an open source project, the validation process is performed to ensure that the patch does not add new errors or is not malicious. Of course, the review process can never be perfect and catch all possible errors that the patch may have, but due diligence is usually observed in venerable open source projects.

Answer 3

There was a classic example of how open source projects store malicious code when an attacker tried to send a patch to the Linux kernel containing a backdoor that allowed root access. There is an article here about an attack attempt.

This was discovered when observers noticed suspicious changes. See here . Malicious changes were detected before they entered the official version of the kernel.

Answer 4

See the code overview section in the excellent book Producing Open Source Software.

It comes to the existence of a version control system (for tracking and, if necessary, returning changes) and the availability of a good system for checking code on the spot.