Normally, a CS conference will give a clear indication in its Call For Paper (or in its submittion instruction) if the peer review of the conference is blind. However, if an author forget the anonymity requirement and submit a named paper, the result may be a merciless rejection even without any peer review. I have encountered such a thing, so I am just wondering is it acceptable to submit a anonymous paper to a conference even if the conference does not ask for it ? whether an anonymous paper would cause some trouble to the reviewers or the Program Committee of a conference if the peer review is not blind? Of course, the submittion system of a conference would record the information of all authors, including their names.
This is not a good idea. You just have to read every Call for Papers carefully and submit according to the instructions. That being said, I'm surprised that your program chair didn't ask you to resubmit with proper blinding instead of outright rejecting your paper. Submission errors happen, and authors are frequently given the opportunity to resubmit as long as the content isn't changed. The latter fact might be a little hard if the self-references weren't originally blinded, but I would think that most program chairs would have worked with you on this.