I would appreciate you very much if you would share your own tips for effective peer review and the ethics of reviewers. Well-written articles to address them are also welcome.
Dear Michael, you described it perfectly. That should be the way how the reviewing process look like...But, in practice it can look a way too different. At one point I was thinking of publicly showing the list of most irritable suggestions from reviewres very often proposing the amount of change of the manuscript making it unrecognizable. I will mention just two cases, like asking to cite three papers of no significance (guess who was the author), to statement 'I am not impressed'. Btw how can you be impressed with an application of machine learning, known for quite a long time? I am just saying that I hope the process should become totally transparent soon...otherwise arena for those who are doing concurent research, or those who sre just mean people.
For an effective review of the publishable works it is important to have experts in the area, and with a sufficient number two or three so that their evaluations can serve as support to approve or not the work. In addition, an evaluation guide and ethical standards in writing must be submitted to the evaluator or expert judge. There is the link to the magazine dedicated to Philosophy that I lead in the university.
The question here is, who decides whether the person is an 'expert' in certain field. The fact the someone holds position on esteemed/very well pisitioned institution (famous univesrity) is not a guarantee that someone is an expert in comparison with the author (whose publication he is in a position to review). So, the order of things are in reality, very problematic. We all know what should 'good practice' look like, but noone is in charge to control that. My opinion is that is going to normalize when reviewing process become transparent and when we stop oaying three times for reading every paper, meaning tha everything become open access. But I do not beleive it to hapoen anytime soon.