This issue is not new in our everyday scientific publishing. However, I would like to know your experience and your suggestions to minimize this unfair behavior.
I know that this can be a sensitive issue. First, the authors should provide a list of references which give a snapshot of a status of the selected field and substantiate their claims with selected references. Second, the reviewers should evaluate the adequacy of the list. It's really difficult to list ALL references relevant for the topic. Usually, it's up to the authors to decide what is relevant (and can represent well their points) and what is not. Unless the article is a review article, the list will show only few selected publications - and this is normal. Given the immense numbers of publications produced every single day, week, month etc. it's impossible and impractical to list all. If the reviewers think that something crucial has been missed or not mentioned they should suggest some references to include. Again, this is really up to reviewers to decide.
The system usually works. However, when authors are not diligent and the reviewers are sloppy, it does not work. I'd add that it also depends on a pool of reviewers - so publish in reputable journals which tend to keep the plank high.
You see, the authors must select the most relevant and recent articles/book chapters, according to the topic that they want to explore/describe. Likewise, the reviewers must evaluate whether references are suitable to substantiate the study. However, important references could be left out if both authors and reviewers, fail in their work. Fortunately, I think, this is the exception to the rule.
On the other hand, failing to include the appropriate references in an article may suggest a lack of clarity/focus to approach the subject.
I agree with that it should be so from authors and reviewers. Even I agree with Serge that this is a very difficult task for review papers, but it should not be for ordinary papers.
I mean, I have experienced as non-referenced author and journal editor that some references to a topic are subtly omitted against others by authors when citing relevant works, even when all of them are published in the same period of time and related with the topic discussed. I mean, why to cited A work and not B work when both deal with the problem faced in the current paper?
Further, I agree that contemporary works should be cited, but it is also true that the general/pioneering theory/experiment must be cited and not only those updated works which already cite them. Some updated works only adds minor contributions, but authors consider that because of these already cite those original works, they do not needed to do it. Interestingly, the cited works are indeed "self references". Moreover, sometimes, alternative and, in some cases, limited approaches are done to avoid citing some papers and discussing them, thus limiting, in my opinion, the conclusions provided, and this is permitted by some reviewers.
I have also found that books are referenced when these books are feed from those pioneering works, instead of citing these.
A more objective work from reviewers should be requested by editorial board of any journal.
Maybe, some feedback for already published papers should be allowed by journals, or would this be the subject of a "comment" or a "letter to editor" ? If so, I am wondering why this type of papers are not so prolific in scientific journals.
I understand your point and even though I share your concerns, I'm going with the flow because I don't see a viable alternative to the current system. You're right that when you cite work A and don't cite works (!) B, C, D etc. you may offend those authors. But when would you stop in citing: A, B or A,B,C or ...what is wrong with adding D to the list? You as the author have to make a decision. Personally, I prefer going a bit overboard which may be too excessive perhaps. Most of my recent publications have 50+ references while being not review articles. I'm trying to widen the scope of my articles by addressing a broader audience but at the same time I'm facing a dilemma of selection of representative articles. Why? Because using even three references per point would make the list >100. Not sure what the editor would say in this case, besides some publications have pretty steep cost per page.
I do see published comments or letters to editors in some journals. However, such small contributions discuss usually more scientific rather than organizational aspects of the targeted publication. I'd say such follow-ups are relatively rare because it has to be something really significant that is worthy of publishing as a separate small contribution. Why not summarize it in the introduction or discussion sections in the new full-length manuscript? Then it would become a valid but minor point.
If you can't submit very narrowly focused short contribution (letter, comment etc.) or address the issue in a broader manner in a full-length manuscript, you don't have many options and then... just go with a flow. It means that the issue is not worth of fighting for it (whether you like it or not). So, move on: there are much more exciting things ahead!!
I agree with Jose, and "some feedback for already published papers should be allowed by journals" is a good suggestion. I believe these issues will be solved in the future.
I am currently involved in a journal from Nature Publishing Group as a guest editor, in which, me and the referees are in the head of it once published, as an additional information. This still maintains the anonymous reviewing.
I am wondering whether this kind of initiative in scientific journals also forces to referees to be more professional on reviewing papers.
Let's see how it develops...Probably, it would put more pressure on you as the editor to a) scrutinize a pool of possible reviewers and b) be more involved at a later stage once reviews have been submitted. For a relatively low flow of manuscripts/reviews at your responsibility, you should be able to maintain the standards you want. This is actually the advantage of being a guest editor. As you may imagine, it becomes a more difficult task for "full-time" editors.