Why I write this blog?

I guess most people write for the purpose that somebody would read what they have written. Some of them may publish text because it is their job, or because they think they have something to say that others should or might want to read, or because they see publishing the results as an essential part of research, or making the claim of being the first to invent something on some topic, or because of any motive of this type. Not me, I do not have any such illusions.

           I used to study well-known problems long before Internet and any self-publishing was possible. I just put those studies to my drawer: I had an X-files folder for them. I knew very well that those studies cannot be published anywhere, yet I considered them as my real research. Occasionally I did mention of those results to some scientist, and they told me that I should publish the results so that other people could benefit from them. I was sure nobody would benefit anything from them: I had tried to get one or two of them published much earlier and it had been quite impossible to find anybody who would read any of these studies and say anything at all of them. It was very much so that everybody tried to stay as far away as possible from those papers of mine. It was a ludicrous idea to even suggest that somebody would like to read my studies if I would in some way get them published.

            Of course, these scientists did not honestly think that reading my studies would benefit anybody. They were quite convinced that if I did not get those studies published, it was because they were of low scientific quality. Two or three of these scientists wanted to explain to me where the problem was. So, they gave me some recent research paper they had written with a Ph.D. student. And in these cases I had shown these scientists a paper of mine, one of those that cannot be published. So, they apparently asked me to compare these two papers.

            The comparison was very clear: their paper had an extensive literature and the text showed deep knowledge of the field. The result was, to taken a concrete example, a new algorithm, which was evaluated by simulations or numerical calculations. Plots were shown in the paper. The results fitted very well to existing research: the new algorithm was a quite original improvement of some earlier methods and the plots showed that it was a significant advantage in this particular issue.  English language if the paper was polished, in general that article gave an impression of very good quality. I do not think the referees remade the extensive numeric calculations in the study in order to check the plots of the paper, but there was every reason to assume that the plots have been correctly calculated. So, this was a good paper and it naturally also was published.

           My paper from the X-folder was somewhat different. It was a solution to a well-known problem on a field that was not mine. As the field was not my own field I did not try to give a false impression of knowing the field deeply. Indeed, any expert could see that the paper was written by someone, who is not in the field. Because literature was not used for impressing the reader but only for the actual purpose of referring to some fact or method that is used in the solution, I gave maybe 1-2 or zero literature references. The claimed solution in my paper did not seem to fit nicely to existing research. Indeed, a claim that an old disputed problem was solved in the paper must have been irritating to many referees. Furthermore, my paper did not comment or mention other efforts to solve this problem, so while my paper nowhere claimed that all steps in the paper are new and invented by me, a malicious referee could argue that as the author does not mention any other authors, he must claim these results are all his. Finally, English language in my paper was far from elegant, and the paper was not in a polished form. Clearly, for these scientists my paper was low quality and their paper was high quality and the high quality paper was published, while the low quality paper was not.

           Fine, I tried to explain to them where their error is, but naturally they were very sure they are correct and they did not accept anything from my explanation. The case is that a referee evaluates submitted papers exactly as they think they do, and in this evaluation my paper is of low quality and their paper is of high quality. I knew this as well as they, and the reason why my papers would have been rejected if submitted (most were never submitted) was quite simple and clear. But this is not the correct evaluation of which research is better and which is worse.

           When we started our research career, we did not dream of solving such problems as in the joint paper by the supervisor and the Ph.D. student, giving a new algorithm for some very specific problem that nobody outside of that narrow field had ever heard of. What we originally thought as interesting research questions were the well-known old questions. As the original goal was to study such important and interesting questions, it is totally wrong to replace the original research question by rather irrelevant questions just because irrelevant questions are much easier to study and you can make a Ph.D and get a post in the university.

           We might want to know the meaning of the life, or if we live after the death, or what was before the Big Bang, or who did the 911, if there was the Holocaust, or whatever, but instead we can develop a new algorithm for, say, image compression. The new algorithm may indeed be a remarkable advance in the field, but it is not that question that was to be answered. Such more important questions are spread on a large number of fields. There is no possibility for any single researcher of becoming an expert on all of these fields. Thus, if a researcher would like to try several of those important problems (let’s say, even seven Clay’s millennium problems, though they are all mathematics, not to say of a wider field), he will necessarily have to make research on a field that is not his. Therefore he should not try to give the impression of an expert on the field for two reasons. One of these reasons is that a good research paper should not try to give a false impression and the author is not an expert on that topic. The other reason is even deeper: no research paper should try to make any impression on a reader. It should only provide an argument that the reader can check and verify or discard. Every impression that the paper makes by showing that the author knows the field, or that the English language is good, or that the text is carefully checked for typos, pictures are neat, and all this is just simply wrong: no such aspect should have any importance on the reader (the judge should not look at the person). The paper should ideally be given in the raw form so that the argument is plainly seen.

           Finally, the referee should check the argument, not trust that the author has done everything correctly. Very often it is very difficult and nearly impossible for a referee to check the argument. Let us say e.g. that the results use simulations. Is the referee assumed to do his own simulations to check that everything is correct? Usually he is not expected to do so. A referee can with reason feel justified in accepting results of a paper that gives the feeling of being solid science. In reality it means that a paper solves a problem using a way that the referee cannot completely check, but despite of this the referee accepts the paper for publication. Clearly, this can only happen if the problem solved is fairly unimportant. If the result of the paper is solving an old and important problem, then the referee feels that this time he should check everything. As he cannot realistically check everything, he should refuse to be a referee, and that is exactly what happens in solutions to well-known problems. It is nearly impossible to find a referee, and if you find one, he will just reject, not review or give correct reasons. It is like trying to pass a hot potato.

           It is not only a positive sign if the paper shows that the author is an expert of the topic. It shows that the supervisor and the author have spent years in studying unimportant problems instead of even trying the important problems.

           So, reconsidering which paper belongs to higher level research and which to lower level research the ordering is reversed. Now, returning to the question why I write these posts to this blog. The posts address questions that I myself think are interesting, I could not really care if some reader thinks so. The results in these posts cannot be published because no publication forum would accept them. I have mostly not even submitted them, but I know it, as I have published research papers and evaluated papers of other people. I am not going to put these results in such a form that they could be published, because I doubt it could work and if it would, it would be dishonest. Do I claim something like having invented everything in my solutions? No, actually I am fairly sure that all well-known problems have been solved correctly many times but the solutions have not been accepted. There are some difficult problems that really have not been solved, like the Riemann Hypothesis, but with many others it may be that someone is sitting on the problem and discarding the solutions, and there is nothing you can do about it. If you notice that somebody else has solved these problems in the same way as me but I do not mention anything of it, it is most probably not because of copying but because it is the correct solution: you get to the same correct solution any way you take, if you go straight, and there is no priority for finding truths that were never lost.

           Do I imagine that some people would gain anything from reading my posts? No, and I could not care less. The only reason these my results are at this time not in my drawers, as they have been most of the time, is because I decided to buy a site for a blog and keep it for a few years.