This is a summary of the rough road to publication of the paper entitled ``Properties of blueshifted light rays in quasi-spherical Szekeres metrics''

The paper discussed here (hereafter called "Paper 3") is a continuation of A. Krasinski, Cosmological blueshifting may explain the gamma ray bursts. Phys. Rev. D93, 043525 (2016) (hereafter "Paper 1"). That one was based on the finding that radial light rays emitted from a non-constant Big Bang (BB) in a Lemaitre -- Tolman (L--T) model would reach all later observers with infinite blueshift (z = -1). Consequently, rays emitted at last scattering might reach the present observers with z close to -1. In Paper 1 I proposed that those blueshifted rays might be seen among the gamma-ray bursts (GRBs), a phenomenon familiar to astronomers. I managed to account for several observed properties of the GRBs. One of the unsolved problems was that my sources of the GRBs would have a too large angular diameter in the sky (2 degrees in my model vs. 1 degree, which is the current resolution of the detectors).

In the next paper (A. Krasinski, Existence of blueshifts in quasi-spherical Szekeres spacetimes. Phys. Rev. D94, 023515 (2016), hereafter Paper 2) I showed that a quasi-spherical Szekeres (QSS) deformation superimposed on an L--T background would make the blueshifting more efficient, i.e. the same BB profile would result in z being closer to -1. Consequently, the blueshift required to account for a GRB source could be achieved with a lower "hump" on the BB, and the hump would be further away from the observer, resulting in a smaller angular diameter. The problem was that the QSS model used in Paper 2 was only a geometric example, unrelated to cosmological reality.

Paper 3 was meant to be an improvement over Paper 1 using the idea of Paper 2. The L--T background was the same as in Paper 1, and the QSS deformation of it was adjusted by trial and error so as to produce the largest reduction in z. The paper, initially with a different title ("Modeling sources of the gamma-ray bursts using quasi-spherical Szekeres metrics"), was put on arXiv, see here , and the same text was submitted to Phys. Rev. D on the next day, April 27, 2017. The referee report reached me on June 22, 2017 and read as follows:
(If your web browser cannot open the window below, then click here )



Rather than extensively comment on the report, I copy my reply to it below. I suppose the PRD editors just looked at the title of the paper, ignored my recommendation as to the referees, and chose one whom they believed to be an expert on GRBs. Maybe she was (the last sentence of the second report, see further down below, suggests that this had to be a woman. A male referee would ignore such petty detail. So in the following I refer to the referee as ``she''). But evidently she was no expert on relativity -- some of my statements were not understood, and in other places she asked irrelevant questions. A reader who will be patient enough to read through my reply will spot several examples of her insufficient knowledge. The reply copied below was sent to PRD on July 12, 2017.
(If your web browser cannot open the window below, then click here )





The referee did not accept any of my explanations. In her second report she re-emphasised all of her critical remarks, adding a new erroneous bit (in point [4d] where she said that inhomogeneous dust models necessarily develop shell crossings). This is what the second report, received by me on August 11, 2017, said:
(If your web browser cannot open the window below, then click here )





Composing the reply took me nearly 3 weeks and spoiled my holiday. I revised it many times to make it as diplomatic as possible, and to avoid offending the referee. But it is extremely difficult to discuss with someone with whom I have no common ground of knowledge. So, with my second reply I sent the following message, only to the editors' knowledge:
(If your web browser cannot open the window below, then click here )





Please note (anyone who still reads this): I asked the editor to employ another referee IF the present one rejects my arguments. I still hoped I might convince her.

My second reply to the referee, sent to PRD on August 30, 2017, was unfortunately twice as long as the previous one. This is because of the chaotic composition of the referee's text, so I had to quote the pieces being replied to. It read as follows:
(If your web browser cannot open the window below, then click here )

(*****) (This is just a mark, which will be referred to below.)





I waited for an answer for 3 months, but nothing was happening. So, on November 30, 2017, I sent the following letter to PRD:
(If your web browser cannot open the window below, then click here )





The reply was rather unexpected. The editors assumed that my letter of August 30 was just an expression of anger and frustration and did not need to be acted upon. Here is the PRD letter:
(If your web browser cannot open the window below, then click here )





I resubmitted the paper on the next day. In the long meantime between August 30 and November 30 I noticed a technical error in it, so of course I had to correct it. The correction required several numbers to be recalculated and a few figures to be redrawn. The corrections did not influence the conclusions, so they are unimportant for the present text. But, to prove that I am not cheating, here is the link to the list of changes made in the paper. The resubmittal was accompanied by the reply to the referee reproduced above, at (*****).



Surprisingly, the PRD editors did not care to read my letter, but immediately sent the paper and the correspondence with the referee to another referee, claiming that this was what I wished. Here is the PRD letter, sent to me on December 13, 2017, together with the "report" of the second referee (which was of course negative):
(If your web browser cannot open the window below, then click here )





My opinion on the "report" is expressed in the letter that I sent to PRD on the next day, here is a copy:
(If your web browser cannot open the window below, then click here )





At this point, I lost all hope that my paper would be published in PRD, and prepared an amended version that I intended to submit to Gen. Rel. Grav.. The amendment consisted in removing as many references to GRBs as possible (I could not remove all of them because then my main motivation would become a mystery or even a nonsense). I did this to avoid the same old difficulties with new referees. And, of course, I changed the title in order to make the editors believe that my paper is just on relativity, with no relation to astrophysics or, God forbid, to the GRBs. The new title was "Properties of blueshifted light rays in quasi-spherical Szekeres metrics".

But, to my astonishment, the PRD editors sent to me, on December 22, 2017, the following conciliatory note:
(If your web browser cannot open the window below, then click here )





Of course, I took up the offer. This seemed to give me a better chance for success than starting the whole painful process ab initio at another journal. My reply, reproduced above at (*****), was this time forwarded to the original referee. Her new report came back on January 23, 2018, and is reproduced below. The novelty in it was the suggestion that my paper would become acceptable if I remove the references to the GRBs (her point (2)).
(If your web browser cannot open the window below, then click here )





The proposal to remove the references to the GRBs suited me just fine -- I had already done this while preparing the paper for the GRG journal. This is my resubmittal letter, together with the part intended for the editors only, sent to PRD on January 26, 2018:
(If your web browser cannot open the window below, then click here )





The last referee report came back on January 31, 2018, and it was positive, with one suggestion that was in fact right. Here it is:
(If your web browser cannot open the window below, then click here )





Note that the editors of PRD apparently agreed that the second referee did an extremely sloppy work -- they did not consult him again. The reply to the last remark of the referee was added as the last paragraph of section 1 of the paper and it read as follows:

"Most results of numerical calculations are quoted up to 17 decimal digits. Such precision is needed to capture time intervals of ~ 10 min at the observer, which is ~ $2 \times 10^{-16}$ in the units used here, see Sec. III. (The 10 min is a representative time during which GRBs are visible to the detectors [9].)"

Most of the many changes done in my text concerned style. The references to GRBs had been just a few from the beginning. The main problem was created by the old title, which glaringly suggested that a new model of GRB sources is my main topic. Once I changed the title, the referee's problems seem to have disappeared.

With this last correction, the paper was finally accepted for publication.

This whole story highlights a serious problem, which is pervasive in the astrophysical community: people with superficial knowledge gleaned from accidental sources pose to be experts and are taken seriously. With no common basis of knowledge between the author and the referees, discussions are concluded by means of rhetoric: the winner is the one who speaks louder and more persistently. I was perfectly aware that my proposed class of models of GRB sources is not yet good enough and needs several further refinements. But I was forced to hide the meaning of my results for the GRBs. The referees knew in advance that my approach has no future. This is not how science was supposed operate.