Too long title: On the possible use of Dissenter browser in order to enable an organised and durable platform where performing high level mathematical discussions about papers.
This is not a blog but I think that this question deserves an explanatory introduction. My week was horrible! I had to read a paper (published in peer reviewed Proceedings) that makes an interesting claim for my field of expertise even though their techniques are a bit far from the techniques that I mostly use. First of all I want to say that I respect the authors and appreciate and acknowledge their great work but there writing style was awful. The results in the paper are important and correct but the proofs are all really bad written (full of erratas and typos that make the reading much more complicated than it should be) and some other parts are so sketchy that it is plainly impossible to understand what they are doing without spending hours looking at every equation. These passages are so bad written that I made myself the mental joke that each author of the paper suffers a different kind of dyslexia and they put all their dyslexic effort to make each proof as confused as possible without making them completely incorrect. It is a pity. I lost hours and could even say days because of that.
I imagine that this happens quite often as it has happened already to me before (although this time was by far the worst) and maybe some of you think that it is not that important and that making it sound so bad I am being quite exaggerated. Well, I do not think so at all! But let me develop on this a bit.
I respect and acknowledge as a healthy practice the publication of mathematical works that have mistakes with their mistakes to show that mathematicians are not always right and make mistakes as every other human being in the spirit of the well-known point of view expressed by Grothendieck in conversation with Lawvere about the publication of Pursuing stacks
But in turn, Lawvere felt that the text needed to be thoroughly edited, and numerous errors
corrected. Grothendieck would not hear of it: students should see that even a great
mathematician makes mistakes.
However, I truly believe that there is no incompatibility in publishing works with mistakes and keeping public track of these mistakes so other researches can choose their preferred approach. I personally see the positive side of both totally compatible options: on the one hand publishing works with mistakes tell to the student that they do not have to kill their good ideas searching for a perfection that they are never going to truly find and on the other hand having an open and easily accesible repository for those mistakes can save the time of thousands of researchers that are maybe not connected directly to the circles were the mistakes are well known are folklore.
During this week trying to traverse through the puzzle of the mentioned paper I could not stop thinking on how much valuable and precious time we (the research mathematical community) are losing as a human group because of that kind of think. If our researchers have re-read, re-find and rediscover all the published mistakes or typos of whatever everytime that they have to undertand something new or approach anew field this is something absolutely bad and improductive as we are losing time and money (and any other scarce resources necessary for research) on something could be done one and forever implementing some really easy and handy mechanisms. In this case the mechanism that I contemplate is that already so popular and known one of comments.
Comments and commentaries have been always so important in pursuing knowledge in so many different disciplines that in some fields (like, philosophy, for example) is pretty common to find books make just our of comments from another book and these works are well known to have their own value. Of course, in the case that I am trying to make in this question I am advocating for an even smaller role of comments but having actually a role in Mathematics.
I am well aware of the problems and concerns pointed out by others about this topic. Coincidentally, I found one of the most bitter discussion about this in the comments section of this old post in the blog of Gowers. I followed the discussion there between Gil Kalai, Izabella Laba and Mike Taylor, and after hours of reading I acknowledge that while disagreeing and defending different points of view they all are right. Nevertheless, I was reading that in a situation where I had just finished suffering the pain of reading the mentioned paper and I have to recognise that I find thus myself in the same opinion that Mike Taylor: public comments are necessary to avoid make science (and specially mathematics going slower than it should because of these small or sometimes big stupid typos, erratas and mistakes). Science (if we truly pursue truth) has and can only be an open, public, transparent endeavour where we have to let everybody participate to allow a kind of self-correcting mechanism inside our fields. In this case, I think that math would benefit much more than we might think at first from such a perspective.
Said that, nonetheless, I think that we have to understand and respect the choice of those people who are not willing to receive comments (there are also good points to defend that position and I see and recognise them: hate speech maybe being the biggest problem that, although I think that it is a correctable one, some might think that it is not, and that is completely okay). Another related (and thus widely criticised) position (read in this related entry) is that of asking arXiv and other publication, repositories or editorials and publication platforms that they include a comments functionality or a discussion section where these comments are allowed. Again the main (and I share it) criticism is that arXiv and journals mission is not that: no they are not supposed to allow discussion but to announce and disseminate discoveries, results, theories or theorems, which is not the same at all. (Of course, when the mistake is too bad and/or affects fundamentally a central result, then withdrawals, corrections and comments in these traditional journals are expected as usual but this is not the main point or purpose of journals and I think that we all can understand this.)
Having all this points in mind shows to me that an optimal solution that can convey all the points of views respecting at the same all the parts implicated and all the different and well reasoned criticisms and opinions about this topic actually exists. It exists already and it is technically easy and feasible as it is already in use: we do not need new technology. Although of course new technology could improve by a lot the effects of proposals similar to this one: thinking in some kind of extended Stacks Project in the form of a well mathematically suited platform allowing TeX editing that could eventually being under constant observance, actualisation and corrections (this would require moderation, which contributes to make it much more difficult) and build over arXiv in some fashion and capable of connecting math topic from very disparate areas is as much hope as it looks certainly a kind of scify (not completely unfeasible with some monetary effort though).
Being more humble the best options are things like the already existing (and of course criticism is also already existing as it always will be) and commented or suggested in the answers to the questions just linked. However the possibilities currently in use by the scientific community are weak in some sense: the scientific community use them so scarcely that these platforms finish dying (and with them all the possible interesting storaged information in their corresponding comments and discussions sections), this is a pity and again improductive. There is not much value in spotting a reparable error for a paper in a platform that could close tomorrow because of financial problems making all the effort put by the person doing that (necessary) comment disappear from one day to another letting not track behind for the future generations to know that this error exists (which is the main point of this question).
The solution to the problems of unstability associated with these sites dedicated mainly to scientific discussions and comments could be easily solved if we choose more resilient platforms having bigger bases of users that can make these platforms more durable and accesible. Platforms like Twitter, YouTube, Facebook or Reddit, having a bigger, wider and more active audience than the (relatively small) scientific community in this planet that makes them more likely to last for decades (at least) without having to close due to financial problems have proven to be very helpful for the endeavour of doing mathematics: more recently platforms like Zoom, WebEx, BBB or Google Meets also joined the party of being very helpful for researchers when they are thought for a wider audience in order to keep them resilient and time-resistent (durable) enough (having a bigger audience also makes these platforms more easy to introduce and usable in general as these companies have more incentives to invest in the development of these platforms, apps and software than they would have if this one would be dedicated to the small audience willing to share scientific or academic comments). However, after all, the previously mentioned platforms are not so great for the comment capability implementation that I ask and wish for: they present several problems for that (the most important one from my perspective that they do not contribute to make the possible comments done there very accesible and easy to find as there are better attached somehow to the paper in question that requires a clarification comment) while being great for the tasks they are being used already. But it turns out that the platform perfect for this purpose already exists! Although some will not like which platform it is.
The browser Dissenter perfectly fits the requirements (only lacking some TeX capabilities for being actually perfect). Let me develop. The software is free and optional so those people not willing to see comments at all can completely ignore and avoid them while the users willing to see the discussion only need to access through the browser. With the Dissenter’s capability to comment every internet page each paper could be directly commented in their journal download page allowing thus an easy-to-find-and-reach place to have the discussion related to that paper. This comment feature is completely independent of the journal page or arXiv as it works completely through the Dissenter and Gab platform, which do not need to have any connection with the editorial leaving thus a clear and neutral whiteboard to talk about the paper freely, being free speech the main concern of Gab. The comments in each page using Dissenter can be rated and this is fundamental as this capability would allow some kind of automatic moderation in the comments: those cranky or insulting comments that do not give anything to the conversation would eventually be sent to the really low part of the whiteboard (sort of like in Reddit or this same forum). Additionally, a point in favour of mathematics using it is that the problem of trolls is not very high in a field like mathematics and if so they are easily spotted and ignored subsequently. Looking at the Dissenter whiteboard in arXiv I see not a single comment at all, which makes me think that trolls in that platform would not be very interesting in (excessively) annoying us; this would of course (and sadly) not be true in the case of other disciplines more open to the action of typical far-right trolling like Gender Studies and hence, for this reason, I uniquely see this as an available option for really technical and abstract subjects like math (or hard, pure sciences) and related.
I expect some criticism for this proposal because I am here defending the good parts and possible virtuous use of a platform that have been accused repeatedly of allowing hate speech and far-right movements. I again understand these reticences but I at least hope that they are as elaborated as mine. I took a great effort writing this and reasoning how and why the platform and the fundamental idea behind Dissenter is partly rescuable and could be extensively useful for our field. If the idea and the implementation are good it does not take such a long shot to try to change the platform base at least to use it to do mathematics for bigger and more respectable purposes as, at the end of the day, we all use internet while we know that hate speech happens in internet but we try to put all together our effort to make internet a bit better because this could only contribute to make the world a better place (perhaps just infinitesimally though).
After all the arguments shown here, my question is easy: do you consider possible and viable (in the medium and short term) feeding and pursuing this idea (maybe disseminating it among our students) of making (the whiteboard of) Dissenter in math sites behave as a place to collect scientific relevant data produced through the interaction between the several readers of a paper (that could be improved somehow) that could save time, money and resources through letting our papers (and thus math in general) being more suitable to be autocorrected by the mere action of the performers/paper-readers (aka mathematicians and students)? In short although less specific so you understand me better: do you think that the use of Dissenter in the way I just described could be beneficial for the development of mathematics and should be publicise as such among professionals and students?
P.S.: Thanks for forgiving me if there are too many typos but I have to go to sleep now as writing this took me finally longer that I expected. I may correct these typos in a subsequent revision.