Math has a culture where author affiliations are often mentioned only as a footnote, at the end of the paper, and not at the front of the paper. For a field with such culture, where it has repeatedly been shown that quality work can come from many places, this kind of coarse gatekeeping creates a very unpleasant environment. I'm trying to imagine the motivation of the platform creators, but drawing a blank. Maybe it's a ham-handed attempt at spam filtering.

Their about page says that they use an algorithm to verify if you are a PhD candidate in math/stat. I don't really think that is a good requirement for the exact reasons you say, but I would guess it's set up like that to make this automatic verification more practical.

Wow, yeah. Came here to say the same. I could understand somewhat making a distinction between emails that end in .edu and those who don't -- at least during the early days -- but this system does seem excessive.

I do think the ideas in the paper are pretty cool! One question: have you considered or attempted some sort of generative models based on this system? You'd have to do some clever symbolic manipulations to deal with the "infinities" that crop up, but it seems like it could be interesting.

> emails that end in .edu

Only universities in the United States have emails that end in .edu. Most of the universities in the world have domain names that end with their respective country code.

> Only universities in the United States

Not strictly speaking true. Univeristy of Toronto (Canada) will give you a toronto.edu email if you're a CS student (even undergrads) or faculty.

Not sure what you mean exactly by "generative models" but it reminds me of Rohde et al 2004 (https://www.nature.com/articles/nature02842) which is all about modelling this sort of thing. In fact, some of the results of Rohde's paper are quantitative things which Rohde et al discovered by simulation and which my paper provides theoretical explanation for (see my paper's Proposition 6 and the last sentence of the abstract of Rohde et al).

Their guidelines [1] state that they need your e-mail because "an algorithm uses it to determine if you are a genuine PhD candidate in math/stat." It seems that the site is meant for PhD students rather than PhD recipients.

[1] https://www.hessix.com/guide

If the algorithm figured out from my email address alone that I'm not a student anymore, I'm pretty impressed. But I doubt that's what happened. :)

I was responding more to the idea that they're discriminating against PhDs who've left academia. If it's not meant for anyone who's graduated, then it's discriminating equally against those who've stayed.

I was sort of kidding earlier. I tried registering with my current, academic, address and was rejected. I'm not surprised because it ends in "mx". The "joke" was that a PhD student at my university would have a similar looking email and I don't think they would have been accepted either. (I did have an "edu" when I was a student in the US and that address would probably have been accepted.)

You should ask a PhD student at your university to try. It'll be interesting if they get accepted.

It might not be too hard to cross-reference that email address against published papers to get an estimate of how long since you started the program, or even potentially find online records of your degree and/or thesis.

PhD recipients are very welcome too. The registration is put on hold for me to check manually. The algo to auto verify works for PhD candidates in several universities (US and abroad, not only .edu)

The application and approval process, filtering, certification, authentication, etc. sounds like a separation problem. Since the site is about math, IIRC there's some math for separation problems??

Uh, since the site is for "math/stat" and there have to be rates of false positives, that people are already complaining about, and false negatives, I don't see people complaining about spam, then we're into statistical hypothesis testing, right? Sooo, to do better on the rates of false negatives, we want more data for a more powerful test. Sooo, we need a multi-variate test. Since no way can we justify assuming probability distributions for all the relevant data, we need a distribution-free test. So, where can we find one of those???

Disclosure: This question is just an exercise. For an answer, I published one of those. So, it's an applied probability calculation based on an algebraic group of measure preserving transformations! It may be a rationalization of resampling theory. Crudely the result is obvious, but a proof is tricky. It may be that the work is a stimulation for and or connection with approximate independence, e.g., maybe as in some work of Choquet student M. Talagrand.

Could you elaborate on the vision behind the "selection" criteria? Also, it would be best to clearly explain that in the Hessix user guide, in the spirit of communicating clearly with your audience :-)

You should probably make it more clear on that page. If the email address is only required to verify current candidates, why is it required for all registrants?

This is now clarified in the user guide

> PhD's who are no longer in academia.

And what about those that never were in academia? Mathematics has a long history of participants that were never formally educated (ironically Ram is even mentioned in the username section). Plus, being such a niche topic (especially one that a lot of people were taught to fear), I doubt that there'd be much abuse. Evidence of this belief is that Wikipedia has great math pages. Ones that many times are better than Wolfram's MathWorld, which can be too concise.

I've only just skimmed the first section, and already I feel like I'm in for a trip. The math is indeed as approachable as you suggest. I'm only a philosophy major (though our logic department is seriously the bees knees, shout-out to Tennant!) yet I'm still able to pick up what you're putting down. Very fascinating paper, thank you for sharing.

Also, one of the many nerds across the street in University Hall says hello :)

Hi, hope you'll enjoy the paper! I'm not at OSU any more unfortunately (miss it a lot). Agreed the philosophy dept is great there, I even had Stewart Shapiro cross departments to be on my thesis committee :) How far along are you in your program and is Tennant your advisor?

Shapiro! I'm actually in Advanced Symbolic Logic (PHILOS 5500) this semester and he happens to be teaching it. Talk about Hacker News serendipity.

Tennant isn't my advisor, but I've taken a handful of classes with him, and his "holistic" (that's the best descriptor I can come up with atm) approach to teaching logic and philosophy of science makes everything fall into place in my head. His pleasant, informal teaching style combined with a skill for connecting big ideas across disciplines is seriously pedagogical magic.

And I'm currently a senior, hoping to wrap everything up by this summer. :)

> you're discriminating against many math PhD's who are no longer in academia

FWIW, they gave me an account when I emailed them and told them I'm in that situation.