Nobody gets a PhD by publication with 0 publications. This is usually a backdoor for people who have done a lot of work in a field, certainly far more than a PhD thesis, and have just never gotten the credential.
It's amazing how many people on hn are experts on things they do not have the qualifications to be expert on.
> It's literally axiomatic.
You've made up some axiomatic definition of "by publication" that does not bear any resemblance to the actual definition. Consider that it's possible to
1. Submit a preprint to arxiv and have it count
2. Submit a preprint to a journal and defend before it accepted (or rejected)
3. Not submit anything anywhere and have the PhD itself count (almost all PhDs get an ORCID)
Are you aware that "PhD by publication" is a real thing that is a separate path than a normal PhD? It is relatively common for schools in some European countries to offer these, but not that common outside Europe.
This is a process where you can write your "dissertation" by putting an intro and a conclusion on ~3 papers you have already published and get a PhD that way. You enroll in the school for ~3 months, write the missing parts, and that's it. This is a flexible path to a PhD for industry researchers or other people who have a lot of expertise and have pushed the boundaries of a field but did not do a formal PhD program.
I have never heard of anyone doing this with ArXiV preprints or any school accepting this path if they are not referreed papers. I would love to see an actual counterexample if you have one.
>You enroll in the school for ~3 months, write the missing parts, and that's it.
There are degree mills that do what you describe.
There is also the format in countries such as Germany or the Netherlands where one typically "bundles" one's publications into a thesis. However, the work is typically done in the context of supervised doctoral programmes and no less rigorous than that done under different PhD studies formats.
I agree, you seem to be claiming to be an expert on something.
"PhD by publication" is a specific thing, it's in italics in the previous post.
My Universities offer a "PhD by publication". You basically staple together a bunch of your publications, and write a brief intro. It saves you writing a full PhD document. But, the standard on those publications is quite high -- you certainly wouldn't get one from preprints to arxiv at any University I've ever worked at.
Of course, you can get a PhD with no publications, just write a good PhD. Lots of students do taht.
Did you know that there are universities outside of Europe? And that in those universities, "3 publications plus intro and conclusion for a PhD" is also called (usually) PhD by publication (sometimes it is called a kitchen-sink PhD). And my point was that that rule of thumb does not apply to theory students.
Note the italics please.
So I'm sorry you're right I'm not in expert in Europe's monopoly on the phrase "PhD by publication" but who would want to be an expert in trivial bs like that (answer: apparently you and the other guy!)