a57721

I get the same message for every particular artifact that I look up. If almost everything is behind a paywall, I'm fine with javadoc.io.

An average paper from arXiv has 10 people in the world who can appreciate every detail of the text, half of them are the authors, their advisors, and paper's referee, the other half will be mentioned in the acknowledgements. This is especially true e.g. for mathematics. I want to point this out because papertalk.xyz features many texts like that.

Moderating some kind of "official" internet forum for papers will require a lot of time, and most researchers won't bother to follow it, because messages will be of low value, from crackpots and occasional ill-spirited anonymous commenters.

In 2016 arXiv run a survey about new features. Comment section, votes on submissions etc. were not popular. Many people were against this stuff, so it was not implemented.

There have been projects like PubPeer that tried to explore this territory and never really took off.

I see a value in a website like HN with links to papers of bigger interest and discussions for the general public.

E.g. someone posts a link to a new publication/preprint about cancer research, a supposed superconductor discovery, massive high energy physics experiments, etc. and users are free to discuss the meaning, beauty, impact, and credibility of the results.

But I'm not sure if this can work for mundane papers that are >99% of published research. I also think it will be much harder to market to the researchers. As someone who read, wrote, reviewed papers.

> And don't get me started on the price for publication and for readers that supposedly fund this "detailed peer review process" that misses so many obvious flaws.

Readers don't fund the peer review process. People on the editorial board and peer reviewers do their job as a community service and have nothing to do with the inflated subscription prices.

This text mentions some valid and well-known points about SQL, but it lost me when it proceeded to compare SQL to GraphQL. The latter as an API spec/framework doesn't magically solve any fundamental problems, and it is not a data model by itself. Also the suggestion that an application may be reduced to a database with some kind of auto generated API - this is only true for simple cases.

I've been following Quanta for some time, I'm sure they don't care about SEO and number of visitors, they care about the quality of texts. They write for the general audience, and even though they try to preserve the scientific accuracy, sometimes their explanations may seem oversimplified and even a bit confusing when you come from the same field. It's not clickbait, it's their popular science style.

If a passenger is denied entry at the destination, the airline is normally responsible for returning the passenger. This is why they care.

The RFCs for HTTP require headers to be in US-ASCII.

Kotlin supports nullability on the level of the type system, it is similar to TypeScript in this respect.

The problem with nullability annotations in Java is that they are not enforced, and there is no commonly adopted standard. There are like ten competing libraries with similar annotations. There was JSR 305 ("Annotations for Software Defect Detection"), but it has been dormant. When you import a third party library, you never know what kind of nullability annotations it uses and if it uses them at all.