Re: boost.sqlite seeking endorsement

From: Kostas Savvidis (kotika98_at_[hidden])
Date: 2024-03-24 22:05:47

> On Mar 23, 2024, at 19:37, Vinnie Falco via Boost <boost_at_[hidden]> wrote:
> 
> What
> makes this library better than others? Discussion is needed to determine
> this.

Let's make this stronger. It should be a requirement on the author to provide

1) a review of the state of the art in a given space
2) if there are computer-science algorithm(s) in use by the library, specify them and
    include links to academic papers which describe the algorithm
3) if not an algorithm, then is it a language extension-like proposal? Even in this case i feel that links to academic literature should be mandatory.
4) Is it an API to yet another software? Please explain if the proposal merely aims to save on typing boilerplate or does it do some computation as well?

Not all of these items apply to the proposed boost.sqlite, but the reason I post this is because
some recent submissions were completely unclear about what computational model or algorithm was being implemented. E.g. there clearly was 
an algorithm involved in boost.parser, but what was this algorithm? I could not figure out even after reading the entire documentation, 
though the impression I got (probably completely wrong) that it was "regex with callbacks".

Cheers,
Kostas