On Fri, Sep 15, 2023 at 1:34 AM Sagar Acharya <sagaracharya@xxxxxxxxxxxx> wrote:
This sounds amazing. Are you saying that there is already an RFC for this or by your link, you're saying that one should have this system?
My question is is there a problem with existing softwares implementing an already public standard?
We are close to having a standard, the existing RFCs and .well-known conventions (almost) completely constrain the design space. What we lack is recognizing it as such.
Standards are all about making choices that don't matter, such as the choice of protocol identifier. The one area in which there is wiggle room today is that in theory, you can today register IANA protocol MINE giving you the SRV prefix _mine._tcp and then register .well-known tag OTHER so your URI is /.well-known/other.
I don't think it takes any particular insight to see that the convention should be that both tags are the same.
>If your Web Service benefits from caching, it is an information retrieval, not a Web Service.
This is profound, thank you for this. Exactly the case with cdns, why should one cache on cdns if their own server is online!!
