[Discuss] A Web Server is No Longer a Web Server

[markw at mohawksoft.com (markw at mohawksoft.com)]

> On Tue, 11 Nov 2025 08:23:21 -0500
> markw at mohawksoft.com wrote:
>
>> Http was originally a document format and a file transfer protocol.
>> The idea that it could be a dynamic page generation engine and
>> eventually a working RPC mechanism (SOAP, REST, etc) came afterward.

I was going by this:
https://www.w3.org/Protocols/HTTP/AsImplemented.html

HTTP and HTML were kind of two parts of the same idea.

"Dynamic pages, SOAP, REST and friends are not HTTP. They are content
rendered by the browser"

This isn't exactly true, I understand we are more or less dealing with
definitions so we may not actually be debating.

Dynamic pages, and RPC mechanisms actually have nothing to do with the
browsers. I'm sure I can find more of the history, but off the top of my
head I don't have names and dates, someone had the brilliant idea of
intercepting the "file open" logic of an HTTP server and passing that to a
spawned process and creating pipe to the standard output (The old CGI
interface) to produce content.

Then they embedded processing engines like ASP and PHP into the HTTP
servers for efficiency.

Then, some time in the mid to late 90s someone made a HTTP compatible RPC
format called XML. It was to convey the data and the local reader could
interpret how to display it. Over the years, style sheets, cascading style
sheets, and so on.

Then prior to the "SOAP" spec, devs started using HTTP and dynamic handler
modules to make rudimentary RPC mechanisms. I know I did this at a startup
in 1998. Then SOAP became a thing, I think it was really a bad/ugly spec.
but that often happens.

The exact steps between CERN's first HTTP server and what we have today is
very blurry with lots of different standards floating around.

>
> I think you're conflating HTML, the document format, with HTTP, the
> transport protocol. HTML has changed drastically over 4 major versions,
> but HTTP remains largely the same. The changes from v0.9/1.0 to v2 to
> v3 mostly are protocol optimization (v2) and use of QUIC instead of TCP
> (v3). HTTP is content agnostic, and it addresses common problems with
> earlier protocols like FTP and Gopher, which are major factors why it's
> become ubiquitous.
>
> Dynamic pages, SOAP, REST and friends are not HTTP. They are content
> rendered by the browser, which may be HTML or may be some other
> mechanism such as WebAssembly, or handled by an application server.
>
> --
> \m/ (--) \m/
> _______________________________________________
> Discuss mailing list
> Discuss at lists.blu.org
> https://lists.blu.org/mailman/listinfo/discuss
>