On Tue, 28 Aug 2018 06:54:06 -0400, Paul wrote:

Nil wrote:
On 26 Aug 2018, Peter Jason wrote in
alt.comp.os.windows-10:

The domain ".website" seems to hold ads & junk.

That is a top-level domain designation, not a domain. There are many
domains with that extension, not just one. Their various contents could
include anything, not just "ads & junk".


This article will put you off to sleep on the topic
of TLDs, but I did like them naming the worst one.

https://en.wikipedia.org/wiki/Generic_top-level_domain

"In 2018, Spamhaus rated .men as the worst top-level domain
in terms of spam and scamming. .men comes top with 60.6 per cent
of its 73,000 domains identified as "bad", resulting in a
badness index of 6.48.

The company that runs .men, Famous Four Media also runs the
third worst registry - .loan – with 59 per cent bad domains
and a 6.22 index.
"

https://en.wikipedia.org/wiki/Domain_Name_System

"For example, the domain name www.example.com

translates to the addresses 93.184.216.34 (IPv4)"

I copied those two, to show the difference between a TLD
and a "domain name". I wasn't really aware of how dangerously
similar the terminology is.

The diagram in the DNS article, shows some of the steps.

https://upload.wikimedia.org/wikiped...S_resolver.svg

An organization can run their own nameserver. For example,
Wikipedia can have a wikipedia.org nameserver, and if
a query like "cucumber.wikipedia.org" came in, it could
be that nameserver which answers the query. And names
like that, also allow an IP number to be shared

www.wikipedia.org 22.33.44.55
cucumber.wikipedia.org 22.33.44.55

A server on 22.33.44.55 can look at the symbolic
name and figure out what "service" the user was looking
for. If I do this

http://22.33.44.55

I wouldn't get a reasonable answer. Whereas if I do this

http://cucumber.wikipedia.org

then equipment in the Wikipedia building can figure
out which server I want ("cucumber"), even though the same numeric
IP address happens to be associated with it. And I only
continue getting reasonable answers, as long as I use
symbolic addresses.

http://cucumber.wikipedia.org/index.htm

I only wanted to mention that, so people could understand
where "brokenness" comes from sometimes. It's not always
"Internet equipment" which is broken, and the broke part
can be inside a business.

FYI, there's nothing broken about your example above. All of that is
intentional and working as it was designed. The days of being limited to
one site per IP address ended nearly 30 years ago. Also, you can access
any site by its IP address, even sites using the shared hosting model,
if you insert the proper headers so that the destination server can
determine which site you're looking for, but web browsers won't do that
for you automatically, so it's easier to use the name versus the IP. By
using a fully qualified domain name (FQDN) as the destination, your
browser will know how to create and insert the headers referred to
above. Bottom line, it's those headers that do the magic, not the fact
that you requested a site by its FQDN. The destination site doesn't see
what you typed into your browser.