The Mudcat Café TM
Thread #159035   Message #3772173
Posted By: GUEST,DaveRo
12-Feb-16 - 05:10 AM
Thread Name: Tech: Mudcat browser tools
Subject: RE: Tech: Mudcat browser tools
History lesson, it's time to remember...

In the early days of the World Wide Web, the www referred to an actual webserver, a physical box. So a company owning the domain widget.com might have two servers - mail.widget.com for email and www.widget.com for its website. The www was only a convention, easy to remember for a public unused to internet addresses, but it became so familiar that a lot of people thought that it was a necessary part of a web address.

These days many sites do not need the www. Big websites work out what actual server to use based on the user's location, device, and the current traffic. Also the domain name is now being advertised as the name of the service: "Confused? Come to confused.com!" So the www is becoming obsolete, a historical relic. But the original server-name.domain-name system is still in use, especially on small sites or ones that have been around for years.

Some browsers did indeed add the www. - and .com on the end - maybe some still do. A few years back if you typed 'dog' into Firefox it would look up www.dog.com - which annoyed many people, and caused shysters to register thousands of mis-spelled domain names. These says Firefox and most other browsers have amalgamamated the address bar and the search bar so they will look up 'dog' in a search engine instead.

But even sites that don't need the www often add it themselves. (It may look as if the browser is adding it, but it's not.) I suspect that's because the 'www' is so ingrained that people type it anyway, or expect to see it: so confused.com displays here as www.confused.com. Firefox no longer displays the www part of a non-secure web address, which is an indication of its irrelevance.

I referred in a recent post to the browser not sharing data between domains, and the browser treats the www as part of the domain (technically it's an 'origin'). It's important that the browser does not share data and cookies between, say, shop.widgets.com and forums.widgets.com. So sites usually present a single domain - which is often www.something. But they could equally take the www off, or use something else.

mudcat works with or without the www and as far as I can see they both refer to the same site - at least for a guest. It doesn't add the www to the address it presents to the browser, and it mostly seems to stay in the domain - mudcat.org or www.mudcat.org - it starts in. But sometimes it doesn't - which is when my addon stops working. I doubt if they're using different physical servers - I believe the servers are etta.mudcat.org and awe.mudcat.org but I don't know how it assigns traffic to each.

When I wrote the scripts that became this addon I didn't even consider www - I've always used just mudcat.org
I've never taken any notice of Trace before, but now that I try it it takes me to my personal page. Is that what's supposed to happen? And the DeTrace links do exactly the same.
I don't know anything about Trace and DeTrace - as a guest I've never seen it.