I use Lynx on a regular basis 1) to check the accessibility of my web designs and 2) when I break my X. And I sort of like it!
I also used it years ago to post (stupid) comments on this very blog…

After an update, I lost my graphical interface and I needed to use a browser to find something. So I used Lynx.
It really sucked!

Also if you want to see real design failure look at The Verge. It just blows up on links2/lynx.

Hello from consoleland! I am pretty surprised to see that this article is inspired by me. I’d just like to make few corrections:
1. I am using links2, not lynx.
2. I don’t use it in text mode but rather in framebuffer 1280×1024 graphics mode.
3. It is my main, not backup, machine and it isn’t that old. It is from 2000 and has 700MHz Pentium III and 64MB of RAM.
4. It can run X server but all the software that was usable under x11 works faster on framebuffer.
Also if you are wondering how I am commenting on this I compiled older version of links (2.1pre21) with javascript enabled and small patch to prevent your server from sending it gziped html. I works pretty well.

Re: HN, instead of using alt text, they should just use the proper unicode character of the “black up-pointing triangle”, which Lynx can actually show perfectly well, and it scales better too.

@ ths:
That’s not quite so true anymore: https://twitter.com/mattcutts/status/131425949597179904

Personally, while I made sure my personal landing page looked good on it (and it wasn’t hard, since it’s extremely simple and JS-free), I don’t remember the last time I used it. I usually just use SSH port fowarding and point Firefox to the local port.

I love lynx very much. my server machine is LFS without X, so when I’m doing something directly on the server I always use lynx or wget. lynx -dump or lynx -source are quite handy. Before I start writing a perl script with some HTML-module for parsing I simply use lynx -dump | grep (or awk or perl) to have a look at the HTML text without tags.

And all you SEO types out there: remember that GoogleBot does understand neither javascript nor flash or other crap, only pure HTML. So if you want to be found and cleanly indexed by all means make the page readable with lynx and you have success with Google automatically.

In emacs, I use emacs-w3m through pylookup addon, so Alt+Enter on any function gives a page from actual online html docs (ok, I’ve cached them locally now, for performance reasons, but I was browsing python.org at first).

So that’s an actual text browser accessing the internet not-just-for-lulz in the second decade of 21st century, though it’s probably not what most people use web browsers for.

I guess Lynx users don’t frequent sites like Facebook and Twitter to a great degree. As a Lynx user on a pure command line laptop I tend to frequent sites that have good info and navigational layout.

Results vary depending on the text mode web browser used too, try w3m and links to and see how sites render.

Nice article btw :)