TL;DR, for the tip:

If you can't or don't want to install a tool only for checking up
gopher, you can curl.

curl gopher://gopher.black/1/phlog/20220316-sensory-deprivation

(chances are that you have curl on your machine)

Sure, you won't be able to "navigate" the gopherhole easily, but you can
at least read the page on the link.

Some more ideas to improve readability, you can cut the first character,
which is the item type.

curl ... | cut -c2-

If it's a long page to read, you might want to use awk or sed to replace
the first character with a tab instead of only cutting it.

curl ... | awk '{gsub(/^i/, "\t")}1'
curl ... | sed 's/^i/\t/g'

You might want to use `^.` instead of `^i` if you want to replace all
the item types.

I've said that you can't navigate `easily`, but if you have history in
your shell (who doesn't?), use your uparrow/^p and change the link for
the next page you want to visit.

---

So, I'm a big fan of technology minimalism.

Ok, I just made up this, what I mean is, as much as possible, I try to
use builting tools instead of installing extra software.

This have the advantage of learning the Unix tools better, as it forces
me to look if I can do something else with the tools I already have, so
I'm constantly knowing more and more tips and tricks on how to use the
shell without the need to install extra software.

Also, whenever I need to change computer, or format it, I don't have a
long time setting up anything, as everything (or almost) I use is right
on there on a fresh installation.

No dotfiles.

I confess that sometimes I overdo it, some time ago I was using ed(1) frenetically for writing software. For some reason I felt really
productive with it, but now I'm back to a more robust text editor. haha

Anyways, I don't know why I never thought about using curl before.
I mean... right?

So much trouble looking for software y or x just to have a little bit of gopher.

Now I'll be thinking, how to use NNTP with Unix builtin tools. lol
--- Synchronet 3.19a-Linux NewsLink 1.113

On Thu, 14 Apr 2022, Jack wrote:

So, I'm a big fan of technology minimalism.

Ok, I just made up this

You didn't made that up, or at least you're not the only one that did.
As a related tangent, there is a book (which I have read):

Cal Newport, "Digital Minimalism - On Living Better with Less Technology"; Penguin Publishing, 2019; ISBN 978-0-241-34113-1.

For me, it means a use of technology in by-need basis,
while also being selective about which technology to use
that fits one's circumstances while having minimal resource,
technical requirements, and inflict less negative effect
on ourselves and the society.

This also includes a decision of using low-tech or no-tech solution
over high-tech ones when the circumstance is appropriate.

Anyways, I don't know why I never thought about using curl before.
I mean... right?

Well, I use cURL with `gopher://` occasionally, when I have to play
A/V media or livestream (!) on that protocol, as media players on my setup don't usually support them. [1]

Now I'll be thinking, how to use NNTP with Unix builtin tools. lol

I occasionally do this on Tildeverse Netnews for debugging reasons.
But when I actually read the posts, I often just use `less`--
which was the exact tool I used when I first read your post I'm replying to; but this is only applicable since I logged in to Tilde.club
to read it from site-local spool.

When it's NNTP, Netcat (any variant, or functionally-similar program
like Ncat or Socat) will do- as long as you know some basic formalities
of the NNTP protocol [2] and the USENET/Netnews message format. [3]
Following though reply threads manually can be a bit tough however.

(cURL, as far as I have read, doesn't suport `nntp://` and `news:`)

But personally, since I normally post from within Tilde.club
(as opposed to using NNTP to post from somewhere out there),
I simply run `inews -hS` there, paste/redirect the drafted message in,
and kthxbye. [4] It is also the way I posted this very reply.

I confess that sometimes I overdo it, some time ago I was using ed(1) frenetically for writing software. For some reason I felt really
productive with it, but now I'm back to a more robust text editor. haha

Well, I have become reasonably proficient in using GNU Ed
from a journey of exploring ways to use GNU/Linux (or Unix in general)
with a dumb glass/printout terminal that has only pure ASCII
but not ANSI/VT100-compatible control codes.

(I had also learned things like GNU Ed, Sed, and AWK in an offline way,
thanks to their GNU TexInfo manuals- which are full-book user guides
available on the go, and don't require Internet connection
or anything graphical to view)

I still use Ed semi-regularly for editing configuration files
and for hammering very laser-focused edits on huge mostly-text file
that I can't afford having text editor chew off not-text bytes inside
(like GNU RCS archive), but I had not gone as far
as using it for actual programming.

Side note: Ed is pretty much the original reason behind Kernighan's
"one sentence per line" advice [5]; which I also apply it today
in many kinds of text markup I wrote: including HTML, DocBook XML,
Troff-Man, and _original_ Markdown. [6] Though the main reason
that me and other people do this nowadays is that it made
version control and diff'ing much easier with stock tools.

To think of it, this is another prime example of changing
a bit of (life)style to made mediocre low tech super effective.

Regards,
~xwindows

-----

[1] In the process, I have discovered that FFplay and MPV
support `gopher://` out of the box.

[2] RFC 3977: Network News Transfer Protocol specification (2006 edition)
<https://tools.ietf.org/html/rfc3977>

When posting, it could be as simple as connecting to TCP port 119,
and type...

MODE READER
POST

Keep an eye on the status reply on each command. If they're all good,
then you paste your drafted message with full USENET header in [3];
and terminate the message with a line that has only dot (`.`).
And finally, after you finished the deeds, issue:

QUIT

to terminate the NNTP session.

[3] RFC 1036: USENET message format specification (1987 version)
<https://tools.ietf.org/html/rfc1036>
RFC 5536: USENET message format specification (2009 modern variant)
<https://tools.ietf.org/html/rfc5536>

Spoiler: it's like Email, but with comma-separated `Newsgroups:`
header value instead of `To:`, `References:` being the main
thread-tracking method; and lines that started with dot
have to be escaped by prefixing another dot.

[4] It's mostly the same as when you (or your email client)
run `sendmail -t` on shared Unix-like system to send an email.
Both `sendmail` and `inews` will automatically stamp `Date`
and `Message-ID` header for you too, if you did't already provide one.

[5] See <https://rhodesmill.org/brandon/2012/one-sentence-per-line/>

But this technique actually predated Unix for very long:
apparently it was invented by Buckminster Fuller in circa 1930,
which he called it "ventilated prose"; as a way to make dense research
report more readable.
<https://vanemden.wordpress.com/2009/01/01/ventilated-prose/>

[6] I emphasized original Markdown, because many other Markdown variants
eschewed this feature.
--
xwindows' gallery of freely-licensed artworks
https://tilde.club/~xwindows/ http://tilde.club/~xwindows/ gopher://tilde.club/1/~xwindows/
--- Synchronet 3.19a-Linux NewsLink 1.113