IN DEFENCE OF UMN GOPHER
There's been discussion amongst some phloggers about how strictly
Gopher content should adheare to the Gopher standard in the modern
age. With these things I generally prefer to practice than preach.
It seems a bit of a silly discussion to me really, but perhaps I'm
being ignorant. Who'd have thought originally that the Web would
turn partly into a virtualised application platform with HTML just
used as a sort of launcher? Or that the browsing of its more modern
content would demand many multiples of the computing power that I
personally require for any other personal computing task? Tacking
on functionality is the path that it took, and my own stubborn
attempts to quitely practice otherwise with websites I made sure
didn't change any of that.
But anyway for now I'll leave that discussion be. What I did pick
up on was a bit of "UMN Gopher? Who'd use that old thing anyway?"
sentiement in response to IanJ's original post which pointed out
issues browsing some Gopher holes with it. I do mainly use UMN
Gopher, and I'm not alone according to IanJ's poll here:
gopher://gopher.icu/1/poll?poll=1707390454
Maybe it's just because I'm a stick-in-the-mud using a PC from the
mid 90s for my recreational internet browsing (which is actually
possible on its own for all my recreational activities _except_
browsing the modern Web and watching videos - even HTML and TLS in
email can actually be coped with by a Pentium 1), but I think it
has some worthy features besides just being the first Gopher
client. Navigation in UMN Gopher is quite similar to Tin, which I
use for Usenet. It's easy to navigate around using just the arrow
keys, especially _because_ most text content isn't in gophermaps so
menu options are more concisely listed and don't require bouncing
all around multiple screen-fulls of text which is what puts me off
keyboard-orientated navigation of Web page links in browsers like
Lynx. As in Tin, the line numbers assigned to each menu option also
allow quickly jumping down a long list of options simply by typing
in the assigned number.
Other key assignments are admittedly wacky, and sometimes
contradict the displayed instructions. I wont deny that UMN Gopher
clearly suffers from being typical university-developed software
where too many people were trying to get their own great ideas into
it. But I think the text file viewer with its display of file size
and percent of page position is very well suited to the job.
Launchers for other file types can also be configured quite
flexibly, although with a pretty wacky syntax.
On the down side it does hide some selectors that it doesn't
understand, such as 'd' for documents (PDF/Postscript). Also for
browsing Git repos through GophHub I prefer to use a Gopher plug-in
for Dillo because I like the HTML syntax-highlighting and Markdown
rendering and UMN Gopher doesn't pass Gopher-downloaded HTML over
to a web browser for viewing, just the gopher:// URL for it to open
(or, more likely now, not). The tabs in Dillo are usually pretty
handy for browsing source code via GophHub too.
So no it's not ideal, but it's good enough for me, certainly
lightweight enough to run on my old PCs, and better in some
respects than any of the alternatives I've tried. I think there's
good cause for me to use UMN Gopher, not just pretending it's the
best simply because it was the first. Also it is still being
maintained to the extent of still compiling for modern Linux with
its Debian package, which is as much maintenance as I expect for
such software anyway:
https://tracker.debian.org/pkg/gopher
Actually a lot of the software I use is in a similar sort of state
(the text editer I'm writing this with, for example), and I'm
perfectly fine with that too. The most annoying thing about the
internet is that it allows other people to make their dislike of
software I'm using _my_ problem, because they eventually find
excuses for excluding it. On the phlog-posts-as-gophermaps
navigation issue in UMN Gopher though, there is an easy
work-around: Simply hit Shift-S while viewing the gophermap and
save it to a (rendered) text file, then switch to another terminal
and open that file using less, deleting it when you're finished.
It's not so hard for the few times that it's required at the
moment, though I would certainly get tired of doing it everywhere.
- The Free Thinker
Prior discussion:
gopher://gopher.icu/0/phlog/Computing/The-state-of-gopher.md
gopher://gopher.unixlore.net/0/glog/response-gopher-icu-state-of-gopher.md
gopher://gopher.black/1/phlog/20240205-re-the-state-of-gopher
gopher://zaibatsu.circumlunar.space/0/~tfurrows/phlog/2024-02-07_reStateofGopher.txt
gopher://sdf.org/0/users/gallowsgryph/phlog/2024-02-08_state_of_gopher.txt
gopher://gopher.icu/0/phlog/Computing/Responses-to-The-state-of-gopher.md