Path: ...!news.mixmin.net!eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Ivan Shmakov <ivan@siamics.netNOSPAM.invalid>
Newsgroups: comp.os.linux.misc
Subject: good (old) free software
Date: Thu, 15 Jun 2023 20:57:08 -0000 (UTC)
Organization: A noiseless patient Spider
Lines: 220
Message-ID: <u6ftv3$h90a$1@dont-email.me>
References: <wFFdUuFN21aDDknIA@bongo-ra.co> <64865aea@news.ausics.net> <u66c54$310kc$1@dont-email.me> <6486bf9f@news.ausics.net> <u6bg30$3tvrt$1@dont-email.me> <957rljxosj.ln2@Telcontar.valinor>
Injection-Date: Thu, 15 Jun 2023 20:57:08 -0000 (UTC)
Injection-Info: dont-email.me; posting-host="04891589515d5924aa5d0d4f32ce95a7";
	logging-data="566282"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1+n4F59TF+J4D3CtaoQQQm7"
Cancel-Lock: sha1:+woXi//ievdkfmuMaC6S8V6P4OY=
License: CC0-1.0 (original contributions only)
Bytes: 11218

>>>>> On 2023-06-14, Computer Nerd Kev wrote:
>>>>> Ivan Shmakov <ivan@siamics.netnospam.invalid> wrote:
>>>>> On 2023-06-12, Computer Nerd Kev wrote:

 >>> I'm posting from Debian version 3 right now, so that makes sense
 >>> to me, but it did occur to me afterwards that the OP may have
 >>> meant an old but still supported Debian version.

 >> While I'm no stranger to running unmaintained software (or
 >> versions thereof) myself, I'm curious what could be the reason
 >> to run a no longer supported version of Debian specifically?
 >> (With i686 in User-Agent:, I'd venture to guess it's not a matter
 >> of having hardware no longer supported by Debian?)

 > It's set up how I like, runs old software I like without having to
 > patch it myself to make it build for later GCC or libraries, and
 > runs much faster (possibly also better, in terms of supported
 > drivers) on old computers that I still use.  Actually I don't think
 > current Debian would run on the PC I'm posting from now either
 > (that i686 has changed to an i586).

	To me, using software is largely means to an end; namely,
	software allows me to contribute to society.  Were I able to
	grow my own oats, my reasoning would likely be different, but
	as it is, I need to do something of value, so that society will
	give me food, clothing, shelter, and so on.  And of course I
	feel responsible for contributing back to free software projects
	that enable me to do my work, to learn new things (that may or
	may not become opportunities in the future: there's no way to
	know in advance), and to find inspiration in the work of others.

	Hence, from where I stand, I'd be inclined to ask myself, is
	the software I use /necessary/ in my choosen field of work?
	and if so, will the others benefit from it as well?

	I don't generally believe that the problems I solve are
	unique.  If some software is useful to me, then there're
	likely other people who will benefit from its availability.
	If the software I came to prefer is old and unmaintained, and
	not compatible with contemporary platforms, I will consider
	adopting it and updating it for such compatibility.

	At the same time, I will consider the lack of interest
	in software I use as a possible indication that there're
	alternatives I should consider learning and switching to.

	As to the examples...

	I've used GNU Emacs from c. 2002 to c. 2020.  When it grew too
	big for me to wrap my mind around, I've looked for possible
	alternatives, and before long switched back to Vim.  Should
	there be an indication of sufficient interest in maintaining
	GNU Emacs 24, I'd probably join the effort.  I don't find the
	idea of sticking to a version that probably has lots of by now
	known security issues particularly enticing.

	I've used Firefox from about the time it appeared until Quantum.
	I've used a number of plugins that the new APIs made impossible
	to implement (e. g., Classic Theme Restorer, Certificate Partol;
	I've also tried but never mastered Vimperator and Pentadactyl.)
	I /did/ note that there is a project dedicated to maintaining
	pre-Quantum APIs, but found the effort needed to become involved
	prohibitive.  As such, I've switched to Chromium for my second
	Web user agent.  (My preferred Web browser since about the time
	I've got Internet connectivity at home was and remains Lynx.)

	Alongside, I've found that I can use HTML and CSS for preparing
	printable documents.  Instead of having to care about Chromium
	/and/ LaTeX, I'm now using only Chromium (which I need for other
	purposes anyway.  See also news:u656kk$2pm6s$1@dont-email.me .)

	I've also found that if someone asks me to send them something
	or another "in Microsoft Word format," I can often send an HTML
	with <style /> instead and the person will be just as happy.
	This way, Chromium allows me to avoid having to deal with the
	likes of Libreoffice, too.

	The last version of Debian to support i586 was Buster, so when
	time came to retire a Squeeze install, I've unfortunately had
	to retire the i586 box that it ran on as well.  (The Jessie's
	version of Bash has particularly atrocious bugs, so by that
	time I've already decided against using Jessie there.)

	My intent is to revive the machine with a NetBSD install.  The
	'base' NetBSD system is developed rather conservatively and the
	system's x86-32 port claims to support "486DX or later."  There's
	some effort involved in learning pkgsrc, but it seems rather
	modest (as compared to becoming involved in UXP development.)

	There's a couple of no longer maintained packages that I rely
	on: Polipo HTTP proxy and hddtemp(8).  Frankly, I've found the
	reasoning given for removing the latter from Debian somewhat
	surprising (it can be read as suggesting that it's in the best
	interest of system stability to have an in-kernel implementation
	of a feature rather than one that needs root; like, really?)
	Nevertheless, I intend to update my scripts to rely on /sys or
	(and) smartd(8) .state files instead.  Eventually.

	I'm considering taking over Polipo maintenance.

[...]

 > OK, fair enough, although using virtual LANs on a supported router
 > might make keeping separate LANs easier than you think.

	My understanding was that VLANs are a feature of switches.
	(Not that it's impossible for a router to implement a switch
	/ bridge function as well.)

	Either way, so far I've been requiring strong cryptography
	for nearly everything but accessing 'public' resources on my
	servers, which once again takes the (mental) 'LAN vs. Internet'
	specialcasing out of consideration.  Thankfully.

 >> And of course I use Rsync over SSH extensively, be that for backups
 >> or for pushing new versions of ~/.bashrc et al. from my primary box
 >> to every other *nix home directory I have.

 >> I suppose with some 'necessary work' I can do the things above with
 >> Telnet as well, but I'd think that by that point, resurrecting RSH
 >> would be a more straightforward solution.

 > Sure, I don't try to use Telnet for anything but terminal access.
 > But the other tools without encryption do the same job with much
 > less to go wrong.  You've got RSH, also Rexec, FTP, Rsync (without
 > SSH), etc.

	There's a difference between Rsync over TCP vs. Rsync over
	remote shell, though.  In the latter case, remote shell allows
	access to arbitrary accounts (subject to access restrictions.)
	In addition to my 'user' account, I also run Rsync over SSH
	as 'backup' and 'root'.

 > The latest SSH annoyance I've had is a system I set up with current
 > software only three years ago now needing a redesign because of the
 > present switch from SCP to SFTP.  If I'd used RCP, or more likely
 > FTP, instead, no problem.

	I've pretty much always used Rsync over SSH to copy files around.
	Sure, Rsync is limited to Unix-like systems, but that's what
	I'm using for any serious work for over two decades now anyway.

	I recall the issue with scp(1) and while I find it unfortunate,
	I personally wasn't affected in the slightest.

 >>> (especially because SSH isn't very helpful with its error
 >>> messages, and old versions don't support the -Q option).

 >> Well, cannot quite argue with that.  If anything, I haven't yet
 >> figured out how to connect to my OpenSSH instances with SSH2DOS.

 > Yes well I'd definately look for a Telnet client there, but I never
 > have used networking in DOS.

	It's relatively straightforward, so long as you keep in mind
	that while a packet driver provides a standardized ABI to send
	and receive Ethernet frames, implementing any protocol on top
	of that is a responsibility of the application.

	There're (static) libraries that implement TCP/IP (such as
	the "classic" Waterloo AKA WATTCP, and the much newer mTCP,
	http://www.brutman.com/mTCP/ ), but you need to make sure
	that if you use applications that rely on different libraries,
	you configure /all/ such libraries.  And you'll probably want
	to keep those configurations in sync, too.  Like:

### MTCP.CFG
PACKETINT 0x7E
IPADDR 192.0.2.34
NETMASK 255.255.255.0
NAMESERVER 192.0.2.53

### WATTCP.CFG
pkt.vector = 0x7e
my_ip = 192.0.2.34
netmask = 255.255.255.0
gateway = 0.0.0.0
nameserver = 192.0.2.53

	That said, while the ability to run an SSH client on FreeDOS
	would've been handy occasionally, it's hardly a necessity.
	For me, Telnet is no substitute for SSH, and if I can't run
	an SSH client there, then so be it.

========== REMAINDER OF ARTICLE TRUNCATED ==========