[unpythonic]

When perl came out we were living in horrific times. You had the choice of either Bourne, C or Korn shell. Automation was glued together in one of these with a series of grep, awk, sed, ls, test, commands glued together. Anything more complicated was written in C and called from one of these things.

Perl in one stroke collapsed the programming of C, text manipulation, the capabilities of all of the Unix utilities, and data structures into one system. For anything which wasn't subsumed into the monolith of Perl, you could easy access via backticks. It was very friendly in dealing with text streams, and that's what those call-outs in those back ticks spoke.

Yes, awk and sed were replaced by Perl, but more importantly, the unmaintainable nightmare that glued all of it together was wiped out.

[imiric]

> Automation was glued together in one of these with a series of grep, awk, sed, ls, test, commands glued together. Anything more complicated was written in C and called from one of these things.

This doesn't sound that horrific to me. It's the classic Unix approach of building small tools that do one thing well, and composing them in novel ways to solve problems. For any problem that can't be solved this way you write another small tool using your programming language of choice. Rinse and repeat.

But occasionally Unix attracts users and programmers who reject this approach, and who prefer building a monolithic tool, or in the case of Larry Wall, new programming languages. To be clear, I'm a fan of Perl and think it has its place, especially in the era it came out. It inspired many modern languages, and its impact is undeniable.

Personally, I find solutions you refer to as "unmaintainable nightmare" to be simple and elegant, if used correctly. No, you probably shouldn't abuse shell scripts to build a complex system, and beyond a certain level of complexity, a programming language is the better tool. But for most simple processing pipelines, Unix tools are perfectly capable and can be used to build maintainable solutions.

The classic Knuth-Mcllroy bout[1] comes to mind. Would you rather maintain Knuth's solution or Mcllroy's?

[1]: https://matt-rickard.com/instinct-and-culture

[Scubabear68]

I don’t think you’ve seen the kind of scripts the person you are responding to is talking about.

I have, and mentioned one lower down in the comments. Unix philosophy was great but does not scale well in terms of maintainability or efficiency. Invoking processes over and over again loops is godawful slow. And the horror of complicated shell scripts is legendary.

[sigg3]

As a self-taught coder, I've experienced many times how highly skilled software engineers groan and sweat when they encounter shell scripts. I don't understand why, but it seems like people with a CS background are never really taught shell scripts and have come to irrationally fear them. It's sort of taboo.

This results in weird behavior, such as writing a groovy (Java?) script for Jenkins to execute bazel in order to build a go binary that runs the very same commands in an exec.Command() construct. Or people who download and import pandas to grab the third field in a csv file.

During the course of learning, I've naturally written code in bash that should have been written in another language. I replaced if statements with case because they turned out to be more performant. It's a great learning experience and why I got into python and go.

IMO we should use the right tool for the job. Sometimes that tool is a combination of unix utilities that you can put in a shell script for easier maintenance. It's just procedural execution of (usually very efficient) binaries, akin to a jenkins script or gitlab pipeline. Just mind the exceptions and use exit codes.

[Scene_Cast2]

I'm the type of person that would grab pandas to parse a CSV. Here's my reasoning

* often times, it's not just the third column I want. Sometimes it becomes "third column unless the first column is 'b' then instead grab the fourth column". Having a good data representation makes sure that I'm not mixing logic code with representation parsing code

* I don't have to care about CSV parsing edge cases. Escaped comma? Quotes? I don't care, the library will either handle it or throw an explicit error. With custom parsing code, instead of an error, I'll get some mangled result in the middle of the file that I won't even catch / notice until later down the line

* when working with CSVs, in my area (ML / scientific compute), Python is often the right context to be in.

[stanleydrew]

> I've experienced many times how highly skilled software engineers groan and sweat when they encounter shell scripts. I don't understand why, but it seems like people with a CS background are never really taught shell scripts and have come to irrationally fear them. It's sort of taboo.

It's not a lack of being "taught" shell scripts. It's the fact that shell programming constructs aren't well documented, your "standard library" is basically dependent on whatever binaries happen to be available on the filesystem, error handling is almost non-existent, etc.

It's very easy to write a bad shell script that "solves" a problem as long as a bunch of assumptions aren't violated. In my experience, senior software engineers are extremely averse to hidden assumptions and very concerned with reliability of the systems they build.

[pixl97]

Yea... "Well the script works fine in MY SYSTEM" was the most common issue with said scripts. Running across different versions of Linux was fraught with issues, much less any other operating system that could execute a shell script.

Of course this can happen with any language, especially as it ages and adds complexity.

[imiric]

I certainly have, and might have written a few of those myself.

But this doesn't make this approach inherently wrong or obsolete. The programmer is wrong for trying to use the tools beyond their capabilities. Where that line is drawn is subjective, as is the concept of maintainability, but if you feel that you're struggling to accomplish something, and that it's becoming a chore to maintain, the path forward is choosing a more capable tool, like a programming language.

[Scubabear68]

I think we are actually mostly in agreement there then.

Perl was invented because the gap from shell to more capable languages was (and is) really big. Languages like Python and Ruby didn’t exist yet, and Perl had a really, really strong sweet spot in text processing.

[ryl00]

> Perl had a really, really strong sweet spot in text processing.

Still does.

[cutler]

Since Ruby took the best bits of Perl what advantage does Perl retain?

[moomin]

Yeah, the actual experience of leaky abstractions and non-portable code is forgotten. Perl solved a very real problem in the 90s. Grief, I shudder to think back to the sheer complexity of my bashrc file back then.

[beebeepka]

Yes, exactly. I have seen entire backend systems written in bash. Everything was shell script, sed and awk. The owner didn't want python or perl because he only knew bash and the related tools.

Everything was needlessly hard because these tools were not built for that. Easy to talk about philosophy and the "classic Unix approach" if you don't have to build modern applications this way.

[unpythonic]

Horror is apt.

You hit it on the head with the slowness of loops when the body comprises a series of program invocations. The horror really seeps in when you realize the original author wasn't stopped by the lack of data structures: they could get around that with some creative variable names.

[mulmen]

Programming environments including shells and operating systems are just tools. And every tool can been misused. I opened a can of beans with a screwdriver once. Reality is messy. That doesn’t make the tool bad.

[stubish]

At the time, before everything became Linux, all these tools and the shells used to glue them together were an incoherent mess. Was your glue sh, ksh, csh, tcsh, bash or something uncommon like zsh? Did your grep, awk and sed use the same regexp syntax as your text editor? Single letter command line options, all meaning something different to each tool. Dozens of domain specific languages (shells, awk, sed etc.) meant dozens to learn and keep in your head. And you needed it in your head, because finding the information you needed in the massive single document man file reference was a pain because hypertext links had not been invented yet (well, probably in Emacs, which was another tool like Perl that people used to avoid the command line nightmare).

[chx]

> or something uncommon like zsh

By the time our lord and savior zsh appeared on the scene Perl was already at Perl 3. And, to be fair, I do not think many used zsh before 2.1 which was some time 1991 fall and by then the Camel Book was out for half a year or something like that. So the pre-Perl and the we-use-zsh days do not really overlap.

When Perl appeared the absolute hotness was the version of Korn shell what later became known as ksh88. https://github.com/weiss/original-bsd/tree/master/local/tool...

[timthorn]

> And you needed it in your head, because finding the information you needed in the massive single document man file reference was a pain

That's what the O'Reilly books were for, especially the Nutshell series.

[jghn]

I remember capturing every password at my university via "methods". Because we had a printer quota. In the summer when everyone was gone I printed out all the man pages (all the mans, the system libraries, etc) so I'd have a nice reference book. I made sure to make it so no one was charged any money.

The one thing people can't possibly fathom if they started coding after the mid-late 90s was how much we relied on the printed medium.

[chx]

I still remember when we measured the documentation IBM shipped with the mainframes not in pages but in yards it occupied on the shelves. It was a lot.

[cafard]

Oracle used to be hard on your lower back, within the last 25 years.

[fuzztester]

The Unleashed and Bible series of books come to mind. Waite group or Sams or so, publishers.

Rather Waite-y.

[jancsika]

Honest question-- why weren't the tools glued together or perhaps replaced entirely with the lisp inside Emacs? What was it missing?

[turndown]

While there are many possible answers, it eventually boils down to Unix being a C runtime and thus has a C culture. Lisp is from outside of this section of the world, so it simply had less adoption and support inside Unix land. Other languages, like sed, awk, and shell are not C but share its heritage(essentially, they were made by people close to the making of C.)

[shrubble]

Memory and CPU efficiency, since most systems were memory constrained. My first server had 128MB of RAM...

[chx]

> My first server had 128MB of RAM

Whippersnappers! :D

The first big iron I had the luck to work with was an IBM 3090 , essentially a gift from IBM, it handled the university entrance exams of the entire country of some ten million people and it had 64 MB of RAM. (It was also the first computer in Hungary permanently connected to the Internet via a leased line to Austria so it had an Austrian IP address. Hungary didn't have its IP region for two more years.)

I think the first machine with 128MB was a VAX 6510 a year or two later at another university. A little bit later, in 1994, CERN had gifted a VAX 9000 with an astounding 256MB of RAM.

To compare, the first server I installed Linux on had a grand total of 4MB RAM -- and that was one of the largest computers a small department at the university had.

It would be a long, long time before "128MB" and "mine" entered the same sentence.

[raiph]

Shades of the Monty Python sketch here, but the following is true...

4MB?!?

My first encounter with IBM kit was a, er, darn I'm not sure cuz I'm getting old, but I think it was a 4300? Not big iron in some senses, but still with a box that was something like 6-8 feet long iirc and definitely several feet wide and high. (And a bank of about 6-8 tape decks, each as tall as me, and two disk units, each the size of a washing machine, and so on.)

Its RAM? A massive 1 MB.

That IBM kit was the heart of the super new expensive upgrade in 1980 that cost something like 5-10 million pounds iirc to build, including a brand new building to house it and a team of programmers.

The older setup, which is where I was until its last days, was an ICL system that was expanded at the end of its life to a whopping 48KB -- yes, KB -- of RAM.

And that kit ran all the systems, internal (payroll, accounting, etc., etc.) and external (sales etc.) for the largest car dealership in the UK.

128MB? 4MB? Even 1MB? That was an unimaginably insanely large amount of RAM!

(Yes, it was very weird to be working with this physically enormous setup, and dealing with keeping it all cool enough not to halt for a half hour or so, through super human efforts when the A/C broke down, when the likes of PETs, Sinclair Z80s, and Acorn Atoms were a thing...)

[shaftoe]

There's an old joke that emacs stands for Eight Megs And Constantly Swapping.

[fuzztester]

Eventually Munches All Computer Memory.

Other backronyms too.

[cpach]

Elisp is very much a niche language. For whatever reasons, the use of Elisp outside of Emacs is basically non-existent. Elisp is quite clunky, and AFAIK there hasn’t really been any big efforts to make it usable outside of Emacs. People who wanted Lisp outside of Emacs already had Common Lisp. (And Chez Scheme, and Scheme 48, etc etc.)

[em-bee]

you probably shouldn't abuse shell scripts to build a complex system, and beyond a certain level of complexity, a programming language is the better tool

but the only free programming languages available at the time were C/C++, various shells, and awk. everything else was expensive or not generally usable for other reasons. all the really useful languages to build complex systems didn't really appear or become freely available until the 90s. and perl was first among those.

[imiric]

I'm not saying that Perl didn't have its time and place. It certainly fulfilled a need at the time for a language more capable than shell scripts, but less cumbersome than C/C++.

But the thing is that today the shell landscape is much more mature for solving simple problems, and we have C/C++ alternatives that are saner and more capable than Perl (e.g. Go). So it arguably has lost its place, as shell tools are still in widespread use, while Perl is mostly underused. Raku is interesting, but it goes in a different direction, and its adoption is practically zero.

[lolive]

Python killed the game.

[cutler]

Unfortunately. Ruby should have been Perl's natural successor. Python is the VHS of scripting languages. For a start it doesn't have a decent answer to Perl or Ruby's one-liners. Then there's the crippled lambda implementation. Python is a sad case of worse is better.

[tharne]

> For a start it doesn't have a decent answer to Perl or Ruby's one-liners.

This is by design. Readability is core to the design and philosophy of python. One liners are cool and fun to write, but trying to decipher someone else's incredibly dense bash or perl one-liner is absolutely awful.

[j16sdiz]

> This doesn't sound that horrific to me. It's the classic Unix approach of building small tools that do one thing well, and composing them in novel ways to solve problems.

This works really well if your problem can be solved in one or two liners.

It go bad very quickly when, say, you have two CSV files and want to join them the sql-way. In sed, you have to use positional variables and think about shell escaping. In perl, you can at least name those variables and use \Q

[imiric]

> This works really well if your problem can be solved in one or two liners.

My personal comfort threshold is around the 100-line mark. It's even possible to write maintainable shell scripts up to 500 lines, but it mostly depends on the problem you're trying to solve, and the discipline of the programmer to follow best practices (use sane defaults, ShellCheck, etc.).

> It go bad very quickly when, say, you have two CSV files and want to join them the sql-way.

In that case we're talking about structured data, and, yeah, Perl or Python would be easier to work with. That said, depending on the complexity of the CSV, you can still go a long way with plain Bash with IFS/read(1) or tr(1) to split CSV columns. This wouldn't be very robust, but there are tools that handle CSV specifically[1], which can be composed in a shell script just fine.

So it's always a balancing act of being productive quickly with a shell script, or reaching out for a programming language once the tools aren't a good fit, or maintenance becomes an issue.

[1]: https://miller.readthedocs.io/

[hnlmorg]

You’re discussing modern tooling in a conversation about early UNIX tooling. Back in the period being discussed, even ‘read’ was less functional. Ksh introduced a lot of the stuff we now take for granted, some of which wasn’t even available until the Ksh93 (long after Perl was released). Bash itself is a younger project than Perl. Albeit not by much.

[imiric]

Fair point. I'm not arguing that Perl wasn't an improvement back then, but that the approach of composing Unix tools is not inherently bad. And as the shell ecosystem evolved since then, and more capable programming languages appeared, Perl has been left by the wayside as a historical relic, rather than the replacement of Unix tools that Wall envisioned.

So I don't disagree that it was needed back then, but it's important to mention the modern context it struggles to exist in.

[hnlmorg]

Perl is still a commonly used tool chain. It is far from being a “historic relic”.

I agree that there’s nothing wrong with composing UNIX tools. I mean, that was one of its key selling points. if you watch any early promo videos for UNIX you’ll see them talk heavily about the composability of the command line and shell scripting. It wasn’t an accident — it was designed that way.

The point of the conversation wasn’t to say that one shouldn’t write shell scripts, it was just to say that there was a massive and unfilled gulf between what was easy to do in Ksh, awk and sed, and what could be done in C.

[mulmen]

> It go bad very quickly when, say, you have two CSV files and want to join them the sql-way.

Then just put them in a database and write a simple SQL query. If you use Perl it’s really very simple to do.

[yesbabyyes]

> It go bad very quickly when, say, you have two CSV files and want to join them the sql-way.

That sounds like a perfect use case for `join`.

[NoMoreNicksLeft]

> This doesn't sound that horrific to me.

I think it's possible that things that seem normal and inoffensive can become horrific simply from scale. You'll climb the stepladder without complaint, but then there's that radio tower in Canada...

Others like the idea of the family cow, then they see the 10,000 head feedlot from the highway.

Scale is sometimes sufficient by itself to induce horror.

[aidenn0]

csh was a decent interactive tool, but not great for scripting. Bourne shell had the right idea but there were so many bugs in various corners of it (I still sometimes end up writing "test "x$foo" = "xbar" even though shells that need that are long gone).

If you can depend on a recent bash and use shellcheck, then it's actually quite a pleasant programming environment, with fewer footguns than one might think. (I want a @#$@# "set -e" equivalent that returns non-zero from a function if any statement in the function results in non-zero).

There are some things that are more awkward than they should be though (e.g. given a glob, does it match 0, 1, or many files, or the way array expansions work).

Also, there's no builtin way to manage libraries (I don't know about Perl, but Python suffers from this as well). This results in me pasting a few dozen lines of shell at the top of any of my significant shell scripts, for quality-of-life functions. Then I have to use "command -v" to check if the various external programs I'm going to use are present. Say what you will about C, but a statically-linked C program can be dropped in anywhere.

[imiric]

Mostly agree. The modern shell scripting environment is much more robust than 30 years ago, with ShellCheck and some sane defaults, as you say. I also find it pleasant, once you get over some of its quirks.

As for managing libraries, that's true, but you can certainly import and reuse some common util functions.

For example, this is at the top of most of my scripts:

    set -eEuxo pipefail

    _scriptdir="$(dirname "$(readlink -f "${BASH_SOURCE[0]}")")"
    source "${_scriptdir}/lib.sh"

This loads `lib.sh` from a common directory where my shell scripts live, which has some logging and error handling functions, so it cuts down on repetition just like a programming language would.

[orra]

I haven't seen -E (aka -o errtrace) before, but it looks like a useful addition to the standard-ish -euxo pipefail.

Moreover, it's existence being required explains why my error handling, recently, wasn't working as expected.

[unpythonic]

The effortless composition of complex commands out of simple standalone programs is one of the best features of Unix. And yes, I admire and love it as well.

That said, imagine a metrics system for a huge networking company that used these methods to cover all automated testing or defect analysis. Those inner loops were made of greps and seds and so forth, and each one is the invocation of a new program. It wasn't uncommon for these runs to take almost a day.

Besides performance, the other nightmare was was someone described below: each script was a one-off that didn't leverage the work from others. If the author only new C shell, then you know you're going to be doing gymnastics to catch the stderr of some of those programs (you can't capture it in the same manner that Bourne variants do).

Anyway, yes, we all adore the Unix philosophy, but there are limits.

[pjmlp]

The "philosophy" that I never saw in commercial UNIXes, starting with my first experience with Xenix in 1993, beyond being endless repeated in FOSS circles, which is ironic given how GNU and BSD applications work, with their endless amount of command line parameters.

[_2z1p]

Not just the language itself but the whole ecosystem it brought with it was revolutionary. CPAN was incredible. There seemed like there was a module for just about anything! Perldoc and a testing framework were built right in. Regexes, backticks to shell out to the system, reporting built in. It really was the whole package.

[adastra22]

Yeah CPAN is easy to take for granted now that every language has its own package manager (npm, cargo, pip), and even for C/C++ you can find what you want on GitHub. At the time though CPAN was revolutionary and no one else had it. To be able to just search for a module, find one that did what you wanted, then download it to your project was pure magic.

[dhosek]

There is some question whether CTAN (for TeX) or CPAN was the first big library. It was pretty close. I was involved tangentially with the guys in the UK who were setting up the first pass at CTAN as the administrator of the ymir.claremont.edu archive and I remember one of the things that they came up with back then was that you could do an FTP get of any directory and get back a zip archive of its contents which was pretty fancy in the 90s. That said, both CPAN and CTAN definitely show their age in that they assume a single version of dependencies will be installed on any system which causes untold nightmares when it turns out that there’s a non-backwards compatible update to something three dependencies deep that you weren’t even aware you were using.

[jcynix]

> There is some question whether CTAN (for TeX) or CPAN was the first big library.

There isn't "some question", there is a definite answer, namely in "Programming Perl" and I quote the 4th edition, page 629:

History

Toward the end of 1993, Tim Bunce, Jarkko Hietaniemi, and Andreas König set up the perl-packrats mailing list to discuss the idea of an archive for all the Perl 4 stuff floating around the Internet. Perl 5 development had started that year, and one of its main features would be an extensible module system that would allow people to extend the language without changing perl. Jared Rhine suggested he idea of a central repository, but nothing much happened. His idea had come from CTAN (http://www.ctan.org), the Comprehensive TeX Archive Network.

Edit: fix markup

[sverhagen]

>There is some question whether CTAN (for TeX) or CPAN was the first big library.

I'm not in this space, but the similarity in names made me wonder if these would really have started around the same time, with an unclear "dependency order", so I reached for a search engine, and according to Wikipedia, CPAN (1993) is based on CTAN (1992).

[tod222]

See also The Timeline of Perl and its Culture https://history.perl.org/PerlTimeline.html

[justinator]

>assume a single version of dependencies will be installed on any system which causes untold nightmares when it turns out that there’s a non-backwards compatible update to something three dependencies deep that you weren’t even aware you were using.

I may not completely understand what you're describing, but with perlbrew managing different Perl versions, and various CPAN clients installing modules in versioned libraries - as well as the test/smoke servers that test CPAN on tons of versions of Perl for Perl CPAN authors automatically, I don't know if I've come across what you're describing.

What's still a pita is when there's an outside dependency from the Perl ecosystem that breaks a module, but I'm not sure that's Perl's fault.

[dhosek]

It happens more with CTAN where there is no possibility of a testing framework and there is no ability for multiple versions of an artifact to coexist.

[swills]

As far as I know, https://cpantesters.org/ is still unrivaled though. None of those have the level of testing that CPAN Testers provides.

[imron]

> Yes, awk and sed were replaced by Perl,

I still use awk and sed semi regularly. I haven’t used Perl in over a decade.

[kristopolous]

1980s awk and sed aren't the same as the 2020s version.

They've become much more useful because of the influence of things like perl.

To see what I mean, here's SunOS 4.1.1's man page for awk and sed as grabbed from https://github.com/ambiamber/Run-Sun3-SunOS-4.1.1/blob/main/... and ran through groffer(1) . These are from 1989 and 1987 respectively. The core stuff is there but not much else.

https://9ol.es/sed.1v.pdf

https://9ol.es/awk.1.pdf

[tannhaeuser]

Awk today is very much the same language it was in 1977 when it was introduced. Comparing your links with today's POSIX awk manpage doesn't yield a single thing Perl brought to awk. At best, while JavaScript syntax is even based on awk, you could say JavaScript regexpes are based on PCRE, and it wasn't clear whether use of capture groups/back references makes Perl regexpes accidentally Turing.

[kristopolous]

POSIX, right. POSIX sucks. POSIX anything is stuck 30 years ago. Nobody in 2023 really uses POSIX awk on some commercial UNIX like IBM AIX, they use GNU awk, that's the modern one I'm talking about https://www.gnu.org/software/gawk/manual/html_node/index.htm...

[kdmccormick]

Well, I look at the POSIX standards whenever I want to write shell scripts and Makefiles that work on both Linux and macOS. Which, in my current role, is often enough that I am driven crazy by how far POSIX is behind GNU tooling...

[kristopolous]

Yep. Funny enough, Linux is actually not POSIX certified

POSIX was initially a bunch of the Unix vendors getting together in the 80s seeing a rise of incompatibility saying "this shit is crazy let's find something we can agree on" and then the conflicts of trying to still have a differentiating But Also compatible product. They knew that if they cannibalized themselves, IBM Novell and DEC were there to snatch up the customers

As a result, POSIX was intentionally kneecapped by all the players but not enough to make it completely worthless.

By the 2010s I decided to simply ignore it and if something breaks on FreeBSD or whatever, make the decision then whether to support it or not. Mostly it's platform testing at the top and then subbing the gnu version of things and bailing out if it's not there.

[tannhaeuser]

What are you talking about? There is no POSIX awk implementation, just a language spec and nawk AKA the one true awk. As shipped on Mac OS, whereas Debian ships mawk in addition to gawk. gawk is just much slower and unfortunately also buggy (crashes on a non-"C" locale with complex regexpes).

[kristopolous]

Right, "the one true awk" corresponds to a book written in 1988, very explicity. https://github.com/onetrueawk/awk

I was incorrect about 30 years. It's actually 35.

You were the one that that said POSIX awk to begin with, I was using your terms. I understand what you meant, why don't you?

As far as shitting on the GMU tools, I don't think I've seen someone do that in decades, especially just referencing the bottom of the man page like that.

As far as speed, I don't know how you're using these tools, but if gawk is your bottleneck and not i/o, there's probably smarter ways to do things.

This is not a productive conversation. You can live life however you want and if you're happy than I'm happy. We are definitely at an impasse. I'm off to bed.

[eigenvalue]

Agreed. I would turn to Python for anything more involved that couldn’t be done very directly and efficiently using awk/sed. I would never use Perl for anything.

[kdmccormick]

Same, but if there were no Python/Ruby/etc, I would probably be reaching for Perl quite a lot.

[idlewords]

Sounds like you were also replaced by Perl.

[natch]

…that you’re aware of.

[tharne]

> I still use awk and sed semi regularly. I haven’t used Perl in over a decade.

Same. Awk and Sed are delightful little tools that have aged exceptionally well.

[grumpyprole]

They are also constrained languages which gives many advantages. POSIX regex will always execute in O(n), Perl "regex" are potentially exponential and can be a security risk in certain situations.

[2colours]

Well, that's at least an actual argument, even though POSIX regex would have never proven so useful in a variety of use cases.

Other than that, I'm completely shocked that somebody would praise sed or awk in well into the 21st century. Turing-complete languages barely capable of properly solving any problem that has any sort of algorithmic complexity, let alone one that requires proper data types, variables or interacting with any sort of system. Really, you can do that in Perl (even in Raku) without your couple-of-liner horribly breaking down once you need some custom logic to it. Oh, and it will even resemble programming languages, not some hieroglyphs left behind by aliens.

[2colours]

What is the reason for that? I think they should be long gone for the lack of proper data structures, if nothing else.

[asah]

Ignore the haters - they're too young to remember.

I remember different CLI tools working differently on SysV and other variants. I remember AIX and HPUX.

Perl meant one thing that just goddamned worked mostly the same way. That and vi and emacs. :-)

[stubish]

Back in the 90s I got my first 'real' job, and needed to learn well grep, awk, sed, col et. al. Or learn Perl. Perl kept us afloat until we had too much Perl and needed to switch to Python. And I still don't know awk or sed; to this day my fingers just go with 'perl -pie'.

[gm3dmo]

The dozens of different versions of UNIX did not help either. A bug in their sed grep awk toolchain meant that you had to write workarounds for things like "Sequioa Unix" to support is single customer. as long as the UNIX flavor had a PERL port all of that could go away.

PERL being open source meant you were a "make" away from having the same environment on all those unixes.

even today when the dozens of UNIX distributions have collapsed into a couple of Lunux flavours, getting a shell based script to work reliably between say macOS and redhat can take some effort.

[tod222]

Yes, writing shell scripts calling awk and sed was awful, just an aesthetic nightmare to work with. Properly quoting strings was always problematic, and while Usenet was available there wasn't a plethora of information like nowadays. This was a few years before Bash was released.

Through my use of Usenet I learned that a better reader program was available called 'rn', which I downloaded and built. It had an amazing handwritten install script (autoconf was years off) which could automatically configure and build rn on any *NIX system. All developed by a fellow named Larry Wall.

rn was truly a joy to use and made reading Usenet swift and efficient. It would get updated with the fixes that came across Usenet that I'd apply using the clever 'patch' program, also written by Larry Wall.

Based on my experience with his other software, when Larry Wall released Perl on Usenet I immediately downloaded, built, and started using it. As promised, for scripting things not requiring a C program, it was massive improvement. Version 2.0 came out and brought many great new capabilities.

I wasn't writing software while versions 3 and 4 came out; I started using it again after version 5 and the appearance of CPAN. Over the years I've used Perl extensively for task automation and data wrangling.

Python now dominates Perl's niche because it's easier to learn and interfaces better with C. It's also less flexible, which compared to Perl is a virtue. One of Perl's mottos is TMTOWTDI—there's more than one way to do it. But many of them are bad. Much of Perl's poor reputation ("line noise") stems from this.

But when Perl was released it was a revelation, like a drab day when the clouds suddenly part letting warm sunlight pour down on the land.

[slavik81]

    # Sometimes I wake up screaming.  Famous figures are gathered in the nightmare,
    # Steve Bourne, Larry Wall, the whole of the ANSI C committee.  They're just
    # standing there, waiting, but the truely terrifying thing is what they carry
    # in their hands.  At first sight each seems to bear the same thing, but it is
    # not so for the forms in their grasp are ever so slightly different one from
    # the other.  Each is twisted in some grotesque way from the other to make each
    # an unspeakable perversion impossible to perceive without the onset of madness.
    # True insanity awaits anyone who perceives all of these horrors together.

https://github.com/openembedded/openembedded/blob/fabd8e6d07...

[SoftTalker]

sed and awk are still around and still used a lot especially for one-liners. Anything more than that is typically not done as much anymore, though is certainly possible.

[LinuxBender]

sed and awk are still around and still used a lot especially for one-liners.

Adding to that some popular examples are here [1][2]

[1] - https://www.commandlinefu.com/commands/matching/awk/YXdr/sor...

[2] - https://www.commandlinefu.com/commands/matching/sed/c2Vk/sor...

[opan]

For me in the current day, perl is too much like "real programming", while I'm fairly comfortable with grep and sed, and to a lesser extent, awk. Piping a bunch of stuff together is just easier to understand and gradually expand. One thing I do in a lot of scripts is read from and write to the clipboard, usually transforming text with sed and co. Stuff like turning YouTube shorts URLs into regular ones, taking just the last directory name from a long file path on my clipboard, making my clipboard lowercase, etc.

I barely get into loops and lists when trying to learn Python before losing interest/focus and not trying again for months. It feels very abstract and like it's not useful for day-to-day stuff without learning and writing a lot. Shell stuff is very short and powerful and feels more grounded in reality. To be clear, I am not using if/else and the like in my shell scripts either. Usually they're just a few lines with little to no logic, but they're immensely useful. I can't imagine learning perl and then using it for all this stuff.

[BirAdam]

All software eventually becomes an unmaintainable nightmare.

[yjftsjthsd-h]

> Yes, awk and sed were replaced by Perl, but more importantly, the unmaintainable nightmare that glued all of it together was wiped out.

Er, Perl replaced an unmaintainable nightmare? Perl, the language infamous for being indistinguishable from line noise?

[margalabargala]

Yes. It's comparative. The absolute value of Perl's readability, maintainability, and general intuitiveness is massively better than all comparable tools that existed when Perl came out. Perl is a massive improvement over those tools.

If you have doubts, try writing a complicated production software stack in awk. Then hand it off to a coworker.

One need not be irreplaceably good, if one is already beating the current state of the art by an order of magnitude. Perl did this.

[BaseballPhysics]

Moreover, as often as people joke about the readability of Perl code, that's entirely a function of the developer.

I've easily written tens of thousands of lines of Perl, and not a single person has complained about difficulty reading or maintaining that code. Why? Because I apply all the usual best practices for code hygiene that apply to any language.

Frankly, I think most people are just repeating a meme they heard once, and the rest just get put off by a) sigils, and b) the use of implicit variables (which I tend to use only very sparingly, and mostly in quick one-liners).

But a developer that poorly names their functions or variables, fails to modularize appropriately, fails to document their code, abuses language features because they want to be excessively terse or clever, that kind of person is gonna write crappy, difficult-to-maintain code no matter what language they use.

In fact, I'd argue the advent and popularity of Python--which was pretty radical in how opinionated it was at the time--is a direct response to languages like Perl and C that were a lot more free-form and easier to abuse by poor coders.

[zgluck]

and not a single person has complained about difficulty reading or maintaining that code

Developers extremely seldom complain about other developers' code to their faces.

Instead they go to lenghts to avoid it if they dislike it severely; right or wrong.

[yjftsjthsd-h]

> that's entirely a function of the developer.

I would argue the same of sed/awk; you can write an unmaintainable mess, but you don't have to.

[BaseballPhysics]

Oh sure. IME Perl just pulls it all together into a much more convenient, all-in-one package that has all the libraries and tools you need to build anything from small scripts all the way up to large projects.

[tribby]

absolutely a response to perl. “TOOWTDI” has been an expression in the python community for a long time :)

[thesuperbigfrog]

TIMTOWTDI is Python's motto for managing dependencies and deployments:

https://packaging.python.org/en/latest/overview/

<kidding> OCI (Docker) Containers were invented to overcome the problems inherent with deploying Python applications. </kidding>

[anthk]

Gen-Zer spotted. No, Perl is not line noise.

Get the Perl source code for PangZero and you'll understand proper Perl code.

[gpvos]

All nontrivial sed programs are line noise, there is no other way to program in it. In Perl you at least have a chance. Awk is not much more readable than Perl actually.

[unpythonic]

Oh yeah! Perl sure did have a bunch of ugly programs, but I believe that was because it was so easy for non-experts to program in it. There have been many companies successfully built atop of a Perl code base, and I've seen fantastic systems built with it. I've also seen one-off programs handed from manager to manager which would scare you to death.

[lonelyasacloud]

Yep, that’s how bad it was before.

[marttt]

A single-executable perl5 (like awk, but with advanced string manipulation + sigils + timtowdy madness) would be great.

[zenlot]

And now you can just use PowerShell which combines it all in one cross platform scripting language.

[csydas]

assuming that the security administrators of the systems you want to run the script on haven't followed Microsoft's 'recommendations' and basically made powershell unusable on the systems due to random GPOs ;)

PS is 'okay' as a scripting language, but it's very frustrating how Microsoft's security defaults seem to be dead-set on ensuring you can't use scripts on systems without jumping through a bunch of hoops. I get how MS got to this conclusion that it needs to be locked down, but it makes me think twice about pumping out a script since I know I might need to walk my colleagues through how to actually run the damn thing depending on what the Security Admins decided to lock down that week.

[doliveira]

Cross-platform? I guess technically it is...

[tannhaeuser]

Perl didn't replace anything. It was replaced by PHP for CGI programming (which says a lot!) and botched itself by the second worst version migration of all times.

[bonzini]

Perl 4 to 5 was quite successful. Perl 5 to 6 was definitely worse than Python 2 to 3, since the latter actually worked out in the end.