Lorinda Cherry, author of dc, bc, eqn has died | Hacker News Reader

Lorinda Cherry, author of dc, bc, eqn has died

ncwit.org

1,291 pointsggm185 comments

Lorinda helped me massively with my attempts to bring back online some hardware that had escaped early 90s Bell Labs so that, as a pre-teenager trapped in rural Illinois, I could run Plan 9 and build some software for it. She did not stop at connecting me to the right people to locate the hardware I needed to restore my computer to the living. She also went above and beyond by introducing me to individuals whom I did not deserve to talk with nor did I appreciate how lucky I was to be sharing e-mails with these people at the age of 12. These unlucky (former) Labs employees included both Rob Pike and Ken Thompson.

Lorinda, thank you for taking such extensive efforts to encourage my passion and interest in obscure operating systems. I have not lost my drive to explore this area of computing. My only wish is that I had sent you more than one of the coffee cakes from Morton, IL that you loved and said made us even. Hopefully, wherever you are now, you can have as much coffee cake as possible. Thank you again, truly.

wow, that's kind of crazy cool. 1) the help was offered at all. 2) to put you in touch with others (lots of personal cred at risk there) 3) they helped you out in kind 4) you were only 12.

things sure were different in the 90s.

Indeed. These days you're lucky if you can get a project maintainer to respond to a PR within a few years.

I don't think a colleague and friend would disrespect someone for making an introduction to a kid, even if the kid turned out less cool than thread OP.

Nowadays kids can befriend celebs on Twitter. It's easier, not harder.

Good grief. There's a huge difference in following someone on Twitter vs a personal email account. You think someone you meet that knows $famousCelebrity is just going to give you their email account? Really?

The favorable reading is that friending a celebrity on Twitter is a historically low effort way to make contact, even if it is correspondingly low value. Everyone is “close”. This means there are more opportunities for connection.

It smells like a natural progression from the “old” internet of newsgroups and email.

Is friending a celebrity on Twitter the same as getting an email or phone call from Ken Thompson on how to fix your old Bell hardware just because Lorinda Cherry put him up to it? No, of course not. But that connection could happen over Twitter.

I agree. I think she was so chuffed that someone so young was interested in her work that she was willing to break social norms. I also think it definitely helped that I didn't mention a big reason I chose to pursue Plan 9 was that Glenda was such a cool mascot.

While I'm mostly joking about that last part, Lorinda did always mention that my childishness was expressed in curiosity rather than blatant immaturity. She said that was the key to being able to exist in these professional spaces without drawing undue attention to myself. It also helped that I steeled myself by participating in Usenet, where often times it seemed people were competing to be the most toxic person in the collective.

The closest to that for me was being a just out of high school employee at an interesting post production facility that had some talented people in the engineering staff. I was a kid in a candy store, but all electronics and video/film based stuff. The engineers were probably taken off guard by my incessant questions of why/how/huh until they eventually realized that I wan't being a pest but actually learning what they were sharing. At some point, the questions were advanced enough they wanted me to switch departments. Definitely the best OJT situation I've ever been in, but that was 30 years ago now. damn.

Have you ever read “I Am A Strange Loop”? You may enjoy Chapter 5, “On Video Feedback” if I recall correctly.

I have a hunch you’d enjoy it :)

Things were different. I had a similar experience with Michael Hart from Project Gutenberg. I stayed in his house! He was ridiculously warm and welcoming.

You can't imagine what scarcity of nerds there was back then. These days, every city has hundreds or thousands of programmers. But in the 90s -- for any area of computer interest -- there were only a few hundred of you around the world. It was wonderful to meet someone with the same interests.

Being interested in programming/hardware/security AUTOMATICALLY made you a member of a small club. Most of the other members were glad to help you. And there wasn't a surplus of people all trying to stand out and be noticed on social media. (In fact, because you were into computers for the love of them, you probably weren't even thinking along the lines of "I must get noticed by $celeb so I can ask them for a job". The odious phrase "personal brand" had not been coined yet, and the concept was foreign to most in our computer nerd world.)

It lasted until around 2008 or 2010, I reckon -- about the point at which "go into computers, it has good money", the success of YC, and the rapid growth of FAANGs really made nerds ubiquitous. Then Marvel/Disney turned our niche interests into the mainstream, and we became culturally adrift. We were in sitcoms (Silicon Valley, even Big Bang Theory to a great extent) and fully mainstreamed. At that point, you could ask someone for something but you were just one of an ocean of unremarkable others.

That's how things were different in the 90s and (as a white English-speaking het guy who had Internet access) gosh I miss those times.

Makes you think how not ridiculous the scene from War Games where they get stuck on an island and the guy just lets them stay in his house when they miss the ferry. Makes for convenient story telling that in today's standards just seems like would be a 'awhellznaw' kind of response.

Or the Neil Degrasse Tyson story of how he met Carl Sagan.

That easily happens also today in less densely populated areas.

I think a few hundred programmers in the 90s is underestimating the numbers by a factor of at least 1000 and (probably much much) more. Of course such numbers are difficult to estimate, but here some pointers: It was introduced in school in several countries in the early 80s: https://en.m.wikipedia.org/wiki/Computer_science_education

And there was vast literature. Petzolds programming windows first edition is from 1988 and second from 1990. While the first might have been a niche publication usually there is only a second edition if there is substantial interest. (number of prints would be nice to know).

I'm not sure the OP was making this claim exactly -- I think they just meant for any particular interest (like running Plan 9 on some old hardware) there were probably only a few hundred people around who shared that particular interest.

Could be. But any particular interest is def not true (windows programming), so more like some niche interest. But this is the definition of niche interest, isnt it.

I can count on one hand the number of people I’ve met that did windows programming, both before or after 2005.

You're right. I said "for any interest" but obviously, like all things attention there's a power law distribution. Many Windows programmers, or anything else that had massive interest. But if you were a Unix kernel nerd, there weren't many Unix kernel nerds around. If you were a Perl programmer, there weren't that many Perl programmers.

But even given the prevalence of Windows development, it was still rare to encounter another programmer socially unless you lived in the occasional city where they were common. My point is that nerds finding nerds was a much rarer thing back then.

Computer programming was a nationally available subject in secondary schools in the UK in the mid '70s.

I co-wrote a noughts and crosses program to run on a Busicom in 1973; and continued programming at university ('74-'77).

See, inter alia, https://en.wikipedia.org/wiki/Cesil, https://en.wikipedia.org/wiki/Busicom.

>>> really made nerds ubiquitous

It did not make nerds ubiquitous. It made nerd wannabes ubiquitous. The ppl with OG nerd-like qualities are actually frowned upon in the industry now. Somehow a SWE needs to be a "well rounded" individual. Technical skills don't matter much anymore.

This.

Way back in the 90's I reached a conclusion that a programmer or sysadmin or hardware guy is now just the new auto mechanic.

'98 or so I was shopping for replacement software for my uncles business (y2k prep) and we went to this firms office, and basically everyone in the office, the owner, all the developers, the sales guy, the guy who would come out to install the hardware... every one of them could have been in their high school football team 10 years earlier. They were doing this kind of job because there was money in it full stop. They took us out to dinner and the conversation only comfirmed that snap perception. You could substitute "propane and propane products" in place of software and IT hardware without changing a thing about any of them.

Perversely today I probably wouldn't use the same comparison because auto mechanic is probably more nerdly and interesing persuit today than it used to be.

I think what I mean to say it's now a trade.

You might be a nerd in that trade just like you might be a nerd welder, but it's no longer required. You don't even have to be a wannabe nerd although many are.

Countless people in those jobs with no special love of it, and no fundamental curiosity.

School counselors told them it was going to be in demand, so they did that.

I think it's both better and worse today since then.

Obviously the sheer mass of IT work needed today, and the sheer mass population of people needed to do it, means that there are a many many people doing IT work that aten't good at it and don't love it.

Obviously today it's no longer special to to work with computers.

And there is also the ever present dislike for intellectuals in general that hurts nerds even at the same time when there is some nerd cache from making money. (actually I bet the nerds don't even make the real money any more. they get harnessed by the business types)

But I think it's also true there are more actual nerds today and they are more accepted than in the past. Still outnumbered by wannabes and everyone else, and the real nerds still not really liked or taken seriously by most, but more and better than 20 years ago.

I agree, the Tech field has exploded and lotsa people are into it which is a good thing. But when anything goes "mainstream" it loses its charm, it loses the X-Factor which made it special.

Tech companies have lost that charm now. I have worked at companies where doin a good job does not even matter anymore. Its all about "projection" and "perception"

There will always be a place for those of us who like tinkering with software and/or computer hardware, the nerds. We like using the CLI or Vim or Emacs. We go deep on compilers and PLT. But I personally welcome those who aren't nerds in the same way. I like showing them how to be more efficient in their work, build more robust/reliable software, and helping them understand the job they are doing better. Maybe, I'm nerding out about mentorship but I find it rewarding to help the non-nerd become just a little more nerdy in their career. It's an opportunity not a problem.

Niche and Nerdery is fractal, I find. I also find sharing my findings from exploring random niche corners of obscure topics to be rewarding.

What I dislike now is that previously I could assume that anyone with the same interest had some form of passion for the craft, but now they’re swallowed up in a sea of people that are there -like you said- just for the money.

They still want to know enough to not get fired, but that’s where it ends.

I kind of feel it pulls the average down though, and it makes my work environment less ideal.

I've heard it described as "social gentrification", but that particular article tried to defend offensive behaviour (think 4chan) as a personality trait or a way to keep 'normies' at bay. I don't think that's the way to go either, because the very same type that defends their offensive and antagonistic behaviour will suffer from loneliness as well. The venn diagram overlaps a lot there.

But that particular blog post aside, you do make a good point. The 'nerd' in my head (and / or the one that I am) doesn't score too well on EQ and social skills, or to be even more blunt, is often on the autism spectrum and has to mask to match those expectations from the more neurotypical folk. But that said, I think one has to mask even with other neurodiverse people, because else they will just rub each other the wrong way.

Computer nerds suddenly became attractive when Bill Gates made his first $billion. Before that we were pariahs <g>.

I was and still am a computer nerd and I was never a pariah. I'm sure some nerds out there felt like it. But, anecdotally, that was not my experience.

Consider movies at the time that had computer people in it. They were always portrayed as bumbling, inept people with bad greasy haircuts, odd clothing, pimples, and tape holding their glasses together. The hero would always berate them with "speak English, please".

You can also see it in the Seattle sketch comedy "Almost Live" series, whenever they did a bit on Microsoft.

Myself, I learned to avoid mention of my profession when meeting women.

Even at Caltech at the time, being a computer nerd was not popular. Popular was physics (possibly because Feynman was there at the time). Astronomy was popular, too. Back then, however, nobody had any inkling of how much money could be made from computers.

Yes. I am not sure if Bill Gate made the difference in the public image of nerd. May be in the State. Even the Google era in 00s didn't change much. ( May be very briefly during Dot Com bubble ) My first realisation of Tech really going into Mass Market was the Smartphone era in 2010. When you over heard conversion from everyday life people having lunch talking about ISO ( it should be iOS, but they got it wrong ). And Apple, or iPhone became a yearly event that all mainstream media publish. By around 2012 App Developers became "cool".

In some ways I think that was the revenge of the "nerd". Although in another way most of these new "nerd" never experience the 80s, 90s and 00s. And they are very different nerds to those in that era. Too much hype, too little engineering.

I had an old IBM XT clone sitting in my bedroom at the age of 12-13. I brought a friend over once and he called me a nerd. Not the "nerd" of today. Back then it stung. He said it with a bit of disgust, as if he learned some dark secret I had been harboring. I even loaded up Catacomb 3-D (a very early id Software precursor to Wolfenstein 3D) to prove I can be cool too. It didn't really work.

I can definitely relate with this. Grew up hacking IBM PCs in the 90s, got heavily involved in BBSs as a young kid, even wrote letters to random software developers (remember scorched earth?).

I really dislike the modern tech industry. I think that money ruined everything. Vast majority of my coworkers over the last decade or so are the type that used to train to be doctors, lawyers and other lucrative careers, with no real passion in computers or tech. It's all about money.

I do dream about changing careers. I know at least one other thing I'd love to do but it would pay a fraction of what I can make now and simply isn't feasible.

Last time I was in Silicon Valley, I was shocked at how the nerd aesthetic had gone away.

Lots of athletic folks.

I think it’s because Ivy League and Wall Street folks have been invading (or taken as captives), and the Ivy League/Wall Street aesthetic is not exactly geeky.

Nerds can be athletic. Loving tech doesn't mean you can't also take care of your body.

I’m struggling to find the right word for “type of person” … words like “nerds” and “jocks” can be limiting, since plenty of people are both.

The guy who got me into tech and computers was a former green beret. Most athletic guy I’ve ever known, and a genius with tech.

I don’t think he was a nerd, but this is really subjective.

Is it so different today? Kids are still curious and reliant on their elders to both provide and mentor. It’s our turn to be the Lorinda (or Ken Thompson) to some other curious kid.

The 90s were like that. Tech was a much smaller community and you could cold email just about anyone with a decent probability of a response because being on the Internet was a highly selective filter in its own right. In its own way, it was a golden age in terms of accessibility to really smart people.

It was a brilliant moment in time. I was able to routinely chat with an astonishing number of excellent and famous minds when I was young and impressionable in a way that would never happen today. I feel very fortunate to have lived that, it had an enormous impact on my life I think. And it was safe for the people I corresponded with to be accessible back then.

Yes and no. Things were different, for sure. Hell end 80's I hung around a random polytechnic school at age 14, where I played to my hearts content with their professional CAD stations and programed their robotic arm. I had no reason being there, where it not that we visited once because my mom knew someone there, and the IT lab guy took a liking to me.

But I'm reasonably sure that if today I bumped into some random kid that was obsessed with anything 'geeky' that makes me fondly remember when I was that age, I'd jump on the opportunity. Wouldn't we all?

It was different prior to the mid-90's and the 80's (despite much more primitive tech) were even more fun. Two things happened around the same time: Bill Gates became the worlds richest man (and a household name) and a bunch of college dropouts started becoming overnight multi-millionaires (at least) with the dot com explosion which was well publicized. Prior to that, sure there was good money to be made in some areas of tech[1], but many were driven by interest rather than a career path and most non-tech people just regarded computers as those things they didn't understand. Most business people regarded computers as fancy calculators that did the accounting and it was hella-hard to even get them to learn about spreadsheets. So geeks were pretty much left alone and programming wasn't seen as the mail room job on the path to getting rich it is today.

In that environment, a lot of adults were willing to give technically minded kids at least some amount of their time because the only reason most of us (kids, at the time) were asking questions was because we just wanted to understand how all this stuff worked for fun rather than working on a get rich quick scheme. I also suspect the adults found this enthusiasm more interesting than the general disdain they probably experienced at the office.

[1] As in, put in some time and develop an area of expertise that someone valued... then you could start making money. Most college grads were viewed as fairly useless for their first few years out of school and their salaries reflected that.

Yeah there was this odd change over. I'm not exactly sure when but probably around the dotcom boom.

I remember that if you said you mentioned computers people in general would just parrot that back to you in this "nerd" voice as if it was a joke.

It was rare to even read about what was happening in the world of computers in a mainstream newspaper.

To me, a kid at the time, I felt like I was seeing the gateway to a new kind future unrolling but there was just no way to discuss that with "normal" people at the time.

I guess any adult working in that industry would be keen to share that with anyone interested regardless of age because it was all changing so fast and was so damn exciting.

I don't know - it also went away with the bust. I picked CS as my major shortly after the bust and everyone advised me against it. Between the bust and outsourcing I was told I would be making minimum wage.

And, welp, here we are.

You're actually reinforcing my point about the change in the industry. Many of the people who came up in tech prior to the mid-90's didn't care of the industry was in boom or bust mode: they just wanted to work in/with tech. I completely understand those who got into it because they saw it purely through a financial lens, but that's a different subset of people in tech.

Things were different. Michel Gien (co-founder and CEO of Chorus) would couch surf at my place when he was in Austin on business.

You can get access to people at age 12 that you wouldn't get in a million years at age 20.

You shouldn’t stop trying at age 20 (or 30, or 40...), as long as you are earnest, not too much bother, and aren’t worried about not always getting a response.

If you show interest in people’s work, especially if you have questions, corrections, etc. that demonstrate that you are engaging with it seriously, even “famous” people are often happy to respond (but people certainly also often don’t respond to cold emails, for a wide variety of reasons).

I'm 40 now! I'd better start emailing people :D

RMS is pretty famous for responding to pretty much any query, for better or worse. There's a treasure trove of fantastic email responses to questions like "do you watch anime?" and "what do you think of my birthday card?"

I've seen a number of people rag on him, espousing that he's got nothing else to do (they may well be right) but it makes me respect the guy even more to know that he's replying to emails, even when they're dumb, instead of watching TV or browsing Twitter. He's a real consummate professional, even when the people around him aren't.

I dimly recall receiving a call from RMS one Saturday when I must have been in high school. I was working (on my own) on porting the ETH Modula-2 compiler from VAX/VMS to a NS32032 based "co-processor" board that plugged into the pc isa bus (an early hardware design from Trevor Marshall of YARC Systems). I think RMS asked if I would port the compiler to GNU, but alas I told him I knew nothing about Unix back then. The NS32032 went no where, but my compiler porting effort got me my first job with an interesting (to me) startup!

> RMS is pretty famous for responding to pretty much any query

Can confirm. In 1990, I managed to send him a letter from a borrowed Internet account (I didn't have an email address at the time.) He tracked me down and telephoned me(!!), I still remember his friendly demeanor, and at the end of the conversation he said "Happy hacking!".

i still get access at age 45; I think what changes is not the propensity of people to answer but the audacity of people to ask.

The motivations of adults are, usually, not the same as those of younger people, so that some people may be more skeptical to respond, too. Obviously, I'm generalizing here. But there are logical reasons behind that.

I think also people just instinctively feel bad turning down a child's request.

There's also that if you're 45 years old, people expect to see some accomplishments, and are skeptical if you don't have any, whereas there is no such bar for 12-year-olds other than perhaps curiosity and ambition.

It’s more that I’m happy to help people if they show me they’ve done their homework. The homework required of a 12 year old is obviously zero, because if they’re asking me they already know more than they should.

Ever actually tried? I've contacted well-known people in their field (most recently Hugh Darwen for an obscure database question) and I find them very willing to help if you don't waste their time. Stop with the it's-all-stacked-against-us-don't-even-try bullshit.

It's crazy to realise that these people you look up to are actually other human beings and with just a little bit of luck you can strike up a normal conversation with them and they won't hate you for it!

I am scared to ask another team question about my work without checking documentation twice because I'm scared I'll ask something that's already covered in detail somewhere.

I am very good at asking stupid questions. Thankfully, my teammates are patient with me.

Can you imagine wasting the time of one of the brilliant minds of our times and it turns out to be user error?

I would say try to sleep on a problem before reaching out for help if time permits. Probably isn't good for business but it is good for my personal development.

Exactly what you say; sleep on it first. Make sure you've done your background work. But then, don't be afraid to contact them.

> I am very good at asking stupid questions. Thankfully, my teammates are patient with me.

Are your questions really so stupid, are theirs really so much smarter when they ask, or are you being harsh on yourself perhaps.

It's not about being stacked against one, it's more that one understands the constraints other might have and is more hesitant to reach out, instead of doing a round more of research yourself and using more general forums first.

I dunno, I feel like I'm pretty close to the aforementioned Rob Pike and Ken Thompson through working with Go; I feel like if I ever had a deep question about or issue with the language I could reach out to them through the mailing list. It's more a matter of having something to talk to them about than finding and reaching them I think.

One of my achievements in life is a email exchange I had at around the same age with someone from Cray, a conversation sparked after a newsgroup post about the model of computers used in the book version of Jurassic Park.

It wasn’t anything as extensive as what op is talking about but it made a deep impression on me. Something like the nerdling equivalent to getting a baseball hat signed. The potential with email to speak directly with people working in rarefied places was eye opening. It also gave a real confidence boost that I could stumble on interesting things to say.

The ghost of Steve Jobs would disagree - "Make the call". Anyway it's awesome you can form spontaneous human connection with other people from most surprising of places if you only try. I suppose the pathological twist to this is all of those phone-center scams...

It depends. Most famous people in tech aren't going to answer easily Googleable questions from a 20-year-old (or a 40 year old) But if you've got a genuine reason to be asking them, they respond more frequently than you'd think.

I wanted to use a purported quote from Maurice Wilkes that I couldn't track down a citation for in the introduction to my PhD thesis, so I emailed him. He must have been well into his eighties by that point. I got a very helpful reply and was able to use the quotation, properly cited (as well as add a "Wilkes, personal communication" to my citation list, which still makes me smile).

But what was the quote!?

Indeed, that tends to change again in your 1000030’s.

> Lorinda helped me massively with my attempts to bring back online some hardware that had escaped early 90s Bell Labs so that, as a pre-teenager trapped in rural Illinois, I could run Plan 9 and build some software for it.

Gnot or Blit terminals? Oh man if its a Gnot....

I'm Gnot saying it isn't one... ;)

The nice thing about celebrities in academic worlds is that they're so much more accessible than celebrities in big-money industries like acting or business. Their email is publicly available on their university pages, and IME they respond kindly and quickly.

The tool `eqn`, which she wrote (joined by Brian Kernighan) is an "ancestor" of TeX: The mathematical syntax of TeX (design started 1977) was based on eqn (1975). For example, the eqn paper (https://research.swtch.com/eqn.pdf) has the expression

    sum from i=0 to infinity x sub i = pi over 2

which would have been entered as something pretty close to that in the first draft design of TeX (the second draft introduced the backslashes), and today as:

    \sum_{i=0}^{\infty} x_i = {\pi \over 2}

and you can see the evolution. The Wikipedia page has more examples (https://en.wikipedia.org/w/index.php?title=Eqn_(software)&ol... ). Knuth credits Lorinda Cherry explicitly a couple of times in the first draft design (https://www.saildart.org/TEXDR.AFT[1,DEK]):

> A special syntax is used in formulas, modeled on that of Kernighan and Cherry, Comm. ACM 18 (March 1975), 151-157. For example, ``sup 9'' in line 41 specifies a superscript 9, ``sub {n+1}'' in line 74 specifies a subscript n+1.

and in the last page/paragraph:

> To conclude this memo, I should explain how TEX is going to work on math formulas. However, I will have to sketch out the code in more detail and it is only fuzzy in my mind at the moment. […] it may be necessary to build the parse tree first as Kernighan and Cherry do.

(Yes indeed TeX builds the parse tree first, as we can see from the second draft: https://www.saildart.org/TEX.ONE[1,DEK] )

The eqn system was a pioneering and capable one: although Knuth did not use any of the code of troff/eqn as-is (not sure if it was available to him; in any case he was targeting "book" quality), clearly it influenced the design, and I imagine it inspired him about what was possible in the first place. Even after TeX became widely used, there have been some math books typeset with troff and eqn.

Reading about Lorinda Cherry's other accomplishments like `typo` and the Writer's Workbench, it's clear we've lost someone who was a pioneer in multiple respects.

Completely agree, 'eqn' was a pioneering achievement followed by TeX and Mathematica

After having a look at: https://www.oreilly.com/library/view/unix-text-processing/97... and a few man pages, I figured I could play with eqn like so:

    eqn -Tpdf - <<eof | groff -me -T pdf > eqn.pdf
    .EQ
    sum from i=0 to infinity x sub i = pi over 2
    .EN
    eof

    xdg-open eqn.pdf

I did not have much success trying to force it throug groff/troff and render pretty in a (wayland) terminal - with neither `--text` (as below) or `--tty`. But perhaps there is no such "magic" renderer?

    eqn -Troff -Tutf8 - <<eof | groffer --text |less -R
    kEQ
    sum from i=0 to infinity x sub i = pi over 2
    .EN
    eof

I am not aware of a powerful tty renderer. Groff looks like it's aimed at paper outputs first. Also eqn can be called with `groff -e` (other preprocessors also have their flag) and inf (instead of infinity) produces a correct infinity glyph.

groff -step -k < input.groff > output.pdf

> (G)roff looks like it's aimed at paper outputs first

Indeed - seem to remember (?Brian) comment one journal turned down a paper due to how well typeset is was believing it'd been printed elsewhere before

I somehow feel it is a pitty that Knuth went on to design his own method for describing mathematical expressions, as his seems to less semantic and more based on representation. Note how in his notation the two '_' have a total different semantic meaning, while in eqn different syntax is used. His method is a more consize, but even not that much, but her method is much closer to how the resulting expression is read. I did have a look at the paper mentioned and note that terms like 'sub' and 'sup' are also related to position. An even more semantic approach probably would use 'x power 2' instead of 'x sup 2'.

I've so often wanted to replace LaTeX snippets in my org-mode documents with machine-readable `eqn` snippets (or maybe SageMath snippets). I'm sure someone out there must be sitting on an unpublished groundbreaking setup that pulls this off.

`eqn` looks way more readable than TeX.. All those \ are making my eyes bleed.

As one of few authors of an implementation of `dc` and `bc`, but one who never actually met Lorinda Cherry, perhaps I have a slightly different perspective of her work.

I've read the closest thing we have to the original sources of `dc` and `bc`: the source bundled with Plan 9. It didn't take me long to read the entirety of the source because they were simple and concise. That immensely impressed me.

However, I hate to say that my youth (compared to Ms. Cherry, at least) caused me to look at the source with disdain, mostly from a lack of handling errors caused by user mistakes.

It took several days for me to think more carefully about the context. She was writing for herself and other programmers, who would probably be able to recognize when they made a mistake and fix it.

The code has a simple elegance that mine will never have. Sure, you might call mine "industrial strength," but I think a quote by ** Gabriel sums up the difference between Ms. Cherry's code and mine:

"I'm always delighted by the light touch and stillness of early programming languages. Not much text; a lot gets done. Old programs read like quiet conversations between a well-spoken research worker and a well-studied mechanical colleague, not as a debate with a compiler. Who'd have guessed sophistication bought such noise?" [1]

And that says nothing of the design of her software.

`dc` was, and in many ways still is, the simplest calculator that could ever exist. It was the simplest shell too, with the `!` command. I personally believe that it was for this reason that `dc` was the first program Bell Labs made run on the PDP-11. [2] `bc`, while more complicated, is also a great design (for the time).

In short, Ms. Cherry was a master of her trade, and I only recognized that from afar.

One of the items I had in my bucket list was to meet Ms. Cherry; because of her work on `dc` and `bc`, I felt a kinship to her having written my own. It is sad to know that item will never happen. Oh, well.

[1]: https://people.csail.mit.edu/alinush/6.824-spring-2015/l07-g...

[2]: https://youtu.be/EY6q5dv_B-o?t=1767

Here's a copy of dc: https://github.com/dspinellis/unix-history-repo/blob/Researc...

Thank you. It does look familiar.

That's a very insightful quote.

She would be the first to tell you that you have met her, by using her programs. Her programming style, which you've done a superb job of relaying, highly reflected her personality and the values she held dear.

It sounds like you two would have gotten along splendidly had you met physically, from what I know from the time I spent working with her (albeit mostly virtually).

Thank you for your kind words. I hope you are right.

That's a pretty awesome tribute. Thank you for sharing!

> I'm always delighted by the light touch and stillness of early programming languages. Not much text; a lot gets done. Old programs read like quiet conversations between a well-spoken research worker and a well-studied mechanical colleague

Much of this was of course due to the lower performance of computing at the time - compilers just weren't highly sophisticated, and it made sense to code simply so as to leverage them as much as possible. That same attitude also extends to old-style "system design", which was often bespoke in a way that would not be very well regarded today. A hardware-portable compiled system like early Unix, designed for practicality and real use, was a huge novelty back then.

fwiw the original dc/bc source was on the V7 tape and can still be found there:

https://www.tuhs.org/cgi-bin/utree.pl?file=V7/usr/src/cmd/dc

https://www.tuhs.org/cgi-bin/utree.pl?file=V7/usr/src/cmd/bc...

Here she is demoing Unix, pipelines, etc. (1982)

https://www.youtube.com/watch?v=XvDZLjaCJuw&t=828s

such a cool video

even now it is all pretty slick especially for people who have always been in GUI operating systems, at the time it must have seemed like wizardry.

I love how she takes a sip of her coffee/tea when she has to wait for the computer

It's interesting that uniq is spelled unique in that video. I wonder who changed that!

I was wondering what "mismatch" was too, turns out it's comparing to the system dictionary

https://github.com/watson/old-unix-spell-checker/blob/master...

Thanks for the video! How come we lost the `lowercase` command? I know you can use `tr` or `sed` among others to accomplish the same but I would still love to have a `lowercase` command :)

Possibly the most amazing thing is that speak(1) is actually functioning in 1982 !! And remember, it's not just the software, you need some sort of DAC connected as well. Granted, Max Matthews had done the first digital synthesis of sound nearly 30 years earlier, but there's no sign that Cherry is working in an digital audio lab. She's on a terminal, probably connected to what was then called a "minicomputer" ... with a DAC!

Found a transcript of an interview with her, not sure how old. Talks a great deal about the statistical analysis of text that McIlroy praised her for.

https://www.princeton.edu/~hos/mike/transcripts/cherry.htm

I've used all three of these programs and never knew that the same person had written all of them. I still occasionally use bc--I started using it decades ago and prefer it to any calculator for basic use. For more complicated calculations I now tend to use the python RPL at the command line, but I'll always remember bc, dc, and eqn fondly. Thank you Lorinda Cherry.

There may be some people out there that don't realize these commands are available on your machines. The macOS operating system on every contemporary Apple computer has a core that is build on Unix. Just open a the terminal app and you now have access to the only user interface that all of us old Unix users used for many years. At the command prompt you can learn how to use bc and dc simply by using the man command (short for manual). Type man bc to get the brief page from the Unix manual describing bc (or any other shell command in the same fashion).

One of the next commands you should look up in the manual is the man command itself (type man man). Somewhere, I've got the original thick printed Unix Manual. I used to carry it around just to read before the days when I had a dial-up terminal or personal computer at home.

This article is a profile, not an obituary (I'd expect an obituary to state that she died, and give a date). What's the source that reported her death? I searched the web, but couldn't find any. At the moment, her Wikipedia article refers to her in the past tense, but doesn't cite a source for her death either.

(In case the title changes, this comment refers to the original HN title: "Lorinda Cherry, author of dc, bc, eqn has died".)

It was reported on a mailing list of ex-Bell labs staff and other early UNIX users. This write up of her career was the best one I could find, sourced from information there. It's from 2018 when she got an award. If you have a better one I'd welcome it being posted.

Thanks for the confirmation. I first saw her mentioned when I started using Unix in 1979, and am sorry to hear of her passing.

Rest in Peace!

Her demonstration of the capabilities of the (then novel) pipes and scripts of the Unix Operating System is one of the best parts of this documentary:

https://youtu.be/tc4ROCJYbm0?t=935

I love what she does here: https://youtu.be/XvDZLjaCJuw?t=949, also demonstrating pipes and scripts. The input she chooses makes the whole segment pretty damn funny.

Less technical and detailed (but with some distinct stories and quotations):

https://ncwit.org/profile/lorinda-cherry/

Some other material:

+ An interview between Michael S. Mahoney and Lorinda Cherry: https://gist.github.com/telemachus/c01e3a213574a7bdcf79a4802...

+ Some mentions and quotes in this oral history of Unix: https://www.princeton.edu/~hos/frs122/unixhist/finalhis.htm

I’d be curious to see her appearance on the Today Show that McIlroy mentions. She and [Nina Macdonald](https://www.ninamacdonald.com/pub.htm) were on in May of 1981 to talk about Writer’s Workbench, but so far I can’t find any video.

Using dc on a regular base whenever a small calculation is required - it's just so much more convenient compared to starting up a graphic ^H^H^H GUI calculator when you're in a shell+keyboard flow. And eqn is also much, much easier to use for basic high school math and casual use than all-mighty TeX. Used to document entire app suites using troff/eqn/tbl/pic in the 90s, order receipts and bills even, as well as preparing moderately math-heavy course material.

RIP

I pretty much always have an open terminal window running dc for doing quick calculations throughout the day. And I do my taxes every year in dc!

Does anyone have a link to her dog show judging bias paper? I'd love to read it.

I knew some people on the dog show circuit a long time ago, and each judge having obvious bias for random dog aspects / ages was well known. Since dog breeders and serious dog show people would have many active dogs simultaneously, for each judge at an event, they would enter their closest matching dog to what the judge liked, and skip entirely if that wasn't your dogs. Thus if you had eight dogs you were currently showing, and a two day weekend event with two judges, you would bring two dogs, one picked for the Saturday judge and one picked for the Sunday judge.

All yall going on about her programming and that’s great but she was a rallycross racer for 21 years!!!

I remember her from my summer job in 1980 in the Computer Library at Bell Labs in Murray Hill, NJ. She was one of the few women in computer science department. Nice bunch of brilliant people who patiently answered my newbie questions about Unix, Shell, C, etc.

I love the fact that HN celebrates the largely unknown tech heroes of our times :-)

I've used "bc" quit e a few times, thanks & RIP Lorinda.

Here's a good interview with her [1] which I happened across indirectly via a discussion here about the most surprising Unix programs [2]

[1] https://www.princeton.edu/~hos/mike/transcripts/cherry.htm

[2] https://news.ycombinator.com/item?id=22574603

Hey! She was having the same problems we're having:

> 'Cherry: That's hard to say. If you look at the Berkeley Unix system and some of the commands that are similar, the same in Berkeley as what we have here but you look at the Berkeley manual they've added 85 flags to the Cat command or something. It was a very simple elegant thing that did a very simple job. I guess we've always had the attitude that it has to be really useful to be worthwhile putting in. Maybe just 'cause it was a smaller group than at Berkeley or maybe people in Berkeley, everybody needs to find a niche so they've got to put a flag on something, I don't know what the environment is there. But I think it was here to prevent featurism. I think that's the difference between the two systems. And I think that undoubtedly has to do with the university environment where everybody has to do something as opposed to the environment where in some sense everybody had to justify what it is they were doing to your cause. And there is also some hesitancy 'cause it you touched it you owned it, you thought hard about whether you needed to add that flag or whether there was some other way around it. Whether there was some program. You said "I'll find some other way to do this 'cause I don't want to own this program."'

I just did `man dc` and then tried it out to run 1 + 2. I tried only entering the numbers, then tried writing "push 1", then guessed, that maybe it is just "p 1" and then guessed, that I also need to "p +" and hit "=" and return. Voila! Figured it out intuitively! Great!

It's still one of the quickest ways to do math on the command line.

dc -e '3 k 57 47 / p'

Or what have you.

It's pretty useful. I recommend spending an hour or two with it. You'll use it.

One of the nice things about rpn is if you forget something or want to modify things, you'll quickly find out that it's more convenient.

A little syntax weirdness apparently goes a long way in making things ridiculously easier to manipulate.

You can write programs with it btw. Hobby away: https://rosettacode.org/wiki/Category:Dc

Things get delightfully cryptic: https://rosettacode.org/wiki/Generate_lower_case_ASCII_alpha...

Go through a Forth tutorial to get the hang of RPN style programming if that's truly gibberish to you. It won't make it easy, but it will make it readable

Haven't used a language with RPN as default, but Forth is somewhere on my endless "to look at at some point" list : )

Unless there's a difference between yours and mine `dc` that's not quite right:

   10 # pushes 10
   20 # pushes 20
   +  # pops 20 10 sums, pushes results
   p  # prints top of stack (peek)

Equal (`=`) is a conditonal macro invocation `=r`: pops two values off stack, invokes (the contents of register) `r` if they are equal.

  [p]sa # store peek stack top as macro in a
  1337
  10
  20
  +
  30
  =a # Stack should be 1337, 30 (20+10) and 30, so should output 1337
     # pops 30 and 30; 30==30;p -> top is 1337

I use dc almost every day. I even do my taxes with it. I will forever be in her (and others) debt...

Glad I'm not the only one.

I started using dc for shell scripts before I knew that expr existed, and when some UNIX systems I logged on to spat out a bunch of noise on the console whenever bc was invoked. As a result I started thinking in the stack arithmetic way, and just adopted it as my standard way of doing math on UNIX. Eventually I got so used to it that I also preferred using it on Windows (via Cygwin) over calc.exe. It was a sad day when the developer world moved to Git Bash which either never included dc or stopped including it at some point. After a few years of grumbling with calc.exe and ancient versions of GNU dc that didn't work quite right in a Git Bash console, I discovered Gavin Howard's dc, which works well in Git Bash, and now it's once again one of the first things I install on a new computer.

Remarkable how this program written ~50 years ago is still useful today, in whatever rebirthed incarnation. I can't think of many others with such staying power.

It's good to hear of another person using my `dc`! You're actually the second confirmed user; the other is a die-hard original Unix user that now uses NetBSD.

That said, I'm glad my `dc` works for you under Git Bash, but it's weird that GNU `dc` wouldn't work.

I can't remember exactly the problem - it was some time ago - but I think it might have been related to console input. Either it was built for Windows console and didn't quite handle mintty LF, EOF and line wrapping, or the other way around. Usually you can wrangle programs into working by using winpty, but it still didn't operate quite as smoothly as dc did under UNIX (or Cygwin). In the end I mostly ended up invoking it in a pipeline instead of interactively.

Fortunately your dc worked out of the box with winpty a couple years back, and - to my delight - I just discovered the latest version works both in a modern Windows console and Git Bash mintty as well, without any winpty wrapping. Great work, thank you!

You're very welcome.

vi, in its various incarnations, comes to mind.

Given the antiquity of dc, ed would probably be a better match. And nobody willingly uses ed anymore...

I <3 RPN and depend on dc every day and use it for calculations in shell pipelines! RIP.

> In these years, Cherry recalls, the potential of the computer had barely been tapped, and if asked what she did for a living, she would say that her job was to “see what kind of neat new things I can make the computer do, and in those days the computer wasn’t doing a lot, but it was super interesting and there was a lot more stuff you could make it do.”

It would have been so different. What we have now must feel like an Eternal September to anyone who was around back then.

Don’t you just ache to have missed it?

Oh to have been forging that frontier!

Wonder what frontier we'll look back on in the same way in 50 years.

Yeah, I was wondering the same, I know for sure it's not the industry I'm in, which is now "mature" and boring. It's probably something we're not even familiar with because it's so niche. Or maybe it is something that we know, but to our eyes it doesn't seem that powerful. In the same way that people used to think of computers as just tools for doing multiplications and other arithmetic operations and they could have never imagined everything else that computers do.

Wavey memories of typesetting my graduate thesis with eqn, nroff and dvi to postscript things all in a long Unix pipe chain. Thanks Lorinda.

"The eqn program was created in 1974 by Brian Kernighan and Lorinda Cherry. It was implemented using yacc compiler-compiler.[1]" eqn Wikipedia.

Interesting to know that eqn pre-dates TeX, TeX was released in 1978

bc is still my favorite calculator, 30 years after I first learned *nix (SunOS for me back then).

Similarly, I have used dc as my desktop calculator of choice since the mid-90s. It has been useful in many one-liner scripts too. A simple, powerful tool.

One-liners? dc is using reverse Polish, with newlines to separate tokens, or is there an alternate separator I didn't know all these years?

tokens are separated on parsed input, where the token needs to be explicit, like two numbers in a row a space can separate them. An example from the manual. the first ten values of n!

[la1+dsa*pla10>y]sy0sa1lyx

I like her quote - “ see what kind of neat new things I can make the computer do, and in those days the computer wasn’t doing a lot, but it was super interesting and there was a lot more stuff you could make it do.”

Still true today!

She's a hero to me, I use "bc" at least a couple of times per week.

Thank you to Lorinda and everyone else working on UNIX operatingsystems and their tools, you make my life better.

I use bc all the time to do quick calculations. I never knew who the author was. I wonder how the software world will change as a generation of authors or maintainers of everyday programs dies off. Not all software package maintainers/developers have planned how things will continue after their death.

The U.S. Government could fund some work into this, seeing as how a lot of open-source software is essentially a public good.

Is there a runnable version of her "typo" program anywhere? From the description at least, it sounds like a really cool idea.

There are some links to various versions here: https://news.ycombinator.com/item?id=22582644

Many thanks Lorinda for bc. A super useful tool. Rest in peace.

RIP. I use bc in production for calculations in the shell.

> She worked on several influential mathematical tools, including a desk-calculator language (bc); TeX and eqn, both typesetting systems for publishing mathematical formulae…

Saying she worked on TeX seems incorrect. Can anyone confirm either way?

I think the lovely comment by svat covers the extent of her contribution to TeX: the eqn language informed the design of TeX's math-mode syntax. But that's all I'm aware of; it seems the much-copied quote you mention is off base. Keep in mind that TeX was initially written in Sail, which had no Unix compiler, and also assumed 36-bit words, which no Unix system had. The later rewrite of TeX that ran on 32-bit machines, and finally became available on Unix, pretty much retained the Sail version's existing math syntax.

Thanks for everything, I love bc, use it relentlessly. Rest in peace.

> She then joined the (self-organized) Unix team, collaborating on several applications with Bob Morris.

Bob Morris Sr, the grandfather of The Worm, I assume?

I think the thing in her obit that most delighted me was the note that she developed an analysis of judges' decisions for dog competitions.

for me it was the rally car racing:

https://web.archive.org/web/20120624081721/http://www.scca-n...

Rip, Lorinda. I'm using bc almost daily.

I use those every week of every year.

really a lovely thread, as an aside here is nice Nimplementation of dc called ad: https://github.com/subsetpark/ad

I use bc a hundred times a day, it's invaluable. Thank you; RIP

Still use bc

i use bc just about every day, thank you!!!

big fan of bc. love that program

scale=0

Brian Kernighan is not mentioned in the obituary and nor should he be. Can we please change the HN title?

Sure. I changed it. I put it there because I expected some pedant to say "She didn't write eqn, Brian did"

There is a very good non-obit by Doug McIlroy, written when Lorinda got an award in 2018. I was going to post it, but it's in a small listserv archive and I think the list members prefer not to be cited into these kinds of thing. I can't find the text elsewhere which is a shame, it has a really good rundown on her time at Bell Labs and the amazing things she did.

I guess i could cut-paste it here, but that would exceed a 10% quoting rule I try to stick to. They're Doug's words, not mine.

Thank you.

It'd definitely be an interesting read, could you pastebin it for us?

4 levels down a comment threat is probably deep enough it doesn't cause problems. The list archive is actually fully publicly visible:

    https://minnie.tuhs.org/pipermail/tuhs/2022-February/025390.html

That’s really good. Thanks again.

Hear hear!

All of our giants are passing away. Makes me a sad.

Who is the next generation? I can't think of any names.

I have two thoughts on this:

1) our “giants” will be Zuckerberg, Elon, and Bezos. They have taken all the talent that would have gone to bell labs and used it to generate ad revenue.

2) our “giants” are still making their name and working, and we will recognize their accomplishments in 30-40 years.

9front guys with plan9's legacy being modernized. Gemini. NNCP as a disaster-resistant network. Solar powered minicomputers and devices.

The future won't be like the 20th century one. It will be less power hungry.

I got solar a few months ago and I am shipping 7kWh back each day and trying to find useful stuff for the electricity (may even give in the Air Conditioning crowd prior to temps hitting 35 Celsius). I am all for efficiency but cheap solar changes the material conditions significantly.

Most giants I've met never make a name for themselves and prefer it that way.

How can you be a giant if you never make a name for yourself?

You must learn that people working on UNIX or on Xerox Park were not there looking for fame. There was no indication early on that UNIX would be revolutionary. They just did their job as good as they could and the future took care of itself. People who are just trying to make a name for themselves usually go nowhere.

There are probably several definitions of "giant" in play here.

> Elon ... have taken all the talent that would have gone to bell labs and used it to generate ad revenue

This is pretty unfair to Elon. He took all that talent and started putting things into space. And building the next generation of cars. He seems to be doing amazing work.

Bezos and Zuckerberg, sure. Musk is in a different league, possibly an entirely different game.

It’s perfectly fair (for some value of fair). Both SpaceX and Tesla would not exist today if not for US government money, yet Musk bitches endlessly about government interference, as if his own personal fortune wasn’t down to simply cashing out at the right time during the dot-com boom. Peter Thiel is right about one thing only: Musk is an entitled emerald scion braggart.

Ehh, lots of folks get billions of dollars of government money and have zero to show for it. At least with Elon we get cheap, reliable lift vehicles and 1-2 million EVs per year being built. It’s easy to handwave away his hard work because it costs someone nothing but contempt, but if it was so easy why was he the one to do it? Because it’s hard.

He’s obnoxious and yet surgically effective. Seems fair for the results, even when considering how much luck was a component (PayPal).

What kind of talent are we speaking of here? I would not consider either of them talented in SWE.

Musk is more of a great entrepreneur. His track record is quite impressive.

Your first point seems to be conflating workplaces and individual talents in order to make a snarky point about tech CEOs. I mean, I get it, but the "giants" that sergiotapia was referring to clearly aren't the people who ran Bell Labs, Xerox PARC, SRI and BBN. (Also, to be persnickety, Tesla and SpaceX don't make any of their revenue from advertising sales, and while Amazon does have a growing advertising services business, it's still a pretty small sliver of their overall revenue stream.)

Your second point seems more reasonable, although I think klyrs makes a cogent point: the "giants passing away" now are primarily the ones who made their names in the pre-internet age. The next generation after them are the folks who were making their names circa 1980–2000, and we can hazard some plausible guesses. It may be too early to say yet who the anointed giants of 2000–2020 will be.

I can’t edit my post, but I think there are good points here.

The “ad revenue” comment was snarky and for Elon and Bezos may not be fair because they have had some major impacts to quality of life - Elon with electric vehicles and Bezos with availability of goods, AWS, etc. They both have given really smart people some really interesting problems to work, and we may recognize that in stories as they come out in 30-40 years.

Thinking a bit more about it, I think it takes a company today to make an impact given the complexity of technology. Things like dc and bc have been coded and discovered, and now we’re building advanced products to do advanced things.

Underlying my original comment is my concern for the allocation of brain power in technology today. I don’t know what the right ratio is, but I do feel like we’ve got too many people going toward taking users’ attention and discretionary income rather than working more valuable (on a societal level) problems.

> our “giants” will be Zuckerberg, Elon, and Bezos. They have taken all the talent that would have gone to bell labs and used it to generate ad revenue.

If we compare these to people who contributed to basic science and technology, I think we must draw a line between industry giants and enterprise giants.

Morgan, Ford, Edison, Carnegie, Rockefeller, were enterprise giants.

Tesla, Holonyak, Salk, were industry giants.

Torvalds, Stroustrup, von Rossum, Larry Wall, Gates, Wozniak, to name a few.

Lorinda Cherry earned a Masters degree in 1969, so guessing she was born around 1948, meaning she'd be about 74 now. Steve Wozniak is 71, Stroustrup is 71, Larry Wall is 67, Bill Gates is 66, van Rossum is 66.

Torvalds at 52 is the most clearly "next generation" of those people.

John Carmack? Fabrice Bellard?

Dan Bernstein, Bryan Cantrill, Russ Cox, Udi Manber.

And of course we lost one of the brightest lights way too early in Aaron Swartz.

The only thing I know about Bryan Cantrill is that 'have you ever kissed a girl?' post he made to a Linux dev in the late 90s.

Funnily enough another hero, Miguel de Icaza, popped up in that thread too.

Carmack for sure, if we're talking about prolific producers who really advanced things outside of academia.

Can't help but think that those people mostly redid what their parent's generation invented, but in a more practical/commercial way, which is an achievement in itself, but of a completely different kind.

One of the oldest math precision calculator was written by a women programmer. That cool. What other well know programs were written by the women authors.

There are not few. Margaret Hamilton * was lead Apollo flight software designer at Nasa for example.

*: https://en.wikipedia.org/wiki/Margaret_Hamilton_(software_en...

Look up Ada Lovelace and Grace Hopper.

Reply on news.ycombinator.com