Gnuplot is not GNU | Hacker News Reader

Gnuplot is not GNU

gnuplot.info

232 pointscab404199 comments

Reading the comments, I get the impression that many people are somewhat lacking perspective.

I've been writing software professionally for more than 25 years, > 30 years in total. gnuplot has been around all that time. Developed in an academic environment, people probably (certainly) had other things to do than argue about name comparisons to other projects. And nobody cared.

There are things that are not an invariant of time. And frankly - open source along with the ideological battle over it is certainly one of those things. There were times with a much healthier way of dealing with zealots (Oh, this will cost me karma, but fortunately I don't care :-)

Embarrassing story time.

I once introduced Richard Stallman to a fellow grad student in Japan like this (in Japanese): "He is the creator of GNU. You know, as in gnuplot."

Stallman understood what I said and immediately corrected me: "gnuplot is not a part of GNU".

That sounds exactly like him.

I'm not sure if this is obvious from the gnuplot.info web page - not only is gnuplot not gnu (new), it is also not GNU (gnu project)

"Despite its name, this software is not part of the GNU Project." https://en.wikipedia.org/wiki/Gnuplot

Stallman speaks English, French, Spanish, and “somewhat flawed” Indonesian. — certainly he could not understand the difference in Japanese between that and saying that the two are unrelated?

He probably just understood "gnuplot", and was used to the confusion.

That is what I inferred, but it seems a bit ungentlemanly to jump to that conclusion after merely picking “GNU” and “gnuplot” up in a sentence and nothing more, which is what most likely happened.

But its reasonable to:

1. Conclude that its 90% likely to be what happened.

2. Not burden someone by asking a question you don’t need the answer to.

3. Disclaim credit for work you don’t lead.

I submit that adding "In case you just said gnuplot is part of GNU,” in front of it is not too long, and far less præsumptive.

In context of adding burden of having someone to translate for you in Japanese, the statement he made seems good enough and to the point.

Maybe you are far too pre-assumtpive that there was any ill effect, or even ill intent, in the statement being made, especially if the same fact is accounted for in gnuplot FAQ.

I mean what other reason would there be a reason to bring up gnuplot, if not for it's assumed connection to GNU?

I guess "GNU, but gnuplot is not GNU" could be one, but perhaps at that hectic moment RMS didn't consider that option!

> "In case you just said gnuplot is part of GNU,” in front of it is not too long

No longer than, say, adding GNU/to the front of GNU/Linux.

Of all the things I've heard RMS accused of, being "gentlemanly" has never been on the list

Perhaps it's only fair, then, that it's pointed out when he actually seems to have been?

You can have a very low level of understanding in a language, as in no more than a few hours of experience, and understand what people are talking about. You don't get the details or have the most basic conversation, but you can piece together the few words you understand.

It is a cautionary tale: never assume that because someone doesn't speak your language, he won't get it if you are bad mouthing him.

Here "blah blah, Stallman, blah GNU blah blah gnuplot" is enough for some not skilled in blah to understand what is going on.

If you have to keep correcting people about your product's name, you made a poor name choice. Names, like everything else, have a U/X aspect to them.

> If you have to keep correcting people about your product's name

What product? It was created by two people in a lab that didn't know whether it would ever have more than a dozen users. Gnuplot as a name is fine imho, barely anyone cares whether it's a GNU project or not. JavaScript still has Java in the name and all you see about it are jokes.

GNU cares.

I think this applies to LaTeX as well. The fact that people pronounce it "lateks" is entirely the creator's fault.

Yeah, but pronouncing TeX "tek" is a bit of a shibboleth among mathematicians. The use of Greek letters is common enough in math papers to make it a natural inference (it's not an English "X", it's a capital Greek χ [0])—at least within the field. And if you don't know it, when one of your professors explains it to you you'll feel like they let you in on a little secret.

[0]: https://en.wikipedia.org/wiki/Chi_(letter)

Also PNG, which in an attempt to prevent another gif/jif debacle wrote in the specification that its pronounced "ping"

WTF. There's probably a lot of things in the world for which most users have talked to someone who has read the manual. But did the authors expect PNG to belong in this league?

I’ve always pronounced it like that. Do some people pronounce it differently?

I've always pronounced "png" as "pee-en-gee", while pronouncing "gif" as in "gift"

I have some friends who strongly assert only the "jif" pronunciation is correct and the "gif" pronunciation is incorrect due to the stated intention of the creator of the format. I can't wait to experience their transition from PNG to "ping" when they learn they've been "mispronouncing" it all along. I wonder if I can convince them to use aluminium?

Wait, but an argument from original intent on aluminum/aluminium would go towards using aluminum [0]. That was the original published name, and the change to "-ium" only happened shortly after. We could go even further on that route and argue that the entire pattern of ending metals in "-ium" rather than "-um" is a neologism based on a misunderstanding. Of the classical metals [1] such as gold (aurum), silver (argentum), lead (plumbum), none of them end in "-ium". So where does that come from?

The best explanation I've heard is that it comes from the timeline of when different metals were discovered [2], back in the 1700s. Magnesium was discovered, and was named after Magnesia, a region in Greece. Basically, "magnesi-" + "-um". Then Barium was discovered, and was named from the Greek "baryta", meaning "heavy". Molybdenum followed the "-um" trend, and "tungsten"/"wolfram" ignored the discussion altogether.

And now we get to the first mistake. Tellurium was named from the Greek "tellus", and should have been called "tellurum". Instead, by analogy from magnesium and barium, the suffix "-ium" was added instead. Strontium came next, after the town of Strontian. This reinforced the trend of the "-ium" suffix, even though it came from "stronti-" + "-um". By the time Zirconium was discovered from the mineral "zircon", the association was cemented. So many metals had ended with "-ium" by sheer coincidence, that people assumed that that was the correct Latin suffix.

Bringing it back to the "aluminum" vs "aluminium", this means that there's neither authorial intent nor historical consistency in the use of "aluminium". I'm fine if people think it sounds better, and argue from that, but for the love of all that is holy, people need to quit pretended that it has any more validity or correctness to it.

[0] https://en.wikipedia.org/wiki/Aluminium#Coinage [1] https://en.wikipedia.org/wiki/Metals_of_antiquity [2] https://en.wikipedia.org/wiki/Timeline_of_chemical_element_d...

I don't think it has more authority or correctness. That's part of the joke. I've actually read various incarnations of this argument on the internet (mostly carried out on Reddit), and at this point I consider it to be somewhat of a meme. What I remember from all of the times I've seen it brought up is that some international standards committee declared and it to be "aluminium" and so some internet denizens (especially Europeans who already tend to pronounce it that way) proudly declare this to be the authoritative be-all-end-all resolution to the subject, whereas the other side (typically those speaking American English) remind everyone that the person who made the original discovery considers "aluminum" to be correct.

Why I am grouping "gif", "png", and "alumin[i]um" together here is that people consider some nebulous authority to be the prescriber of all things pronunciation. I personally think that pronunciation is derived from how it is commonly spoken, and not the other way around. Hence its inclusion in my playful ribbing. If I appeal to the correct authority, can I make them change their pronunciation? Maybe it wasn't very funny.

Ah, got it. I thought that the joke was that you were tricking them into using a less favorable pronunciation on the basis of some perceived authority, rather than the joke being that there would be such an authority at all.

Same here on both counts!

I've mostly only heard Pee-En-Gee

Fair enough. The comparison to GIF was leading me to the idea of a different pronunciation rather than just simply reading the letters.

Other than friendly banter over a beer I've never given af how someone pronounced as long as they used one of the commonly "accepted" versions. Also never heard -anyone- call png "ping"

I pronounce it “ping”, but that’s because I’ve read the spec. I e never met anyone else that pronounces or that way.

Then you haven't met me (or if you have, PNG hasn't come up), because I also read the spec.

pee-enn-gee?

When I was at university there was a discount food store in that city called Matex (mat is Swedish for food). Me and my study friends we pronounced the store as MaTeX similar as LaTeX instead of the normal eks sound at then end, techies giggled and everyone else looked confused.

Heh, we did much same at my (Danish) university with the store FøTeX.

Even Google gets it wrong: https://imgur.com/a/nswxueq

It gets TeX just as wrong. Interestingly enough, GNU and SCSI get reasonable "featured snippets" instead.

I almost tend to think these are intentional to tell the in-crowd apart, like a shibboleth. Including pronouncing SQL as sequel and so on.

SQL is commonly pronounced as "sequel" or "S-Q-L". Is one really more correct at this point?

When David DeWitt (of "DeWitt Clause" fame) introduced SQL in while teaching CS564, he made it clear that SQL was pronounced as three letters.

I recall him saying that Sequel was a distinct language. The info I can find on SEQUEL today suggests he was trying to distinguish modern (1994) versions of SQL from the early versions. Or maybe he was afraid someone in a suit[1] would show up to admonish him for improper use of a trademark[2].

1. https://news.ycombinator.com/item?id=15887145 2. https://en.wikipedia.org/wiki/SQL

Obviously the only correct pronunciation is "squirrel"

And the only correct spelling of "squirrel" is sqwrl.

To be truly one of the cool kids, and not a trend follower at all it is:

"Structured Query Language"

For the super cool who use the one true vendor it is:

"Microsoft Structured Query Language Server"

S-Q-L is a lame normie version, if you are really a leet haxor you know about the obscure information that it's sequel.

I hate that I always have to look these things up to not look out of place. For example nginx is engine-ex not en-jinks. They often seem intentionally hard or ambiguous to pronounce.

It's kind of ironic because I only learned the "correct" pronunciation when I got a boring office job at an American tech company. When I used the n00b pronunciation I was in college, and way more of a 1337 h4x0r.

So as a shibboleth it mostly functions to distinguish those who discuss tech in English with their colleagues from those who do not. The latter group is quite diverse; it includes self-taught people, people communicating mostly in writing, and people who speak other languages than English with their colleagues.

> For example nginx is engine-ex not en-jinks. They often seem intentionally hard or ambiguous to pronounce.

I think it's a side-effect of the desire to create a name for your project that is both short and unique.

>I think it's a side-effect of the desire to create a name for your project that is both short and unique.

Wheel Of Fortune has taught us that vowels are expensive, so best not have any in your name.

That is kind of funny to me because where I live the S-Q-L vs sequel divide is more along the linux/microsoft divide.

Meaning if you encounter someone here that pronounces SQL as sequel you probably found a DotNet developer or at least someone that primarily focuses on Microsoft technologies.

S-Q-L is used by pretty much everyone else here.

I have only ever heard it pronounced as "S-Q-L". When I had a customer on the phone going on about their "sequel" years ago I genuinely had no idea what they were on about, only only when they said (after a few minutes) "Microsoft sequel" did it finally click it was SQL.

I mostly have a background in Linux/Unix-y stuff, and I've hardly ever heard someone say "sequel" since, so this seems about right.

Not that I care either way, pronounce things however you like, but it sure can be confusing!

Have you met anyone who is confused about how to pronounce nginx? I haven’t, but I have met _a lot_ of people who have trouble writing it correctly.

It all started with all good .com domains being taken, and now, here we are.

I did not know how to pronounce it correctly up until last year, when I first heard it spoken by a colleague.

I assumed it was n-jinx for years, until I happened to discuss it with a colleague.

Same with Apache, since in Europe it typically gets pronounced appa-sha (as it would be in French)

Apache is an awkward one, because it's a Spanish loanword from Nahuatl, which is further distorted by English speakers.

I can't find a website that does the Nahuatl pronunciation, but my guess is that it's closer to the Spanish than English

https://www.spanishdict.com/pronunciation/apache

> Apache is an awkward one, because it's a Spanish loanword from Nahuatl

Probably North American native language (though there is a less-accepted theory of entirely-Spanish origin), but almost certainly not Nahuatl. Per Wikipedia, the dominant theories are Zuni and Yavapai.

The correct pronunciation of Apache is "a patchy" as in "a patchy HTTPD server". The pronunciation of the Spanish and Nahuatl etymological origins of the other side of the pun are irrelevant.

Unless you read the very next paragraph on the wikipedia article...

> Brian Behlendorf, one of the Apache's creators, asserted that the origins of Apache were not a pun, stating:

> > The name literally came out of the blue. I wish I could say that it was something fantastic, but it was out of the blue. I put it on a page and then a few months later when this project started, I pointed people to this page and said: "Hey, what do you think of that idea?" ... Someone said they liked the name and that it was a really good pun. And I was like, "A pun? What do you mean?" He said, "Well, we're building a server out of a bunch of software patches, right? So it's a patchy Web server." I went, "Oh, all right." ... When I thought of the name, no. It just sort of connotated: "Take no prisoners. Be kind of aggressive and kick some ass."

So, no, it's based on a stereotype of a group of tribes indigenous to North America. The pun endures because the indigenous culture did not.

Maybe we should break down the semantics of WYSIWYG while we're at it.

SQL is one that I regularly flip flop just to annoy people :) .

I’m on the S-Q-L side of saying that both is correct, and the accepted answer below agrees with both of us.

https://english.stackexchange.com/questions/7231/how-is-sql-...

I pronounce it "squeal like a pig". "Squeal" is as good a pronunciation for SQL as "sequel", and indeed better because it saves a syllable. But after "squeal", who can resist adding "like a pig"?

"as good as", I guess, other than the point of language is to transmit ideas, and if you're the only one that says something in a certain way for some pedantic "it's as good as" reason, you've failed. I guess reasonable people can disagree.

I think that a point of language is to express ideas. Bullseye.

There are many points to language - quite a porcupine....

I wonder what the split is between “O-S-Ten” and “O-S-X”.

I suspect -X is more common.

thankfully apple solved that for us by switching to "macOS"

ma-cos?

Is that Klingon for "shit"?

I do hope the LaTeX folks remember to always spell Yahoo! with an “!” !

Even Yahoo forgot about the ! :(

I have jokingly pronounced it Yay-hoo for so long that I am now unable to stop.

The correct pronunciation hardly matters for the marketability of the product. — it is usually understandable.

In Linux' case, Linus Torvalds has said that he doesn't really care and admits that he instinctively pronounces it differently depending on the language that he's speaking.

TIL that I do the same and each of the five letters sounds subtly differently in English and Russian.

Oh noes. I'm one of those people who keep calling it that. What is the correct pronunciation?

La(like law)-tech

> La(like law)-tech

That's definitely not how you pronounce LaTeX.

Wikipedia recommends LAH-tekh or LAY-tekh.

https://www.youtube.com/watch?v=LJYBqvTllgY

https://www.youtube.com/watch?v=5nQeXvwx6gE

IPA is really needed here. When you wrote LAH I wasn't sure if you meant /lɑ/ or /læ/.

Likewise for "law": is it meant to be /lɑ/ or /lɔ/? It's not clear because the word is pronounced differently in different dialects.

For reference, Wikipedia recommends /ˈlɑtɛx/ and /ˈleɪtɛx/, and the two linked videos use /lə'tɛk/ and /ˈleɪtɛk/ respectively.

I’m baffled that IPA is ever needed. I’ve always preferred this style pro-nun-see-AYEE-shun.

What is lacks in perfect clarity it makes up for in instant readability.

How on earth could LAH be ambiguous now?

Right, so whoever wrote the software decided to give it a name identical to an existing English word, but would then balk at it being 'mispronounced'? If the developer didn't care enough about avoiding confusion then why the hell should I?

Case-sensitive naming is dumb.

I'd say "la (like law)-tech" is pretty much the same as "LAH-tekh", minus the indicated emphasis.

Your accent exhibits the "cot-caught" merger, then. Those are two different sounds in most accents.

Guess it depends how you pronounce 'law' then ha! I can't start 'law' with 'lah' no matter how hard I try - 'lah-ore' maybe?

Ah, glancing at your profile I think you're hitting the traditional English problem of pronouncing letters which aren't there :) "To be forced to sawr into such rawr meat should be against the lawr"

It's nicely counterpointed by the other habit of not pronouncing letters which are there, like in Worchestershire.

I always say latex just because it makes pedants grit their teeth, and then just for fun randomly pronounce it correctly. A lot of CS people need to learn to chill a bit.

Yeah but in English, if everyone pronounces it wrong and you wait 100 years (now more like 20), it will become the default pronunciation via a special rule used by speakers and codified nowhere.

The first release of gnuplot was only three years after the GNU operating system was first announced to start development, and long ere it became widely known.

Look a bit lower in the FAQ, apparently they were at some point associated with the FSF. Their license also seems copy left to me. So the name wasn't that wrong and since most end users don't care about these details it probably doesn't make sense to rename it now.

> So the name wasn't wrong

I'm not sure what you mean by this. The name is a pun on the name of the gnu animal. Just because GNU also made the same pun doesn't mean they are related, or that they have anything to do with each other.

gnuplot's history goes back to 1986, when GNU was in its infancy. The name was fine then, and is probably fine now.

> Their license also seems copy left to me.

The license is interesting! I'm surprised they've stuck with it. But yeah, it does appear copyleftish.

Naming things so that people can try to understand what they are and do from the name is the most fundamental act of UX. I feel like too much software goes for a cute brand name, marketing approach.

Do you think the same applies to other products too or just software?

What is a Linux? Android? Google? We are not living in a neatly planned world, can’t just name things General Kernel(tm), General Mobile OS and Search Engine. And people like to give names to even inanimate objects, it is marketing, but at the same time you do remember them better that way. Would you rather start Image Editor, or Photoshop? In the former case it may very well just be Paint, while you do know what to expect from the latter. Also, should we as well get rid of icons/logos?

As a product name, MacOS did not exist until version 7.5 if I recall correctly. Prior to that, it was Macintosh System Software, the system software, or simply System <version number>. Neither icon, logo, nor typeface were associated with the system software throughout most of that period. That seems very out of place for a company that is very much conscious of the value of brand image.

I’m fairly sure people referred to it as Macintosh, simply.

Quite possibly. The 80's and 90's were an awfully a long time ago, so I can only refresh my memory from what's on the screen and in the manuals.

For what it's worth, the about box on my older PowerBook shows "System Software 7.1" and the manual simply says "system software" (i.e. not a proper noun).

When I started using the Mac, System and Finder had their own versions. So power users would say something like I’m running System 3.2 and Finder 5.3.

I've always wondered what would happen if you gave a popular product an impossible name.

Can you name a company with Unicode symbols?

My guess is that a popular made up name by the user base would emerge.

Which makes me guess that a bad name is worse than no name. Gnuplot misleads.

There must be other examples, but you're right in the famous case of Prince changing his name to an abstract symbol to get round a contractual issue he was having with his record label. Apparently journalists started saying "The Artist Formerly Known as Prince" though apparently it didn't stick? Unfortunately I'm not enough of a Prince fan to know what his fans did?

https://www.independent.co.uk/arts-entertainment/music/featu...

It stuck enough. I'm not really a Prince fan, but "The Artist Formerly Known as Prince" is what people called him for pretty much my whole childhood. He didn't use the name "Prince" 1993-2000, but it's my recollection that a lot of people kept calling him "The Artist Formerly Known as Prince" well in to the mid 200x's. Sure, eventually people went back to calling him just "Prince" once he was using that name again, but "The Artist Formerly Known as Prince" absolutely did stick for a number of years.

There is also a band named , or something to that effect. I used to have quite a nice album from them but I can't find it now (I'm not even sure if those are the correct Unicode block symbols).

I think intentionally being obscure is kind of the point for a lot of these things.

A good example of a product with impossible name is the 1726 book "Travels into Several Remote Nations of the World. In Four Parts. By Lemuel Gulliver, First a Surgeon, and then a Captain of Several Ships". Everyone just calls it Gulliver's Travels.

Prince changed his name to a symbol.

“ After the change, Warner Bros. had to mail out floppy disks to news organizations featuring a font that allowed for reproduction of the new Prince symbol, because the symbol itself cannot be replicated in the fonts most publications use.”

https://www.vox.com/2016/4/21/11481686/prince-name-change-sy...

/. was done in this vein.

fp!

Yes, the page with the famous URL http colon slash slash slash dot dot com

FWIW, it took me about 10 years to understand this joke, as a non-native speaker who would read "dot" as "punto".

For a very long time I thought it was a play on "./" in the unix sense.

I the old Presto Opera browser if you entered just /. in the URL bar it would go to slashdot.org

org

My guy, gnuplot came like 3 years after the GNU project. It's like saying Charlie Chaplin chose a poor moustache because Hitler made it look bad

I am a thankful user of Gnuplot. As some before me mentioned, the naming history was complex. What is more important is that the developers provided - and still support - an excellent program that we may freely use under a very reasonable license.

Why not just change the name to GNG (Gnuplot's Not Gnu)?

It probably doesn’t matter enough. There’s limited downside to people thinking this is an official GNU project. This is a minor entry on the FAQ, not a Crusade Against Wrongness like “GNU vs GNU/Linux vs GNU+Linux”.

>... the name "gnuplot" was actually a compromise. I wanted to call it "llamaplot" and Colin wanted to call it "nplot." We agreed that "newplot" was acceptable but, we then discovered that there was an absolutely ghastly pascal program of that name that the Computer Science Dept. occasionally used. I decided that "gnuplot" would make a nice pun and after a fashion Colin agreed.

Sounds like a change to camel plot (cplot) might make more sense (another even-toed ungulate, not easily mistaken for a gnu)? ;)

The moment they announce that we will learn of someone GPG Next Generation (short: GNG) effort

I am sure they will be kind enough to rename themselves to GNGNGNG.

This is, of course, a representation of the noise you make when you think about silly program names and all the chaos they can cause.

Good numerical graphics

Funnily they refer to some people wrongly using "Gnuplot" instead of "gnuplot" yet they use "Gnuplot" 23 times in this FAQ.

They don’t say “Gnuplot” is wrong. They even include themselves (“many of us”) in those people that don’t want to start sentences and title words with lower case. My reading of that paragraph is that while “GNUplot” is definitely incorrect, “Gnuplot” may naturally occur due to capitalization, just like with any lower-case noun.

But only because it's the fist word of a sentence all those times.

“ 1.1 What is gnuplot ?

gnuplot is a command-driven plotting program. It can be used interactively to plot functions and data points in both two- and three-dimensional plots in many different styles and many different output formats. Gnuplot can also be used as a scripting language to automate generation of plots. It is designed primarily for the visual display of scientific data. gnuplot is copyrighted, but freely distributable; you don’t have to pay for it. You are welcome to download the source code.”

If it’s lowercase in title font, it should be lowercase everywhere.

> “ 1.1 What is gnuplot ?

> If it’s lowercase in title font, it should be lowercase everywhere.

If you read closely, you’ll notice that only “x” level headings use title case, “x.y” level headings use sentence case; there is no clear instance of “gnuplot”, capitalized or not, in title case anywhere in the document. The only arguable case is the page header “gnuplot FAQ”, but since it is sui generis there is no way other than inferring from capitalization in the rest of the document of telling whether its titlecase or just the preferred labelling of the product followed by “FAQ” (and the context suggests the latter).

Did you read this part?

You see people use "Gnuplot" quite a bit because many of us have an aversion to starting a sentence with a lower case letter, even in the case of proper nouns and titles. gnuplot is not […]

Most of us don’t care about forcing the first letter to be always lowercase, and in fact feel it’s a lost cause. But I agree that after such an explicit clarification I would expect to see “gnuplot” everywhere in the official documentation, except for this particular excerpt and when they refer to “GNUplot”. Otherwise it is clearly inconsistent with what they are explaining.

> Did you read this part?

Yes, I also read the body text after the numbered heading in the prior excerpt, which makes two different choices in sentence-starting position in sentence case. But what I was responding to was the inference of a rule bases on usage in “title font” (which seems to be a reference to title case), not a complaint about the inconsistent use in sentence case. Hence, the except you provide here isn't really germane to anything I've said.

They very correctly mix it to show that it does not matter. Better make some plots then bikeshed this to death.

If you look at the quotation, they mix capitalization in the body text.

> If it’s lowercase in title font, it should be lowercase everywhere.

The section titles are in sentence case, not title case. You would not expect it to be capitalised unless it were the first word of the title. The inconsistency in that paragraph is in the first sentence.

Because it's the correct way to spell it. Names are spelled with an uppercase letter in the beginning and all the other letters are lowercase. "I saw John" vs. "I saw john" vs. "I saw JOHN". First one is correct. Even if John tells you "My name is spelled with a lower case j". Or if John writes: "this orange thing is a kaRROT", it's still a carrot.

You can write whatever you want. You can also write "Tis is me bunni and he likes KArRoTs" and have a lot of friends agree with you that's the correct spelling. And it is still generally considered wrong :)

There are many names that do not follow your rule, and generally speaking I'd find it very rude if I told you my name, and you'd go "let me fix that to the _correct_ spelling". For example a name as Angus MacGyver, or Armand de la Cour. Both don't follow your rule. I don't see why that wouldn't apply to brand names as well.

Counter examples: “iPhone”, “eBay”, or people like Norm deSilva [1].

[1]: https://en.m.wikipedia.org/wiki/Norm_deSilva

Wikipedia often rejects corporate case styling and uses the sane "first letter upper case, others lowercase" for some brands. For example Wikipedia writes Nvidia instead of NVIDIA, although the company always capitalizes its name without exception.

No, Wikipedia does not make the judgement call on whether to "reject" corporate case styling on its own, it follows what is "in widespread use":

> Follow standard English text formatting and capitalization practices, even if the trademark owner considers nonstandard formatting "official", as long as this is a style already in widespread use, rather than inventing a new one: (But see exception below under § Trademarks that begin with a lowercase letter.)

> use: Time, Kiss, Asus, Sony Mobile. (Capitalize IKEA, IBM, as acronyms/initialisms.)

> avoid: TIME, KISS, ASUS, SONY Mobile

https://en.wikipedia.org/wiki/Wikipedia:Manual_of_Style/Trad...

One thing that entry does not cover is all lower-case names. Although a couple things I looked up (gnuplot, xkcd), Wikipedia does seen to respect all lower-case. I'm not sure how many companies actually have all lower-case names--even if their logo is all lower-case. And my observation is that even many projects that are nominally lower-case, aren't very consistent about it as in the current example.

Relevant Manual of Style entry: https://en.wikipedia.org/wiki/Wikipedia:Manual_of_Style/Trad...

Quite arbitrary. They disallow all caps ASUS and NVIDIA, but allow all-lowercase xkcd, because I guess xkcd is "ours" while big corps are the "Other".

That is not true.

> use: Time, Kiss, Asus, Sony Mobile. (Capitalize IKEA, IBM, as acronyms/initialisms.)

> avoid: TIME, KISS, ASUS, SONY Mobile

https://en.wikipedia.org/wiki/Wikipedia:Manual_of_Style/Trad...

Not sure. There's GNU after all. Wikipedia's "arbitrary" standard is actually pretty common in my experience. But I'm not sure I could come up with a coherent explanation as to why.

GNU is an acronym, that's fine. But if they insist on Nvidia, they should also insist on Xkcd (it's not an acronym, just an unpronounceable artificial word).

And radar is an acronym too and it's not commonly capitalized although SCSI is. And, often, Fortran, not FORTRAN. I'm not making a case for how things should be. I'm just observing how they commonly are in style guides, etc.

I think part of it is a general stylistic distaste for having "unnecessary" caps. See also general shift away from "Open Source," "Big Data," and the like.

Do people actually write eBay? It looks as dated as eMail.

iPhone is definitely an example of marketing winning over grammar, but it's not a given. Plenty of style guides indicate "our company name is to be written in all caps", when they want to stand out, which the rest of us happily ignore.

>Do people actually write eBay?

Yes?

In my experience, publications and the like do tend to respect the capitalization and other choices of a company/project/etc. with perhaps some exceptions like the ! in Yahoo! and perhaps all caps as in NVIDIA. In the case of Wikipedia, I'd go so far as to say that they're simply wrong if they're ignoring the actual way a company name is officially styled.

Mind you, I'm not necessarily a big fan of case-sensitive everything but here we are.

You'll see me even write e-mail!

Names should be written as they were created, grammar regarding them only comes into question in case of suffixes or the like.

I can mostly agree with people's names, especially stuff like cultural variations or things that are basically name prepositions like de/Mac/o etc. That said if you decided to change your name to "JOHN SMITH" don't be surprised if that decision was less respected and you were referred to as "John Smith" following capitalisation norms.

Companies don't get even that assumption for me, sure I'll use "iPhone" because they've won that one and it's the cultural expectation, but the only time I'm using "FREE NOW" is to criticise that decision

Maybe. Probably. But in practice no one really cares other than editors and pedants. I'd probably only bother if I was showcasing something to people from that particular entity.

Not all names:

danah boyd - https://en.wikipedia.org/wiki/Danah_boyd

brian d foy - https://en.wikipedia.org/wiki/Brian_d_foy

https://en.wikipedia.org/wiki/List_of_people_with_lower_case...

People are perfectly free to spell their names however they want. Good luck to them in getting everyone else to spell them that way, though. And even more so, good luck not having other people think they're being pretentious wankers.

And even in cases where the lowercase starting letter is somewhat accepted, the start of the sentence rule is still left untouched by most writers. Iphones come to mind.

How do you spell iPhone when starting a sentence?

>Iphone is a phone made by Apple

In my native language, the general advice from our language council is to NOT listen to corporation’s rules. You would write “an iphone”, and casing as usual. In fact you wouldn’t really use iphone at all, but smartphone. The idea is that the name casing rules are for showing respect, and an inanimate product does not deserve the same respect as a person.

OTOH you would say Apple though, but I guess that avoids some obvious ambiguity

I don't know what your native language is, could it be Swedish? In Sweden, media consistently write Iphone, because it's a name, and names start with upper case.

As you mentioned, what the marketing department says about the case of letters in the name does not affect this generic recommendation.

Of course, you'll find a lot of people writing it in accordance with the marketing style, bit typically you'll see media follow the Swedish spelling rules.

> In fact you wouldn’t really use iphone at all, but smartphone

The sentence "Smartphone is a phone made by Apple" (or maybe "A smartphone is a phone made by Apple") would have a very different and wrong meaning though, so that doesn't make much sense to me.

To be fair, they mostly advise against GNUplot, and mention Gnuplot because of the aversion to start sentences with lowercase (which is also where they use it with uppercase).

Doesn't explain

  8.4 Why does gnuplot ignore my very small numbers?
  For some purposes Gnuplot treats numbers less than 1e-08 as being zero....

Proper nouns should begin with a capital letter?

But only in answers, not in questions.

Later in that same answer:

  The solution is to change gnuplot’s idea of "zero"

:-)

I always refer to the GNU one as "GNUPlotutils."

This blew my mind - for 30ish years I've taken it for granted gnuplot was part of the GNU project.

Sort of like gnutar (gnu tar) which actually now is a gnu project but had that name before the gnu project because gnu is John Gilmore’s uname.

Why are we so bad at naming things

People often think they're the first to think of somewhat obvious things, have many shared cultural experiences, and they like puns.

Gnuplot isn't a bad name in itself, it's a name which describes what it does (plotting system) with a small pun. Just bog standard stuff.

GNU isn't a bad name, it's a cool self-referential acronym which also describes what it is really well (GNU's Not UNIX).

The only problem is that those two names exist in the same cultural context.

what’s the pun? Never knew there is one (non-native English speaker here)

Probably New vs Gnu. (Pronounced identically.)

FWIW, I have never heard anyone pronounce “GNU” (the operating system) without enunciating the “g”. Not sure about gnuplot.

whaaat? i pronounce the g in gnu ("guh-new" but short, more like g'new).

Google now tells me that's wrong.

I guess Gary Gnu on the Great Space Coaster led me wrong in my formative years.

The Wikipedia page on GNU shows the "g" is pronounced.

https://upload.wikimedia.org/wikipedia/commons/2/24/En-gnu.o...

..Unless you mean "gnu" as in the animal, then it's like "new" (which I didn't know before).

> Google now tells me that's wrong.

Google is not an authoritative source on pronouncing words in English. In fact, there are no authoritative sources on pronouncing words in English.

There are references for common usages, but anything that is mutually intellegable is not wrong, even if it's not common usage or listed in the references.

Plenty of people pronounce the g, and plenty of people would know what you mean if you said I graphed this with g'new plot, and not if you said you graphed it with new plot. So, I for one, would prefer if you countinue to voice the g.

Yeah, I'm gonna stick with it. I've never heard _anyone_ pronounce GNU as "new", so my mind was fairly blown.

Sibling comment to yours helpfully points out the "new" pronunciation is only for the animal, not the open source initiative.

Either way, I blame Gary Gnu. Or maybe I thank him, for getting me ahead of the curve on pronouncing GNU.

There are more things to be named than good names, and also what constitutes a good name depends on our individual perception (prior experiences, cultural background, neurological disposition, etc.).

Looking at it from the other side: Why are we so hung up on names? It’s not that we shouldn’t, but it’s an interesting topic. A computer doesn’t care as long as a name is syntactically unambiguous and fits the lengths constraints.

Because it's hard. There are even professionals who think all day long how to name things (advertisement and marketing industry). A little open source project doesn't have the resources for that.

Names should come last (when everything else is known) but usually they come first (when most things are yet undecided). Once set, the name gets quite a lot of inertia and is hard to change. While naming things is hard, renaming things often isn't worth it.

If GNU’s not Unix and Gnuplot is not GNU, is Gnuplot Unix?

Yes, if not GNU is Unix.

Licence is here: https://sourceforge.net/p/gnuplot/gnuplot-main/ci/master/tre... Unusually it requires you to distribute modifications to the released source code as a set of patches.

, plotutils is GNU.

Plotutils comes with vector and rasterizer support, as well as a library.

This is why I hate these stupid OSS naming schemes. Screwing with capitalisation serves no function other than as a shibboleth - signalling that the devs belong to the free software tribe.

I avoid software with dumb-ass case-sensitive names precisely for this reason. Look at the confusion, the time wasted on this topic. Why should it matter if it's octave vs. Octave? Normal users aren't case-sensitive, they just want to get on with their task. It just points to devs with the wrong priorities.

llamaplot would have been a great name

Except in fonts that don’t distinguish between “I” and “l”. ;)

https://en.wikipedia.org/wiki/Ilama

Those ought to be banned, much like fonts which make the distinction between 1 and l difficult or O and 0.

It would still have a fun pronunciation controversy, with some calling it lamaplot and some calling it yamaplot.

I pronounce llamaplot with the voiceless alveolar lateral fricative like in Welsh.

It's incredible to open a source tree and see commits talking about OpenStep. Really shows you how long this has been around and taking BC seriously.

DOS is in there too. I wonder what else lies among the depths?

Quite the achievement to have written software that is still in use and discussed 36 years later.

The FAQ says gnuplot used to be distributed by the FSF. So, perhaps it used to be GPL-licensed? I wonder what changed... is there really that much money to be had from forcing derivative source code to only be distributed as patches over the original? I don't get it.

It's ancient.

For some context: gnuplot was first released in 1986; the GPL v1 was in 1989.

Uh, while that was informative, more important is IMHO that gnuplot is a great tool and I like to take the opportunity to express my gratitude to all who contributed.

gnuplot is not gnu, gnu is not unix, unix is not multics. I wonder what multics was not like.

Pretty idiotic name, I realize this now.

Great software package tho.

gnuplot is not GNU

The plot thickens.

I have to admit that it's pretty easy to build your own plotting software. As long as you have the ability to draw lines and pixels, it's easily done.

Don't forget text plotting, that's the hard part. The technical challenge is not huge, but the amount of small and "easy" features a good plotting lib needs is huge.

It’s easy if you keep it simple, your data is regular, and you don’t handle the multitude of corner cases. Doing it properly is about as difficult as processing Unicode correctly, which is to say quite bloody hard.

Lol if you have tons of free time on your hands. I'd like to see your replacement for gnuplot that has all the same capabilities. HackerNews will be awaiting your contribution!

Reply on news.ycombinator.com