'indexes' has been a just-fine variant of the plural for a very long time -- actually it's the first one listed in Merriam-Webster, so probably the more common variant, though I haven't dug into other references. That said, you're correct that the more common or preferred variant will depend somewhat on the culture/community/industry. The house style guides in finance probably tend toward 'indexes' -- in part because an index is a branded product they produce, so consistency is important -- while software/tech companies (or at least people in them) may tend prefer 'indices'.

I will say that, at least for me, indices and indexes aren’t the same word and mean two different things. This has made me realize that I’ve just naturally made this distinction without really knowing it.

* In a loop i,j,k are the indices. Matrices have indices too.

* Databases have indexes. The things at the back of books are indexes.

Indices are things that represent a position in something. Indexes are the things — the literal “objects” or mappings or trackers.

I dunno why, but I typically use “indexes” when I talk about a set of lookup tables (database indexes) and “indices” when talking about positions within indexes (indices of arrays)

Why do they need a justification? English is under no obligation to imitate Latin plurals, and it adds no functional benefit.

Edit: interestingly, the title of the article uses the English, while the body uses the Latin (although in one case that’s because it’s from a name).

Edit2: Relatedly, I had a "shower thought" that even the most pedantic Latinophiles say "albums" instead of "alba". Although I also know a Latinophile who reluctantly used "alba" for a folder on his music app -- but that was only because "albums" was a reserved word.

I agree, it’s time we stop pushing words like “cacti” and just accept the inevitable English-ication of words and say “cactuses”.

This is a tangent to the article, but along these lines people love correcting “octopuses” as “octopi”, but “octopus” isn’t a Latin word, it’s Greek, so the “-i” ending isn’t correct.

"Virus" is also a word that frequently gets pluralized incorrectly, if you're thinking of it as a Latin word used in an English context. The only plural form that makes sense in English is "viruses," mostly because the Latin form actually has no plural. [0]

That doesn't stop people from writing "virii," however. Both "octopi [1]" and "virii" are examples of linguistic hypercorrection [2], in which a speaker misapplies a known linguistic rule that seems to fit, but does not. Neither of these examples are attested in the source language the word derives from; people literally use them just to sound smart, most of the time.

Other examples are using "whom" in places where the correct word would be "who," not using a preposition to end a sentence or clause with [3], and to not split infinitives [4].

---

[0]: https://en.wiktionary.org/wiki/virus#Etymology

[1]: https://en.wikipedia.org/wiki/Hypercorrection

[2]: I kinda like "octopodes" myself, but that may just be me.

[3]: Winston Churchill is frequently quoted saying something like "this is the kind of nonsense up with which I will not put." This is most likely apocryphal. See https://quoteinvestigator.com/2012/07/04/churchill-prepositi...

[4]: My favorite origin theory for this one is basically "you couldn't do it in Latin, so don't do it in English." Early grammarians frequently took inspiration from Latin, because its grammar was already well-codified, mostly due to it being a dead-ish language. See https://en.wikipedia.org/wiki/Split_infinitive#The_argument_...

I think it’s fine to say or write -es or -I in general usage. I think it’s also fine to prefer one or the other in contextual usage. And I think it’s fair to roll ones eyes at people making a big deal of it.

But it’s incorrect to say octopus is not a Latin word. That’s like saying utensil isn’t a French word. It is, and it came to English from Latin through French. Octopus came to English through Latin from Greek. It’s a Latin word and it’s a Greek word.

So, I don’t think it’s a consistent position to be okay with the Englishification of Latin words, but not be okay at the Latinization of Greek ones!

But, at least in classical Latin, Greek words used the original Greek plural (transliterated into the Lstin alphabet) rather than the Latinized one.

Looks like a super common question - nasdaq has an article about it: https://www.nasdaq.com/articles/indexes-or-indices-whats-the...

I think I agree with that article. I'm used to seeing "indices" in a mathematical context, referring to an object that labels a particular subobject of another object.

It seems a lot like the difference between "vertices" and "nodes" in a graph. Go to the math department, and you'll hear people talking about vertices. Walk down the hall to the CS department, and it's all nodes, all the time.

I have no idea why this is, but I suspect some kind of cultural thing. Maybe the CS term and the math term originated independently? Kind of like how you talk about machine learning in CS, but a good portion of ML would actually be classified as statistics by the math and stats departments.

Thanks! This read was more interesting than the crypto article

The real puzzler is how they used both spellings in one headline.