Hacker News Reader

Comment by labster | Hacker News Reader

Bloom filters prove OP’s point though. The first few times I heard the term, I wondered how a Photoshop filter to blur things could possibly apply to the problem. Maybe if it was called an exclusion filter it would be less jargony, I don’t know; naming things is hard.

View on HN

At least having a word like "filter" in it narrows down the choices even if it doesn't make it unique. If instead it was "Bloom's construction" or "Bloom's algorithm", or "Tim's procedure", we'll be at a total loss to even guess what it was about, which is what happens with a lot of math starting from "Pythagorus theorem", anyone instantly recall "Apollonius' theorem", "Ackerman function", "Euler's function"?. If "Fermat's last theorem" or "Goldbach conjectures" weren't crazy famous I wouldn't have a clue.. The request to at least give us a "Fourier transform", if not "frequency spectrum" is not unreasonable.

I've lamented this for a long time, but on the other side, I doubt if mathematicians would ever get sufficient recognition if their names weren't immortalized thus, since they can't get patents on their works. They totally deserve recognition. Would you even remember Leonard Euler if his work was named factually? Most of us I guess have no idea who came up with sin/cos/exp/log etc. I'm glad for the names of these functions, but lament the loss of knowledge about the one (or many) who discovered them.

Longer names are a candidate .. along the lines of "Einstein's theory of general relativity". "Euler's relative prime counting function" .. but they too will likely, depending on familiarity, collapse over time.

Do they get recognition? I’m an atmospheric scientist, and though I’ve used the equations many, many times, I have no idea who Navier-Stokes was. Or maybe they were two people? Whatever. Presumably they invented the equations and were scientists or mathematicians or something. If real recognition only comes from inside the field, everyone else has pedagogically unuseful name to deal with.

But could you propose a better name for the math terms you mentioned? Fermat last theorem for example is famous because of its history and not significance and I don't think any other name would be better. Pythagorus theorem - how to call it with a short and significant name? The only option I can think of is "a squared plus b squared equals c squared" which is hardly a good name :)

The alternative "Euclidean distance" is already half way there and is better since we at least know it's about "distance". At this point, offering any alternative will feel alien and unfamiliar, but "Linear distance" works for me if I feel the need to push Euclid out as well.

edit: If I want to talk about distance in a curved space, we already have a well named "Geodesic distance".

That sounds like a different theorem. While it conincides with sums of squares of distances for the Euclidean setting, for the case of a sphere or other manifold it is decidedly about triangles, not so much distances.

I was telling a friend about Bloom filters and he'd never looked into them, because he'd assumed from the name that they were some kind of screen space shader.

For a long time I thought the algorithm or data structure used by it had some sort of metaphorical relationship to a flower blooming.

He wasn't assuming - there are Bloom shaders[1] are an item in Computer Graphics, but at least the name has intrinsic logic ("light bloom")

1. https://en.wikipedia.org/wiki/Bloom_(shader_effect)

>Bloom filters prove OP’s point though.

That depends on your native language, though. For Non-English speakers sounds no different than 'Shell sort'.

I'm a non-native English speaker, and I struggle to remember that Bloom filters are named after a person. I have at times, when coming across the term, ended up wondering how they're supposed to relate to whatever "bloom" the name refers to.

I don't think it really "proves the point", though, except for the point that sometimes names are confusing. So maybe it would be nicer if someone had happened to consider whether a specific naming might be confusing, but that's not the same as "names that aren't directly descriptive are automatically bad".

And conversely why are photoshop filters even called filters!

Presumably because these exist: https://en.wikipedia.org/wiki/Photographic_filter

The point being made is that photoshop manipulates matrices and filters are circles made of translucent material you put on your camera; and in fact many of those photographic filters don't primarily filter light but rather distorts it in some desired fashion. "Transformations" would be much more accurate.

In signal processing, those transformations have always been named filters. It goes way back to the early 1900s

I always presumed it came from signal processing[1].

[1]: https://en.wikipedia.org/wiki/Filter_(signal_processing)

Exclusion Set

Reply on news.ycombinator.com