Hacker News Reader

Comment by bluepizza | Hacker News Reader

It's yet another funny trend that goes around in tech circles. As someone originally from math, my personal pet peeve is the way that tech people use "orthogonal" these days.

I guess people just want to communicate the idea that this piece of opinionated text is biased. Of course, everything is biased, so opinionated is supposed to be redundant. So it's a signal that the author understands that they have biases, and expresses them.

EDIT: I am not contesting any downvotes, but feel free to drop a note explaining why.

View on HN

> the way that tech people use "orthogonal" these days.

If you mean the sense which means “independent”, that goes beyond tech. It was used that way by a law professor in front of the Supreme Court in 2010: https://www.abajournal.com/news/article/supreme_court_word_o...

In software, its use in that way dates back at least 30 years. E.g. from https://www.sciencedirect.com/science/article/abs/pii/016560... (1992):

> It is shown that most of the above concepts are orthogonal, i.e., they can be implemented independently.

At that time, its use was already reasonably well established, although apparently not well enough to omit the i.e.

Possibly the first published use of the term in software was Van Wijngaarden, in the design of Algol 68, from 1968:

> The number of independent primitive concepts has been minimized in order that the language be easy to describe, to learn, and to implement. On the other hand, these concepts have been applied “orthogonally” in order to maximize the expressive power of the language while trying to avoid deleterious superfluities.

https://www.softwarepreservation.org/projects/ALGOL/report/A... , section 0.1.2, Orthogonality.

If you come from math, you should know that “orthogonal” means “independent” in statistics and linear algebra.

https://en.m.wikipedia.org/wiki/Linear_independence

https://math.stackexchange.com/questions/1536234/correlation...

It means perpendicular in geometry. But perpendicularity implies connection, and therefore, dependency.

Your links barely prove anything? I already knew that statistics is a fictional field.

You’re using an overly narrow sense of the term in math.

Consider any orthogonal coordinate system (https://en.wikipedia.org/wiki/Orthogonal_coordinates). The whole point of orthogonality in this case is that it makes coordinate values independent of each other - varying only one coordinate cannot result in a change in any other coordinate. The term is used in this same sense in many other areas of math, and elsewhere. This is the sense in which it’s been used in software, for more than half a century.

Forgive my curiosity, but in what sense were you “originally from math” but aren’t familiar with this term beyond a high school level understanding of geometry?

Good explanation!

Sorry, but now I highly doubt that you have any background in math. That's just a non sequitur.

You just need micro-dose so you can grok it from first principles. Here, just read my Medium blog titled “Why I choose X for Y.”

X for Y considered harmful. I have an article listing 10 alternatives for X, number 8 may surprise you!

Reply on news.ycombinator.com