Hacker News new | past | comments | ask | show | jobs | submit login
Nils Nilsson has died (twitter.com)
216 points by villgax 28 days ago | hide | past | web | favorite | 21 comments

Oh man... first Joe Armstrong, now this. What a sad last few days. :-(

For anybody who isn't familiar with his life, Nils was one of the early pioneers in Artificial Intelligence. He was probably best known for creating the A* algorithm.


R.I.P., Mr. Nillson.

Decades ago, I bought a copy of his book, Principles of Artificial Intelligence. The A* algorithm is Dijkstra's search algorithm + heuristics. So you estimate the cost from starting point to goal through a node by adding the cost of getting to the node to an estimate of the cost of going from the node to the goal (the heuristic).

I always had him down as a neat, but he said in a speech to AAAI that AI needed scruffies as well as neats.


I've been meaning to get a proper list of timeless programming / CS type of books I should read. I feel like the original Erlang book is good enough even if Erlang evolves beyond the book it's got so much from Joe that's so great you can enjoy it years from now. I still remember a good amount of what he said about the Actor pattern, I definitely want to buy a fresh copy of that book to have with me. I had originally checked it out of my local library.

He came up the the A* algorithm while working on Shakey the Robot, which is the other project he's most known for (https://en.wikipedia.org/wiki/Shakey_the_robot).

Rough week for giants of computer science. :(

I remember learning A* when trying to make my first game, it was an awful game but A* really changed the way I think about simple algorithms and was the main thing I took away from that whole experience. RIP fellow engineer, you did great things and I hope your work continues to inspire.

Truthfully, I had never heard his name until now. But I have been cherishing A* as one of my favourite algorithms ever since I first learned of it. It's take on injecting sensible heuristics into BFS was an eye-opener for me. In many ways, it represented what I thought algorithms would be, before I studied them, but never encountered during my first class on the subject.


Implementing A* recently I was shocked at how beautifully simple and elegant it was. What a tremendous gift to bequeath. We truly stand on the shoulders of giants. RIP Mr Nillson.

Never knew his name, but A* reached me as one of the first algorithms I implemented during my CS studies, and was the first one introducing me to the concept of heuristic functions.

It is truly impressive how far his work reached, almost every developer I have interacted with have at some point either used or implemented A*.

RIP mr. Nilsson, your legacy will live on

And the top reply is "Can AI predict whose next?". Sometimes I with the rest of the internet was a bit more like HN.

Dupe of https://news.ycombinator.com/item?id=19726411. Makes me wonder if I’m shadowbanned, and why.

I can also see them. (Assuming no vote brigading,) It's fairly random which of multiple stories for a notable news event will be "the one" that gets traction.

Don't sweat it/don't chase fake internet points.

Thanks. I am always a bit miffed when I see stories trending on US time and linger during EU timezones, I guess that's the main factor here.

I can see your comments, so I don't think you are.


Damn. A* was the first cool algorithm I implemented when learning to code games many years ago. I remember feeling so cool having a tower defense game with A* pathing.

I just started reading his Understanding Beliefs yesterday and love it thus far. Recommended.

The post title does not correctly spell his surname. It's "Nilsson" with one "L" and a double "S".

Ugh, I'm sorry we missed that. Fixed now. Thank you.

Can we get a black bar for this?

We should be able to get it... in the meantime, here you go: http://i.imgur.com/4pE9QbE.jpg

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact