Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

John Carmack on Functional Programming (2013) [video]

youtube.com

Post reply on HN
Print view View original

21–30 of 90 posts

Re: John Carmack on Functional Programming (2013) [video]

#21

Post by scott_s »

sololipsist wrote: post #15
ece wrote: post #12

Earlier quoted context omitted.

He talked about a whole FP-powered architecture to solve this.

Motivated reasoning. He prefers imperative programming (or something other than functional programming), so he hears the "criticism" without hearing the solution. I mean, of course he physically hears both, but one is prominent in his consciousness. It doesn't matter that there is a solution, he heard what he needed to hear to get the evidence he needs to maintain his bias, and shuts off his frontal lobe. To be clear…

I think you misunderstand Carmack's position. As of this talk, he had not yet solved this problem, but he is optimistic that it can be solved, and ece mentioned the solutions he outlines. He goes on to say such a system would be beneficial to all game developers.

Re: John Carmack on Functional Programming (2013) [video]

#22

Post by vog »

rootlocus wrote: post #13
vog wrote: post #10

EDIT: Why all those downvotes? Are famous developers not supposed to criticized on HN? While John has a lot of interesting things to say, the presentation is awful, almost an imposition to the audience. There's not a single slide, or any repetition to clarify structure, or any notable gestures to make up for that. A simple overview, just a damn simple list of keywords, would already go a long way. That would add a lo…

This was a keynote at a Quakecon, a conference for game players, not a technical presentation for software developers.

If the audience isn't even technical, isn't that even more a sign that more effort should be put into the presentation, especially for a keynote?

Re: John Carmack on Functional Programming (2013) [video]

#23

Post by vog »

While he says a lot of interesting things, it is too bad he is really just sitting and talking. No slides, gestures or any other facilities. Those would have brought more structure into the talk, making it easier to follow - especially for non-native speakers.

Re: John Carmack on Functional Programming (2013) [video]

#24

Post by savanaly »

vog wrote: post #10

EDIT: Why all those downvotes? Are famous developers not supposed to criticized on HN? While John has a lot of interesting things to say, the presentation is awful, almost an imposition to the audience. There's not a single slide, or any repetition to clarify structure, or any notable gestures to make up for that. A simple overview, just a damn simple list of keywords, would already go a long way. That would add a lo…

Personally, I watched it with rapt attention throughout and I'm eagerly anticipating having time later to do so for the other 6 parts of the keynote. Can't remember the last presentation that captured me so. I think he is a very good public speaker and the talk did not meander and was not hard to follow in any way.

I did crank the playback speed up a lot though, which helps considerably (I'm not sure I would enjoy it half as much if it were live, where of course I have to hear it at 1x speed).

Re: John Carmack on Functional Programming (2013) [video]

#25

Post by epigramx »

vog wrote: post #23

While he says a lot of interesting things, it is too bad he is really just sitting and talking. No slides, gestures or any other facilities. Those would have brought more structure into the talk, making it easier to follow - especially for non-native speakers.

Or more distracting and annoying.

Re: John Carmack on Functional Programming (2013) [video]

#26

Post by saganus »

vog wrote: post #10

EDIT: Why all those downvotes? Are famous developers not supposed to criticized on HN? While John has a lot of interesting things to say, the presentation is awful, almost an imposition to the audience. There's not a single slide, or any repetition to clarify structure, or any notable gestures to make up for that. A simple overview, just a damn simple list of keywords, would already go a long way. That would add a lo…

HN's downvote mechanism is a funny one.

As far as I know there's no clear meaning to what a downvote means and it seems like everyone has their own definition.

For some, downvoting is a way of "flagging" out of place comments, aggressive ones, etc. However for others it's just a way of disagreeing.

The guidelines don't seem to establish any particular definition to it, so it's kind of a community-driven thing.

At first I was certain that downvoting was a way for the crowd to silence unwanted comments, but "unwanted" has many values depending on the person. Personally, I prefer to downvote when there are clearly aggressive or out-of-place comments, and if I disagree I rather respond with my disagreement. That's the way to have a civil discussion in my opinion.

However, like I said, other people give downvoting a different meaning, so it's not so much that you can't criticize famous celebrities, but that the community seems to disagree with you. I don't see anything "flaggable" about your post so that's why I assume it's the reason.

But again, downvoting is an "undefined" behavior in HN. Guidelines don't mention what shoul or should not be downvoted so it's kind of up for debate.

Don't worry too much about it, as long as you are not clearly being uncivil, downvotes are probably just a lazy way of saying "I don't agree with you" :)

Re: John Carmack on Functional Programming (2013) [video]

#27

Post by craigsmansion »

jandrese wrote: post #7

Interesting that he ran into friction exactly where you would expect him to. He was talking about having the AI scripts running in different purely functional threads. Later (around 21 minutes in) he mentions what happens when two AIs decide to move to the same place at the same time, and has not figured out the solution. Of course parallel programming is easy if you ignore the data dependencies, but eventually you r…

As I heard it, he talked about independent agents which traverse a world by taking it as an input and reporting their own state as an output, and how easy to code such a model was in functional languages because the world was immutable. You don't have to mess around with the world data to keep the actions in sync.

Of course there's a level of conflict resolution needed at some point, which wasn't something he had not figured out, but something that was out of line with the elegance and simplicity of how the basic game mechanics were implemented so far.

As I understood if, it's wasn't so much about easy parallel programming as it was his discovery of clean models and elegant solutions as encouraged by functional programming.

Re: John Carmack on Functional Programming (2013) [video]

#28

Post by jandrese »

ece wrote: post #12
jandrese wrote: post #7

Interesting that he ran into friction exactly where you would expect him to. He was talking about having the AI scripts running in different purely functional threads. Later (around 21 minutes in) he mentions what happens when two AIs decide to move to the same place at the same time, and has not figured out the solution. Of course parallel programming is easy if you ignore the data dependencies, but eventually you r…

He talked about a whole FP-powered architecture to solve this.

Yes, and I'm curious how it turned out. Are there any followup talks on this that anybody knows about? It's certainly a solvable problem, but how many of the advantages remain once the data dependency problem is solved?

Re: John Carmack on Functional Programming (2013) [video]

#29

Post by pmarreck »

seanalltogether wrote: post #16
pmarreck wrote: post #8

He wrote a very good piece here: http://www.gamasutra.com/view/news/169296/Indepth_Functional... on the same topic (is it in fact the same?)

I come back to this quote over and over when talking about desktop and mobile apps/games, especially since they tend to be highly state driven view collections and devs are always trying to come up with DRY patterns to bury important features in subclasses or helper utils. > A large fraction of the flaws in software development are due to programmers not fully understanding all the possible states their code may exec…

This is also simultaneously a strong argument for unit tests (which encode knowledge of that state into a proof of sorts).

The limit of a programmer is his/her brain's ability to contain all these possible states. Bugs always come from missing some mental modeling of state or having a flawed conception of it, either at the point of design, the version 1, or the rewrite.

OO just buries that state "elsewhere", so things seem easier superficially to contain mentally (but the state is still there, ready to get corrupted and pounce on you). FP makes the state explicit, so you're forced to deal with it at all times (this perhaps not uncoincidentally also makes FP easier to unit-test AND reason about). If managing that state upfront becomes unwieldy, then that becomes a good code smell/indicator that your design is suboptimal.

The upshot of all this is that I think that using FP in conjunction with unit-testing reduces bug production by some statistically-significant amount, especially as a codebase grows. We definitely need more empirical data about this, though. But that's what my intuition says.

Re: John Carmack on Functional Programming (2013) [video]

#30

Post by pjmlp »

keymone wrote: post #5

this is a best video to explain to your C++ friend why functional programming is worth it even in gamedev world.

C++ has been getting functional programming goodies since C++11.

Nowadays with C++17, functional programming in C++ is a common talk subject at C++ conferences.

Post reply on HN