HN Theater @HNTheaterMonth

The best talks and videos of Hacker News.

Hacker News Comments on
John Carmack's Keynote at Oculus 4 Live Stream · 353 HN points · 0 HN comments
HN Theater has aggregated all Hacker News stories and comments that mention's video "John Carmack's Keynote at Oculus 4 Live Stream".
Watch on [↗] Summary
Tune in to hear John Carmack, Oculus CTO, talk about the state of mobile VR and more!
HN Theater Rankings

Hacker News Stories and Comments

All the comments and stories posted to Hacker News that reference this video.
Oct 12, 2017 · 353 points, 112 comments · submitted by staunch
Huge fan of John Carmack!

These quotes from the keynote really stand out to me:

"If you think you've got that new killer app by all means go ahead and work on it but otherwise you can always spend time improving the existing applications and those are good muscles to exercise so even if it turns out that it wasn't the magic app - going through the disciplined work of making it as good as it can possibly be is whats going to need to be applied when you eventually get the magical application"

"Embrace the grind. You've all shown that you're bold by starting to work on an emerging platform that is not mainstream yet but it takes more than just being bold - you have to actually work really hard and you've got to fill your products with give a damn, to really care about every aspect of them."

"Success isn't about that one brilliant idea. Its about doing the one thousand little things right and getting it all done."

Thanks for this I wonder if someone is compiling all of his quotes from other lectures into a handbook.
Here's a huge archive of them. I shipped it to PrintMe1 and had it spiral bound for reading.

How much runway does Oculus have left? They're 5 years in.

VR headset unit sales to March, 2017:

- Sony: 915,000

- HTC: 420,000

- Oculus: 243,000

Will Facebook keep pouring money in, or pull the plug?


First, you didn't count the 5,000,000 Gear VR headsets, which are co-produced with Oculus.

These numbers, if accurate, are prior to the massive price drops from this summer. There was a huge spike in sales, with Oculus struggling to keep up supply (it took a few weeks to actually get my own order).

Unless the numbers were an order of magnitude lower, I doubt Facebook would be pulling the plug any time soon.

Those are low numbers for everyone involved.

I think this is a longterm space that Facebook is definitely not interested in missing out on, so they'll keep it going.

How much does it cost really, in the scheme of things? If anything, I don't know why Facebook simply doesn't ante up and steal some of the talent away from the competitors, they have the cash to do it and I'm sure people would be happy to work under Carmack.

they tried throwing around their weight in the VR game market by attempting to lock developers into Oculus Exclusive deals in exchange for a fat facebook check, backfired.
> I don't know why Facebook simply doesn't ante up and steal some of the talent away

To what end though? Say they could spend $10 million and quadruple the resolution, halve the price, and reduce the barf factor of a Rift v2 in a year. What then?

They would probably be better off funding a bunch of people to try to find something compelling to do with the current generation of hardware. It's cool for some types of games, but a market that's a subset of a subset isn't going to set the world on fire. It's easy to come up with a cool demo and hard to come up with a good application.

> They would probably be better off funding a bunch of people to try to find something compelling to do with the current generation of hardware.

As a completely separate peripheral? Generally speaking most external peripherals sell to niche markets.

I was thinking more about some compelling application. The current generation of headsets is pretty neat.

In the current form factor, I just can't see it moving beyond a gaming peripheral anytime soon and the market for that is limited. I was excited when Facebook got into this because appealing to everybody is what they do. But there's been nothing.

I'm starting to think VR is a dead end. There is a market for it, but maybe it's not very big.

Check out this blog post by Michael Abrash, Oculus' chief scientist, that was posted a few days ago:

It's clear that Oculus are spending a lot of money on research, and have recruited a lot of people. A key quote from the article is:

"I signed up to build a 30-50 person research team. Oops. Orders of magnitude matter."

So there's at least a few hundred people just in the research team, and probably a lot others in regular product engineering as well. I think it's clear Facebook are pouring a lot of money into Oculus.

They made the horizontal strategy clear.

Expect VR assets in the Facebook feed, and this to slowly bleed into AR items.

That's the thin end of the wedge for wider adoption.

> Expect VR assets in the Facebook feed

Like what? I've been thinking about this and I'm having a hard time coming up with something that would get my relatives to buy a headset.

They have a new 3D object content type for the feed, which can be rotated and such by tilting your phone; has some support for AR using the phone camera. The assets come from their modelling app and such.

Full WebVR content will follow. They're easing the masses into VR via changes to the feed.

These numbers are as of March 2017, this summer has seen rapid growth (fuzzy memory but I saw numbers that indicated roughly double this) due to the major price cuts of the oculus rift (and also HTC vive to a lesser extent). Adoption appears to be quickly ramping up
For facebook this isn't and never was about VR, it's always been about AR. Zuckerburg has talked about how the problems that have to be solved with VR (head tracking, frame rate etc) are the problems that have to be solved with AR before it can fully take off. Facebook was late on mobile, they aren't going to be late on AR so don't expect them to stop investing in Oculus anytime soon.
Sony's numbers are impressive, I wonder if they could create a VR portable next, i'm sure chips like the Nvidia k1/2 could push VR and also work as a portable and tv console.
Sony did that much with PS VR? It's really unfair. I've been following VR from the beginning.

Oculus made it possible, and it's not working much -> that's unfair

HTC Vive has produced the best headset that made me, and many other, love VR. And they're not dominating the market, and they're not going to, and VR is not becoming this huge thing. It's the future really... I feel like that's unfair.

If you put walls around your garden, don't be surprised when many people don't come in.
Don't expect them to pull the plug. Zuckerberg recently stated that he didn't expect VR to be profitable for some time.

He also committed to spending 3 Billion more on VR development over the 10 years!

Any restreams that don't require a login?

Edit found one:

Here's the upload of just his talk:
Nice, that one's a 360-degree livestream, too! Thanks :)
Just had an interesting experience with that. I moved the frame so that it showed the timer countdown and screens as well as John. After it hit 0, they put up "Please go to Q&A". After a minute or so they started flashing it from the white text on black background to a bright white screen to get his attention. After a couple rounds of that, John goes "Please stop flashing the go to Q&A, I'll get to it" and they stopped the flashing. Certainly something I wouldn't have seen otherwise.

Also interesting seeing now what is close to John's view of the queue for the Q&A.

I wanted to see this also. It occurs at about the 1:39 to 1:40 mark.

Edit: 1:40:58

Yeah, I saw (and heard) that as well! Quite funny :) Cool to see the crowd and everything from his perspective, too.
Huh, really good quality. Wonder what kind of hardware they are using for that stream
Try using a private window. It only makes you log in if it detects certain FB cookies and thinks you're a user.
Interestingly, this works on desktop but not on mobile
John Carmack is the most natural presenter i have ever seen he just dumps his thoughts in a clear informative way.

Not afraid of anything.

It certainly helps that he thinks about this topic from dusk till dawn. He has interesting comments and thoughts about everything the questioners mentions.

John Carmack once wrestled a bear for 6 hours until the bear tired out and admitted defeat.

After the contest, he picked two bushels of berries and fed them to the bear just to show there were no hard feelings.

They are still friends to this day....

Edit: Since this bothered some people, you will absolutely love this...

They call me CarmackDaddy, the original G,

and I like to pop n lock it while I'm coding for my rocket and you just cant stop it

Hit it!

Carmacks back with a brand new edition

The killa coding wizard, Algorithmic Magician

Check my ID, i'm the Keen Commander

There can only be one, just like the last Highlander

Hittin straight to the kernel, Eliminate the latency

Hit ya where it hertz, Swivel in yo chair aimlessly

I own all the mobile, releasing shit global

‘Mack aint playin, you know what Im serzziziziz Sayin?

Srsrszrz Saying..

Lol you should lace a beat to these bars
Sounds likely given the events of
my favorite Facebook comment from the live stream: Artificial Intelligence does exist. This guy is clearly not human
Facebook, I don’t want to log into your fucking website. Can you please get that fucking button out of the way, so I can watch the video?
"I do my morning trawl of Twitter and Hacker News inside the browser, in Oculus, inside the Go"
The record:
Is there any good software yet for Gear VR (which could thus be used on Oculus Go) for emulating multiple monitors?
Might be worth asking why you want it? Simulated monitor(s) in VR on the first generation of VR hardware run into blurriness problems from having low resolution displays relative to the monitors they're simulating.
I work remotely and travel all the time. It would be great to have a VR headset that doesn't require a desktop and provides expanded screen real estate. I know this is something that many people besides myself want.

The oculus go sounds like it solves the major blocking issue since it is stand alone and thus doesn't depend on the computer for the 3d rendering capacity. There is still the question of if using the VR device for work is feasible/productive, but at least it is now possible which I why I was curious as to if any such software exists.

Some of the resolution mismatch could be solved by simply expanding the virtual screen and/or setting the computer to display at a lower resolution so that the resolutions match closer. Some of this is probably helped by using a cylindrical surface in the VR environment.

The resolution you would have to use to do this without excessive head movement would probably be more appropriate for mobile UI than desktop UI. I don't think the current gen headsets can provide a satisfactory Excel experience for example. Also, the relatively small 'sweet spot' that current gen headsets provide make extended reading uncomfortable since it requires head movement for what would usually be accomplished by eye movement.
What do you mean, like streamed through VNC? You can't run windows on the Go obviously so do you mean android monitors? I don't even think android supports multiple monitors.

Basically the answer is nope

Well, I did my own research and thus does more or less what I would want, but only works on Rift and Vive:

This rather complicated process does work with Gear VR (and so should work with Oculus Go), but seems to only support one monitor (and streams only games?):

So no new higher resolution devices yet? That's a bit disappointing. Any other events in the pipeline where such may be announced?
Having tried a couple VR headsets (including the Vive), I'm pretty sure resolution is good enough for now. Yes, it needs improvement, but it wouldn't be my top priority.

I currently have two pet peeves: wireless latencies, and hand tracking. I want to cut the cable, which is cumbersome, causes problems if I spin too much, and tether me to a small area. I also want to track free hands (perhaps with gloves), for fine grained hand gestures (the Vive controller is good, but still a bit intrusive).

I'm mainly into simulators such as DCS, Elite Dangerous and iRacing for VR. My understanding is that increased resolution would make those titles better; conversely, none require / would benefit greatly from hand tracking, so it's all about different needs. :-)

I was hoping to see a new Rift to address this, but as such I'm happily staying on my 95hz 34" 3440x1440 with TrackIR a bit longer.

Here's the recording:
This stuff is compelling, especially after having recently read Ready Player One. However, how do you type? There's all this talk of how this tech is great for programmers, with tons more screen space, etc... but unless I just missed it somehow, don't see where a normal keyboard fits into the picture.
I already use a Das Ultimate keyboard. No key caps. It's pretty straightforward to learn the keyboard if you don't fall back on looking. It's like full immersion in a foreign language. If you _have_ remember things, then you will.

So I can type while wearing an Oculus Rift. It's not a problem. And I'm sure you can do it too.

That's not exactly what I mean. It looks like most VR rigs these days are using some sort of control thing that you hold in your hand. Doesn't look like you could type (with more than one finger of each hand) and hold it at the same time.

On the other hand, maybe lightweight haptic gloves are a thing already and I just don't know about it.

Maybe some kind of (improved) Leap Motion finger tracking combined with a virtual keyboard would be acceptable?
Gloves aren't really a mainstream consumer product yet in a way that you'd expect in the future. When I'm typing in VR, I have to put my controllers down to the side of the keyboard. Easy to pick them back up though. It certainly would be nice to have the keyboard tracked so I don't have to slide my hands on the desk to look for it, but once I grab it, typing is fine.
>No key caps

I think you mean "blank key caps". The alternative could be pretty painful, depending on the type of switch.

Bow to the king.
Tl;dw anyone?
Noes. Only you.
Would you be so kind to give a summary of what you've seen?
There are some excerpts in other comments, but it's essentially a dense brain dump for almost 2 hours. I dont think you can get a meaningful "tl;dr"-style summary short enough that you wouldn't just ask tl;dr again.
"In game development, you talk about how the second 90% is always the hardest"
I think John Carmack is one of my favorite people to watch present. If you haven't watched his long rants about using FP in game dev @ QuakeCon a few years back you should.

It was QuakeCon 2013, although I'm not sure which part:

Yeah, he is really quite good at this. (I wonder if it came naturally to him, or if it's just a matter of practicing very regularly for like 20 years?)

I wish he would give Elon Musk some mentoring on how to communicate with a technical audience. Or just a regular audience, I guess.

Carmack used to have serious speech tick where he would always say hmmm at the end of the sentence. It has gradually wanished

Often when something looks easy and efortless there is lots of work behind it.

He still has it, he just says "aye" instead.

I remember he said he had some kind of training on how to deal with the "hmmm" and it got better, but the "aye" remains.

(Not complaining, I loved his talks before with all "hmmms" and still do, "aye" and all)

but the "aye" remains

It's an engineer thing. Like Scotty from ST:TOS.

Linked video is still a great talk. Carmack comes off as someone who could talk about anything for a literal fortnight, so the 'mmm' may've been a tick where he was forcing himself to shutup & let the interview continue
Probably helps that he is in a field that is so esoteric that he really doesn't have a lot of other people to talk to this stuff about.

It's not like people go home and talk to their partners about vertex shaders for a half hour.

Although in his case he probably could, I believe his wife is very bright.
From what I gather she is, my only point is that when you are in a dense field, generally your partner isn't too enthusiastic when you get into the minutia you've spent a 1000+ hours in that year, so when you get the chance to really talk about it, you gush.

Granted I've watched every QuakeCon keynote that was posted and enjoyed them all (along with reading every .plan). His enthusiasm is infectious. Yet even then he is breaking it down to the big things that happened that year, not the day-to-day stuff.

If it were over the dinner table every night, I'd probably lose my mind.


This has reminded me that I used to have the same tick.

I was mocked a bit because of it in middle school and set on to remove it.

it took a conscious effort at first but I have not thought about this tick since that time.

For a while he was saying "on there" all the time too. But I think that has been reduced in his recent talks.

Public speaking well is hard. Most are not good at it, myself included.

That said, I'll take a poor speaker (Musk, older Carmack) over a PR drone any day.

I'm completely the opposite. I really enjoy watching Elon Musk spatter his nervous stream-of-consciousness at an audience.

No-one should be forced into attempting something unnatural to them, to satisfy the whinging of the the peanut gallery.

Carmack really should have quit Oculus when Facebook bought it and gone on to work for Musk at SpaceX. Both are workaholics, too, so they'd probably get along quite well. Plus, Carmack is a big fan of space rockets.

Perhaps Musk needs to go to him and tell him, Steve Jobs-style:

"Do you want to work for a company that invades people's privacy for the rest of your life, or do you want to come with me and change the world?"

Carmack's skills are being wasted at Facebook.

There's no way they purchased Oculus without buying Carmack for a set period of retention. That's a major part of the attraction for FB.
Musk has already offered Carmack a job a couple of times on twitter. Not verbatim but it was close to that line.
>Both are workaholics, too, so they'd probably get along quite well.

Or a clash of wills of epic proportions...

Perhaps he got burnt out on space...

Last history bit claims

> As of August 2013, Carmack was "actively looking for outside investors to restart work on the company’s rockets"

Before stating assets were sold, likely due to lack of investment. SpaceX has the investment. Carmack & Musk also exchange tweets re rocketry at times

I found a really odd video of him speaking in 1998:

So yeah, he was good back then too.

>>I wish he would give Elon Musk some mentoring on how to communicate with a technical audience. Or just a regular audience, I guess.

Musk is an effective speaker in my opinion. To be effective, you need not be eloquent, engaging, and charming. Musk has his foibles and stumbles around a lot, but that is part of who he is. I appreciate it.

You're being downvoted, but I do agree. Musk isn't a very good speaker and he seemed very nervous at his IAC talk about colonizing Mars.
> Musk isn't a very good speaker and he seemed very nervous ...

Sort of.

Looking back, he sounded much more confident when he was younger [0] perhaps because he hadn't yet experienced life altering events like the loss of a child or coming close to bankruptcy at the end of 2008 after sinking his networth into SpaceX and Tesla.

I think his seeming lack of confidence when he speaks publicly today is a latent acknowledgement that the stakes are much higher relative to his PayPal days when the Internet was incipient, and that failure is still a possibility, regardless of the amount of resources he may have sunk into his investments.


He didn’t strike me as very nervous, but if he was, he had good reasons: He announced that they would halt production of Falcon rockets after building a big enough batch of them and completely focus on building spaceships. He announced to endanger the only big revenue stream SpaceX has in favor of an unproven technology in the hope to solve all problems and start production before the last Falcon rockets seize to be reusable.
I'm under the hypothesis that (good) game developers know how to talk to people (or at least how to present things to people) because part of game design is thinking about how to manipulate people's emotions into wanting to (or even being eager to!) face a challenge you've set out for them.

It's similar but not identical to the reasoning for actors and directors to be good entertainers and interviewers. Those folks constantly think about manipulating people's emotions in general, but game design is completely focussed on just manipulating people into excitement or flow states — and that just so happens to be the emotion you (usually) want people to have in reaction to a presentation: the feeling of "I'm going to go out— and buy their thing —and change the world."

Carmack is more of a game engine engineer than a game designer, as evidenced by the lacklustre design of id games after Romero left.
Quake III Arena (Graeme Devine) was a fantastic game, although Doom 3 and RAGE weren't particularly inspiring.

Doom (2016) was released after Carmack left, but it's one of the best shooters they've made.

Q3 always suffered from the fact that it was released right around Unreal Tournament which was a far superior game, especially in level design.
Its a matter of preference, I think. I was only able to play UT growing up, thanks to its software renderer, and finally played Q3 (Quake Live) in college.

Q3 is more minimalist, which makes competitive play more interesting. Q3 has also maintained a competitive scene to this day where UTs seems to have fizzled out.

Quake 3 Arena is still one of the best multiplayer shooters ever, so I can't agree with your statement.
While surely being a marvel of software engineering and level design, it had nothing in common with Quake I and II. It is not my intent to be critical here, but QA3 felt like a circus, especially compared with other multiplayer shooters such as the original Unreal Tournament, which had a wonderful mysterious atmosphere to it.
Q3 is intentionally stripped down. Primarily due to Carmack's direction. Iirc from his old plan files.
I was just playing something recently and thought "This feels like an Unreal Tournament Space Castle." Can't remember what game it was.

UT99 had some fantastic level design. Facing Worlds, Phobos, HyperBlast, Deck 16, Coret, etc.

It has great game mechanics, but in all other manners of design it feels like a hodgepodge of unrelated assets thrown together. Quake 1 had a similar problem early in development, but they managed to tie it together. Quake 3 didn't even bother with a true singleplayer campaign, which seems like a winning move in hindsight, but at the time it was a departure from the norm and I can't help but feel like it was a move they did out of necessity.
Romero did not leave, Carmack fired him, in his own words, for "not working hard enough."
Eh, judging from the Masters of Doom book, i wouldn't take his words seriously considering the tension that existed at the time between the two and how Carmack basically burned out himself during Quake's development. Romero wasn't the only one who left id at the time.
Why the downvotes?
I think it's less this and more that being able to present this information is the mark of a really good technical leader.
Great perspective.
I like that idea. It's also true that a field like game development, which is both collaborative and interdisciplinary, pretty much requires excellent communication skills. Being a game designer is less about having amazing "ideas" or "vision" (everyone has those) and more about your ability to align everyone's ideas and vision in the same direction.

Personally, I prefer to describe "thinking about manipulating people's emotions" as _empathy_, but that's just me.

Empathy = understanding the other, including on emotional level in such a way that the other feels understood.

Nothing to do with changing them.

If the other did not feel that deep understanding, empathy did not take place.

Source: I teach empathy with a worldview rooted in nonviolent communication. This is the definition I use; it is not universal.

> Personally, I prefer to describe "thinking about manipulating people's emotions" as _empathy_, but that's just me.

Empathy is the core skill, yes, but there's a sort of... I almost want to say an instinctual disgust? that people also have to overcome, when they want to turn empathy around to use it to change someone else's mind, rather than just using it to predict someone else's mind. You have to become at least a little bit of a sociopath, is maybe the problem.

See, other people have different foundational beliefs—different axioms they're working from. To use "empathy aikido" on them—to come up with arguments that will convince them of something, not slowly and laboriously from logical first principles, but by building up quickly from what they already assume to be true—you have to be willing to make arguments that are true under their axioms, but not under yours. That is, you have to be willing to use arguments that you think are false, just because the person you're trying to convince will believe them.

It feels weirdly like lying; like you're a politician swaying the populace with empty rhetoric. But you're not saying things that nobody would believe (if given long enough to think about them); you're instead just getting into the head of—empathizing with—the person who holds those axioms, and then saying things that you—as that person—really do believe.

This is why, I think, there's a big divide between people who like or hate the idea of "salesmanship": some people fundamentally see it as lying, while other people fundamentally see it as empathizing.

Personally, I think it can end up either way—some people "sell" an idea while holding back a bunch of facts that, under their axioms, are total deal-breakers. Others, though, "build a bridge" between their interlocutor's world-model and their own, using their arguments to help the other person build a world-model enough like their own that they can then present the facts that they believe to the listener, and the listener can understand them through the "consensus schema†" they built.

People who are said to have "reality distortion fields", I'm guessing, are just good at making those kind of points that build a consensus schema, that they can then state plain "facts" against which will seem—within the consensus schema—to be obvious, rather than having to convince you of each fact through argument. Despite the gnawing feeling that accepting that consensus schema into your brain is sort of an indoctrination into a cult, it's really the less ethically questionable of the two options, in my mind: the speaker doesn't have to say anything they don't actually believe (other than the arguments required to build the consensus schema.)

Thank you for this comment - that's a really striking way of looking at these things. Makes me wonder just how many human interactions involve being that little bit of a sociopath. E.g., "putting your best foot forward" for a job interview or first date can feel like a kind of dishonesty, although it's expected in those cases. I wonder how you could start drawing a definitive line between "good" empathic manipulation and "bad" sociopathic manipulation, when even just smiling at someone can be manipulation of a sort?
> "That is, you have to be willing to use arguments that you think are false, just because the person you're trying to convince will believe them.

It feels weirdly like lying; like you're a politician swaying the populace with empty rhetoric. But you're not saying things that nobody would believe (if given long enough to think about them); you're instead just getting into the head of—empathizing with—the person who holds those axioms, and then saying things that you—as that person—really do believe." -derefr

Response to derefr: If you are willing to use an argument that you think is false then, by definition, you are lying [even in situations where the lie can be mistaken for truth by others].


> "E.g., 'putting your best foot forward' for a job interview or first date can feel like a kind of dishonesty, although it's expected in those cases. I wonder how you could start drawing a definitive line between "good" empathic manipulation and "bad" sociopathic manipulation, when even just smiling at someone can be manipulation of a sort?" -zazen

Response to zazen: If it feels like dishonesty then it most definitely is. Consider cases where the interviewee simply lacks confidence but puts on a facade to appear otherwise. As for drawing a line between "good" and "bad" manipulation, reference derefr's note:

> "To use "empathy aikido" on them—to come up with arguments that will convince them of something, not slowly and laboriously from logical first principles, but by building up quickly from what they already assume to be true—you have to be willing to make arguments that are true under their axioms, but not under yours. That is, you have to be willing to use arguments that you think are false, just because the person you're trying to convince will believe them." - derefr

The addition of "but not under yours" constitutes manipulation on grounds which are aside from truth.

I think you're misunderstanding what I mean by "axioms" here—a lot of these fundamental beliefs that inform which arguments you have to use with people to convince them, are "orthogonal to truth"—that is, things that aren't part of a causal graph, like normative or theological beliefs.

It takes a different thought process to convince someone that e.g. climate change is happening, if they're a Young Earth Creationist, than it does if they're a paleontologist. It takes different arguments to convince someone to donate money to charity if they're a deontologist than if they're a consequentialist. None of these axiomatic positions affect what (empirically discoverable) facts are true, per se; they just affect what facts are relevant to changing one's mind about what one should do—that is, these axioms influence how the "is" statements† a person hears will affect their confidence in various "ought" statements.–ought_problem

So, this kind of "empathy aikido" is less about modelling a person who believes different facts are true, as it is about modelling a person who cares about the truth-value of different facts than you do. It's not that you might have to believe [X thing you believe is true] to be false; it's that you will have to pretend that [X thing you believe being true] is not a compelling argument, and might be so unimportant that you've never even thought about it and never will. Whereas the truth-value of [Y thing you don't care about] might be, to the person you're talking to, the most important thing in the world; the "trick" is figuring this out and then using a (true) argument about Y to convince them, despite thinking personally that only X, and not Y, holds any real sway over the truth-value of your conclusion C.

It can still feel bad, but I hope you can see how that intuition is less grounded here in any real injustice you're doing. Telling a virtue-ethicist that it is "noble" to e.g. donate to the Against Malaria Foundation, when you think that "nobility" is complete poppycock and the only thing that matters is that those donations mean people won't die, isn't an example of a lie. It is a manipulation, but not an illegitimate one—because, from the other person's perspective, it's just the honest truth.

> That is, you have to be willing to use arguments that you think are false, just because the person you're trying to convince will believe them.

I think genuinely arguing over whether something is true or false, is rarer than it seems. More often, I think we find ourselves arguing over which things are important, or how we should feel about something. In which case, I think you can put yourself in someone else's shoes and still be perfectly sincere.

I 've watched most of his talks - the ones I could find - and its really such a pleasure to listen to him; he just shows up and rants about all kinds of subjects, without breaking a beat, and rarely if ever following a script or notes he compiled earlier.

I am not sold on the premise of VR myself; nowadays it comes down to novelty byte-sized demos or "games", with the occasional random mainstream game offering "VR support" (whatever that means, which usually doesn't mean much).

There's also the practical issue of strapping yourself to gadgets and wires, and on top of that, this is still expensive. That said though, I trust Carmack, Abrash and other smart folks elsewhere will solve those problems, and wind up building better alternatives and "next gen" killer apps that do justice to the term(VR) will be developed and we 'll all eventually experience it.

Agree. Whenever I come across a keynote of his, I tell myself "ok, just watching this for 10 min, then get back to real work", and then usually what happens is that I find myself 2 h later after the presentation, totally amazed by all the cool stuff that's going on in Carmack's lab.
This is the part... John discusses his functional programming adventures in Haskell at QuakeCon 2013.

"So what I set out to do was take the original Wolfenstein 3D, and re-implement it in Haskell."


"I've got a few conclusions coming from it. One of them is that, there's still the question about static vs dynamic. I know that there was a survey just coming out recently where the majority of programmers are still really not behind static typing. I know that there's the two orthogonal axes about whether types are strong or weak, and whether it's static or dynamic. I come down really pretty firmly, all my experience continues to push me towards this way, that strong, static typing has really significant benefits. Sometimes it's not comfortable, sometimes you have to build up a tight scaffolding to do something that should be really easy, but there are real, strong wins to it."

(Starts at approximately 1:21:16 in case the direct link doesn't work correctly.)

This is great when you're re-implementing Wolfenstein.

Not so much when you're doing exploratory computer science, or blue-sky prototyping. To build upon his analogy, scaffolding is most helpful when you have a reasonable knowledge of the shape of the building.

"It depends on the context" isn't exactly a shocking discovery however.

I love Carmack, love his presentation style, and when he talks from experience, I listen.

> Not so much when you're doing exploratory computer science

I'm not sure. The following experiment is outdated (I'd love to see it redone more rigorously and with modern languages) and has several methodology flaws, but "An Experiment In Software Prototyping Productivity" (1994, Paul Hudak et al) shows that Haskell and static types are actually great for rapid prototyping and exploratory programming, even in the face of vague or incomplete requirements. This runs contrary to common sense, which is why the experiment was fascinating.

Thanks very much for sharing this. Awesome.
Another great quote from that same talk is akong the lines of

"Any syntactically valid code, that the compiler will accept, will eventually make it into your code base."

HN Theater is an independent project and is not operated by Y Combinator or any of the video hosting platforms linked to on this site.
~ yaj@
;laksdfhjdhksalkfj more things ~ Privacy Policy ~
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.