Pioneers: Jonathan Blow

Interview transcript: www.notion.so/blog/jonathan-blow
For more episodes of the Pioneers series including photos and transcripts, please visit: www.notion.so/blog/topic/pion...
Jonathan Blow is an independent video game designer and programmer. He created Braid (2008) and The Witness (2016) - two puzzle games that, when released, received widespread acclaim. Jonathan also created a new programming language called Jai, intended specifically for game development.
Jonathan: / jonathan_blow
Devon: / devonzuegel
Notion: / notionhq

[@mesofreak]

Honestly one of jon blows better interviews. Good job to whoever was interviewing him, kept the conversation up well.

[@fennecbesixdouze1794]

The interviewer was excellent but I think what you're noticing is the editor more than anything.

[@____uncompetative]

I agree. This interview is the best one I have seen him do, all due to the domain knowledge of the interviewer.

[@alexnoman1498]

You did an incredible job at diving straight in the deep water and staying there, just exploring the sea floor immediately. Amazing, thank you very much!

[@delcanprobably]

scuba diver yelp review

[@user-tt4dy1ti5v]

(See the video description for the full transcript.) 00:00 Introduction 01:10 What have you learned from playing and designing video games? 01:16 On appreciation for the complexity of systems 02:03 On economics, unintended consequences and evolving systems 02:47 On feedback loops during design 03:38 What are some examples of unintended consequences in video game design? 05:31 On minimalism and aesthetic in design of games and technology 08:29 Do you iterate over the design ideas you had in your mind? 10:25 On being critical over design 10:56 On trade-offs in engineering 12:06 On good programming style 13:22 On hearing advice from others 14:01 How do you structure your game development team? 14:30 On preferring to under-hire rather than over-hire 14:42 On hiring people who do not require micromanagement 16:14 On the management hierarchies in larger companies 16:36 On the management structure of pirate ships 17:55 On a flat corporate structure 18:55 On leaving a company when not satisfied with the rules 20:01 On people switching companies every few years 20:17 Why do you think large companies tend to over-hire? 21:40 On the need for process in large companies 22:38 On large companies choosing growth and money over production efficiency 23:31 What is the limiting factor for your current development projects? 23:49 On the difficulty of hiring due to being in Silicon Valley with many large companies 25:13 On small companies lacking an HR department 25:55 On HR staff and recruiters likely not understanding video game development 27:28 Hypothetically, if you had unlimited resources, what would you build? 29:18 On programmers and depression 29:50 On simplifying all of computing 30:12 On programmers' guilt due to regular fixing of broken systems 30:55 On complex systems and keeping the status quo 31:44 On the notion that previous generations were smarter and, hence, nobody today should change the systems made by them. 32:12 On the decline of civilization 33:39 What can be done with software to more easily foster tinkering, maintenance and improvement in the future? 34:26 On knowledge transfer and scary complicated features in the Jai compiler 35:25 On setting a high quality bar while designing the Jai compiler 36:25 On iterating Jai compiler features several times 37:25 On Pixar's animation software, Presto 38:09 On demo-driven development 38:43 On reviews for the best app / tool 39:51 Are there any heuristics that have guided you in a better direction? 40:23 On cutting some puzzles from the _The_ _Witness_ 41:08 On writing essays 41:27 On 'footnotes' in a video game 43:08 What does video game design unlock by virtue of being a spatial medium? 43:52 On interactive dashboards on websites 45:20 Can you talk about your in-development game series that will be episodic over twenty years? 45:50 On programming fun side projects 46:17 On exploring a space of ideas for a game 46:48 On the episodic game series' combinatorial explosion of interactions 48:33 On the interaction between the Earth and the Moon 49:03 How do you prototype or sketch out your games? 50:22 On the growth and stabilization of raw computing power from the 1990's till now 51:29 On the proliferation of web frameworks 52:22 On game development without schedules 52:44 Why does game performance matter? 53:06 On action games that drop frames with timing-based events 53:48 On games that don't aim to be maximally enjoyable experiences 54:20 On mentorships and the mentor-mentee dynamic 55:06 On achieving a cohesive high-quality experience in games 56:40 On not creating specific experiences for gamers 57:43 On sensory deprivation of computers 58:17 On building better program visualization 1:00:22 On ankle bracelets and cardinal direction adaptation 1:01:00 On camera devices for the blind people that approximate vision via sensations on the arm 1:01:37 On cognitive science and developing intuition 1:03:17 How did incorporating architects and landscapers while developing _The_ _Witness_ affect how you viewed the game world? 1:06:13 Wrap up (See the video description for the full transcript.)

[@fennecbesixdouze1794]

Fantastic job on the part of the interviewer keeping things on track, but also kudos to the editor (maybe the same person?). They clearly spent a great deal of effort editing out lengthy pauses etc. Helps things massively over other interviews I've seen with Blow, and I really appreciate that respect for my time as a listener. Will be interested to listen to more of this podcast if it has other interesting guests on.

[@____uncompetative]

Blow's brain is running in debug mode. Lots of breakpoints slows him down.

[@Nick_fb]

One of the best interviews out there. The sensory deprivation aspect of computing is unfortunately true. 51:00 New computers every six months being completely better than the previous one, was a weird time to be a pre-teenager (tweenager, it was once called?). Set my expectations up for super-awesome-califragilistic and explosive computing growth in the future. It was such a rush! Nothing was exciting as the whole new revolution setting in. Of course the pace of increase didn't last. ahh memories.

[@sergesolkatt]

Devon makes this interview much more interesting by being an insightful interlocutor

[@____uncompetative]

Yes, she did a great job.

[@majorhumbert676]

Great guest, great interviewer, great interview.

[@bernoulli9047]

This episode was my introduction to the podcast and I've stuck around and subscribed. I'd make the argument (right now) based on sheer subscriber numbers/view counts, that this is by far the most underrated podcast in my feed.

[@georgeokello8620]

This particular comment on 19:03 about confusing the boundaries of equality of a workplace versus in general society resonates so greatly with me. The current mentality that is decaying the progress of skill and feedback to improve in workplaces in the modern day has been perverted with politics where people tend to apply certain rent seeking dynamics in the workplace by penalizing risk of making and testing different ideas that do not conform to groups by using political imagery to buy consensus and force ppl whom are talented or want to pursuit excellence by conforming them to group decisions no matter how suboptimal. This dynamic is especially entrenched in very large companies whom have economies of scale where they is absorption of loss for terrible decisions while at the same time innovative ideas at localized teams do not project positive signaling to influence allocation of resources and dynamic decision making. Thus we have systematic production of suboptimal terrible products that consumers have to pay for adversely.

[@LouPeabody]

Great conversation! Yes Jon voices super critical thoughts a lot of the time, but he's so insightful. Nice questions, brought out some really interesting discussion 👍👍

[@UncoveredTruths]

thanks for this, finally got around to listening to it :) was a nice interview

[@TGX999]

Great conversation. The program visualization problem has been something I've thought about for a while. Flame Graphs man.

[@hkva1859]

Good stuff, thanks for uploading this. In the future can you leave the pauses between dialog in? The conversation is a little hard to follow with the gaps cut out.

[@sirfrancis2220]

One of the best indipendent developer

[@4Exp4]

Great stuff

[@Technomancr]

One of the best interviews I've heard.

[@PabloGnesutta]

This is edited kinda weirdly. Anyway, great interview. I really love this man

[@user-sl6gn1ss8p]

On horizontal organization, a lot of the criticism made is directed to experiences which are naive - in the sense that the very problems discussed aren't taken in account when structuring it - and/or misaligned - in the sense that the context they're on imposes pressures which are adversarial to horizontal cooperation. Things like consensus vs consent, competition vs cooperation, "separation of concerns", etc, can all work to mitigate a lot of the common objections

[@timkrueger1179]

Subscribed

[@0xCAFEF00D]

Is this the free version of the podcast where you cut out half the words? What's going on?

[@clickrush]

Paraphrasing: "No successful army in the world was flat". This is only kind of half true. There are some impressive armies that won battles being outnumbered. This is a rare and notable feat and goes against both common sense (more is better) and well-established strategic thinking (don't fight if you're outnumbered). Many of them had a close to complete set of following properties: - elite training and veterancy - leaders that were men of the people, not detached from the reality of the soldier but close to it - officers with very high agency and great ad-hoc communication between them and within their groups - smaller groupings/teams with high cohesion that operate more autonomously within the larger army - ability and possibility for daring tactical decisions Under the toughest and most competitive circumstances you want to be very adaptable and autonomous. You want people who have learned to take responsibility and make decisions, meaning you let them make mistakes (there is no way around this part). You want them to be highly collaborative and engaged with their task. Rigid, hierarchical systems are a sure way to suppress all of the requirements for elite teams to exist. Autonomy is learned, through hard work, mistakes and freedom.

[@sergesolkatt]

Where can I learn more about what you said in this comment? Examples, books, etc.

[@clickrush]

@@sergesolkatt I was thinking specifically about Napoleon (especially infantry) and Alexander - but I'm sure you'll find many more examples. Contemporary ones would include either guerilla warfare or special "commando"-type divisions where small units, high cohesion, high trust and high autonomy are applied.

[@sergesolkatt]

@@clickrush Thank you so much, Denis! 💯🤟🔥

[@urbankoistinen5688]

Ouch: "as you get that n-squared exponential growth".

[@WibottaZu]

Watching this hour-long video while I wait for the mobile app to load. Jk, Notion's great

[@Heater-v1.0.0]

So can I gather that the likes of Facebook hire as many people as they can afford so as to ensure that talent does not end up working for someone else who may be a competitor? Basically paying talent not to work elsewhere, it does not matter if they don't actually do anything useful.

[@raj3shv]

Good questions but the interviewer cut off Blow's thoughts most of the time and it's annoying

[@jack_crawford]

It's an editing problem. They cut out a ton of what he said. Very jarring and kind of hard to follow sometimes.

[@MemoryDealer]

@@jack_crawford Agreed.