Hacker News Reader

Comment by Daub | Hacker News Reader

The article says... "that research effort is rising substantially while research productivity is declining sharply". No surprise there. Nowadays, Universities have to become paper mills in order to get funding. Also... Risky research and cross disciplinary research is a sure fire way not to get tenure.

As to the question, where new ideas come from... I teach creativity... the literature is in near agreement on this topic. New ideas come mostly from recombining old ideas. Essentially, this is an act of playful association. New Ideas also come from new technologies. For example... the neodymium magnet is the reason we had the Sony Walkman.

View on HN

> I teach creativity... the literature is in near agreement on this topic.

Can you point me towards some good papers to start with? I don't really know my way around any of the academic literature outside of engineering.

They replied to the second person requesting this: https://news.ycombinator.com/item?id=22951757

For those curious about the impact of neodymium magnets in the creation of the Walkman, as I was, I found this: https://hackaday.com/2019/07/22/forget-the-walkman-its-the-h...

As one of the Hackaday comments mentioned, the original Walkman must have used samarium cobalt magnets rather than neodymium magnets. SmCo magnets were known since the early 1960s but neodymium magnets were discovered in 1984, 5 years after the Walkman launched:

https://en.wikipedia.org/wiki/Samarium%E2%80%93cobalt_magnet

https://en.wikipedia.org/wiki/Neodymium_magnet#History

Cross disciplinary conceptual mining is a great way to realize the "low hanging fruit" waiting for disruption in any industry. The silos of knowledge are high and thick, just like the skulls of the industry thought leaders who say things can't be done in some different manner.

I've worked as a developer a number of industries, and as an MBA consultant in 3 times as many more. I'm in a developer job at the moment, but when I was actively consulting I'd just learn how they run operations, and the glaring opportunities for substantial expense eliminating and productivity enhancing changes glow like the sun.

> the literature is in near agreement on this topic. New ideas come mostly from recombining old ideas.

Absolutely. Not just in literature/arts but science as well. The ideas of atoms, physics, etc are a "rehash" of ancient greek ideas. Though obviously not the same, but there is no doubt that the study of pre-socratics influenced science.

> New Ideas also come from new technologies.

Even more true. Everything from the invention of letters, numbers to telescopes and microscopes led to new ideas, new arts and new sciences.

Also, another major component is wealth. Research, new ideas, etc require wealth so that you have the time and resources to pursue and expand on these ideas. Ancient greece was a wealthy slave owning society which allowed them the freedom to pursue noble ideas rather than slaving away to make ends meet. This is where we get the idea of "liberal" education. Liberal here doesn't mean the education is free, but rather education is for those who are free ( aka wealth slave owners ). Wealth frees you from life's mundane tasks. It's also why every major civilization or flourishing of knowledge/ideas always was preceded by immense wealth ( america, ancient greece, mesotopamia, ancient egypt, italian renaissance, the englightment, etc ). People mistakenly assume you get knowledge and then wealth, but historically, it's always been the other way around. You get wealthy first and then you pursue knowledge.

Synthesis!

So, take all the ideas, compute their cross-product, sort, and voila! Found all of them, though as it is itself a new technology, recurse... this will take a while...

"New ideas come mostly from recombining old ideas."

Does this suggest that new ideas would become easier to find over time because there are more combinations of old ideas?

It also means that patents restrict the flow of new ideas by locking up the source material for invention.

The idea that invention is a isolated creative force that pops up without dependency on anything else has always been a myth without foundation.

I’ve become extremely interested in this notion lately. When I have discussions about intellectual property with random people the dominant idea is that patent promote innovation. People get angry if I suggest that might not be true, as if they think I want all creative people to go bankrupt or something.

But its increasingly questionable to me. Open source software has shown the great wealth of innovation that occurs when we collaborate instead of hiding our work.

I think we could (probably) sharply increase the rate of innovation if we let go of intellectual property restrictions. I really want economists and other thinkers to take seriously this notion that intellectual property restrictions are deeply harmful to society and study it in detail. Unfortunately entrenched interests benefit from the status quo (naturally) and I feel they would be in opposition here. But we can still work to open source our output. It’s harder when open source competes with closed source, but if we can learn how to use open source to our advantage we might be able to win.

In any case I think it is desperately important that we take seriously the idea that IP laws might be doing more harm than good. Most people I speak to take it as a given that we need them, and I think we’re shooting ourselves in the foot.

> I think we could (probably) sharply increase the rate of innovation...

As long as we're looking at deeply held assumptions, it's probably also a good idea to figure out what the ideal pace of innovation is. None is certainly bad, but too fast also causes problems for people: technological products are expensive and will be shorter-lived as innovation speeds up. At some point, the cost to society of all that churn probably balances against the benefits.

It’s a good point. I would argue however that reducing or eliminating intellectual property restrictions also significantly eases the difficulty in repairing and upgrading old hardware. So obsolete technology is more easily upgraded or repurposed.

How do you solve for 'work for 5 years on open source project, get it to some degree of popularity, watch AWS making a service out of your code without paying you a dime'?

A new licensing model where it's prohibitive to make a profit off of this work.

The more restrictive the license, the more it drives potential users and collaborators away.

That being said, do you have a concrete licensing model in mind?

The Unity licease works.

I think any licease that takes 50% of revenue would be fair to both parties with a minimum for the small user.

50% of revenue might make sense for Unity, where Unity provides clear value. But it absolutely would not make sense for some small library that gets used for a few operations. It also would not make sense for high profit work. Personally I’m seeking an end where everyone shares basically everything they do and no one extracts anything out of the duplication and re use of information.

That end is incredibly sad. We still have to pay our mortgage and our dentists, but can't ask for a fair pay for our highly skilled work. I guess we deserve the ad-based monetization model that has taken over the world. Brave new world, let's shill for megacorps with all the clickbait we can possibly muster.

This is a misunderstanding of my goal. Even if we share everything we can there is always novel work someone needs done, and we can get paid to do it. But also if we share everything we can then most of life will become much cheaper. My end goal is to reduce the marginal cost of living to near zero and then arrange society to provide for everyone. In that case you wouldn’t need to pay the dentist or the mortgage. But IP restrictions add costs everywhere, and those costs must be eliminated. Perhaps we can’t eliminate a mortgage, but we can make your car cheaper to repair and we can make the goods you need much cheaper. Remember my goal is to make life better for all. And I know it is counter intuitive. That’s why I say we need economists to study it. But the goal isn’t sad, the goal is to thrive. The status quo, where a majority of people slave away their entire lives even in the wealthiest nation on earth... that is sad.

> But also if we share everything we can then most of life will become much cheaper.

This is an extraordinary claim that demands actual evidence. Eliminating IP costs might make things cheaper, but physical goods (like food) will still need material, energy, and labor inputs to produce for the forseeable future. How do you plan to bring these costs down to “near zero”?

If you want economists to study this, you’ll need to convince them it’s plausible. That requires at least some back-of-the-envelope reasoning from data that already exists, even if it’s not a perfect example of the policies you’re advocating. There are records from times & places where IP law was either nonexistent or unenforced; what do they say about your theories? If you expect things to be different this time, what leads you to think that (with specifics)?

You're looking for the GPL. Richard Stallman shared your end goal. The GPL has been in decline for a long time though; most people concluded it was a bad idea.

The GPL does not solve our cultural problems. We’ve got to be doing advocacy for open collaboration and study its effects on our economic systems. If people do not understand the importance of open source the license won’t do much to help. And you’re right that GPL has its detractors. It’s highly restrictive. I use CC0/BSD licenses now. It’s perhaps overly permissive but it’s good to experiment.

And without any intellectual property rights, what prevents someone from simply ignoring the license?

Exactly! Only because of IP rights you can enforce a license.

Thank you for your question. I want to challenge your assumptions without sounding combative, and provide what I think is a solution that prevents the conflict from existing in the first place.

First of all I will point out that if one individual creates something useful and then a major service provider picks it up, this can mean very good things for society. If the service is good, hopefully AWS can provide it to many customers and help many people gain access to the utility this creative person has helped realize. There are situations where the big company disturbs the project so much it fails, but I assume you’re actually asking what happens when someone else profits off of your work.

Well in the theoretical I am also imagining that all parties share. This would mean that even if AWS takes on the code and sells cloud services with it, they would only be competing against other cloud providers who could provide an identical service. That is to say there is no one way source code exchange, it should be mutual. In this case, I would argue society only stands to gain from AWS entering the fray. It is a pure positive and not a dilemma directly.

Then there is the question of the individual. This one individual worked hard on something for five years and then a big company swooped in and took the profit they could have made. But if it is open source in my theoretical then anyone could have swooped in at any time, and the creator should have known that was a possibility. They never should have sunk more effort in to the project than they were comfortable with knowing an alternative provider could have taken over the market. That creative person should instead have joined groups of other engineers working on projects that people need. Perhaps they contribute to a project that eventually gets taken up by AWS, but it would not damage their livelihood as they would just shift focus to another.

We play by the rules we’re given. In a society where everyone shares, creative projects do not strictly belong to one individual. We would not make choices that exposed us to risks we did not in some way accept.

Some would say that without this person being willing to sink five years in to something for their own profit, they would never be creative and this would be a loss for society. But I don’t believe that. I’m two and a half years in to a project that I’ve licensed CC0 & BSD. Anyone could swoop in at any time and bring it to production, potentially earning great profit. And I would find it a great honor if someone did that. So you see, the creative drive exists outside of the need for profit.

As far as how people survive - there is a great deal of work that cannot be digitally duplicated. You can pirate a musicians albums but you can’t pirate a live show. In a world where there was no IP restrictions and we all shared, people would find income doing novel work, instead of duplicating something someone had already done. One good hack to this is to automatically provide for everyone. Universal basic services. Then there’s no starving artists.

Sidebar: IP laws are not strictly necessary to enforce this. You are not decent to your neighbor because the law requires it. If society values open source and collaboration and boycotts those who would use IP law selfishly, we can accomplish this all with no change in laws. Some kind of union system can help enforce these boycotts and formalize their terms. Though some laws do need to change to reduce the benefits of playing selfishly.

I think you'd lose a lot of people with the open source argument. Open source software is many wonderful things but it's not innovative. Been using and writing it for over 20 years: the projects that fundamentally change the conversation are nearly always proprietary (or sometimes open sourced for strategic reasons but developed as proprietary). Open source hits are usually clones.

Linux -> clone of UNIX, on desktop sometimes macOS

Git -> clone of BitKeeper

Android -> developed in a proprietary way with minimal community, heavily iOS inspired

GIMP -> clone of Photoshop

OpenOffice -> clone of MS Office

qemu -> clone of VMWare

etc.

It makes sense. Open source projects exist to 'scratch an itch' and are rarely optimised to maximise the size of the userbase. There's little incentive for open source developers to innovate: more users can be as much of a downside as an upside. Mostly it just equals more support headaches. Proprietary products are all about getting lots more users, so there's an incentive to try new things.

Or they would become harder to find when there are too many of them. For example, it's probably easier to find one thing among 3 than among 1 million.

Hmmm. Maybe. But also, some ideas get 'ploughed under' by time. Also, many new ideas require re-stating as they get older.

Yes. How many novels nowadays riff off of history and paint it as fiction?

Another term for this is the Hegelian dialectic [0]. Very similar to your term playful association but it describes that the "first" old idea, the thesis, allows there to be a counter idea, the antithesis. These two counter ideas morph into one idea, the synthesis. A consequence of this term over playful association is that these New Ideas are more tangible and have less sense of randomness.

An example of this playing out - First came Disco, the thesis. People who didn't like Disco came up with Rock, the antithesis. The combining of those two became Pop, the synthesis. And the music timeline continues (don't overthink the music choices I'm completely guessing)

[0]: https://en.wikipedia.org/wiki/Dialectic#Hegelian_dialectic

I’m seconding the request for any literature on the subject! I’m very interested in the harm intellectual property restrictions might be doing to our rate of innovation.

Tons of books out there, very difficult to sort the wheat from the chaff. The following is what I found useful. Its a mixed bag, academic and coffee table.

- Beveridge W. I. B., 1950, The Art of Scientific Investigation, New York. https://archive.org/details/artofscientifici00beve

A charming and easy-to-read enquiry into how scientists conduct research. Old but (like all good science) still applicable.

Kaufman, J. C., & Sternberg, R. J. (2010). The Cambridge handbook of creativity.

Tons of great material. Section 1, chapter 2 gives a literature review of the topic.

- Johnson. S., 2010, Where Good Ideas Come From, Penguin UK.

Packed with lots of good anecdotes, this is a very accessible book.

- Currey, M. 2013, Daily Rituals: How Artists Work, Knopf. Blog avaiable here: http://dailyroutines.typepad.com/.

This started off as a blog and developed into a wonderful and useful little book. In my opinion, creative rituals are a very under-estimated component of creativity.

- Sternberg, R. J. (2001). What is the common thread of creativity? Its dialectical relation to intelligence and wisdom. American Psychologist.

A very readable 3 page essay.

- Christensen, C. (2013). The innovator's dilemma: when new technologies cause great firms to fail.

Perhaps the most well-known book on this list. It was this book that addressed the idea of 'disruptive innovation'... innovation that completely destroys some businesses, and creates others. He uses the railroad as an example, but clearly the internet falls into this category as well.

- Seelig, T. (2012). inGenius: A crash course on creativity.

This evolved from the course notes from one of the first university courses on the subject of creativity. A bit annoying to navigate (her chapter headings are 'too creative') but it has good stuff.

On top of this, there is almost everything that Edward De Bono wrote (the lateral thinking guru).

Wonderful, thank you so much!

> New Ideas also come from new technologies.

This may be why it seems like it's harder to find new ideas now. It was relatively easy during the waves of "do X on a computer", "do X on the Web", and "do X on a phone", but now we're waiting on the next transformative idea that will allow everything old to be reinvented again.

"do X using AI/ML", "do X using robotics" seem to be a pretty effective template.

I will throw out that AI/Ml is less platform-y than 'on a phone', 'on the web'. But I would also suggest the following likely to be platforms

Do X on Alexa (sorry voice enabled UI)

Do X on Augmented Reality

Do X on Unreal Engine (seriously read up on fortnite and the metaverse)

How do you teach creativity? What does such a class look like?

I'm curious what a syllabus for your class looks like.

Author of "Borrowing Brilliance" (Borrowing Brilliance: The Six Steps to Business Innovation by Building on the Ideas of Others) proposes some thoughts about creativity - like, it's a combination of ideas from another area.

Well... I designed the course for the school, but only taught it once. Frankly, it was not a roaring success. One reason was that I was asked to deliver it online. With my recent (obliged) experience in online teaching, I think I could do a better job if I did it again. I incorporated much of the material into other courses, to some success.

Here are my course notes, which include assignment descriptions: https://www.dropbox.com/s/m3s83x036makvhu/Creativitity.pdf?d...

To answer your question directly, like all teaching, it requires that the assignments and exercises be appropriately designed. No matter how much I tried to design an assignment that embodied the ideas I was teaching, students would still come up with really un-creative responses. I came to the conclusion that the only way to guarantee success was to mentor them individually.

Most creativity courses just deliver the 'received wisdoms' (i.e. they are theory courses). Very few actually try to foster creativity. The book inGenius: A crash course on creativity (Tina Seelig) was written by someone who did.

Here is the course description to a creativity course delivered at the University of New South Wales. Its not bad, but for a creativity course, the assignments seems a bit un-imaginative.

https://www.dropbox.com/s/mnbw3v4qzrtpx14/D52C5C9E-1043-4261...

This is amazing work, thank you!

Reply on news.ycombinator.com