00:00hi everyone welcome to day 6 and Z
00:02podcast I'm sonal today's episode is on
00:04crypto networks and decentralization as
00:07a means to an end and on what building
00:08blocks are needed for us to get to
00:10web-scale as it's still just the
00:12beginning the discussion is moderated by
00:14guest host Devin's oogle in conversation
00:17with Chris Dixon and Ali aya both of
00:20Asics in secret oh speaking of please
00:22note that a six and Z crypto is an
00:24independent fund managed by cnk Capital
00:26Management a registered investment
00:27advisor at the SEC it's separate from
00:30the andreessen horowitz family of funds
00:31the content here is for information only
00:33and should not be taken as legal
00:35business tax or investment advice it
00:37does not constitute an offer or
00:39solicitation to purchase any investment
00:41or a recommendation to buy or sell a
00:42security and in fact the content is not
00:44directed to any investor a potential
00:45investor it may not be used to evaluate
00:47or make any investment C&K is not
00:50seeking investors an investment in any
00:52fund advised by C and K may not be made
00:54with that first getting final offering
00:55docs and diligence materials anyway for
00:58more details please also see a 6 and C
01:00crypto comm slash disclosures and a 6
01:03and Z crypto comm slash disclaimers
01:05finally you can find this episode as a
01:07video as well with links graphs and more
01:10on Asics and Z's YouTube channel hi I'm
01:13Devin I am a software engineer and
01:15writer based out of San Francisco today
01:17I'm talking to Chris and Ollie from a
01:2016z crypto about potential crypto
01:23business models and the building blocks
01:25that we need to put in place to take
01:27advantage of that full potential so for
01:29starters can you guys tell me a bit
01:31about why you're excited about crypto
01:33and where you see the potential is going
01:36you know it's a very this the crypto
01:39currency kind of movement is a very
01:40interesting thing because it can be
01:41looked at through multiple lenses so for
01:43example from one lens it's it's a it's a
01:46movement of software developers who are
01:47you know kind of analogous to the open
01:49source software movement that started 20
01:5120 plus years ago of you know it's a
01:54movement to people that are
01:55collaboratively creating communities and
01:57building projects together and building
01:59networks you know you can look at it as
02:02a lot of the kind of press does from
02:04more of a financial aspect and like you
02:06know the prices of Bitcoin and and just
02:09like as a you know store value and a new
02:10kind of alternative financial
02:13you know being kind of being
02:15technology-focused we kind of we like to
02:17look at it or one way we like to look at
02:19it as as a as you can think of a crypto
02:23network like aetherium
02:24or Bitcoin or one of the many other new
02:26innovative you know networks out there
02:28as as a type of computer really and so
02:31whenever you have a new computer if you
02:32kind of look at the history of computing
02:33right you had you had different kind of
02:35new computers introduced along the way
02:36so you started off with like mainframe
02:38computers back in the you know 40s and
02:4050s and then you had the sixties I guess
02:44and then you had the PC you really kind
02:45of started 70s and rise to prominence in
02:47the 80s you know obviously had the
02:49internet in the 90s you had smartphones
02:51in the late 2000s and each time you had
02:53a new computer you had new kind of
02:55things you could capability to unlock to
02:57new applications so let's just take the
02:59smartphone as an example like that was
03:01the first time we had a computer where
03:02you would actually have it on your you
03:03know on your body with you all the time
03:06as you go around right it has a built-in
03:09high quality camera my iPhone does or
03:11your Android phone does right it has GPS
03:13it has like all these other interesting
03:15interesting things which you didn't have
03:16in computers before and that and that
03:19then unlocked if you think about like
03:20some of the some of the most popular
03:22applications on smartphones it's like
03:23for example uber and lyft right take
03:25advantage of the fact that you have the
03:26phone with you all the time and have GPS
03:28and the driver has GPS or you know
03:30snapchat and Instagram take advantage of
03:32the camera that you always have with you
03:34and you can do interesting things with
03:35it right so you had a new kind of
03:37computer and had new capabilities and
03:39that unlocked new applications that kind
03:41of uniquely exploited those capabilities
03:43right and so if you think of if you
03:45think of a blockchain computer like a
03:46theorem as a new computer through this
03:49lens the way you think of it is what are
03:51the new capabilities of this computer
03:52right and I think one of the things that
03:54kind of throws people off in the space
03:55when they think about it is they're used
03:56to thinking about new computers just
03:58sort of being better in terms of
03:59performance and size and other things
04:02crypto networks like a theorem I'm
04:04actually kind of worse than some of
04:05those dimensions so for example as in
04:07the performance side it's not as
04:08performant as let's say AWS or Google
04:12cloud services or something if you
04:13wanted to write a video game or
04:14something it's not as performant but it
04:16does it is it is it has had this other
04:17very very powerful new property which is
04:21right and so the idea that through
04:23cryptographic and game theoretic
04:25mechanisms a blockchain computer can
04:28provide you know provable guarantees
04:33that for example if it says if I if you
04:37know if you have a currency like
04:39aetherium and if then if that computer
04:41says that I you know by owning a private
04:43key own a piece of that currency I can
04:47trust that that that is the case all
04:49right and that's oh I know it so it
04:51allows you to do things like financial
04:53things so if you think about the basics
04:54of any kind of financial operation
04:55whether it be something simple like
04:57owning something or having property
04:59rights or owning a piece of currency
05:01you're owning a digital good or
05:03it requires trust they requires trust
05:05that that you do indeed own it right and
05:09so that's so there's so there's
05:10different dimensions in which sort of a
05:12blockchain computer enables new trust
05:14one is with users so users can feel like
05:16they own something whether it be a like
05:18a piece of property or a piece of
05:20currency it also creates new trust with
05:23developers so software developers you
05:26know when they're building on a on a
05:28network like Facebook or Twitter you
05:30know there's a long history of a whole
05:32bunch of software developers to try to
05:33do that so you know Zynga is a very
05:36prominent company that built games on
05:37top of Facebook there were all these you
05:39know Tweety and all these I kind of
05:42Twitter clients that built on top of
05:43Twitter there you know they had to trust
05:46the centralized organization ie Facebook
05:48and Twitter those developers did and you
05:52know and what and what has happened and
05:54with many of those large large networks
05:56is that over time their business
05:58business interest change so Facebook's
05:59did and you know and Twitter's did and
06:02and and that was that ended up having
06:04you know negative effects for those
06:06developers right contrasted to the first
06:08year of the internet when developers
06:12were building on an open network that
06:14you know if you think about the web the
06:15web standards of HTTP in email SMTP and
06:18all these other kind of open protocols
06:19that were kind of governed by
06:20decentralized protocols and nonprofits
06:22and things like this back then
06:23developers could build on it and know
06:24they trusted it right and so blockchain
06:28computers reintroduced this idea that
06:30developers can build things and trust
06:31them which by the way has a kind of a
06:34means not just the developer can trust
06:36them but an entrepreneur can trust them
06:38who builds a company around the
06:39developer the investor behind the
06:41developer can trust as well and know
06:44that if you invest in this thing that
06:46you know you'll have sort of guaranteed
06:47kind of developer rights right it's the
06:50same reason all if you connect economic
06:52studies like one of the top predictors
06:54as to whether a country receives outside
06:56investment is whether they have strong
06:58property rights and so a similar idea
07:00here that if you have a network that has
07:02strong kind of rights for the users and
07:04developers people are more likely to
07:05invest in it and they're more likely to
07:07trust it and build on top of it and the
07:09users are more likely to adopt it and so
07:11I think what's very interesting now is
07:12we're going to we're in a rut you know
07:13we're ten years into cryptocurrency in
07:16the sense of Bitcoin was and you know
07:17the paper first came out in 2008 but
07:20really the design space really
07:21dramatically expanded especially sort of
07:24the rise of a theory around 2015 and
07:26since then many more smart people
07:28entered the space and have been having
07:30all those creative ideas and we're in
07:31this very interesting period now where
07:32people saying we're saying what can you
07:34do you have this new kind of primitive
07:35this concept of trust they didn't have
07:37before with large-scale computer
07:39networks and now that you have that what
07:41are all the things you can create and so
07:43you know you can create for example a
07:45pretty new idea it just sort of happened
07:46last year was idea of crypto
07:47collectables crypto goods and this came
07:49out of a theory UM's called the ERC 7:21
07:52is sort of a new idea that someone came
07:54up with which is you could have instead
07:55of just having sort of digital currency
07:57fungible goods you could have non
07:59fungible goods and they can have
07:59graphics and all these other things
08:00things like crypto kitties and you're
08:02just were there very very beginning of
08:03people sort of thinking about what you
08:04can do with this new concept one of the
08:06most exciting features of this new
08:08building block is that collaboration
08:12tends to be bottlenecked on the
08:14requirements for trust and so if you
08:16have a technological building block that
08:19reduces the activation energy for
08:20collaboration to happen then that can
08:23open the space of possibilities
08:24dramatically and I think that there are
08:26a number of different things that people
08:28tend to look for or a number like there
08:31are three great facilitators of
08:33collaboration and they're all they all
08:35revolve around trust one of them is the
08:39requirement of trust in mechanisms of
08:41enforcement that there are guardrails
08:43around the collaboration that people
08:46old and who prevents certain about bad
08:49outcomes from come from happening and
08:50there's also the existence of an
08:53incentive structure that enables the
08:55participants to believe that their
08:56counterparty or counterparties will not
08:59behave in a certain in a certain bad way
09:01because it's not in their interest to do
09:03so and then there's also trust in social
09:06signals if there is some mechanism of
09:09reputation that exists and is in place
09:11then people are able to rely on those to
09:15feel good about the way that a
09:17collaboration is likely to go and I
09:19think everything that Chris is talking
09:23they like blockchain technology
09:25facilitates two of those very well so
09:28it provides a new mechanism of
09:31enforcement because now all you have to
09:33do is trust math you trust the
09:35algorithms that support the blockchain
09:38to to feel good about the kinds of
09:42interactions that will occur in the kind
09:43of the kinds of outcomes that will be
09:44possible when you transact with someone
09:46on the blockchain you don't have to rely
09:49on some external some third party to
09:53believe that the transactions gonna go
09:55the way that it was that it was sort of
09:57constructed from the start and it also
09:59helps with trusting incentive structures
10:05you're able to build sophisticated
10:06incentive structures that are simply
10:09just code that give you confidence that
10:12participants in any one particular
10:14interaction will behave any one way the
10:16one that is still missing is that of
10:19social signals there's still no real
10:20good mechanism for reputation or for
10:24identity on the blockchain and as one of
10:26the one of the things that many people
10:28are working on and that's a building
10:29block that will be needed going forward
10:31and it we're so we're still very much at
10:33the very beginning of this this new
10:35there's no technology so even though
10:36it's been 10 years it's only sent in
10:38that it's--it adds that one of the
10:39challenges in building you know these
10:41systems on either blockchain systems or
10:44systems on top of block chains is is all
10:47of these systems today have to assume
10:49that any participant of the network
10:51could be adversarial meaning you don't
10:54know who that person is so you just have
10:55to assume the worst and therefore you
10:58through through the kind of game
11:00theoretic mechanisms and the
11:01cryptographic mechanisms that the system
11:03as a whole has these emergent trust
11:05properties even those individual actors
11:07could be bad if one were to have a
11:09better kind of system of reputation or
11:13you could you could relax that
11:15constraint have know that participants
11:18have some history you're playing kind of
11:20an iterative game and it would enrich
11:24and I think it would it would widen the
11:26the design space of possible
11:28applications you could build what would
11:30these identity or reputation systems
11:32look like well I think that one of the
11:35hardest problems in the space is mapping
11:37cleanly one public key or address on the
11:40blockchain to an actual human being so
11:42this is known as the civil problem the
11:44problem that any one participant can
11:46create multiple addresses for themselves
11:48and pretend to be many different people
11:49as they're having a good way to map an
11:52address to a real person it's a very
11:54it's a very difficult problem would
11:56solve a lot of these problems because
11:57then we could associate some metric of
11:59reputation with that particular identity
12:01and so I think in the in the and the old
12:03web in the web web 2.0 and even what 1.0
12:05this was solved via DNS where you have a
12:09thority certify that particular domain
12:13name on the internet corresponds to a
12:14particular organization and to their IP
12:16address and their server and that allows
12:19people who are interacting with their
12:21websites to know that they're actually
12:23talking to the right people and that you
12:25know they were sort of the reputation is
12:28actually theirs and so some equivalent
12:31in this in his new crypto world is
12:33necessary and you can imagine many
12:35different solutions you can imagine
12:37people trying to build something like
12:39the certificate authority model for
12:41crypto where you have some authority
12:44that vouchers for the mapping between a
12:45public key or an address and an actual
12:48human being or it could even be more
12:50decentralized than that and it could be
12:52that that people vouch for each other's
12:53identity on the blockchain you can end
12:55up with a web of trust kind of system
12:57where the the trust is emergent in his
13:01bottoms-up as opposed to top-down that's
13:03really interesting and something that's
13:04newly possible with block chains what
13:07are other ways that block chains enable
13:11sort of treated in different ways from
13:13what one and 2.0 well I think there's a
13:16there's no there's a related topic which
13:17is kind of control of user data or what
13:20people sometimes called data sovereignty
13:22and so you know as we see in the news
13:24this is a very hot topic
13:26you know gdpr was a new european law
13:29which which is going to impose some some
13:32more regulations on tech companies and
13:33just this general kind of meme out there
13:35that you know you sort of had made these
13:40agreements with social networks and
13:41other kinds of search engines and other
13:43things users had sort of used these
13:46services expecting their you know their
13:48data to be kept private of course the
13:49business models of these companies is to
13:50use that data to target ads and things
13:52like that and there there's a there's a
13:56sentiment out there that that maybe that
13:59kind of has gone too far and that we
14:01need to kind of swing the pendulum back
14:03and as technologists think more about
14:06architectures that allow users to have
14:09more control of their data I think this
14:11will become more important over time
14:14just number one as obviously the
14:17Internet becomes you know continues it
14:19will continue to grow will continue to
14:22work its way and deep more deeply into
14:25other way to put parts of the economy
14:29you know I think also you know we have
14:32these issues like you know healthcare
14:34data you know who's gonna hold the
14:36healthcare data who's gonna hold AI data
14:39so you know one of the big big sort of
14:41other mega trends in the world right now
14:42is that AI is getting better better the
14:44algorithms are getting better and tech
14:46companies are investing do you know tens
14:48of billions of dollars into this the you
14:50know the really valuable thing that
14:52people have that that need for these to
14:55create AI systems is they need
14:57algorithms they need computers but they
14:59also need data and data is really kind
15:01of a scarce resource and you you create
15:04data I'm creating a idea right now as
15:06I'm speaking and we all are right so
15:08throughout our life we drive around
15:09we're creating AI data and like that's
15:10what companies like Google and Tesla and
15:13things are trying to capture is that
15:14data to get better and better than their
15:15algorithms you know and one question is
15:17should that data just only reside at
15:19these large tech companies or should it
15:20be an open asset that a book
15:22should the users own them like I don't
15:24know what the answer is but it does seem
15:27like we shouldn't just sort of blindly
15:29assume that the old model for how we
15:31kind of gave data away for free to tech
15:34companies just oh it should always be
15:35the right model going forward but there
15:37are a lot of business models on the web
15:38days but one of the most dominant ones
15:41is that users will do behavior on the
15:46web they will input posts on Twitter on
15:49Facebook they'll click on ads and
15:52they're getting the product for free but
15:55it's not truly free in the sense that a
15:57lot of these services then use the
15:59information that they're getting from
16:00them what are some new business models
16:03that crypto know makes possible yeah I
16:04mean the most that's a great question I
16:06think the one of the most interesting
16:07ones is sort of the quote business model
16:09of let's say a theorem or Bitcoin and
16:11it's a brand-new model and let me
16:12explain it so the model of it lets us
16:14take etherium is they have a certain
16:17number of a certain amount of ether
16:18which is the kind of native currency to
16:20the network and the way it works is sort
16:23of the company company be the it's
16:24really the foundation nonprofit
16:26foundation the developers behind it have
16:29a certain amount of ether which they use
16:31to you know fund the operations of the
16:34company right so let's just say like the
16:35rule of thumb is 10 to 20 percent of the
16:37currencies kind of held back for funding
16:39the operations of the company and that's
16:41kind of the business model of the
16:42company right if and the idea is if they
16:44go out and build a great Network and it
16:46becomes really popular and successful
16:47demand for that currency will go up that
16:49that curd the the value of their
16:51holdings will go up and that's their
16:53quote business model what's so nice
16:55about that model is it also lets the
16:57users participate in the upside of the
16:59network right so you know I was an early
17:01I know whatever more than ten years ago
17:04you know what if all of the early users
17:06who kind of helps kind of you know
17:08galvanize the you know the early network
17:11before it really had crossed over and
17:13and and gotten kind of major mainstream
17:16network effects what if those users
17:17could have just paid it in the upside of
17:19that network right you know what if you
17:23know the developers who were building
17:25stuff on the side of it could have also
17:27had you know participated and they could
17:29have held tokens as well it would have
17:32it would have been a great thing because
17:33everyone would have been aligned
17:35the the you know the Twitter itself will
17:39save had it had this model they were
17:42just want to see the network at used
17:43they wouldn't need to say hey I needs it
17:44like take away you know if this if
17:46Tweety is getting more users and that
17:48makes the demand for my token go up like
17:50that's great right I don't need to why
17:52would I cut them off it's the opposite I
17:53want to help them right the users would
17:56say hey you know I want to go tell my
17:57friend about this thing I want to get
17:58them on there cuz I own part of it and
18:00all my friend own part of it and we can
18:01all like kind of rise together right and
18:04so it's really it's a really just sort
18:06of elegant kind of beautiful business
18:08model where it aligns all the
18:11participants the network whether they be
18:12the core development team the users of
18:15the network the developers you know
18:17other kinds of parties so it's really I
18:18think it's one of the great business
18:21model innovations of the last few
18:22decades and and I believe you know
18:24having worked on the internet for almost
18:2620 years now to me it's finally like the
18:28native business model of the internet
18:30like I think it's up until now we've
18:31been crafting really what are just like
18:35you know like you think about the ad
18:37model it's like you know it's an old
18:38model it's billboards and TV ads and
18:40they're intrusive and you know there's a
18:42whole bunch of issues with them and most
18:44people don't really like advertising but
18:46you know that's what we knew and that's
18:48what we sort of took and so in the early
18:49being I think when you look back on the
18:51multi hundred year history of the
18:52internet at first it's sort of like
18:53whatever you have a new media you know
18:55you tend to sort of impose you know
18:58older ideas on to it you know and if you
19:00go back and watch the first movies that
19:02were made they they had these they were
19:04like plays and they film them and they
19:05didn't understand that you could do it's
19:06a whole new grammar and a whole new kind
19:08of thing and so I think you know we're I
19:10think we're kind of still pretty early
19:11in the development of the internet and
19:12we've now discovered this great business
19:14model which I think will be the dominant
19:16business model of the internet and is
19:17the true kind of native business model
19:19of of networks and then the court that
19:22in the corollary is that crypto tokens
19:25coins assets are the native kind of
19:28asset of the Internet
19:30and as more and more networks are built
19:32and grow and become more important their
19:34lives those assets will become more
19:35important and more valuable the Twitter
19:37examples fascinating to me because like
19:39you were describing before Twitter
19:42currently it has to monetize directly
19:44through the platform they have to put
19:47ads on there something
19:48they have to have Promoted Tweets and
19:51for them having these third-party apps
19:54that may be nicer clients that do
19:56different things for users that they
19:58want it's just something that they have
20:00to maintain and like sure it does like
20:02on net increase the value of the network
20:04which eventually comes back to them but
20:06mostly it's an externality yep if it
20:09were a crypto network they be able to
20:11capture it that's right maybe they would
20:13all be aligned then they'd all be
20:14helping each other and it's a really
20:15beautiful thing I think one thing people
20:17underestimate its if you've been in the
20:18tech business you've seen this but I
20:20like to say the most vicious battles in
20:22tech are not between what economists
20:23call substitutes there between
20:25complements and so what I mean by that
20:27so a substitute is you know you're at a
20:29picnic and you have you with our hot dog
20:30or a hamburger right as a compliment is
20:32when you have a hot dog you also want a
20:34hot dog bun it's a thing that goes along
20:35with it right in technology you know you
20:37have so you know the equivalent the hot
20:39dog and hot dog bun is you know is
20:41windows and you know quicken or
20:43something back in the old days and like
20:44in the modern world it would be Twitter
20:46and tweet deck or something right
20:48those are compliments they each make
20:50each other better right but the most if
20:53you actually look at the history of
20:54Technology the really vicious battles
20:56are between compliments not between I
20:58mean substitutes have battles too but
20:59really is the compliments and what
21:00because what happens is there's this
21:02kind of natural s-curve development of
21:05platforms and networks and so but I'm
21:08including by the way operating systems I
21:09think if it's their kind of forms of
21:11networks there there are their platforms
21:13of developer Network so Windows iOS
21:14Android Twitter Facebook I think Google
21:17search in a lot of ways could be seen as
21:19a platform because you build websites
21:20they get you know they get shown on the
21:23search results and if you look at the
21:26history of it like you know like Yelp is
21:28up you know going and testifying in
21:30front of Congress talking about how
21:31unfair it is Google is listing their own
21:33restaurants ahead of Yelp right they're
21:36not they're not go up on the front of
21:37Congress complaining about the
21:38substitute whatever you know the
21:40competitor to Yelp right you know the
21:42their famous their books written about
21:44all the stuff that Microsoft did and
21:45huge antitrust case and everything else
21:47around all the things they did to fight
21:49with their compliments right you know
21:51the Facebook Zynga's and the tech
21:53world's famous and the you know to the
21:54Twitter and it's ecosystem is a famous
21:56battle and so we spent so much it's
21:59really kind of a shame I feel like you
22:00know someone who works in the industry
22:02how much time and energy and money and
22:04and you know and just suffering is
22:07around like intranet work fighting okay
22:09and and that has to do with the fact
22:12that we have this I feel like we've had
22:13that we've had this kind of like
22:15distorted business model all along which
22:17is like you've got to like control
22:18everything and only and everything and
22:19it's kind of like hoarding kind of
22:20mentality because it's just it's forced
22:22I'm not criticizing these companies
22:24necessarily it's just it's the way the
22:25business the logic of the business model
22:27and the way the capital markets work and
22:28all these other kind of forces that kind
22:30of forced people to do that but now we
22:33like now we have a way where these these
22:35people can be aligned and there will
22:37still be competition but it'll be
22:39competition between networks and more
22:40much more like substitutes and you know
22:43the competition is a part of any kind of
22:45market system you want it but you want
22:48to be healthy right you want it to be
22:49like for the right reasons you want
22:50people to be competing by creating
22:51better products and more useful things
22:53and like as opposed to leveraging my
22:56existing incumbency to you know hurt the
22:59user experience to get more profits
23:01right so I think you know you will still
23:03be competition about sort of being you
23:05know naive about that but I think it
23:06could lead to sort of more healthy
23:08competition and more internet work kind
23:10of harmony incentive structures end up
23:13defining almost everything and at the
23:15end of the day that's what Charlie
23:16Munger famously said show me the
23:17incentives I'll show you the outcome
23:18yeah you know and crypto I think has
23:20created a whole new set of incentives
23:22it's turned things like the Twitter
23:24situation from a one-off prisoner's
23:27dilemma into an iterative game what are
23:29some other places where you guys see
23:31this dynamic playing out right now or or
23:33on the horizon well to some extent I
23:36think that the technology enables broad
23:38experimentation with different in
23:40incentive structures simply because it's
23:41now software so you can now use code to
23:43express all sorts of different kinds of
23:45incentive structures that participants
23:47can engage with and whilst cerium is one
23:51manifestation of this think there I like
23:53a very large number of different
23:54different experiments are currently
23:56being run yeah I mean as well as alia
23:59was saying the you know one of the
24:02amazing things I think about software is
24:03that you know I think it's I think it's
24:06software is sort of misunderstood
24:08sometimes to be sort of a field of
24:10engineering that's sort of a relatively
24:12you know constrained set of decisions
24:15when in fact I mean you know as we know
24:17from like mathematics and philosophy and
24:19things that if if something can be
24:20expressed in logical form it can be
24:22expressed in a turing-complete
24:23programming language and so I think
24:26people really underestimate kind of you
24:29know the the breadth of the
24:30possibilities of what you can build
24:31software and one of the beautiful things
24:33about this this world of crypto world is
24:35that you know the Internet is basically
24:37I mean it'll always get better but it's
24:38basically the hardware side is basically
24:40built out right we've got broadband in a
24:41lot of places a movie need more I did
24:43not you know there's lots of place in
24:44the world we're developing world for
24:45example we need to get more internet
24:47access and more smartphones out there
24:49and things but it's but it's pretty well
24:50built out like a lot of you know three
24:51billion three a half billion people off
24:52smartphones in access and don't continue
24:55to grow and so right now it's mostly
24:57about software and you know and the
24:59beautiful thing with you mentioned
25:00incentives is that if you think you know
25:02the way the internet Arkansas
25:03architected right it's one of the genius
25:05one of the many genius things about the
25:07Internet is the the idea that the the
25:09core protocols are relatively simple the
25:11brains sit at the nodes not in the edges
25:15right they sit at the at the computers
25:17the protocol itself like tcp/ip and all
25:19these things are UDP whatever they're
25:21all very relatively simple and that
25:23means the Internet has the ability to
25:24kind of upgrade itself with the right
25:26incentives so if you build you know if
25:28you look at there's two ways to look at
25:29Bitcoin mining and you'll see some
25:31articles the criticize it for taking too
25:33much electricity and everything else and
25:35that some of those critiques are valid
25:37on the other hand like it's kind of
25:39amazing that a you know eight nine page
25:42whatever the version wipe nine page
25:43white paper it someone can write nine
25:46page white paper and ten years later
25:48there's you know there there are
25:51thousands of data centers filled with
25:52computers running software right I mean
25:54that's just shows you the power of
25:55incentives right a whole new economy on
25:57top of this way that's right and it all
25:59came from it literally is you say I
26:01think the best way to think of the
26:02Satoshi's original paper is nine pages
26:04of incentives I mean there's not such
26:06there's computer science and games there
26:07and other important things but it but
26:08the incentives are kind of the key part
26:09and and those incentives created this
26:12whole new industry and so now what if
26:14you take that you kind of riff on it and
26:15like what else can you do when you
26:17design those incentives what if instead
26:18of those miners just kind of performing
26:21idle computations they were actually
26:22storing useful information and what if
26:25you took the same incentives and you and
26:27you and you had them do things that
26:29would that would provide a purpose so
26:30that does caching and redundant seen all
26:33sorts of other things you might want
26:34from a storage Network and then there's
26:36a whole bunch just the beginning I mean
26:37there's a thousand interesting ideas of
26:39riffing on how you can use this kind of
26:41incentive designs and to other things
26:42yeah because ipfs and and that whole
26:45ecosystem of tools is seems to be sort
26:49of the next step of things that need to
26:51be built out before we can build out the
26:52application layer so you mentioned
26:54broadband you know if you don't have
26:56internet access you're not gonna be
26:57doing very much on the Internet yep um
26:58once that's fully deployed and we're
27:01getting closer and closer then there's
27:03sort of another level of infrastructure
27:05right now we have Amazon AWS but we also
27:08have other things for the web 2.0 for
27:12crypto it seems like ipfs is sort of the
27:15AWS of of crypto or at least s3 can you
27:19dive into some of the other building
27:21blocks that need to be implemented yeah
27:22well we can that's a great question I
27:24mean so yeah I think there's also just
27:26on the compute layer there's a whole
27:28bunch of things need to be done to make
27:29these computers more scalable it's a
27:31whole bunch of work going on at
27:32different levels both the core
27:34blockchain work so for example a theorem
27:35is working on moving to what's called
27:38proof of stake but also sharding and a
27:39bunch of other things that sort of
27:40so-called layer 1 scaling and there's
27:41layer 2 scaling which is like what
27:43they're called state channels and Plasma
27:44chains and all sorts of other stuff and
27:45they're just and then there's sort of
27:46sidechain stuff there's like this this
27:48interesting work and could have been
27:49like fundamental cryptography around
27:51these are these techniques called like
27:53Starks for example that let you do more
27:55computation off chain and only do
27:57verification on chain which gives you
27:59effective in performance improvements
28:00Intel has a technology called SGX which
28:02also lets you do something similar which
28:04you can have do more computations off
28:05chain there's there's really interesting
28:08kind of side network design so there's
28:10all sorts of stuff to sort of just
28:11generally scale and make these things
28:12web scale so web scale is a term we use
28:14in our business to mean like something
28:16used by billions of people right right
28:18now like the these networks are not on
28:20that scale so you need you need the core
28:21kind of network to scale and then you
28:23need other things you that allows you to
28:25do kind of private secure transactions
28:27because most block chains are most of
28:30it's public there's as we were talking
28:33before like things around like
28:35reputation and identity around the user
28:36side there's just a lot of basic tooling
28:39that like you're used to I'm sure as a
28:41software developer I think you you would
28:43know this better than I would but you
28:44know better IDs and debuggers and
28:46whatever just like the whole thing you
28:48know libraries and you know uh you know
28:51verification you know security review
28:53and like a whole bunch of other kind of
28:55stuff that you just want you know in a
28:57modern development environment that just
28:59doesn't exist yet cuz it's still
29:00relatively early um so it's really a
29:03long list of things I mean you can look
29:04at that as like okay we a lot of work to
29:05do you know that's true on the other
29:07hand it's also a great opportunity you
29:10know it's there's just so much more it
29:12sort of this you know we just discovered
29:13this new continent and yes we hadn't won
29:16up houses yet and we don't have roads
29:17yet and you can look at it that as a
29:18negative where you can flip it around
29:19and say you know what this is like a
29:21great place to go and have a career and
29:24we're seeing more and more of that
29:25happen and that's one of the really
29:26exciting things is that we're seeing
29:27just a real wave of talent of people
29:31coming out of colleges and tech
29:33companies and other places and saying
29:34you know what I'm sort of tired of
29:36living on this other continent it's all
29:38filled up and it's gotten kind of mature
29:40and mainstream and something exciting
29:41anymore and I want to go I'm gonna go
29:43get in the you know go go in a Wild West
29:45go in the pioneer be a pioneer and that
29:48and that's how it feels so there's a lot
29:49of work to do but it's also like a lot
29:50of opportunity and there's a ton of
29:51smart people working on stuff plus all
29:53this stuff had to be build out built out
29:55for the web to begin with so it's not
29:56like it can't be done I think there's a
29:58lot of revisionism today I would argue
30:00having lived through the web filled out
30:02like people sort of I think sometimes
30:05imagine that it kind of popped out fully
30:06built or something the experience of
30:09using the internet the 90s it was
30:10literally like you literally had to wait
30:12for a single image to load and it would
30:15go like this and actually you would
30:16change your browser setting systems you
30:17would have like different rendering
30:18methods because it was that slow and
30:20video did not work you know YouTube
30:22finally like was 2005 when you finally
30:25had like enough broadband out there
30:27the websites always funny to look at
30:29those nineties websites things you know
30:31chips and like all the other ridiculous
30:34then the dead didn't have like
30:35responsive design and they couldn't get
30:37to reload the page it was just like a
30:39fun thing to do for people that think
30:41the web came out fully built is to go
30:43watch movies for the nineties so you
30:46watch movies from the 90s it's like a
30:47good historical record of the Internet
30:49in people's daily lives and the nineties
30:51and the answer is there was no internet
30:53except every once in a while you'll see
30:55somebody go and though there were no
30:56smartphones no other thing it wasn't the
30:58thing you didn't even talk about like
31:00LinkedIn and Facebook and Twitter and
31:02tinder and whatever and they were out at
31:03dinner and they were like who as in that
31:05movie and like I guess we'll never know
31:06you know I guess we'll just never know
31:07who's in that movie that's how it was
31:08right like that was what you did in the
31:10nineties and no one thought you could
31:12just well let me take out my magic
31:13supercomputer my you know and that's
31:15connected to all information in the
31:17world and type you know a few words and
31:18then know the answer to everything like
31:19it didn't happen and what when they do
31:21connect to the Internet the 90s they
31:22would be this thing where they would go
31:23and it's like a ceremony and they would
31:25go in front of the computer and of
31:27course the screen was like this thing
31:28was like you know 20 inches deep and
31:29then there would be the sounds and like
31:31as your modem is connecting and then
31:33you've got mail and then you'd been
31:34basically only application of email the
31:36nineties really that people have used
31:37besides like you know developer types or
31:40nerds or whatever like us was uh was
31:44email in fact there was a famous
31:46McKinsey study is like late 90s or
31:48something like would you want broadband
31:49and no one said and they surveyed people
31:51and were like why would I pay and it
31:53basically but people said no and they
31:55the people included the problem when you
31:57get wide adoption because of the time
31:58broadband but just meant faster email
32:00and really has pretty good you know
32:01whatever I need to get faster so so the
32:05web you know and by the way like the
32:07other thing about the web is that you
32:09know the idea that today it was a fait
32:12accompli and like it's obvious like Bill
32:14Gates wrote a book in 1995 called the
32:16road ahead predicting what's gonna
32:18happen the futures 95 this is you know I
32:20think it was well after Yahoo and
32:22Netscape and things like this and the
32:24word the web did not appear in the book
32:25right they actually released an edited
32:27version the year after they added it
32:29back in he did however talk about at the
32:31information superhighway so there was
32:33this there was two revisions of the
32:34world back then people sort of you know
32:37some people just didn't believe in any
32:38of it like and then lot people don't
32:39believe me but and then some people
32:40believed in the web and some who
32:41information superhighway and information
32:43superhighway was gonna be like it was
32:45like Comcast and Disney and Microsoft
32:46and they were they were like running
32:47broadband fibers into like a town in
32:49Florida I forgot which town and they
32:50were gonna have and their argument was
32:52like look it's gonna be this top-down
32:53centralized thing and we're gonna have
32:55but it's gonna be awesome because we're
32:57gonna have high-speed video and day one
32:58and we're gonna ball these are the
32:59things that downside is you literally
33:00had to installing a twenty thousand
33:01dollar like whatever was like a SGI
33:04computer in your house or something but
33:06um and then the web was this ragtag
33:08militia of like hackers and developers
33:11who were like yeah it kind of does suck
33:12and it like it takes forever to download
33:13and like even load an image and like
33:15nothing really works but like you can
33:17just like anyone in the world can just
33:19take a computer connect to the internet
33:21and then what happened was you know it
33:23inspired all of these developers who
33:25said you know what this is cool like
33:27this is this I could do something like
33:28what if I created a website this what if
33:30I created services to this what if we
33:31did created you know that back then like
33:33you didn't have encryption on the
33:34internet you didn't have Netscape people
33:36forget sometimes they created JavaScript
33:37programming language they created
33:39JavaScript occurred SSL like reddit
33:40cookies you never way to keep state like
33:42iswhat cookies are it's okay to create
33:45all these things and it and you know
33:46there's sort of two sides to the coin
33:47there always are with these new
33:48technologies one side of the coin is wow
33:50this thing is early this is not much
33:51data said the Queen's oh my god look
33:53like it could make a great career out of
33:54this because there's so much opportunity
33:55so many new things to do and so that you
33:58know so the web I mean really I don't
33:59think the web is seriously usable by
34:01mainstream people until the mid-2000s
34:03and you could even argue even then like
34:05it was still this thing you went and if
34:07you look at the statistics like the time
34:09spent on the web was was not that great
34:12until smartphones really until it became
34:14just like this thing that was always a
34:15whole concept of online offline right
34:17like the very fact you had to like go
34:18and go online was such a hurdle that
34:21most people like I'm gonna be online now
34:22I'm gonna go surf the web it was like
34:24this thing you did it was kind of like
34:25TV but slower or something you know now
34:28it's as sort of this constant state that
34:30we're in for a better or worse where
34:31it's just just you know it's finally
34:33fully really built out so I that's a
34:35long-winded way of saying it took a long
34:37time I'm more optimistic about the
34:39crypto stuff happening sooner because
34:40it's software base meaning the
34:42hardware's already built out and now we
34:43just need to like like a few quick you
34:46know a few clever incentives of some
34:48nice software and you suddenly have like
34:49you know some some rapid adoption
34:51earlier you mentioned the difference
34:54the information superhighway and what
34:55became the web and we all know
34:58information superhighway is not really a
35:00term we use anymore can you dive a
35:02little bit into the difference between
35:04the growth curves for both of those
35:05things there's two types of two ways to
35:08take a protect product to market right
35:10and the first way is there's sort of a
35:13one step process where you have it build
35:14a centralized system and go directly to
35:16end users and there's a second way to do
35:17it which is to first tips or build a
35:20platform for developers and then have
35:23those developers build things for end
35:24users so an example I like to use it's
35:26not it's it's not a strict analogy but
35:28is Wikipedia versus in Cardiff so
35:30Encarta was a microsoft owned
35:32encyclopedia product that was sort of
35:34dominant in the late nineties and so it
35:35was like you go there and it was very
35:36nicely done it was like you look up like
35:38whatever like a tiger and you'd see a
35:40page for a tiger and someone wrote that
35:41out and have a nice graphic and
35:43everything else and then Wikipedia was
35:45started in 2001 and you know it was
35:47basically a bunch of blank pages right
35:49um and but it had this ability for
35:52anyone to come and become a contributor
35:54and to have a discussion about it and do
35:55all sorts of other things and it started
35:57to get some more and more users and but
36:00you know at the it's sort of it had a
36:02different it couldn't just be it wasn't
36:05a one step go to market and by that I
36:06mean it wasn't you built Wikipedia like
36:09Jimmy Wales and his you know co-creators
36:11and then you know regular end users came
36:14to that website and got value out of it
36:16right instead he first had to sort of
36:18say hey let me go get and get this these
36:20contributor communities excited about
36:22contributing to it and then slowly over
36:24time as they'll create some content and
36:26then some users will come and that will
36:28sort of make the contributors happy that
36:29people who read their content and then
36:31as more you know and then other
36:32contributors will see that and then
36:33it'll sort of you get this flywheel
36:35going and if you look at the today you
36:37can look at the growth curves and what
36:39competed took awhile and you know it was
36:41actually very controversial like 2004
36:43and five because it had some you know
36:44decent amount of misinformation and just
36:46gaps and other kinds of things but then
36:48it just like hockey stick tup once it
36:51got going there was a famous nature
36:53article in 2007 that studied its
36:56accuracy against like Britannica and
36:58Carter and things instead it was
36:58actually as Acuras that's kind of what
37:00at that point it was getting banned on
37:01campuses and it was very controversial
37:03today it's kind of funny to realize that
37:05because it's so obviously the you know
37:07gold standard for you know for a lot of
37:09this information so and then it just
37:12sort of hockey stick tup and meanwhile
37:13in Carta you know and started off in a
37:15better position because I'm day one they
37:17could have you know a really nice page
37:20for a whole bunch of content but over
37:21time they just couldn't scale in the
37:22same way and they shut down I think goes
37:242009 Microsoft shut it down finally
37:25because it was just getting crushed
37:26right and so so I would use that as an
37:30analogy where you know Carta is the
37:31equivalent of kind of a centralized
37:32system so like the information
37:34superhighway was an attempt to do a
37:35centralized version where you would
37:37literally have a company or a set of
37:39companies go and lay all these cables
37:40and build all these services and build
37:42content and do the whole thing and end
37:43and the advantages on day one it would
37:45be really nice just like in Carter was
37:47but over time it would never get that
37:48flywheel and that's what that's what the
37:50web one I so the web start off worse but
37:53it got better at a faster rate by
37:55getting this flywheel of entrepreneurs
37:56and developers and users and so and that
38:00you know so the web did that is a great
38:02example like Wikipedia versus Encarta is
38:04a great example in the case of the
38:05information superhighway there was no
38:07real way of scaling it as well wasn't
38:09one of the limiting factors that you
38:10can't really do that for the entire
38:12globe or for the entire yeah I mean they
38:14thought the theory would be you know the
38:16way you did the theory they had was you
38:18would try to scale it like you would any
38:19centralized service was just like get a
38:21whole bunch people to sign up and they
38:22pay more money than it costs and all
38:23these other things and like you would
38:25just kind of fund all the content
38:26development and do all of the things but
38:28you know they had that theory but you
38:31know the practice you know we know how
38:33that that history ended look I don't
38:35think it's one size fit all fits all I
38:36think there are cases where you want you
38:39know centralized services and other
38:41cases we want decentralized and there's
38:42and there's different pros and cons I do
38:44I would argue that the pendulum has
38:47swung too far towards centralized today
38:49so we have you know essentially you know
38:52the vast majority of people's time they
38:54spent on the internet now is spent at
38:56Google Apple using a service by Google
38:58Apple Facebook Amazon Netflix etc and so
39:02you know and I think that I think that
39:04that I believe we're sort of peak
39:06centralization and we're gonna see more
39:07and more interesting things happening
39:09and sort of this kind of you know it's
39:10like a lot of things in life like
39:12politics and you know fashion trends and
39:15other things look there's a natural kind
39:16of equilibrium State and we've sort of
39:18really hit sort of centralization and
39:20there's gonna be more and more and
39:21that's motivate through cultural factors
39:23I mean you know people who want to work
39:25on these stuff and entrepreneurs and all
39:26sorts of other things like don't you
39:27know like the incentives I was talking
39:29about earlier like not having the right
39:31incentives when you build them
39:32centralized systems like it just
39:33naturally pushes a lot of entrepreneur
39:35energy outward and we're seeing that and
39:37that's you know a lot of what we do and
39:39our investment businesses sort of you
39:40know try to vac those people I think
39:42people talk a lot about decentralization
39:43is the thing that is the ultimate code
39:45the thing that we should optimize for
39:47but there is this question as to why why
39:50is it that the centralization matters
39:51what does it actually buy you and one of
39:53the framers that we've been toying
39:55around with here at the firm is trying
39:59to break down what are the what are the
40:00four primary things that actually matter
40:04for end-users and that actually matter
40:05for developers so the for our security
40:09which is sort of the obvious one we so
40:11that's the kind of thing that ultimately
40:13there's scalability because in order to
40:15build anything that is web scale as
40:18squares were saying that billions of
40:20people can get on you you have to have
40:22some mechanism via which technology can
40:24scale and that's not really the case yet
40:26there's usability how do we how do we
40:30interface with end users and what
40:32developers it has to be usable
40:35experience that that is seamless and
40:37that is at least comparable to existing
40:41solutions for perhaps like the same
40:43things that they may be trying to do
40:44yeah on the topic of usability something
40:46you've talked about Chris is the
40:48importance of stable coins can you
40:50unpack that a little bit yeah same core
40:52is a very interesting concept which
40:53we've made a few investments in which is
40:56the idea of having a cryptocurrency
40:58where the value of the currency is
41:00pegged to something like the dollar for
41:02example it depends of different things
41:03but the dollars the obvious thing and
41:05the advantage of that is that if that
41:07can be if that can be properly done is
41:09that it makes you know kind of the
41:12financial use cases for cryptocurrency
41:13much more mainstream accessible you know
41:17most people want you know wanted to
41:20nominate things in their local currency
41:21that's how they think that's what they
41:22pay their taxes in that's you know what
41:24they need so like if you want to have
41:26like a mainstream lending platform a
41:28payment platform other things like I
41:29think I believe that that a stable point
41:32critical ingredient in that and there's
41:34a bunch of really interesting attempts
41:36to do that to kind of how do you the
41:37question that is how do you create that
41:39kind of technically how do you create
41:40that mechanism that keeps the coin
41:42stable right because just in a normal
41:44open market the reason that the price is
41:46a Bitcoin in theory might set or go up
41:47and down is that if they just change
41:49with supply and demand so how do you do
41:51that and so there's been a lot a lot of
41:53interesting work in my mind this is one
41:55of the really interesting thing is that
41:56that I think will likely start to become
41:59more prominent over the next year or two
42:02and could unlock a whole new wave of
42:04kind of applications and just broader
42:06mainstream adoption cryptocurrency
42:08awesome and then there is a process via
42:11which the protocols themselves can
42:14evolve which is governance because given
42:17that we are in this decentralized
42:19context there there isn't really a good
42:22answer as to how you how you evolve
42:24forward it's the case that no one really
42:27controls the crypto Network how do we
42:28how do you change things about it that
42:30don't quite work so there's an important
42:31like last last pillar so for that for
42:36the four of them we talked a lot Chris
42:38Chris alluded to many different building
42:40blocks that are contained within each
42:42for example in the case of scalability
42:44there there are solutions like charting
42:46and like plasma on the etherium and the
42:48theorem blockchain there's this Intel
42:50SGX there's Starks so we covered a lot
42:52of those in the case of security I mean
42:55one thing that's interesting to note is
42:56that there are various different threat
42:59models and different applications
43:01require different levels of security so
43:04certain block chains like say like the
43:05Bitcoin meme is that Bitcoin is fully
43:09decentralized and is extremely secure
43:11secure to the point at which it'll be
43:13resilient even against attacks by nation
43:15states like China but there are
43:17applications out there that may not need
43:19that level of security and so there is
43:21room for other kinds of crypto networks
43:23to emerge that don't necessarily strive
43:25for that level of security so that it's
43:27an access via which developers can can
43:30sort of explore different possibilities
43:32I guess of usability one thing that I
43:35think is fairly underserved by the
43:39current business models we were talking
43:40about earlier the token business model
43:42which is sort of a tempting thing to do
43:45climate the current sort of hype
43:48ecosystem is working on developer tools
43:51working on making making the technology
43:53useable to developers in order for that
43:55flywheel to emerge and similarly
43:57building good interfaces for for
44:00end-users we don't care at all about the
44:02centralization to come on board and to
44:03begin engaging with with the technology
44:05and then finally this is question of
44:08governance and I think there already is
44:10a mechanism of governance which is that
44:12of open source and people people
44:15organize themselves in a normal human
44:19way they decide which projects to work
44:20on they decide whether or not they
44:23should continue to work on those
44:24projects or or fork them so the
44:27basically the the act of taking the code
44:30and creating a new instance of it and
44:32going going on on your separate way but
44:35there are also ideas as to how you can
44:37streamline that process how do you make
44:38it so that humans or a small group of
44:41people don't have like out size control
44:45over the evolution of the network and so
44:47one idea is is you you can also build a
44:50protocol for the evolution of the
44:52protocol itself and that would be like a
44:54non chain governance kind of solution
44:56where people are able to vote they're
44:58able to to express their opinion as to
45:01how the system should evolve and the
45:03system would have take that into account
45:04somehow in order to make a to make a
45:06decision and this depends heavily on
45:08building blocks that I think as we were
45:13talking about earlier identity and
45:15repetition are very important building
45:17block that helps make the the games that
45:19people play with one another iterated as
45:21opposed to one-off prisoner's dilemma
45:23style games so that's maybe a broad
45:26abroad outline of a framework for which
45:29we can think of the many different
45:30building blocks that are needed
45:31decentralization is a means to an end is
45:34a means to each of these four things it
45:36provides you with security if done
45:38correctly it can provide you with
45:40scalability of the of the underlying
45:42world computer and it can help I think I
45:47mean usability and and governance are
45:48things that we have to build
45:50despite this decentralization extent
45:52rotation makes it hard to build those
45:53things but those are absolute absolutely
45:56necessary for any of this to work
45:58well thanks guys I think we'll wrap up
46:00here thanks for taking the time thank
