juan benet

juan benet 2.png

[palo alto]

intro’d to Juan via Thomas


from a blueyard convo in berlin – jun 2016 – decentralized and encrypted – rewiring the internet for decentralization

The mission to upgrade the internet – jun 2016

mission to upgrade internet.. not just our missions.. all our mission

we (ipfs) think this (internet) is most important piece of tech humanity has ever created and has brought about an enormous amount of social change in just few decades

2 min – we have almost 8 bn mobile phones.. but only about 3 bn people on the internet..  *what’s going on there..

*this is where juan captured me.. decided to add a page

3 min – so this machine is important to our everyday lives

internet: wires and connectivity framework

web: the application platform that we humans use..

we don’t speak in bite streams to each other.. so we can’t directly use the internet .. we have to use these applications… properties of how these applications get to us.. how we find/retrieve/use them.. those properties have vast implications on our capabilities as humans.. because the software that we use thru the internet.. the way they function.. is defined thru a whole bunch of protocols underneath the hood.. and the properties of those determine what you can do as a human.. so if you want to give yourself/world a superpower.. you go underneath the hood and change protocols around.. and then you can just patch the internet.. and the world has new things..

4 min – this is amazing.. the ability to upgrade the internet.. the thought that you as a user could go.. wouldn’t it be great if the internet did x y and z.. and actually go thru and patch it.. and suddenly new capabilities start emerging and everyone starts sharing this.. this is what we all should be thinking about..

i’m thinking about.. hosting-life-bits via self-talk as data on some sort of blockchain/deep address [to facil/scrape 7 billion brains.. to some wikipedia like blockchain/platform, ie: io dance]..  so now wondering diff between bandwidth and server farms.. was thinking blockchain took care of that.. but maybe need this ipfs layer/bit..?

5 min – the internet is everywhere and the edge is increasing.. so you have to think very carefully about your users out in the edges as well

problems we think about with internet & web:

1\ has gone from decentralized to centralized.. ie: if we’re all together in room using web.. and central server goes out .. we’re gone.. but why.. this should not be happening.. we have supercomputers in our pockets..

2\ terrible security model.. only encrypting pt to pt links instead of content at rest

3\ data control.. huge orgs (fb, twitter, reddit) kind of own your data and how you access it

4\ latency and bandwidth barriers .. esp in developing spaces .. most websites to heavy in many places.. and so.. those who need internet most can’t use it..

5\ natural disasters.. would bandwidth work

6\ human disasters.. ie: surprise oppression; censorship

7 min – biggest problem in my pov..

7/ web has no notion of giving permanence to links.. links break.. sites disappear

on permanence.. indeed important.. but more important.. 1\ the process of a human being becoming.. 2\ the links are legit

if we don’t have as base.. free people.. get carried away with permanently storing non-legit data…

so these are problems the ipfs project is here to solve..

8 min – it’s really about making sure that it’s these sites we depend on.. can be accessed thru our mobile-phones/browsers.. w/o relying on the backbone all the time..  that’s a model that worked really well to scale up the web.. but over time.. we are no longer able to rely on that… we have been sold this myth that connectivity.. suddenly will achieve perfect thruput/latency.. and that everyone is going to get connected.. and just sit back and wait till everyone gets wired up and everything will be fine.. but reality is very diff .. the applications today.. will not work tons of people on the planet.. unless we change the underlying structure.. unless we go under the hood and patch the protocols.. to change how data flows in the network

9 min – so the ipfs project is about that.. about turning this link structure.. that forces us to go to a specific location into a diff link structure that allows us to go directly to the content… wherever it’s available..

the good news in all this.. the internet has been upgraded many many times before.. this is not an event that happens.. rather .. a continuous stream of happenings.. that sometimes have huge spikes.. for ie the web.. when the web exploded.. we saw an enormous amount of innovation happening.. and again in web 2.0 early days.. we’re seeing yet another massive wave.. but all these changes happen over time.. and they go thru this process of turning ideas into these protocols that form the infrastructure of the applications that we use… pipeline: ideas/specs/code/computers/humans

10 min – so the exciting thing for us is that changing the network/internet.. upgrading with new capabilities.. is really as simple as translating a whole bunch of good ideas into running applications.. the reality is this pipeline gets pretty long and problematic when protocols get stuck somewhere inbetween

11 min – another problem.. most ideas die in research .. pipeline: research/development/deployment/use.. we have to get really good at this pipeline.. of translating ideas to d and d and u.. have to be careful about users.. can cut out whole swatches of users..

why ..hosting-life-bits via self-talk ..as the day [aka: not part\ial.. for (blank)’s sake…] .. is huge..

goes back to where he first captured me.. none of this will work.. unless it’s all of us.. let’s work on that first..

let’s do this firstfree art-ists.

12 min – bitcoin reminded us that it’s actually fairly easy to deploy a thing and get it adopted by people.. if it’s the right thing.. and that change is really around the corner.. if you work hard and are careful about how you construct your protocol..

on leaping.. for (blank)’s sake

a nother way

13 min – it’s always useful to *step back a little and observe what changes are actually happening..

web 1.0: putting content easily accessible on the network.. writing on top of internet.. .being about to address each other’s content

web 2.0: wiring programs to that.. being able to have dynamic content.. able to execute changes in fast pace.. while collabing with applications.. etc… but all still dependent on specific orgs like computers..

web 3.0: (my view).. we have an inversion going on now.. things like bitcoin reminded us.. we don’t need to trust org’s/infrastructure.. we can just have content/programs linked to each other.. forget about the specific orgs/locations.. and that is the power that these new systems ie: bitcoin; ethereum; ipfs; … are bringing about.. this inversion of how the computation actually happens… at the end of the day.. this is all fancy ways of addressing/moving computation over data.. it’s just a whole bunch of data and a whole bunch of programs operating on data..

*zoom dance ness

14 min – think of it as a stack that is emerging around verifiable decentralized applications that don’t need to trust anyone.. and you have things like smart contracts and multi party computations….ledgers and secure consensus protected transaction systems..

questions some verbiage/function/necessity here: trust..contracts..consensus..

underneath that you have all these data structures that are just half linked.. that are making all this possible.. secure data structures.. merkle web/links… and at very bottom layer have this secure high performance connectivity framework.. p2p networking..

15 min – so what ipfs is.. looking at that bottom layer.. making that a really high performance thing and enabling these applications to work better.. and .. we have a model that *we can’t even assume you can have a consensus.. **we can’t assume you’re able to talk to the rest of the network..

good thing.. because *public consensus always oppression someone.. but not dependent on **being able to talk to everyone.. dependent on being able to listen to everyone.. again.. hosting life bits.. everyone becoming indigenous (holmgren indigenous law)..

so.. we really want to make sure that applications work here.. in this room.. ie: you should *be able to chat with the people in this room.. even if the internet is down..

*great .. but again.. to me more about hosting every voice.. in order to facil those chat rooms.. ish.. who’s together in a space.. per choice.. ness.. (back to first encounter w/howard)

16 min – mission to upgrade.. an evolving machine: patches switching networks in 40s; arpa net; proper internet; web; … and now.. we’re ripping the web off of its location and making it all secure.. properly linked.. and so on…

we need to be reminded of something the web/internet people understood really well: standards matter.. agreeing on certain formats/structures really really matter .. to make sure that adoption happens really well throughout the system

takes me back again to the first capture.. why 3 bn on internet with 8 bn mobiles.. i think our formats/structures/basics.. beg being much simpler.. ie: human constitution of a and a.. really doesn’t matter if under the hood is working.. if 4 bn aren’t dancing.. getting all 7 bn plus to dance at once.. will assure that our.. under the hood ness.. is antifragile.. ongoingly suitable for.. all of us..

deep/simple/open enough for all 7bn


panel with talk above

16 min – ability to create something like airbnb/uber did.. precisely where needed most.. might seem contrarian pov.. because cloud computing is such a secret sauce.. but there’s a  huge achilles heal here.. the major corps have to deal with bandwidth being an enormous/expensive thing.. and if you can beat them and go directly to the edge.. that is a killer app


tedxsanfran – dec 2016


juan on long term infra


find/follow Juan:

link twitter

Obsessed with Knowledge, Science, and Technology — Protocol Labs, IPFS, Filecoin

ipfs (inter planetary file system) site:


A peer-to-peer hypermedia protocol
to make the web faster, safer, and more open.