jbenet

Iceland Eruption

Live Video: https://www.youtube.com/watch?v=GCa4EMcWuoI

Night Time

Day Time

jbenet

Fast Research & Development

People tend to believe that R&D takes a long time.

• Most useful: read this list https://patrickcollison.com/fast -- assembled by Patrick Collison
• Elon simplicity: long is wrong, tight is right
  • also in there: remove design parts. the best engineering design improvement is to remove things -- undesign things.
• Max Hodak (transcriptic, neuralink, ...) says "you should be able to design an experiment, go into surgery, get the data, and do the analysis in under a week" (on fast feedback loops for BMI)
• Demis Hassabis - energy, pace, focus, -- blue sky research group w/ the intensity of a startup

jbenet

Messages in Filecoin allow sending arbitrary values through the params field.

Encoding data into params

• data to send through transactions should be base64 encoded when using the message sending APIs.
• note: you'll be sending to method: 0, which should be the default in your tools. method: 0 will ignore the params.

lotus-shed example

jbenet

#!/usr/bin/env emacs -Q --script

;; Sandbox
(setq
 user-emacs-directory (concat (file-name-directory load-file-name) ".emacs/")
 package-user-dir (concat user-emacs-directory "elpa/")
 use-package-always-ensure t
 inhibit-message t) ; if there are errors, remove this.
 ; debug-on-error t) ; if there are errors, add this.

jbenet

/*
large google docs comments
add this to a user stylesheet
like https://github.com/openstyles/stylus
*/

/* make comment boxes large */
@media only screen and (min-width: 1400px) {
    .docos-layout-anchored
    .docos-anchoreddocoview {

jbenet

instructions to install dig's riot-web theme

(Option 1) Mac Binary built by @dignifiedquire

probably owned, yo.

1. Download https://ipfs.io/ipfs/QmWS4FFFwLrzDfzMwAc2JNzKqKnQKFSJEVpEeLQhX6WbEA/DigRiot.dmg
2. Install the app in the dmg.
3. Follow the using the theme section below.
4. Profit.

jbenet

import "context"

type Key interface {}
// concrete type likely a []byte or string

type Record interface {}
// concrete type is a struct
// (in libp2p, will be ipld -- see iprs)
// mitting here for clairty and avoiding distraction
// for now, let's assume it's an opaque value.

jbenet

Keybase proof

I hereby claim:

• I am jbenet on github.
• I am jbenet (https://keybase.io/jbenet) on keybase.
• I have a public key ASAEBwJK1_z8s2R5Pxgl1wteFS85Kaz748RvCh3h8Y9WjQo

To claim this, I am signing this object:

jbenet

#!/bin/bash
# install into your path, as 'ipfs-archive', 'static' or something like that

# OPTIONS

# use this ipfs api
# this is an ipfs cluster for me :)
ipfsApi="/ip4/127.0.0.1/tcp/9095"

# gateway URLs. for printing, and automatic open

jbenet

When people come up with ideas, they often assume that:

• coming up with an idea is the same thing as coming up with an idea first (it isn't, read existing literature)
• coming up with an idea first is dramatically more special than coming up with an idea (it isn't)
• the wealth of existing literature does not include their idea (almost always it does)
• unpublished ideas do not count (they do, lots of important discoveries happen verbally, and unfortunately sometimes only verbally :( )
• coming up with it first is more important than fleshing out a body of work (it isn't)

And these assumptions tend to make people very unhappy. I repeatedly see people succumbing to sadness over multiple discovery. It shouldn't be sad, it should be a happy event, as it confirms our thoughts and presents an opportunity for collaborations.