Å integrere teknologi og samfunn

http://talks.kjernsmo.net/solid/2019/bigthing.html

Kjetil Kjernsmo

Solid-prosjektet

Vi skal gi folk tilbake kontrollen over egne data, og dermed muliggjøre teknologiske økosystemer som styrker demokratiet og gir bredere muligheter

Samfunnet gir ingen rettledning

Diskutert sist uke

Hvordan kan vi forankre utvikling i demokratiet?
Hvilke algoritmer vil fremme vennlig dialog? Respektfull debatt?
Når skal ekstern autoritet kunne overstyre individets valg? Hvordan?
Hvordan man indikere tillit til applikasjoner?
Hva slags institusjoner skal styre utviklingen?
Hvilken effekt vil lovgivning ha?
Hvilken effekt vil teknologivalg ha på mulighet til politisk styring?

Business models that focus on data harvesting and persuation architectures causes

stagnating innovation
alienation amongst citizens
a few winners takes all
general exploitative environment

We need a new social contract

The Solid ecosystem
brings back choice
by decoupling data and applications and
enables personal data control.

People choose where they store every single piece of data they produce.

They can grant applications and people access to very specific parts of their data.

Separating app and storage competition drives permissionless innovation.

Before the Web,
the world was highly heterogeneous.

Exchanging and connecting information was hard.
- different hardware
- different software
Innovation was hard.
- For which target machines should platforms be built?
- Developers depended on software and hardware vendors.

The Web strives to be universal, and thereby enables permissionless innovation.

Anyone can use the Web, regardless of
- hardware
  desktop phone tablet watch …
- software
  operating system browser app …
You don’t depend on these to innovate.
- Build for the Web!
- Standards provide interoperability.

The Web is Normative Software and de facto Standards

The whole normative landscape

I'd like to air a thought with you. I think society has gotten normative all wrong. When we think normative work, we think laws and regulation. Technologists too. I think it has been ineffective at best.

There is little sound science behind studies of effects of technology. Leaves us with trial and error, but there shouldn't be a lot of that up here.

Down here, extremely little work is normative, but it can have very high impact. It can be cheap and quick. Importantly, here we can experiment, we can study effects, and we can see what people can actually adopt.

Once we see adoption and desireable effects, we can standardize. And once we see what standards make it, and with even more hard evidence, government can mandate.

So, down here, we need to live with many failed experiments. Standards fail too, even when backed with adoption. Still, it is the study of effects that should drive upwards.

Emphasise the interaction in the system.

WHAT?

Technocratic Elitism!

What is a right worth?

What is the right to data portability worth if you have nowhere to put and use your data?

Consent in a Persuation Architecture

It's an Arms Race

If you spend more resources than your opponent, you loose

Uklare lover favoriserer de skruppelløse

De ærlige bruker store ressurser på å forstå lovene, og justere adferden

De skruppelløse kjører på, og vinner ofte på det

Introduksjon av teknologi må studeres

Do people really care?

91% say that they have lost control.
People are not giving up data as a tradeoff, but because they have no actual choice.

Choice is essential

But there is no choice unless technology enables people to do what they need to do

Permissionless innovation

Enables people to experiment with solution to their problems.
Counteracts technological elitsm.
Has historically brought unprecedented creativity to the world.
But isn't the whole solution

Linked Data

URL: http://talks.kjernsmo.net/solid/2019/bigthing.html

It is an address to something
You get a document back if you go there
It is also an identifier for that document

URI: https://solid.kjernsmo.net/profile/card#me

Generalized to identify something physical
You will get a description about me if you go there.
Can identify anything, including abstract things, like the type of weather.

Triple

Subject	Predicate	Object
`<https://solid.kjernsmo.net/profile/card#me>`	`<http://xmlns.com/foaf/0.1/name>`	`"Kjetil Kjernsmo"@no .`

Turtle

@prefix foaf: <http://xmlns.com/foaf/0.1/> .
@prefix dc:   <http://purl.org/dc/terms/> .
@prefix xsd:  <http://www.w3.org/2001/XMLSchema#> .
@base         <https://solid.kjernsmo.net/> .

<profile/card#me> a foaf:Person ;
                  foaf:name "Kjetil Kjernsmo"@no ;
	          foaf:publications <http://urn.nb.no/URN:NBN:no-56537>, 
                                    <http://urn.nb.no/URN:NBN:no-2021>.

<http://urn.nb.no/URN:NBN:no-56537> dc:title "SPARQL on the Open, Decentralised Web"@en;
                                    dc:date  "2016-11-22"^^xsd:date .

Solid

Solid means Social Linked Data

Solid gives

Everyone a Pod, where you can store and manage your data
Everyone an identity, a WebID
An authorization system with fine grained control
Takes advantage of various security mechanisms

Personal Data Control

When people control their own data, they are

more resistant to persuation architectures
less dependent on data harvesting business models
more resistant to forms of exploitation
they are free to choose applications that suit them
- use applications that matches their values
- enables competition based on value added
- since people are not locked in, anyone can enter with their innovations

Personal Data Control unlocks Permissionless Innovation

Governance

WebID for everyone using ID-Porten is trivial
Linked Data has some adoption in the Public Sector
Several ways to set up Pods with official records of personal data, whether is based on hjemmel or consent.
But also some limitations we need to work on

Gjeldsregisterlova

With Solid, the user could gather data from all their banks, and the approval algorithm from the loan-giver, and compute a a yes/no answer that could be verified

Grooming

Services to the population over their data

User-centric

People aren't on Altinn

People are on Facebook, Snapchat, TikTok

Public services should be on the same platfor that people normally use

Vi ligger dårlig an...

Lovene er for uklare
...og favoriserer skruppelløse
Sentralisert teknologi er enklere
...og har et enormt forsprang
...og er enormt profitabelt
Forskningsfinansiering gjør at akademia ikke kan delta tilstrekkelig raskt
Det er ingen teknologer som har mandat fra demokratiet til grunnleggende normativ utvikling

Give everyone Solid

Because...

It enhances the public's security
Allows people to truly excercise their rights
Helps free them from exploitative technology companies.
Enhances innovation
Improves their ability to participate in technological development
Public services can avoid centralized architectures
Public services can be where the user is