Re: Mosaic Intro

On Sun, 10 Feb 2019 at 01:25, Adam Lake <adam@mosaic.social> wrote:

> *Hi All, *
>
>
>
>
>
> * It is great to see so many passionate and capable people in this group.
> I am sorry to have missed the kick-off call but hope to make the next one.
> My interest in Solid is the power of its principles to enable a more free,
> open, and cooperative Web and world. To me it represents the promise of
> coming closer to the original vision for the Web, a platform that would
> increase human capacity, our political and economy health, and our
> collective intelligence. My assessment is that individual sovereignty on
> the Web, an extension of civil rights in the digital age, is a fundamental
> requirement to achieving these broader social ends. Solid’s data ownership
> and data portability architecture are critical pieces of the puzzle. I am
> probably preaching to the choir! My role is to help “bring people together
> to build the next generation Web”. You can learn more here
> https://mosaic.social/ <https://mosaic.social/>. The objective of Mosaic is
> to connect teams, technologies, and financing to bring user-centrc (“Self
> Sovereign”) P2P apps to market. My hope is to facilitate the connection
> between disparate parties that may not know about each, but who together
> can provide all of the necessary ingredients required to brings Solid apps
> to market, have a sustainable business model, and designed to maximize
> social well being. The technology is critical, but so are funding, business
> models, marketing, and psychologists who specialize in human-centered
> design. The Mosaic website was launched to communicate a basic technology
> framework (very much still open for debate) and some app concepts to spark
> the imagination and get dialogue going about what app ideas have the most
> support (e.g, existing technology, funding, and public demand). I am
> heavily leaning toward starting with a decentralized Facebook application
> because it exemplifies the struggle for the future of the Web and because
> an alternative, or anti, facebook is a simple concept for people to
> understand. However, choosing this app presents some challenges as there
> are deep problems around distributed search, fake news, and identity to
> solve. These issues may prove to be intractable problems but I think it’s
> worth systematically exploring whether a good decentralized and Solid-based
> Facebook could be designed. Opinions on these topics are most welcome! I
> would gladly engage with others in this group around these high level
> design considerations as well as sustainable and equitable business models
> for bringing Solid apps to market. *
>

Looks very interesting.

I've had done some work on a timeline app but I never got a chance to
complete it, as focus shifted to the server.  The auth doesnt work with the
node solid server in OIDC mode yet, tho.

demo :
http://solid-social.github.io/timeline/?date=recent&profile=https:%2F%2Fmelvincarvalho.com%2F%23me
code : https://github.com/solid-social/timeline
screenshots :
https://melvincarvalho.gitbooks.io/solid-social/content/appendixa.html

I believe darcy is also looking at this route :

https://darcy.is/

So, it helps to understand what facebook did well.  They created their
system based on the idea of giving everything an HTTP URI and growing a
graph around that.

Porting this idea to solid I think is an excellent idea, and would allow a
cross origin social network with strong privacy and everyone controls their
own data.

Seems like a compelling use case, so the question arises as to why no one
in open source has done this yet.

The answer is that resources are scarce.  And those that have tried
inevitably have taken on too much.  Such the very common idea of making it
P2P or creating a new DNS.  Such premature optimizations have never worked,
and if we have learnt from history are not the best strategy.  It becomes
tougher still because advocates of protocol X are widespread trying to
promote their system, whereas solid is just the web with more cross origin
features unlocked.

Doing one thing well which is porting social network functionality via a
graph of URIs can be easily realized if we have someone to code it.  A
small team, or even a single person, could realize this in a few months.
The prototyps of facebook was apparently coded up in about 2 weeks.

Solid is an ideal technology to achieve this.  But the danger is going off
piste and tagging on the latest social protocol du jour and ending up with
an architecture inferior to facebook.  There was a massive opportunity
missed by the social web working group imho when I suggested this route,
that the social web about people, friends and connections.  Whilst I
thought this was self evident, it idea was rejected, in favour of building
a microblogging system.  Well that worked, but we sacrificed social
networks on the altar of the micro blog.

Id suggest doing one task well ie porting social networks to solid, then
think about adding more protocols.

What is needed?  You need a timeline, you need profile management, a
friendship graph.  You need friend requests.  Messages, replies and likes.
It can either be done as one system or in modular parts where different
groups work on different aspects and then a team ties them together in a
single app.

Id suggest working out what is in scope and what is not, and trying to make
something as minimal as possible.  Then trying to achieve a rapid
prototype.  Perhaps work together with darcy if they are going to build a
solid solution


>
> * Kind Regards, Adam Lake *
>
>