varia.website/content/long-article.md

Title: Social Media Critique 
Category: long-read test
Slug: federation
lang: en
status: draft

> This is in the end what Silicon Valley tries to prevent at all cost:
> resistance and exodus. How can such a momentum be unleashed?

So aside from the discussion of who listens (or didn't listen) to whose
opinion it can be interesting to have a closer look at action and momentum.

Three projects caught my attention and I think could be an interesting
case for this 'next steps' discussion:

Mastodon (2016) en Conversations (2014) and Peertube (2015) *

All three are projects that during the past twelve months have somehow
reinvigorated (the work on, attention for) their underlying protocols.
Protocols that have been proclaimed dead or unsuccessful for many years.
And probably will be for more to come.

The first one, Mastodon (https://joinmastodon.org/), you may have read
about or even tried out. It is essentially a twitter clone /
alternative. Technically it is based on Ostatus, which is a protocol to
distribute
status updates across networks. Ostatus is the protocol that powered
early 'alternative 2.0 style' social networks such as Friendica and
Lorea. The latter was a product of and important site of organization
for the Spanish Indignados and 15M movements. Mastodon also supports
ActivityPub which is the likely successor of Ostatus as a protocol for
further ongoing work on so-called federated publishing. The interesting
thing is that Mastodon managed to attract a good chunk of the recent Twitter
refugees. These where mostly voices which aren't white, loud or extreme
right wing and for those reasons felt themselves increasingly out of
place on twitter. Mastodon communities managed to involve so many of
these people by focusing on developing tools for community moderation,
content warnings and the ability to block other instances in the
network. As a result (the english language) Mastodon became a site that
is predominantly populated by the queer, PoC, left and artistic, or
anyone that would otherwise be at risk of being on the receiving end of
the Gamergate-style interactions on twitter. The decentralized nature of
mastodon has created a culture of 'thematic mastodon servers (see
https://instances.social/list) that have become a large part of what
makes the network interesting and relevant to its several hundred
thausand users.


Conversations (https://conversations.im/) is a messaging application
that is based on the very old XMPP protocol. This is a chat protocol
which has at one point also been the underlying technology of both
Google and Facebook chat before they closed it down and made it
proprietary. From the onset Conversations focused on a combination of
user friendliness, security and ultimately visual design to be on par
with mobile messengers such as whatsapp and telegram. The work of
Conversations has reinvigorated the XMPP protocol. Partly because it
focused on implementing the double-ratchett encryption algorithm almost
immediately after it was open-sourced. This is the modern userfriendly
end-to-end encryption algorithm developed by Moxie Marlinspike for
Signal and licensed to companies like Whatsapp. Another effect of the
work of Conversations is that the decades old protocol has been updated
in the span of a few years to work very well for mobile usage. For me
one of the interesting aspects of the development of Conversations is
the role that modern thinking on UIs, design and user friendliness
played in its popularity. This especially becomes apparent in the very
technical and awkward world of XMPP software. The developer has
mentioned multiple times that he 'bases' his design on that of his GAFA
'competitors'. Apropos tactical media, this project's appropriation of
corporate design, yet very clear and
solid political stance (see https://gultsch.de/objection.html) leading
to an increase in popularity and community involvement is an interesting
development.


Lastly, Peertube (https://github.com/Chocobozzz/PeerTube) is an attempt
at making the hosting of video content accessible to small
organizations. The sheer amount of infrastructure and thus capital
required to set up an alternative to the monopoly position of Youtube,
forces any project trying to replace Youtube to use peer-to-peer
technologies. Peertube does so by trying to implement WebTorrents. Like
the older 'BitTorrent' protocol it is based on, WebTorrent tries to
mitigate the sheer amount of data and bandwith involved with exchanging
online media, by making sure these are streamed from many sources at
once. Unlike torrents, which need separate applications, WebTorrents run
in familiar web browsers. One could say the conceptual forbearer of this
approach was a project called Popcorn Time (2014). An app that convinced
many with its good UI and design to do 'Netflix-like' streaming on top
of the torrent network. Again this is something that lead to a
reinvigoration of the decaying (use-wise) torrenting protocol. (I'd also
argue though, that Popcorn Time was simultaneously the nail in the
coffin for torrenting because of the individualistic streaming mentality
built into it. This also meant the definite end of what remained of
-collectivist?- seeding/sharing culture on public trackers.)

The position of the Peertube as a viable alternative or successful
project is the most tenuous of the three. However, one might argue that
our definition of success in this context should also be readjusted -
away from the Silicon Valley, venture capitalist sense of success using
metrics like usage counts, market cap, patent value etc. By nature of
being built upon open, compatible and federative technologies,
developments happening in all three projects could, and probably will,
end up supporting one another. They do so to the extent that one project
could even become an integral part of the other. As an example both
Mastodon and Peertube use the same underlying ActivityPub, allowing one
to become the underlying video delivery function of the other. As was
the case with Friendica, Lorea and Mastodon, projects might stop but
then become stepping stones and inspirations for newer generations of
projects. In this sense definitions of success should consider the
quality of longer term technological ecosystems within larger
socio-political contexts.

So the striking things for me to take away from these projects are:

All three projects have managed to reinvigorate 'decaying' protocols in
large part through their focus on UX, language and interestingly design.
Which seems to me a huge opportunity for the arts which has been left
largely unused in the first round of social media critique. Perhaps the
model of artistic production in this domain should move away from the
artists being on the forefront, sensing out emerging tendencies and
taking the spotlight by creating mostly harmless critical and
speculative works
around these tendencies. Next steps for artistic social media critique
should instead take a much more humble and supportive role contributing
expertise, time and exposure to people working in and with these ecosystems.

All three projects are based on federation. Which is the idea that
various actors making up a network decide to cooperate in a collective
fashion. Distributing responsibility and power as they do so. The future
of social media has to be federated or there won't be any (for those
privileged enough to retreat..). I think the case of Mastodon, where
servers in the Ostatus federation are experimenting with blocking
hostile content altogether from other servers in the federation (while
still maintaining technical compatibility) are interesting experiments.
For one, the debates over on-line harassment and fake news show that the
grand 'electronic agoras', where one can find anyone and everyone
clearly aren't conducive to productive interchange of ideas. Perhaps
smallish, self caring communities are a good answer to the profit driven
model of infinite interconnectedness.

Lastly, I think it is no coincidence that two out of three of the
projects have Germans leading development and all three are European
based projects. I guess the following is anecdotal and partial evidence.
Yet, I've not seen Google and Facebook run full page advertorials in
leading daily newspapers except in the German ones. Ostensibly, part of
an attempt on their side to prevent mass user exodus out of discomfort
with the platform. It is in part German historical sensibility that
leads to this kind of sensitivity on the issues of privacy, but it is
also a sensitivity that is actively nurtured in public discourse. No
Silicon Valley apologies are required for there to be scepsis. At the
same time the European context apparently provides good enough living
conditions for people to risk investing time in this kind of work. Risk
which is also partly mitigated by initiatives such as German Prototype
Fund and other European funding streams. However, testament to the fact
that these projects have healthy communities and are part of wider
ecosystems of support is that all projects finance themselves from
diverse revenue streams, user contributions being the main one.

So I'd say next steps for a social media critique would be to be more
involved in (and involve more) these communities. To use positions of
power to create opportunities for people working on these projects.
While the center of development of these projects is Western-Europe they
have many contributors outside of Europe as well, that could benefit
even more from such opportunities. At the same time, doing close
readings of the technical underpinnings of these media will also improve
understanding of what is (not) going on. Now obviously all this was a
news flash from within a very specific filter bubble, but actually from
there 2017 was a very promising year for alternative media.

> I still believe in vital methods to mass delete Facebook accounts.
I'd say start doing so, but help your friends. Use your network effect
to transition together to different kind of media. This is slow and
laborious so mutual support is important. The time is always right, but
now more than ever.

* these are the dates of the project's source code first appearing in
public, they are still actively updated and used.

greetings,

Roel