[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Cryptography] The Truth Social Network: A Decentralized Social Network
On 11/1/20, Howard Chu <hyc AT symas.com> wrote:
> btc_joe via cryptography wrote:
>> I’ve been working on a new system for engaging in decentralized social
>> media using an open-source protocol.
> We already had that, it was called Usenet on NNTP.
Kinda true, needs an upgrade.
>> The paper is available at:
>> Some context:
>> You can find a very raw proof of concept of the idea at my other site
>> www.beateasyjoe.com <http://www.beateasyjoe.com/>. Take it with a grain of
>> salt, much of
>> what’s in the paper was worked out in recent weeks/months, and will
>> require a lot of work to implement (thus why I figured I would share and
>> hopefully raise
>> some intellectual capital). A little over a year ago the idea began as a
>> way to prove ownership of various traditional social media accounts from
>> my personal
>> website. Then it evolved into how I could tailor content from my site in a
>> manner similar to social media using digital signatures. Then I realized I
>> had to
>> incorporate a way to enable interactions. Then I realized I needed
>> timestamps and some type of protocol. And eventually the goal became to
>> not only enable
>> decentralized social media, but further, to enable the creation of
>> historical records. There are also some potentially huge ideas that could
>> be created on top
>> of this thing.
>> Abstract: A decentralized social network would allow participants to
>> engage in social media directly without relying on trusted third parties.
>> The need for
>> permission-based access to various walled garden platforms would be
>> eliminated. Recreating the vast majority of social media functionality
>> largely depends upon
>> solving the problem of decentralized, pseudonymous identity. We propose a
>> solution to the above by utilizing the cryptocurrency Public Key
>> Infrastructure (PKI).
>> By choosing a single public address, participants can create
>> censorship-resistant historical records of content. Content is provably
>> linked to an identity
>> through the creation of digital fingerprints, which are timestamped and
>> committed to blocks (which are themselves timestamped). Participants can
>> reference their
>> own and others content by simply including it’s unique fingerprint within
>> their message. Any file that can be hashed (audio, video, zips, pdfs, etc)
>> can be
>> included within messages. While files themselves are not stored in these
>> identity blockchains, a record of them is, as their hashes are committed
>> to fingerprints.
> A fully permissionless network will rapidly be overwhelmed with spam if it
> achieves any degree of popularity.
That is a solvable problem that has been repeatedly and diversely
solved. (btw: all problems are solvable)
> Also not sure that there's merit in retaining social media posts forever, as
> on a blockchain. It's nice to have
> a few Usenet archive sites for historical reference, but for the majority of
> use cases periodic expiration is
> preferable. Most social posts are only relevant for a brief span of time,
> and a lot of stuff deserves to be
People only do things that have merit. Somebody did this (spent a
long time on it even); it has merit.
Personally I had an unresolved human trafficking scenario and the
biggest evidence is facebook's archive, some of which is missing from
people involved deleting their accounts.
> There's certainly some merit to the notion of decentralized identity. I.e.,
> we pretty much never care that your
> identity has been verified by some 3rd party authority, or that some
> authority granted you permission to operate
> with a particular account. We only care that you have an identity uniquely
> associated to you, so that no one can
> impersonate you. The cryptocurrency ecosystem seems to have solved this
> problem, yes.
Nice thought. Brief sock puppets can muddy it.
>> Full Paper available at:
The cryptography mailing list
cryptography AT metzdowd.com