Re: Why we really can't use Facebook for technical discussion.

[Phillip Hallam-Baker <[e-mail]>]

There are multiple problems going on here.

One problem that needs urgent attention is a phenomenon I am calling a
profile poisoning attack. Facebook does not allow people to say 'this
content is garbage, do not show this type of content again'. The only
feedback allowed is positive. YouTube and Twitter suffer from the same
basic problem but to a lesser degree. Not being able to tell your racist
uncle that you have no interest in his bullshit is core to the Facebook
brand. It is why it was able to do so much damage.

Limiting people to positive feedback sets the stage for a profile poisoning
attack. The attacker bombards the target with propaganda knowing that any
response from the target will result in the algorithm sending even more
propaganda. There is no escape, there is no way to shut the propaganda off.
The needs of Facebook advertisers means that the product must not be
allowed to say 'stop'.

A colleague pointed out that the QAnon slogan is 'do your research' is in
effect a clever trap to cause the target to be deluged by more nonsensical
conspiracy mongering.


It is time to stop this. Dean points out one of the major problems with
Facebook - there is only one authority. Only King Zuck gets to decide who
can and who cannot speak. His house, his rules. So it is time to make it
our house.

I have given a lot of thought to this problem over the past two years. I
hope folk didn't think I was only thinking about how t do s aslightly
better password vault, secure data at rest and do end to end secure
messaging. The Mesh was always about an end-to-end secure social media
platform.

The absolute bare minimum for a social media platform that gives users full
control is a name that belongs to them, does not come with an artificial
rent, is life-long (except in exceptional circumstances) and can be bound
to the service provider(s) of their choice for messaging, social media etc.

Mathematical Mesh 3.0 Part VII: Mesh Callsign Service (ietf.org)
<https://www.ietf.org/archive/id/draft-hallambaker-mesh-callsign-00.html>

There is also a blockchain in there that does not rely on proof of work for
stability. The ultimate source of authority and trust for every user is
that user themselves. It is kindof a zen thing.

The basic code for phase one is passing all relevant unit tests. Just got
to write the server config interface.


On Sat, Jun 5, 2021 at 8:02 PM Dean Willis <[e-mail]>
wrote:

> Töma Gavrichenkov < [e-mail]> wrote:
> Peace,
>
> The thing is, I don't think we've seen a lot of Facebook engineers on the
> IETF meetings even before COVID.  I'm pretty sure there was someone from
> Facebook doing some work around the rtg area, but I fail to recall anyone
> else.
>
> ------
>
> Ok, you are probably right.
>
> But just a few weeks ago, we had a "tools" survey circulating, and several
> of the questions related to using assorted social media platforms for IETF
> work.
>
> If you can get banned for a kill -9 comment -- well, we really can't be
> using that platform for IETF stuff. Even for jokes, apparently.
>
>
>
>
>