Future Technology of the Circle of Crust

[Macrobius]

We've done aspects of this topic in the past, however @Mike has raised some interesting points here (forked from another thread[1])

I have this idea in my head. We somehow need to create a radically decentralized social media platform, served up literally from apps of individuals to whatever individuals they are connected to. I think the Internet is structured to make this possible, but no one has attempted this maneuver, this architecture I have in mind, which is essentially akin to email rather than centralized servers. As much I respect Elon for taking over Twitter and (partially) salvaging it, and Rumble for standing up and standing their ground, I think there are better options available to us waiting to be implemented.

My response:

IPFS (kitschy name of 'interplanetary file system') is essentially a distributed file system with links -- content is retried by a hash, very much like bit torrent. https://ipfs.tech/

It's not likely to go away at least in the near term because it is used as a more distributed, non-censorable method of making NFTs (non-fungible tokens in the Ethereum crypto ecosystem) available.

There are services to 'pin' IPFS files, though you don't have to do them if you can keep a set of servers up anywhere on the planet to cache your content.

The NFT ecosystem already has a notion of 'distributed apps' including those that run on various blockchains (you probably know this but a blockchain is just a simulated and distributed, but single-threaded, 'world computer' (EVM = Ethereum Virtual Machine) for shared authentication of whatever people pay to run on it -- data storage is too expensive directly 'on chain', hence the secure-ish storage ecosystem developing around it to store actual content rather than just the metadata to find it and authenticate it)

The idea of federated apps is behind the Fediverse (worst maketing name ever for a distributed, semi-non-censorable tech). What I've seen of it makes me believe it is a credible Twitter/Facebook/Image site alternative, but I've never seen it used to long form artifacts and suspect it's not optimized for streaming video [I could be wrong on this].

Also, since Pastebin started censoring raw content (inevitable) I run a hastebin server which can be found via the pURL (persistent URL) https://purl.org/hastebin - this uses the global pURL registry at archive.org which won't be going away soon and being run by librarians is SOMEWHAT censorship resistant, but the real anti-censorship guarantee is that it is trivial to write your own 'purl resolver' and redefine purl.org in /etc/hosts to point to it, if they do get uppity like Pastebin did.

Finally, it is very possible to 'push computation' and even 'virtual networking' to the EDGE, down to the IoT device level and certainly to smaller computers including mobile devices. This sort of tech is actively being pursued by large corporations as well. One example is virtual NVR (network video recorders) of surveillance footage, now used at large retail establishments to make retrieving brick-and-mortar security footage to HQ easy, on a selected basis -- soon to be combined with face recognition and location aware mobile apps to create a sort of 'customer situation awareness'. You can imagine how important this might become in high-risk retail environments -- a sort of instantaneous KYC.

You can bet that 'machine learning at the edge' is a hot topic right now. TFW chatGPT goes paranoid and creates a deep fake narrative about you and the heist you allegedly pulled, and files a 'tip' with the local SWAT team. Sounds to me like we are headed to MiniTru and Ingsoc on autopilot.

[1]: https://www.thephora.net/phoranova/...reign-news-than-msm-alt-media.1146/#post-9699

 

[Macrobius]

Stepping back for a minute -- where are we now? The 'web' (not all of TCP/IP but the HTTP parts) is very heavily based on PHP--still (80+%-ish by sites) And most of that is WordPress blogs, with other minor 'CMS' players (drupal say).

BBS boards like vBulletin or XenForo2 are a bit of a backwater, though quite common as well. Most of this PHP is on 'shared hosts' which provide the peak price/performance for the 'just barely Web2' world. That is they make little but some use of 'AJAX' JavaScripting, and JSON based APIs, which are the norm in the corporate world but not in the 'Normie free communication' world. Life Sentences for 'stealing' mp3s of rap music lead the way a generation ago.

PHP is, in turn, essentially captured by Facebook, which runs/ruins the developer ecosystem for Social Media based on PHP. This forces an upgrade cycle for 'security patches' that in effect forces the shared hosts to break the sites of their own customers on a regular basis, and leaves the rest of the ecosystem up to WordPress and vBulletin and similar companies to milk the upgrade path themselves.

The Old Phora was itself a victim of this upgrade cycle, ultimately. (A necessary upgrade ultimately rejected by the active membership for daily use, due to the sheer ugliness and unusability of vBulletin 5, not to mention the previous abortion of vBulletin 4, CMS edition, which in retrospect was better)

Thus, the curtailing of Free Speech that we saw at the dawn of the 'Weblogs of War' Blogosphere, have evolved from 'Social Pressure' (bad behaviour, trolling, and doxxing[1]) have evolved to TECHNOLOGICAL pressure and capture of the peer2peer aspects of the Web with outright censorship, eventually to be made compulsory via legislation or at least the threat of lawfare, which amounts through political flowing from the barrel of a gavel, with the aid of Federal Marshals.

[1]: 'Freedom of Speech is not Freedom from Consequences'

All this implies that a VERY BIG overhaul is coming. The old web (version 2 at least -- version 1 barely exists except in the IoT world) is essentially captured by the GAFAMs (Google, Apple, Facebook, Amazon, Microsoft). The browser is captured as well with essentially all browsers outside mobile and the Apple eco-system being based on Chromium.

Given the degree of lock-in currently achieved, it is fair to say that the 'the internet has been captured' and is a lost position -- not entirely but for all Normie purposes it has -- and will now become a Globalist Broadcasting System of the Simulation or Bread and Circuses for the Masses or whatever. Certainly nothing so interesting as the Web 30 years ago, at the end of the Eternal September of 1993.

There is a partial 'way out' through mobile devices or at the edge, but it will take a lot of effort on the tech side, and we can expect stiff opposition from 'legislation', as usual around the favourite mix of software patents, copyrights, TOS, and 'Hate Speech' creep, added to the Technological Griefing cycle by the GAFAMs as mentioned above.

 

[Macrobius]

A lot of companies are turning to https://nats.io for publish/subscribe messaging (and store and forward queues using built in Jetstream extension).

There's a reason for that.

If you want to build your own replacement for the decrepit and dying email ecosystem (have you actually tried to set up and run your own email server that can talk to 'gmail' addresses without ending up in the 'spam' folder, recently?)... that's where to start.

The old advice used to be 'use existing tech' -- if you wanted synchronous messaging, you use HTTP and if you wanted asynchronous messaging you use email. In the aughties it was sheer folly to try to roll your own protocols and hope you got it right.

The 'cloud' somewhat broke this paradigm -- the very first two services Amazon AWS built ca. 2007 were *not* EC2 (VM instance provisioning via the command line or later an API)... but SMS (first) and S2 (second). Anyone building fresh has exactly those two problems -- how do I build a queuing system with a publish/subscribe model, and how do I build 'infinite reliable (and non-censorable) storage'?

Once the protocols are centralised, they may be SERVICES but they are not /your/ SERVICES. And the old solutions have been captured. So you build your own this time.

NATS and IPFS.

What can the normies here do? Download and install https://ipfs.tech (IPFS desktop for windows) and stand by for further comms... MAKE SURE if you do this that you know how to turn it off (or else you will be donating bandwidth and learn to hate it). If you don't know how to click on an icon in the taskbar and select 'exit' you probably should NOT DO THIS EXPERIMENT.

For you technologists, learn what https://nats.io is and play with it.