MoonDreamsGG

"Plebbit" Moon Dreams - P2P social media; $3.55mcap

Long
UNISWAP:PLEBWETH_BC628F.USD   PLEBBIT / WRAPPED ETHER on Ethereum (BC628F...B33C84) in USD
A pure P2P, decentralized, global adminless social media with an adaptable front end and no https/dns endpoints, giving ENS, NFTs, DAO, IPFS and pubsub, a long-term usecase and ensuring free speech.

Plebbit market cap: ~$3.55m market cap
Circulating supply: 1.49t

What is it?
Created in late 2021, according to the whitepaper, Plebbit is a social media protocol that uses no HTTPS endpoints/DNS and instead uses "public key-based addressing" and IPFS/pubsub protocols to create a fully peer to peer social media. In other words, it uses similar technology to BitTorrent but for social media.

To run a sub the sub owner needs to leave their computer on 24/7. There are both downloadable clients (which contribute to client speed through seeding) and the web interface which involves leeching off full clients.

The anti-spam measures involve captchas which currently includes a standard captchas and karma. According to the dev road map, the challenge will be entirely arbitrary and customizable by the sub-owner (regular captcha, quiz, balance of x token, reputation/karma points etc).

The protocol itself is text-only. Images/media are only embedded through links through media like Youtube/Odyssey/Imgur etc (not uploaded to the protocol).

The Plebbit protocol is currently being used to develop a Reddit, 4chan, Old Reddit (Seedit) and Barebones client. Other front-end style will likely be created by community members in the future like Vbulletin or Discourse.

Why?
Social media like reddit have become increasingly censorious.

Plebbit solves this by giving full autonomy to the sub/board creator who is the admin of only their sub/board. So long as a board/sub creator runs their full node, then their board/sub is accessible without the hassle of static, public HTTP endpoint which consumer ISPs rarely provide, a domain name and an SSL cert.

Plebbit protocol creates an open-source protocol adaptable to any front end style the developer wants.

Plebbit purports to be serverless and scalable.

Plebbit gives a use case to NFTs (through account profile pictures).

Plebbit gives a use case to ENS (and other DNs) as the plebbit board/sub link can be an ENS.

Each individual sub therein becomes a tradeable NFT that can gain value in proportion to its community value.

Finally, as it is open source, in the future the protocol could be adapted for ads at the sub/board owners discretion meaning each sub/board owner would own 100% of the ad revenue from their specific board or they could choose an ad free browsing experience (as it currently is now).

The plebbit protocol allows for a level of customizability of rarely seen (if ever) in social medias (customizable front end, captchas, administrators, ads/no ads).

Plebbit is potentially the next paradigm shift in crypto. First, there was BTC (decentralized money/gold) Then there was Eth (and all its innovations like NFTs, Defi, Gamefi etc). Then there was Chainlink (real world data into SCs). Now there is Plebbit – the first purely P2P social media experience with an adaptable front end.

How is this different to Lemmy/federated medias?
According to the developer’s telegram and reddit account, Lemmy/federated social medias are not pure p2p as the admins can arbitrarily store some of your data and delete your acc whereas Plebbit has no global admins to do this. Federated social medias enable instances to block others from using other instances.

Federated social medias are also complicated, requiring a board/sub owner acquiring a static, public HTTP endpoint which consumer ISPs rarely provide, a domain name and an SSL cert, which are all vulnerable to censorship and more complicated to set up compared to using running the plebbit client which connects to IPFS/pubsub automatically.

Does it work?
Yes but it is slow and unreliable. Production subs do not seem to be online. Sometimes captchas don’t return for 30 seconds which is a bad UX. However, the testnet of Seedit is surprisingly fast and consistent the past couple days. It will be interesting to see if this translates to other front ends like plebbitapp and plebchan with more optimization. If so, adoption is more realistic.

Disadvantages:
Poor uptime (one seems unable to post currently and I can’t identify any operational subs/boards for 2+ weeks). According to the project’s telegram/reddit, this will likely change in the next few weeks for most clients.

Buggy.

Token isn’t actually needed for posting (this is arguably a benefit as it will lower the barrier to entry for protocol use).

The utility for the token hasn’t been implemented (token tipping and a DAO for feed curation and development trajectory has not been implemented yet).

The token uses a proxy contract which could rug at his discretion/due to hacks. The dev says this was necessary to implement prior changes such as the airdrop, ending the liquidity mining and enabling the chain swap from avax to eth, and may prove useful in the future. The dev asserted that they are not chain loyal and will switch to whatever chain is most secure and used in the next few years, and expand beyond other ENS, which may need a proxy contract. While a proxy contract is concerning, it seems unlikely that the dev will rug as the dev has already self-funded for 2 years and the github is very active with a technically proven, albeit buggy product.
Scalability claims haven’t been proven as there aren’t enough users.

GUI for board creation isn’t released yet. Creating a board/sub seems complicated according to the github command line instructions, which is inaccessible for normal users.
Unless there are lots of seeders using the full client, the UX on the web client will likely be slow.

To run a sub, the sub owner needs to leave their computer on 24/7. If it is not being run, then you cannot comment, however you can still browse, non-updating content. (TPs/VCs have the opportunity to create archiving services to archive and protect downed subs).
According to the devs, if you change your sub from your public address to a ENS, you lose all your prior posts published to your public address. (necessitating better backups/archive service/or protocol improvements).

Tokenomics:
The supply was distributed through airdrops in 2021 on avax (before it migrated to Eth) that were advertised on 4chan and reddit, and then later through liquidity mining.
The token will be used for voting to curate the feed through a DAO and for tipping. However, currently the token use hasn’t been implemented into the protocol.

Funding:
The project is primarily funded by the lead dev, personally, who has already paid out $300k in total to the other 3 devs over the past 2+ years of development according to their github which tracks payments.

Plebbit received small grants from Protocol Labs and Gitcoin.

Technical:
Given the low holder count (652), low liquidity, low volume, low market cap and proven, albeit buggy technical product, I doubt this goes under $3m mcap for long. This is especially true as it has been in accumulation since 2022 (on Avax) and into 2023 on Eth, and sells seem to be limited, evidenced by the price action the past few days. A 50% correction on the weekly would be nice, but not a definite.

Benner cycle:
The meme tier “Agriculture Benner cycle” is also worth considering. This stipulates a crash in 2026. The Benner cycle is often 1 year premature which is why I have drawn a range of 2026-2027 with the two white columns.

Macro:
The 1-2 years proceeding the halving will likely drag all alts up based on previous history. This will likely occur regardless of rates rising or dropping. The global crypto market cap could reach a maximum of x7 to that of Gold’s $13.1t mcap with a ~40% BTC dominance, providing considerable gains to diverse alts like in 2017-2018. This would create a $~250k BTC as the 2026 cycle peak, per the Benner cycle, functioning as a leading indicator to the wider shitcoin market, before entering into a 3y crypto correction, trapping moontards hoping for a mystical $1mm BTC and continued shitcoin growth into 2026.

Crypto is an emerging asset class and in economic recessions, technology performs badly without the infinite M2 hack. Also, Senator Warren’s recently proposed Digital Asset Anti-Money Laundering Act 2022, targeting privacy platforms and strengthening KYC requirements for legally illusory, “unhosted wallet providers,” inducing bears to take a bite out of the crypto bull this past week.

However, this bearishness will likely correct as market regulation clarifies itself and persisting crypto narratives increasingly parrot “digital gold”, “be your own bank”, “lessen your vulnerability to bank bail-ins”, “institutional infrastructure” (ETFs and pensions) and the myriad of shitcoin narratives (“defi, NFTs, gamefi, decentralized social media/storage”), likely overpowering investor trepidation in an increasingly unstable environment (economically, socially, politically), where non-institutional and institutional investors seek fresh memes, dreams and alternative emerging asset classes outside the traditional economic purview, to cope with an increasingly negative worldview.

Also note the recent dovish FED meetings and the reflection of ~4 25bp rate cuts implied by US rate futures; perhaps a cosy foreshadowing of a 2024 BRRRRRRRRRRRRR printing session to help bulls march onwards.

Targets: (Assuming $3.55m mcap)

Conservative prediction 1: If the product doesn’t improve and adoption doesn’t occur = Sept (2024) $12m market cap (x3.3)

Conservative prediction 2: If product doesn’t improve and adoption doesn’t occur = Dec 2024 $50m market cap (x14)

Liberal prediction 1: If product improves, there are 300+ users + tech media coverage = Dec 2025 $100m market cap (x28)

Liberal prediction 2: If product improves, adoption to 1000+ daily users occurs = Dec 2025 $500m market cap (x140)

Liberal prediction 3: If product improves and adoption to 5000 daily daily users occurs + institutional fund’s/TP’s create plebbit infrastructure like archive websites, board hosting services, board backup services for a fee = Dec 2026 $1bil market cap (x281)

Liberal prediction 4: If product improves and adoption to 10000 daily users occurs + aforementioned institutional fund’s/TP’s infrastructure = Dec 2027 $10bil market cap (x2816)

DISCLAIMER: The above is educational/entertainment, not financial advice.
Comment:

TECHNICAL ANALYSIS:
There has been a correction from $3.5m mcap to some expected supports around $2.7m mcap.

It remains to be seen whether this is merely a simple ABC correction and we break out here or it is a more complex sideways or descending broadening wedge pattern.

The Christmas break will likely bring lower volume and boring sideways. Alternatively, lower volume could lead to more volatility.

DEVELOPMENT UPDATE:
Devs have released the full version of the old reddit styled client called seedit.
seedit.eth.limo

Devs have released another version of the 4chan styled client called plebchan.
plebchan.eth.limo

Devs have asserted that he has just finished testing webrtc and the results mean that plebbit can do fully P2P in the web browser which means there is less need to download the app for seeding the plebbit client. This would enable full P2P pubsub with full p2p dht peer discovery, using webtransport + webrtc with no signaling server.

This means that not only is the plebbit applications thereotically fully P2P and scalable but the the web browser clients, too, which would otherwise leech of full clients.
Comment:
There's now a potential breakout on the 4h.
I will wait for a green December and then look for the first 2 weeklies of January being consecutively green, to confirm the end of the correction.
Comment:
There has no been a large correction to the lower horizontal support - from 3.5m mcap to 2.5m mcap.
We have a declining MACD and RSI on the 4h.
I have liberally drawn an ascending trendline on the 4h which has 4 points.
I'd like to see to green weeklies and possible divergences on the MACD and RSI for confirmation of an end to the correction.

Disclaimer

The information and publications are not meant to be, and do not constitute, financial, investment, trading, or other types of advice or recommendations supplied or endorsed by TradingView. Read more in the Terms of Use.