The Daily Builder—On Curation by Moderation I

This post was originally conceived on my main blog on the Paragraph protocol. Please head to the original post if you’d like to have a better experience if you are reading this on mobile via a browser, or care to collect the post on the Base network as an .html object: LINK.

0. Introductory Mumblecore

I do contemplate upon the conceptual space, and contextual architecture of daily information and knowledge input a lot. It is about forming a pattern language for your own well-being.

It is also due to the fact that I need to systemize how I "consume" data on a daily basis. No, I am not talking about productivity. I am just alluding to a personal funnel by which it is probable to make sense of all that epistemological jazz so that I can pursue a day with its ultimate integrity, productivity, joy and serenity—yes, even under duress, and the hardest of times.

I feel lucky to be alive during the early phases of cryptographically self-securing open source, open finance, and consumer grade protocols and platforms; and, am mainly a power user across the Ethereum cluster adjacent social layers. The Farcaster protocol is one of these, and I do appreciate how the ethos of building and development manifests thereabouts where many core building teams accentuate that we need practical solutions to many arising use-case problematics by in actual being practical. Hence, the Farcaster protocol is a sandbox on which people are free to accelerate an evolution of user experience under and through many a domain.

Today, I am writing on a threshold upon which I have been brain-storming either to build a personal app or expand it to the commons space somehow. Yet, I am doing it by direct reference to several products, and toolkits without any affiliation of any of the teams. That is, I am not paid for this, and this is a thought experiment.

1. Problematic

The problematic on which I focus herein is the intuitive act of curation for a specific flow of knowledge. The social media is a blurry agora of the formal and informal serving both as the masters and sub-servients at the same time to use a rough language, that, the tongue strike which the drama thereabouts would appreciate.

What is the problem?

The problem to which we are alluding here is the curatory experience for follow lists on any social media platform. I have at least twenty lists for different domains of expertise on X for example ranging from simply MEV for Flashbots contributors or core developers for the Protocol Guild participants and contributors. This way I can keep up with the latest tidings in the ecosystem, and yes, I do in actual browse those lists every morning and evening. I also subscribe to others' lists such as Solidity Whisperers.

So far, the automation thereabouts on X is not even probable for me, and I need to consume whatever is posted on those lists. People can be speaking about sleep deprivation, or the latest nootropics all the whilst providing valuable information about Solidity, MEV, or Ethereum All Core Developers meeting processes. I am by default bound to the algorithmic flow which adds up to people's habit that they feel the need to share the professional alongside the utterly personal, which in turn buries the high-signal input amongst almost-noise—now do not get me wrong: What I would herein deem as noise may be the high-signal when I am after that type of information.

Can you picture it?

I want to be able to access and retrieve content from someone's personal feed under a curation by moderation approach. If Alice talks about the order flow mechanics at 0600 a.m., I like to be reading what she shares on the differences between the traditional finance and decentralized finance when it comes to latency that she shared some information at 0333 p.m. later that day; and, accordingly I do not want to read how she likes listening to phonk at the gym.

This is the informational retrieve angle 1. However, on my free day, or during leisure hours, I do not want to read about what she has to say about MEV, I wonder what she listened to at the gym that day, and what perks of that particular hotel gym offers that she quote-posted from Bob's feed—the angle 2.

This is basically curating a personal feed into a variety of sub-feeds which should be intuitive, too—not pre-programmed. This is a complex issue that can be solved in not so far from future because of the experiments at the application and client layers across Farcaster.

...and, the reason I am typing these words right after I woke up is that Dan from Icebreaker asked about an evalution of the following cast if I am willing to do it:

2. Insights

1. I am a paying user for Buoy which is a real time keyword search and monitoring application on top of the protocol. Free tier lets you access a single keyword, and if you need to monitor more than a single keyword at ease, you can currently upgrade to the pro tier for 10 $USD a month. As an independent generalist researcher, I needed such a tool for so long, having been used to manually exhuming related data on not so quick apps, some of which are also cool, too. Since I am the owner of /philosophy and /openfinance channels, I am currently tracking topics related thereof. I have been able to cut through the early phase of channel moderation noise thanks to Buoy lately which is tethered to how I utilized Automod.
2. Nook is another Farcaster protocol client such as the orginary Warpcast or über-sleek Supercast, and it is being developed by the team who previously shipped inspiring clients such as now-defunct Flink. Why do I like Nook more than others? Well, I do not like any client more than any others when it comes to these aforementioned. Each offers gradually emergent features specific to their own, and I like to make use of all at once. It is like having Coinbase Wallet for the in-browser experience and a spam filtering XMTP integration, Rainbow for tracking the OP Superchain project tokens, Zerion for an easy mobile cross-chain bridging, and DeBank for a flawless portfolio tracking, Fluidkey for proof-of-innocence privacy, and Daimo for easy USDC transfers. Render unto Caesar, and there are many Caesars around yet it is the needed crowdiness of emergent economies. Each nurture one another.Yet, Nook has a set of features that I like. You can track transactions from the main UI, you have a different tab for Farcaster frames, and you can curate channels and users under lists—just like Supercast has but I want Supercast to be a more wabi-sabi UX and the very same listing options coupled with the schedule feature currently is enough for me to keep supporting their pro tier. Up until recently, Nook had another feature with which you could create Twitter list-deck style clusters called, well, nooks. To be honest, I miss that one feature, and I could dream of a feature of a feature where it would allow us to create super-clusters for channels a la meta-sub-Reddit ecosystems. The only reason why I mentioned Nook in the above-embedded Warpcast cast was those nooks yet it's alright. Hence, we can approach Nook and Supercast as equals in this piece of writing.
3. Automod is a moderation suite specifically designed for the decentralized channel curation needs across the Farcaster protocol. It is one of the most intuitive middle management toolkit that I have ever resorted to. I am utilizing what it has to offer to curate the main feed of the /philosophy channel, and /openfinance channel.For the uninitiated, the Farcaster protocol has recently upgraded so that channels are now decentralized, permissionless social graphs by themselves; and, any visionary builder or developer can come up with heretofore unbeknownst wonders upon them, or merely choose to address the needs these emergent communal information funnels are literally craving for since the implications thereof are unprecedented.Basically, I am running the philosophy channel, and, I do not want the noise to bury the high-signal output of ideas. The main feed in the channels are either curated by the algorithm when owners choose not to uphold the latest moderation opportunities—to roughly articulate it; or, owners now can opt in for the main feed vs recent feed dichotomy where the former being the high quality cast haven whereas the recent might be brimming with spam given the size of the so-called community of a channel. There, I decided to price the access to curation by moderation for the main feed via a Hypersub space where people mint or top-up recurring time on top of ERC721 issuances, that are a type of NFTs, and set a couple of rulesets in the Automod so that those who pay the curation cost can be amplified, or boosted, as long as their casts abide by the additional rules for a civilized discourse with decorum. Here, you can read about it in detail:
4. So far, it's working for me. Yet, as you are aware, the subject matter of this monologue that sounds like a dialogue is that I have curational technique dreams, if not visions, for an inexistent super-app on top of the Farcaster protocol, and all of these apps and products that I mention have some feature to offer towards that end. Hence the question:

What do I mean by postulating that "we'll see just relevant contextual content of a person we follow (in)directly"?

Firstly, we will not need to access each and every crumble of a thought that an individual shares with the rest of their social graph. You cannot prevent people from sharing their own thoughts within the confines of this algorithmic outside that needs an inside. Secondly, people want to read individual takes by otherwise professional figures whose takes on their area of expertise has so far come to define their status. Yet, I do not want to read all of their stream of consciousness. When I am browsing the Solidity Whisperers list, I do not want to read Alice contemplating upon the quotidian as I have asserted.

There will a come phase, a plateux if you prefer where the magick of API—such as the current implementation of direct casts—will tailor to an individual user's holistic social layer experience in terms of the Farcaster protocol, and other future protocols that will be conceived upon it either with a rage-quit, or a convivial survival.

In this ideation, or individuation by techniques, my ideal social media feed will be available on a Daylight Computer, and its HUD emulations on Apple Vision Pro, a Frame computer version of this Apple iPhone 12 mini—till we become disembodied specks of immortal individuations who need no UX/UI for any information a la sensorium—where I will simply be able to read only context-related casts without a preference of the individuals' daily life musings.

Toolkits such as Automod will let me configure the flow of information across not only channels but home cast feeds, too. I will then be able to curate these individuals' pre-curated knowledge flows into RSS-like configurations where applications such as Buoy will re-order in my preferred filter criteria. Many of these applications and toolkits will be enshrined into visionary clients which will also acquire other aspiring ones. Though protocolizing clients upon ever-emergent protocols is akin to protocolizing the self which might have gloomy outcomes, either, which is, however, the topic of another essay.

3. How?

I am coming back to Dan's—of Icebreaker—initial reply-question: How does applications and toolkits such as Buoy and Automod help me curate the current SOTA flow of information?

• On the personal level, I am not able to curate my home feed directly. I approach channels that I own and co-host with different takes when it comes to curation. I own 12 channels, and co-host 1. All of these channels serve for topics I am directly intrigued by—some acting as channels where I cast the noise so that main channels of importance for myself will not get that much of a noise. I make sure that my personal channel /gokhan should act as my thought-book alongside personal artistic sharings, and quoitidian musings. Accordingly, my Automod settings are thus contextualized.
• On the professional level, I approach topically, and set keywords on Buoy that I constantly track which are specific to my interests which are in fact the reason some of channels exist: /philosophy, /openfinance, /atheism, and the like. Hence, it's easier to access quality input from other users that I have no other way of encountering given my social graph, and owing to the exponential expansion of the Farcaster protocol as a social layer.
• As I have already mentioned, I also set the rule sets for the channels via Automod. For example, for the philosophy channel main feed, the casters need either to have a Hypersub pass where their casts also abide by the general ruleset or have their casts favorited by either myself or any other human co-moderator to whom I grant the pre-configured moderation access which also utilized the cast actions—otherwise, in the standard paid access, the automod bot handles the entire process by boosting their inputs.
• Hence, the Buoy to Automod flow that I manually configure help me better the channel experiences in line with the latest genuine trends, and this initial methods of community-bootstrapping modalities specific to Farcaster.
• Yes, there exists many a service or product under these categories, and I make use of almost all of them. I know the best short cut would be to host my own hub, listen to the semantic mempools of Farcaster, and build my own apps; yet, there is still time for me to implement that so that I can recombinantly make use of the features of the above-mentioned products as an inspiration to say the least.

Dan, thank you for the question, and this is how I in fact moderate the curational contextualies of my daily Farcaster consumption. I know the thought process is peculiar to how I think about knowledge economies' infrastructure but I will be diving into more detailed, and properly diagrammed series of such writing pieces. In the meantime, you or anyone can access my personal group chat via this link.

For the reader, this post is a collectible, and you can support independent researchers such as myself by collecting such pieces so that I can keep offering my thought-pieces ungated.