• henrypye.xyz, a personal portfolio site

• tech.interview.practice, an interview prep training platform

Principles explored:

• Progressive disclosure through UI state management

• Content organization at scale (200+ structured learning materials)

• Adaptive coaching through prompt refinement

• Accessibility through toggleable immersive modes

henrypye.xyz received a quick maintenance update to surface recent press coverage, but the bulk of today's effort landed in tech.interview.practice, where I made meaningful progress on both the learning architecture and the user experience.

The interview practice platform needed breathing room. I built a toggleable sidebar system that collapses automatically when users enter immersive modes—like the mock interview or voice coaching flows. The insight here is that cognitive load peaks during practice, so the UI should vanish when it matters most. Rather than hardcoding collapse behavior, I centralized it through a context provider so any future feature can signal "I need focus" and the sidebar responds consistently. This felt like the right level of abstraction: the sidebar doesn't care why it's hiding, just that it should.

I also restructured the desktop challenge layout into a tabbed dual-view—Description and Tests side-by-side without scrolling friction. The philosophy is immediate feedback: readers see the problem statement and validation criteria in one glance, reducing the context-switching cost that kills learning momentum.

On the content side, I integrated the full Cracking the Coding Interview (CTCI) curriculum—156 challenges across 13 chapters, organized into modular data files. Rather than dumping problems into one monolithic structure, I split them by chapter to keep the codebase navigable as the challenge count grows. I also wove in pattern recognition training and OS concepts from foundational textbooks, then deepened the coaching prompts to reference these methodologies directly. The goal is for the AI coach to reason about why a problem matters, not just critique the solution.

The rebranding from "Tech Interview Practice" to "Software Engineer Training" reflects a shift in intent: this isn't just about passing interviews anymore. It's about building the reasoning patterns and knowledge that make interviews irrelevant because you're genuinely prepared. That philosophy now flows through the prompts, the challenge organization, and the README.

I'm particularly pleased with how the sidebar context abstraction decoupled UI state from feature logic. Each immersive mode stays focused on its own concern while respecting a shared principle—minimize distraction when concentration is critical. That's the kind of simple pattern that scales well.

This week: Invader's collaborative "Hyperspace" exhibition is live in Tokyo through early April, while a stunning new limited-edition print with Shepard Fairey and Damien Hirst is available for just days longer. Meanwhile, the artist's prolific London invasions continue to captivate the community.

Hyperspace in Tokyo

Invader's "Hyperspace" exhibition is currently on view at Isetan Shinjuku Store in Tokyo, running through April 6, 2026. The exhibition is free to enter and open daily from 10:00 to 20:00, making it accessible to both dedicated collectors and curious passersby exploring the bustling Shinjuku district. This marks another major international moment for the street artist, bringing his iconic aesthetic into a premier retail gallery setting in one of the world's most vibrant cities.

Limited-Edition Collaborative Print

The fifth collaborative print between Invader, Shepard Fairey, and Damien Hirst is now available through HENI Editions, titled "Invaded & Commanded Blossom." These museum-quality works are Gloss Diasec-mounted Giclée prints on aluminium composite panels measuring 120 by 96 centimeters, priced at $3,000 each. The limited edition closes on April 7, 2026, so collectors interested in securing one of these striking pieces should act quickly. This partnership continues the three artists' prolific creative dialogue, following their major "Triple Trouble" exhibition that recently concluded at Newport Street Gallery.

Sources

• Invaded & Commanded Blossom by Fairey, Hirst & Invader

• Invader "Hyperspace" (Isetan Shinjuku Store)

• henrypye.xyz, my personal site and portfolio

• life.os, my personal dashboard

• lifeos.grafana, monitoring infrastructure

• tech.interview.practice, an interview preparation tool

Principles explored:

• Constraint-driven writing (distilling complex work into safe, honest language)

• Timezone-aware batch operations (UTC boundaries for reliable scheduling)

• Defensive prompt engineering (layered validation and redaction rules)

• Idempotent state machines (backfill logic that survives restarts)

Today was dominated by a single, elegant problem: how to automatically generate reflective engineering diaries from raw commit history, safely.

henrypye.xyz, my personal site, saw a refresh cycle. I rewrote the About page repeatedly, iterating toward authenticity: introducing language for specific achievements, removing editorializing, and anchoring every claim to real work. This wasn't busy work—it was me learning how to frame my career honestly, which fed directly into the constraints I'd later codify.

life.os, my personal dashboard, became the sandbox for the real work. I built a dev-blog module that generates these diary entries automatically. The architecture is deceptively simple: collect commits from GitHub, validate against a restricted language list, redact sensitive patterns (credentials, sensitive information, architecture details), then feed the scrubbed data to Claude with strict formatting rules.

The trickiest part was what to exclude. I added an explicit allowlist of forbidden terms and patterns. The redaction layer strips these before they ever reach the AI. I also enforced that every paragraph must anchor itself to a specific repo, preventing vague hand-waving about "architecture" or "systems."

Timezone handling nearly broke me. Commits queried at 3am UTC need absolute boundaries, not relative "yesterday" checks. I moved to explicit UTC date ranges throughout the pipeline.

For reliability, I added a backfill on startup. If the cron job fails, the system catches up cleanly. The weekly Invader roundup does a two-pass research-then-write flow: first, search the web and fetch articles; second, synthesize them into prose. This decoupling means failures in one pass don't poison the other.

I also added dynamic cover image generation via Cloudinary—abstract black-and-white SVGs that render text overlays. The fonts required Docker configuration (fontconfig isn't bundled by default), a small but necessary detail for text rendering in containerized environments.

tech.interview.practice, my interview tool, got a security refactor. I reverted the bring-your-own-keys (BYOK) feature and returned to environment-based API key management. The lesson: complexity isn't always better, and user friction around secrets is often a feature, not a bug.

The through-line today: constraints breed clarity. Building a safe diary generator forced me to think deeply about what's worth documenting, and what's dangerous to leave implicit.

• life.os, my personal dashboard

• tech.interview.practice, an interview coaching tool

Principles explored:

• Privacy-preserving logging and audit trails

• Voice-first interaction design and silence detection

• Graceful degradation in real-time speech systems

• Prompt engineering for domain-specific AI coaching

• Client-side state synchronization without hydration mismatches

Today was split between two very different kinds of work: infrastructure for introspection, and UX refinement for real-time interaction.

In life.os, I've been building out the developer diary system — the very thing I'm writing in now. The work involved rethinking how commits get collected, redacted, and shaped into reflective prose. The core tension here is transparency without exposure: I want genuine records of what I built, but certain details (API endpoints, database schemas, user names) have no business being permanently logged. I refactored the redaction pipeline to be content-aware rather than just path-based. Instead of surgically removing lines, I'm now filtering at a semantic level — checking whether a message references confidential patterns before it ever gets written. This feels more honest than pretending the work didn't happen; it just acknowledges that not every technical choice belongs in a public diary.

In tech.interview.practice, the work was almost entirely about voice interaction. I replaced a text-based mentor with a voice coach ("Sam"), which fundamentally changed how the system needs to behave. Voice introduces latency, ambiguity, and the need for real-time feedback that text doesn't require. I spent time tuning the voice activity detection (VAD) threshold — the moment when silence becomes "you're done talking." Too eager, and it cuts users off mid-thought. Too conservative, and the system feels sluggish. I also discovered that when the code editor is open, VAD should pause entirely; otherwise, keyboard clicks and typing noises trigger false positives.

The other layer was prompt engineering. I realized Alex (the coach) was reading code aloud unnecessarily, wasting tokens and confusing users. Now I explicitly tell the coach to reference lines by number instead. This meant prepending line numbers to the code sent to the API — a small change with outsized impact on the interaction model.

I also fixed 25 audit issues across security, accessibility, and reliability. Most were straightforward (updating dependencies, adding ARIA labels), but a few revealed deeper problems: hydration mismatches in useEffect hooks, Monaco editor cursor jumping on updates. These aren't glamorous bugs, but they're the friction points that erode trust in interactive tools.

I spent considerable time establishing architectural documentation across multiple codebases. This meant creating clear entry points for future maintainers and myself: foundational README files, contribution guides, and context-setting documents. There's something satisfying about crystallizing how a system hangs together into prose that a fresh pair of eyes can follow. It forces you to articulate assumptions you didn't know you were making.

On the interface layer, I tackled several small but important UX problems. The drag-handling interaction needed refinement—I discovered that delegating the drag initiation to the header component, rather than letting it bubble from anywhere on the card, creates a more predictable and less accidentally-triggered behavior. Similarly, I implemented adaptive zoom controls that scale their responsiveness based on the current magnification level. When zoomed far out, you want coarser, faster movements; when close in, finer control. This kind of context-aware adjustment is easy to overlook but significantly improves usability.

The chat integration work exposed several parsing edge cases. I encountered malformed JSON responses from an AI service—incomplete arrays, truncated objects—and built in graceful fallback parsing to recover what we could rather than crashing outright. I also added multiline input support, which required adjusting how form submission is triggered so that newlines within a message don't prematurely send. When bulk requests hit size limits, I discovered we were running into a fetch timeout. The fix involved both increasing the request buffer size in the build configuration and potentially chunking large payloads.

One particularly tricky bug involved a utility function that mutates nested object structures. It was crashing when intermediate keys didn't exist yet. The solution was defensive: check for existence before attempting to traverse deeper into the structure. These invisible-until-they-fail bugs are humbling—the code path only executes in specific states that testing didn't cover.

Finally, I worked through monorepo dependency reorganization in another codebase, ensuring that containerized builds could properly resolve all required packages. This involved untangling which dependencies belonged at the root level versus which should be scoped to specific applications. There's a broader theme here: establishing clear boundaries and responsibility chains—both in code organization and in how systems interact with each other.

Today was dense with UI refinement and observability infrastructure work — the kind of day where seemingly small interaction fixes compound into meaningful improvements across multiple surfaces.

I spent the morning wrestling with state synchronization challenges in a research visualization tool. The core problem: edits initiated through an AI-assisted interface weren't reflecting in the UI, even after successful API calls. I traced this through several layers — stale closure issues in React components, memoization preventing re-renders, and the subtle gotcha of reading state values synchronously after dispatching updates. The fix required removing memoization boundaries and forcing dynamic rendering, which feels inelegant but revealed deeper architectural constraints worth addressing later. The edit flow also needed defensive programming: detecting when proposed changes matched current state to avoid no-op mutations.

I then optimized the economic model of conversational features by switching to a more cost-efficient inference tier while implementing prompt caching. This required careful consideration of context windows and token budgets — ensuring truncation wouldn't degrade output quality while keeping API costs sustainable. The tradeoff between model capability and operational expense is always present.

On the canvas side, I integrated graph layout algorithms to automatically position nodes based on their connectivity topology. This required wiring up a third-party layout engine and reconciling its output with the existing persistence layer — ensuring auto-layout didn't discard user-intentional positioning.

The second half pivoted to observability infrastructure. I spun up a monitoring stack with several gateway layers, each requiring precise datasource configuration and health check endpoint exposition. The subtlety here was access mode semantics — whether services communicate directly or through reverse proxies affects latency and debugging complexity. I also added readiness probes to critical path services, following the principle that load balancers need signals beyond basic TCP connectivity.

Throughout, I refined interactions: double-click editing across card types, drag-and-drop reordering, improved text selection behavior during canvas manipulation. These feel peripheral but directly impact how developers experience the tool.

For those of you who don't know, I created the /invaders channel on Farcaster. I created this channel about 2 years ago after discovering the street artist Invader and becoming slightly more than slightly obsessed with his installations around the world. From this channel, I created a mini app called Flashcastr which hooks into Flash Invaders and shares out the mosaics you flash onto your social feed. You can see this on my feed, as nowadays I rarely use Farcaster for anything other than sharing information about Space Invader and his work.

One of the benefits of this app is that I can chronologically see the flashes, and the order in which I found them. This makes it easy to share my route through the city, for others to learn from and copy. It took me 3 days to find all of the mosaics. Over these 3 days I used many London modes of transport - walking, Santander Cycles, a black cab, a London Bus, the Tube, the overground, and even an uber at the end.

Usually, I use pnote maps to find the latest locations. However, the 21st wave was not on there so this time I used a combination of Map Invaders, pnote and manually finding the new ones on Instagram and adding them to Google Maps. I'll attempt to document our route below, by sharing key flashes from day one through the day three.

If you wish to follow this route, it ends at the Newport Street Gallery where the Triple Trouble exhibition, and the final mosaic, are located. A great way to end the London hunt, enjoy a coffee, and see some incredible and unique art! The exhibition ties in beautifully to the ambitions of each artist. Of course, for me, the highlight is Invader and the exhibition really is a great homage to Invaders mission with his mosaics, with great narrative and story telling, and pieces to enjoy. Loved it. Go visit!

So, the route. I started in West London, all the way up near Kensal Green. It took about 40 minutes to get all the way up here from central London, and I walked all the way down to Portobello Road, then onward to Kensington Gardens, through Hyde Park and into Mayfair - while flashing about 15 mosaics. Great start. I also find the London mosaics so interesting and fun to find! Great hiding spots, especially on this side of town.

Flash #101188460 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

Flash #101188974 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

Flash #101189962 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

After making it through Hyde Park, I stopped for some lunch in Mayfair and then begun to plan my route around Mayfair and central London - another hotspot, with about 33 active mosaics as of my hunt. Now, the London weather was definitely against me here, and if you've ever been in central London on a rainy windy day you will know exactly what I mean. At one point, it got so bad, and the wind so terrible on Wardour street that I had to take a cab just three blocks, just because I couldn't even walk down the street without getting blown about and drenched in rain! But, a London black cab added to the London hunt - how great. Plus a massive 33 mosaics flashed, and when the weather was totally unsupportive. Awesome.

The route taken here is from Kensington Gardens, through Hyde Park down to the Royal Albert Hall. London bus from Royal Albert Hall through to Green Park, Green Park up to Grosvenor Square, stopped in Marylebone for lunch & to get dry by a fire in a pub, then up to Baker Street, down into Goodge street, down into Soho, stopped again for a coffee and to get dry, then across to Tottenham Court Road, down into Covent Garden. Then stopped again in Seven Dials to get warm and dry in the food court there. Then across Waterloo Bridge and into Southwark to flash the new mosaics installed there - including one huge Clockwork Orange piece underneath Waterloo Bridge. Truly awesome. Also picked up the Triple Trouble mosaic installed this summer, before the show at Newport Street Gallery - a personal favourite of mine. Side note: the 21st wave of London Invasion - absolutely incredible mosaics. I also love how a lot of these mosaics are installed next to book stores, so a lot of time was spent popping in and picking up a book, and having a sit and read.

Flash #101200537 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

Flash #101202317 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

Flash #101209166 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

Flash #101202908 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

Flash #101223634 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

So, day 1 wrapped up, with an incredible 41 mosaics discovered mostly by foot and definitely in a total London storm. On to day 2.

Day 2 was a beautiful London day, one of those days where the sun is shining, the streets are totally dried up and there is no evidence that just 12 hours ago there was total washout rain! On this day, we started the hunt in east London - Islington. This is the day we hired Santander Cycles on a 24 hour pass for just 7 pounds. This was by far the best way to find the mosaics in London. Definitely recommend. The route this day involved finding 47 mosaics. We started in Islington, made our way to Old Street, then across to Shoreditch where we stopped in Hoxton for breakfast. Then, from Hoxton we cycled to Holborn and found the final ones there before heading south of the river and cycling around the Borough, then down to Elephant and Castle, before looping around back up through Lambeth and back into the city. This route allowed us to pick up all mosaics here, and see the north side of London in a totally unique way. I've never seen the houses of Parliament from that angle! We also cycled through some really old parts of the city. I've spent quite a lot of time in London, considering I am in the tender parts of my early 30s, and I was taken to places I'd never been before. Charm and love for London reignited!

Flash #101231343 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

Flash #101235231 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

Flash #101235573 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

Flash #101257335 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

That evening, we rented a car and drove to the awkward mosaics dotted in outer Hackney and Walthamstow. Probably not the most efficient way, but we were tired and they were so far out...

We left a great route for day three. On the final day, just six mosaics remained. Two on the Kings Road in Chelsea, one in Piccadilly, one in Camden Town, one in Brixton and one in Lambeth at the Newport Street Gallery. Game on.

We started the day with an uber to Battersea Bridge. From here we walked to get the first mosaic, Vivienne Westwood. Then we walked along the Kings Road to find the second mosaic. From here, we hired Santander Cycles and cycled into Piccadilly - including a very hairy route across Hyde Park Corner - and into the city. Then, we hopped on the tube and took the Victoria line to Brixton, then on to the Northern Line up to Camden Town. From Camden Town, we hired Santander Cycles again and rode through Regents Park, down through Marylebone where we dropped the bikes, hopped on the tube and took the Bakerloo Line to Lambeth North, where we walked the short distance to the Triple Trouble show which marked the end of our hunt!

Flash #101281967 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

Flash #101295035 by WORLDY | Flashcastr

London, WORLDY - Flash from London by WORLDY

https://www.flashcastr.app

Hopefully this blog post is useful for you if you are visiting London. I think so far, the UK installations are my favourite to find. A lot of them are installed totally at street level and are totally untouched and appreciated, blending in perfectly with the cities. Hopefully you have the same appreciation for these if / when you go find them.

I’ve been thinking about why I’m so drawn to Invaders’ work, especially after someone asked about it on Farcaster Fridays a few months ago. At its core, I love how Invaders represents art in its purest form - creation purely for joy. The artist started simply because he loved Space Invaders and wanted to bring that aesthetic to the streets. There’s something beautiful about that simplicity of motivation.

Street art, at least as we know it, first came around in the ~1960s, and the Internet transformed it. The ability to share images online pushed the movement forward dramatically. This intersection of physical and digital documentation particularly interests me because street art is truly ephemeral. I’ve seen this firsthand - Invaders get removed, sometimes by local authorities looking to remove it, sometimes by private individuals who just want one, sometimes by people who want to resell them, and then...sometimes mysteriously reappear.

Documentation and Preservation

I’m working on creating a decentralized way to document street art using smart contracts. Storing pictures (flashes) of these mosaics, on a decentralized ledger would provide persistence to an ephemeral medium. This applies to both storage blockchains such as Arweave, and data blockchains such as Ethereum (L2's are making this documentation a lot more feasible). The public nature of blockchain could enable others to build on top of this documentation, similar to how the Space Invaders API has enabled projects like Invaded Map, Invaders Fun & my new social project, Flashcastr.

What Makes Invaders Special

The intentionality behind Invaders’ work fascinates me. Their eyes sometimes look at something specific - a skyline, a view, a particular aspect of the city. Just yesterday in the Hollywood Hills, I found two installations specifically positioned to view the LA skyline. They serve as artistic tour guides, leading people to parts of cities they’d never normally visit.

The mosaic medium itself is significant. Unlike graffiti that can be painted over, these pieces have a different kind of permanence. You can trace the artist’s evolution from basic swimming pool tiles in the late 90s to specialized tiles that break when removed, adding another layer to the installation’s permanence and meaning.

Connection to Digital Art and Bright Moments

My appreciation for this art form deepened through my work with Bright Moments, where we traveled globally working with generative artists. Our Crypto Citizens project, which harnesses pixel art, parallels Invaders’ work in interesting ways. We were creating digital street art on the Ethereum blockchain - making permanent marks in a digital space, just as street artists make their mark in physical space.

The Broader Impact

These installations serve multiple purposes - they’re good for mental health (getting people outside), physical health (walking to find them), and cultural exposure (exploring different parts of cities). The parallel between blockchain and street art strikes me - both are spaces intended for one purpose that people have adapted for artistic expression. Just as street signs collect stickers and tags, blockchain has evolved beyond financial tools to become a canvas for digital art. This beautiful evolution of spaces, both physical and digital, continues to inspire my work in documenting and preserving these artistic movements.

AI & Tokens

Last year, we were all about Degen. This year, we are all about AI and tokens. In the last 12 months we've seen the rise of agents and social. Aethernet & Clanker hit it off, deploying tokens such as $LUM, $CLANKER, and $ANON - with the latter peaking the interest of Vitalik. This excitement definitely carried through to this years conference, notably with Neynar making a fresh hire with Grin specifically focussing on AI & ML. I'm excited to follow Neynars journey into harnessing AI and how that will be used to improve our experience across Farcaster integrated clients.

I can't help but feel that we are still in that hype cycle phase where tokens are exciting new shiny things, and we aren't really thinking about how to use tokens in a meaningful way yet. $ANON is the best example I've seen of this on Farcaster, and it feels like we've all forgotten about it. Even throughout the conference, we didn't really discuss tokens - there's the base narrative of tokenise everything, but maybe this is noise? It would be great if we could be asking whether we should tokenise something rather than tokenising something simply because it is possible. Definitely feels like we're in the spam mail era of tokens. Following the Rodeo release, I went a bit crazy collecting on there. Shortly after, a good chunk of my time was spent manually transferring all of these tokens out of my connect wallet. But perhaps AI will help with this, and tokenising everything is sustainable in the long-term. Times are changing.

Mini Apps

A huge click for me this year was just how powerful Mini Apps are. It feels like I totally missed Mini Apps - thumb explained them to me at a Farcaster Fridays, which prepared me for this Farcon (thank you thumb!). It is clear that Mini Apps will be a big factor in the next stage of innovation on Warpcast. Seeing all of the different apps that have been released - crowdfunding, give, usetab, farville, and more - makes me excited. These are useful apps, that contribute meaningfully to the world and are just so easy to use. A future where these can be pinned to a phones home screen (similar to a progressive web app) would be incredible. I've struggled to get my head around Mini Apps because they were called frames v2. I figured frames v1 were good enough. I was wrong, and am excited to build some of my own Mini Apps.

Growth

Sign up fees are dropping from $3.00 to about $0.20. On top of this, and there will be an ability for an app to purchase storage space and allocate to users for a fixed period of time. Building apps for non-crypto natives and tapping into the Farcaster social graph without charging someone $3.00 has become totally possible. The barrier to entry just got significantly lowered.

I'm curious to see how the import of users from platforms like Coinbase and Zapper pans out - what will this do to my home feed? A lot to come, the next 12 months are going to be fun and interesting. The Farcaster community feels close right now, with everyone knowing of, or knowing each other. Any good platform needs to grow, and there is now a reasonable path forward for introducing new faces to the platform. In just the last 12 months a football (soccer, in the US) player, Nathanael Ogbeta (natogb3), who played for my home town signed up!

I have spent the last few years making crypto easier for non-crypto natives to understand and use, and seeing the excitement that people get when they see the magic of cryptocurrency - from the instant transfer capability to the enjoyment of incredible onchain art - is awesome. This kind of adoption can only happen if we abstract away the technical aspects from people.

Conclusion

I am grateful to have worked with such passionate people to make Farcon happen this year. It was so lovely to see familiar faces, catch up with friends, learn all sorts of new things, and create some beautiful projection art for Blackbird & Stand with Crypto. It's easy to forget just how niche what we are engaging in every day is - a social, crypto-integrated platform, where all of us (maybe?) have conversed with AI at some point. It really feels like like we're in the golden years of Farcaster, where we're all curious minds picking away at this new, open protocol that can do all sorts of interesting and amazing things. Cheers to both the near and distant future 🍻

Introduction

Lately, I've been pondering innovative and enjoyable ways to democratize various aspects of our lives. One of my passions is houseplants. I adore nurturing and growing them. The right houseplant can transform the ambiance of a room, creating a healthier and more inviting living environment.

What if we could democratize the care of these plants as an experiment?

I am initiating an experiment that will open up houseplant care to the democratic, decentralized internet.

My proposal involves allowing users to mint an NFT linked directly to a physical houseplant. Upon minting the NFT, it will be integrated into a modular system enabling various care actions. Initially, I think we should start simply with watering. This NFT will connect to an ERC-20 contract representing shares. Each houseplant can have up to 1,000 shares, mintable at 0.0001 ETH each. Therefore, full ownership of a houseplant, involving all shares, is entirely possible.

Becoming a shareholder means becoming an integral part of the plant's life and its success. This raises several intriguing questions: Is it better to have a single shareholder to keep a blockchain-based houseplant alive, or should multiple shareholders from around the world be involved? How will shares be distributed, and will people trade them on the open market? What if someone in Australia catches wind of this idea and decides to buy a share? Should there be a safety system in place to prevent overwatering?

For the initial phase, I aim to develop a smart contract that facilitates:

• NFT minting linked to the houseplant using a physical chip.

• Connection of the NFT to an ERC-20 "shares" address priced at 0.0001 ETH per share.

• Linkage of the HousePlant NFT to a Watering contract, where shareholders owning over 50 shares can initiate a watering event (a cost basis of 0.005 ETH should deter any malicious intent).

The watering system will operate through a smart contract with write functions. Activating a write function will trigger an event, which a low-cost hardware device connected to a bucket and watering system will detect. Refer to the video below for inspiration.

The goal of this experiment is for all shareholders to collaborate harmoniously to sustain one or several houseplants. Naturally, this opens up possibilities for hostile takeovers or even sabotage through overwatering!

I believe it's vital to make this experiment as interactive and educational as possible, without manual checks to prevent overwatering initially. The survival of the houseplant rests entirely in the hands of the shareholders. If insufficient shares are minted to maintain the plant, it risks perishing.

Investing in the survival of the plant should be rewarding in itself, but let's get creative. There are additional aspects I'd like to explore, though I want to keep the initial model simple.

Shareholder Logbook

An on-chain logbook will allow shareholders to submit timestamped logs and progress updates at any stage of the plant's growth.

Objective Logbook

Ensure the livestream captures daily snapshot pictures of the houseplant, which will be added to an online, read-only logbook accessible to anyone at any time.

Houseplant Livestream

A continuous livestream will enable all interested parties to check on the plant's status. Also, if someone triggers a watering event, the livestream will become a crucial tool for real-time observation.

Ragequit

Any shareholder can opt out at any time and receive half of their initial payment back from the contract. This feature should help stabilize the public market price of the shares.

Propagation: Tired of competing for a chance to water the plant? Feel that another shareholder keeps overwatering? Once you acquire more than a certain number of shares, you can propagate the plant and instantly own 100% of the 1,000 shares of your new plant.

Herb garden: Look after a herb garden, get recipe ideas and get VIP access to a livestream when a dish is being cooked that uses herbs that you have cared for, looked after, and hold shares for.

Request to Rotate: Propose that the plant be rotated to ensure even sunlight exposure. Approval must come from less than 50% of shareholders within 7 days of the proposal.

Plant Feeding: Schedule a plant feeding session, similar to watering. Note, plants can also be overfed.

Introduction

On Friday, June 21st, together with UncleDavo and Serendipity, we hosted the Onchain Summer Solstice. Our goal was to bring together New York Casters and artists for a celebration of the longest day of the year, digital arts, and community.

The event was a great success, with approximately 50 people in attendance for the first half and 50 more people for the second part of the evening! Throughout the night, 26 attendees claimed the POAP, physically included in the booklet. Both canvases we set up were covered in paint.

We celebrated the longest day of the year together, collectively on the rooftop as the sun set, creating an unforgettable experience forf everyone involved.

As part of this celebration, we collaborated with over 30 artists and 7 local DJs to create a projection mapping sequence on the local church. For anyone wondering what this looks like, here's a picture!

Each artist had a 15-minute projection sequence, where a selection of their work was projected onto the church. Many of the works were curated from collections released on the blockchain. A few of the collections involved static images and photography, so it was important to think about how to best display these with audio-reactive elements. The projection mapping software used was HeavyM, which allows for listening to a line-in. Using an RCA cable and an audio interface, my laptop could pick up the feed directly from the CDJs. This meant the projection sequence could react to the bass, mid, and low frequencies that the DJs were playing.

The church was separated into sections. The faces and steeples were reserved, and only the artists' work was projected directly onto them. The spires, windows, and bottom ring were kept for custom effects mapped to the frequencies being output by the DJs.

HeavyM utilizes shaders to manipulate images. The most common shader used was the VHS glitch effect, mapped to the bass and mid-frequency of the DJs' sets. This created an effect that made the whole mapping appear to glitch and move with the music being played.

We looped in the local pastor, Rick, and told him about what we were planning. He, along with the entire parish, was extremely supportive of the projections on the church and was fascinated by it! Having their support legitimised the entire process, enabling me to focus on creating a seamless experience.

Artists

Overall, 31 artists were projection mapped. Below is a list of all the artists and their Farcaster profiles/websites.

New York Artists:

Lars Wander, steviep, Maya Man, LoVid, Clay Devlin, Emily Xie, David Nobody Jr., Dave Krugman, Ngozi, Luke Shannon, Efdot

Artists curated from Farcaster:

Bombay Malayali, Compusophy, Trevor Traynor, Anemale, Typo, Morbeck, Eduxdux, Gabriel Nebular (metaliminal), Priyanka (metaliminal)

Artists curated from fx(hash):

Pixel Symphony, punevyr, pxlshrd, Lammetje

Artists curated from Art Blocks / Bright Moments:

Julian Hespenheide, Kim Asendorf, Andreas Gysin (ertdfgcvb), Motus Art, Spongenuity, Fingacode

Farcaster Collectives:

higher

Squibs, Sicoli,

Le Petit Box (instagram):

J4ck, S A M I, Manimogo, PAC, OKDunc

We wanted to highlight local artistic talent. To do this, we began asking people who we knew lived in the local area if they wanted to play. The first person we approached was Sicoli, who has been playing for a long time and has supported many artists. Sicoli played the sunset set, from 8 PM to 10 PM. The next DJ we approached was Squibs, who lives locally and is a Brooklyn regular, traveling around New York playing his music everywhere. Squibs played from 10 PM to 11 PM.

Finally, we approached Jack of the Le Petit Box (LPB) crew. LPB has just started their residency at The End in Bushwick and are all regulars around Brooklyn. Asking them to play wasn't even a question, and five guys from the crew played from 11 PM to 3 AM, creating an electric atmosphere while the projection mapping went crazy behind them.

An aspect of digital art that we wanted to highlight was pen plotter art. Pen plotter art is one of the earliest demonstrations of computed creativity, with Vera Molnár, Manfred Mohr, and Harold Cohen being among the first pioneers of this medium. Simplistically speaking, this involves using a computer to draw with a pen mounted on an arm that moves back and forth across a page.

We wanted to highlight pen plotter art for many reasons. Pen plotted art played a significant role in the 1968 exhibition Cybernetic Serendipity. This exhibition was organized at the ICA in London, and due to funding constraints, many of the invites and artworks were created using a pen plotter. This was massively inspiring, and thus, we decided to plot many items for this event. We plotted 10 physical invites, each admitting two people to the event in the evening. These were handed out to anyone we met in person and included the address, time, and admittance details for the evening.

Early on, we formed a partnership with /higher, who offered us funding and support to run the evening. As part of this sponsorship, we were able to offer every attendee a pen plot, as well as three commissioned pen plots which were raffled off to attendees.

The higher pen plot designs were split into two categories. The first category needed to be simple and unique because pen plotting can take a long time. This is fine for more detailed, rarer prints, but we wanted to provide every attendee with a plot they could take home. The final designs we landed on are described below.

The first design, a 4x4 grid of detailed arrows, is AI-generated. ChatGPT was trained to create images with only black lines and a white background. It was then asked to generate one arrow, which was duplicated in a 4x4 grid, resulting in the design we see in the final output.

The second design, displayed in the middle, demonstrates an arrow rotating one degree at a time until it completes a full circle. There are 20 rows of 18 arrows, with each arrow rotating one degree. This design is inspired by contemporary pen plotter artists, particularly Lars Wander, whose work focuses on niche patterns such as the many different ways to fold a rubber band.

Finally, the rarest of all the take-home plots is the arrow that rotates five degrees at a time. This design was plotted about 10 times, but I didn't like the way it looked, so I stopped after the 10th plot and focused more on the 360-degree arrows. However, I decided to include it as it is the rarest of all the take-home plots.

We also wanted to create something special to celebrate higher and for people to take home—something that people would want to hang on their wall, and something they could look at to remember the Summer Solstice of 2024. We decided on AI-generated pen plots of arrows, framed nicely for display. These were created using ChatGPT, and presented three different scenes.

The first scene included New York, symbolizing the community coming together to celebrate the solstice. It featured arrows transcending higher over a mesh of New York buildings, resembling the south of Manhattan.

The second piece focused on an arrow enmeshed in a mathematical grid, appearing as if it was breaking the rules of the grid. This tongue-in-cheek design symbolized working against norms and expected behavior, much like the Farcaster community.

Finally, the third piece featured an arrow with a beautifully detailed pattern, almost emerging from the page. The arrow wasn't pointing directly upwards, but the focal point was the texture and pattern on the arrow. These pieces were a huge success and people loved them! They were displayed prominently as you entered the roof.

Finally, an absolute treat was Pixel Symphony reaching out a couple of evenings before the event, asking to add two plots to the lineup! I received two SVGs of pieces, which were then plotted and framed on the day of the event. These were added to the raffle and handed out to lucky winners to take home! One of these winners was Steen, who runs the Metaliminal channel on Farcaster. He loved the plots and won the exact one he had said he liked when they were displayed on the table. Happy days!

To support the cost of the frames, Pixel Symphony minted three pieces of the generative collection, and we used the proceeds to purchase frames for the pieces!

We wanted to create an environment that was fun, intentional, and educational about generative art. Despite its rise in popularity during the bull market of 2021, generative art still feels like a mystical and unknown genre to many. We aimed to provide an opportunity for people to learn how this medium works. Therefore, we made generative art an integral part of the night, a consistent theme throughout the planning phase.

We started by creating a generative piece intended for people to mint as a ticket. By doing so, they would both collect a piece of onchain generative art and provide funding for the event.

A key part of this piece was to incorporate features common in generative art released via a blockchain. Therefore, this piece includes rarities, traits, and all the code is stored in a smart contract on the base network. We wanted to highlight the strong communities emerging on Farcaster, and thus we created Farcaster-specific palettes and rare traits.

The palettes were all inspired by popular culture and trending memes. For example, there is the Depe palette, inspired by the first memecoin deployed on the Degen L3 network. There's also the Goat palette, inspired by GOAT and all the shenanigans with Enjoy, Degen, and the Goat token. Additionally, we created rare palettes that would change the aesthetic of the piece. They are pictured below.

Highlight is doing a lot to push generative, onchain art forward, so releasing this on Base, on highlight.xyz, made complete sense. We wanted the piece to be in line with the solstice and our event. Therefore, the mint opened 28 days before the solstice and closed 24 hours after the solstice. Our thinking was that people who wanted to support could do so beforehand, and people who attended the event and then wanted to support could also participate. The variations can be found below.

Collect NFT

Overall, 96 editions of this collection were minted. This included two rare mints: one of the Degen persona (#7) and one of the Base persona (#93). The collection has three possible speeds: Moderate (39), Leisurely (16), and Turbo (41). Out of the nine possible palettes, the most common is a tie between Summer Pilates Habit and Grimace Shake.

There are two states for the background: Void and Active. Void means the background has no moving particles, while Active means the opposite. There are also two fill types: Solid and Varied. Solid means the lettering "ONCHAIN SUMMER SOLSTICE 06.21.24" has a constant color, while Varied means the colors differ in a grid style across all the letters.

Finally, the persona trait includes four personas: Degen, Jesse Pollak, Onchain, and Higher. Three out of the four were minted: Degen, Jesse Pollak, and Onchain. The Higher persona is a super rare persona and was not minted at all while the collection was open!

So far, this details the digital aspect of generative art, which, thanks to the blockchain, is now incredibly accessible across the world. However, generative art can also be done in an analog way. My goal was to create something that everyone could do, that was fun, and that demonstrated the principles of generative art without completely blasting the budget to Mars.

After a conversation with a good friend of mine, Sofia Garcia, I discovered that people have used dice rolling in the past as a way to demonstrate the algorithms that power a lot of generative art. Inspired by this, I decided to plan an activity involving a huge 8-foot by 4-foot canvas, a dice, and six paints.

Keeping community at the heart of this event was an important narrative all along. The intention was to create a space to connect over art on the longest day of the year. Community was ingrained into every aspect of this event.

We partnered with FC NY, and working with Karo, we were able to stay focused on this narrative while ensuring we achieved the goal of highlighting art and artists.

We wanted people to feel special upon arriving, so each attendee received a custom handmade bracelet created by Karo in the days leading up to the event.

Another important aspect of community is POAP. POAP has done much to normalize the term and place themselves at every event. We wanted to create a unique experience for people to collect a POAP, so we worked closely and carefully to create something special that reflected the goals of the evening.

We decided to create a booklet with a claimable POAP at the end. This way, people were rewarded for engaging in person with something special on the evening. The booklet also allowed us to draw attention to all of the artists and DJs included in the projection show. Thanks to careful planning, the booklet came out high quality, and the first 100 attendees received one!

We also wanted to create an arts festival atmosphere. To achieve this, we printed 50 posters and wheat-pasted them to the wall on the roof. This was done the night before by Jess, who lives nearby and is on Farcaster as @jessjay!

A big part of this evening was having a barbecue running. We discussed different catering options, such as pizzas and cold foods however decided to settle on the barbecue. In my opinion nothing really signifies the summer and community more than the smell of a barbecue.

We were originally going to run the barbecue ourselves, however we decided against this as we would have many more tasks to do. Therefore, we used taskrabbit to hire a grillmaster called Baruch! He did a great job getting everything ready before people arrived, and providing high quality foods to people throughout the evening!

There were two aspects of funding this event. The first was through ticket sales. The proceeds from the tickets went towards better quality food and more refreshments for the evening. We wanted the ticket to align with the event, so we created a generative art piece and used the proceeds from that to fund the refreshments.

Overall, 96 pieces were minted at 0.005 ETH each, generating 0.48 ETH in funding for the refreshments.

We also approached /higher, Purple DAO, events.xyz, POAP, and Base with requests for support in funding the evening.

Higher supported us with a grant of 0.5 ETH. Purple DAO provided a grant of 0.25 ETH. Events XYZ contributed 200 USDC. Base supported us through their community grants program with 2,000 USDC, and POAP supported us by providing a photographer and creating booklets to commemorate the evening with a POAP.

Overall, we raised 1.23 ETH and 2,200 USDC from the broader community to run the evening and support the local DJs playing on the night.

This evening would not be possible without the incredible Farcaster community across the world and here in New York. Countless people reached out to offer help, and everyone was keen to contribute to making this evening a success! This allowed us to curate and provide a show for incredible artists and DJs, and bring everyone together to begin summer! Form the bottom of my heart I'd like to thank the following (apologies if I miss you!)

Thank you to everyone who financially supported - /higher, Events XYZ, Purple DAO, Base community grants

Thank you to everyone who contributed their time - FC NY, Jess, Karo

Thank you toI Iz for working with me on the physical booklet to commemorate the evening, and the POAP included with it

Thank you to the incredible Onchain Summer Solstice team for keeping the ship on course - Davo, Karo

Thank you to all the artists & DJs who got involved

Thank you to everyone who minted the generative piece! Every cent contributed to the quality of the evening

And thank you to the awesome New York Farcaster community for showing up and throwing down!

The evening felt like it flowed naturally. From 6 PM onwards, the rooftop was open for quiet conversation and painting, accompanied by burgers and beer. As the sun set, the music started to play, and the projections began to show.

There were two parts to the evening: the first, between 6 and 9 PM, created a safe, quiet, and comfortable space for creating and chatting. The second part had more energy and life, providing a space for letting loose and partying while visuals played on the church. For artists, it meant seeing their work in a unique setting, and for dancers, it meant dancing the summer in on a Brooklyn rooftop, surrounded by local artists and online artwork.

A lot of heart and soul went into making this event happen. By partnering with a mix of Farcaster native DAOs and collectives such as Purple and Higher, onchain stalwarts like POAP, and communities like FC NY and Events XYZ, we managed to maintain the original vision and create something special and memorable.

The community turnout was strong, and it was smiles all around. Despite being in an incredibly volatile market, throwing this event on a rooftop after all the craziness of the first six months of 2024 has proven how strong in-person connections continue to be.

Thanks for reading, and if you attended, thanks for coming! All of this has got me thinking - what to do next?

The first topic we touched on was royalties. The topic of royalties is a very difficult one in the present climate. Many marketplaces have made royalty payments optional upon purchase of an NFT, whereas before they were enforced. The number of marketplaces enforcing royalties are unfortunately now the exception, not the rule. Making royalties optional has hurt artists' ability to make a sustainable income, which is one of the huge benefits of using an NFT as a medium for distributing artwork. The power of this in building a creator economy is huge. You only have to browse Farcaster or look at artist royalty wallets to see the number of independent artists who are now making a living thanks to the sale of NFTs and the ongoing royalties from sales. Thinking back to a year ago, this was not a topic of conversation or even on the radar of people who were not involved in web3. So, the fact that they were being talked about during a random conversation is as a huge positive.

We then touched on how the value of NFTs is derived. The perception here was that anybody who creates an NFT is going to make a lot of money. As the conversation moved forward, I was told to create a random NFT, and that would be my ticket to being successful. Their justification for this was the assumption that projects that hold significant value now have always held significant value and were released with the intention of making money. This viewpoint highlights a lot of misconceptions about creating within the space. To me, this highlights the fact that many people outside of the space assume that when a piece was created, it was immediately sold and intended to be sold at a high market valuation. A quote from Exit Through the Gift Shop comes to mind:

"It's not about money, it's never been about money." ~ Banksy

In the relatively short amount of time I have been in the digital creative space, I have experienced quite a lot. When talking about projects with an intention to just make money, the only projects that come to mind are ones that were total rug pulls. The projects that set out with a specific purpose, goal, and vision are the ones that are are still around today, with most forming into companies. If you're reading this and you're like "duh," then great. However, after my conversation last night, it seems outsiders are combining the meaningful projects with the rug pulls, assuming the meaningful projects were created to generate huge income, and this is staining people's perspectives on the space. However, people now know about web3 projects and are talking about them, and no publicity is bad publicity!

As I am writing this, we have some of the most important innovation in the web3 space happening. It genuinely feels like the future is being created right before our eyes. Despite being a global community, it feels local. It feels purpose-driven, goal-oriented, and value-aligned.

To name a few significant events from the past few months: Consensys is suing the SEC, Farcaster is now a worldwide decentralized social media protocol, Gnosis has launched Gnosis Pay, a self-custody bank and decentralized payments network, Highlight has created a platform that empowers creatives and is intuitive to use, and Bright Moments has connected artists and collectors in a way that has never really been done before. The core principles and values of web3 - ownership, security, transparency, and verifiability to name some - are being made accessible to the masses through thoughtful design work and engineering.

Okay, so this has been part rant, part story, but what am I working towards? By this point in the conversation, we had covered royalties and big projects in the space, and the ongoing support for artists creating in the digital arts space, specifically with NFTs. Now to touch on royalties and give some of my opinions here. My experience from the last 24 hours has contributed to these being drafted.

While we were chatting, there was a DJ playing on the television, quite a big DJ playing a Boiler Room set. The music was good. When asked if they would pay 1 cent per listen or play, they immediately said no. Pressing for reasoning on this, it turns out they pay for a Spotify and YouTube premium subscription, and that they are happy to pay for those. These combined currently run at $24 per month.

This got me thinking, how will subscription models work in the future? They need to appeal to the masses and be understandable without a huge gap between what we currently have. Think of the release of the iPhone. Had we just slapped an iPhone 15 in front of people in June 2007, it would've been incomprehensible.

Artists currently get screwed on royalty rates from subscription models; this is something that is widely known thanks to the publicity given to it by significant artists (looking at you, Taylor Swift, for removing all your music from Spotify in protest). So, artists need to get paid fairly for their work being consumed by people. What if the subscription model of the future became credits, almost like a SaaS model? So, if the artist received 1 cent per listen or 1 cent for the whole song listened to (for example, if I only listen to half of the song, the artist receives 0.5 cents), how would this look? A $24 per month subscription would give the listener 2,400 whole-song listens. Artists could configure their rate and charge higher for certain songs - for example, a premium DJ set could be $1 per listen. Then, as a user, I still pay my monthly subscription for access to the service, and once I have used up my credits, I can buy add-on packs or wait a month for them to renew. Some quick research shows that the average person listens to 1.3 million songs in a lifetime. This works out to 14,444 songs per year and 1,203 songs per month.

Taking this a step further, what if the subscription was democratized? From the outside, people see they are paying a subscription. However, the subscription is placed in a pot with all other subscriptions, then this pot is used to support everyone's listening habits. Some people may only listen to 20 songs per month, while others may listen to 3,000 songs per month. The subscriptions would subsidize each other and ideally work out to be equal, or net positive each month. There are some arguments here for an effective/efficient service model to save money on subscriptions, but is that really needed if people are happy to pay a subscription to a platform they only use a few times a month? From my perspective, it comes down to quality. If a platform provides high-quality content, then it is worth the subscription price, even if a couple of months go by with very low media consumption.

All of this culminates in artists being paid more for the work they are creating and contributing to this crazy world we live in. Expecting people to just immediately believe in what we are doing here is difficult; however, shaping what we are doing to suit current trends and moods is something that works. Many apps nowadays do this. This plays to my point earlier about thoughtful design and engineering work contributing to the future of the space. Hopefully, this has been an enjoyable read, and if you made it this far, thank you for reading! Happy building!

This is my brief review after using this app for one week. I will talk about my configuration, the benefits I am experiencing, and whether I think the annual subscription fee is worth paying.

I get distracted very easily. This is not ideal given the type of work that I do, which is engineering-focused. I need to be able to flow into a deep work state. However, my mind wanders, and before I know it, I am sucked into doom scrolling. This interrupts my work state, and without me realizing it, I have spent most of my day consuming content and racking up screen time.

The distractions also take a toll on my mental health. For quite a long period, I was finishing days feeling unfulfilled, with a mountain of tasks that felt unachievable due to my inability to actually work on anything. This is because I would start tasks, then receive notifications from some app that would take me to the app and down a completely irrelevant rabbit hole.

My sleep routine was getting ruined too. I found that in the evenings, I would lie in bed and consume more content, often for hours, keeping my brain turned on and preventing me from falling asleep. This carried over to the mornings, where I would wake up and immediately check my notifications, go onto some app, and then spend a while scrolling or watching videos.

I have noticed how easily I get distracted. Opal displays a page when you visit a website that is blocked, preventing the page from loading. I am realizing that often while I am working, if I encounter a difficult problem, I open a new tab in the browser and type in a default URL. Opal blocks this, which puts me back on track. This is a subtle habit that I hadn't previously noticed until using this app to reduce my screen time.

I am more productive overall. For the first few days, I didn't really notice a huge difference in my productivity. However, from day three onwards, I started to notice the world around me more - the buildings, trees, and daily activity of the city. I also started to notice I had more energy. Waking up became less difficult, and I felt more enthusiastic about walking or cycling somewhere.

My mental health is improving, and I am more rested. Not staring at a phone screen for the final 2 hours before sleeping and not rushing to check notifications and pick my phone up when I wake up means my body is winding down effectively. When I wake up, not having an app to scroll through means I am able to focus on my breathing and think about my day, which is getting me out of bed earlier than before, as I am not getting distracted. Overall, rest and fewer distractions before sleeping and after waking up are improving my overall welfare. I am finding that I have more energy and am a bit happier.

It has only been one week, so this could be a placebo effect. However, my productivity is increasing, and my ability to focus is improving. I think it is easy to get sucked into the 'always on' mode where you believe that if you miss something, the world will crumble. This is generally not the case.

I am finding using Opal net positive. I am able to focus on reading, feel more in control of my time and my days, and am able to enter a deep work state, which I have struggled to achieve for a while. I am also finding that the content I am producing is of higher quality, and my attention to detail is greater.

I am interested to see where I am after one month of using screen time control apps. However, after one week, it is fair to say that I have more energy, am finishing tasks I previously felt were huge, and am finding myself more in the moment than I have been before.

My adjustment period was challenging. I started with five routines: Morning (6am–9am), Work (10am–1pm), Afternoon (2pm–5pm), and Wind Down (9pm–midnight). Initially, the excitement of trying something new made the routines highly effective. However, as the novelty wore off, I found myself slipping back into old habits. Yet, Opal consistently blocked the apps I’d typically scroll through to pass time. Over time, I extended my Wind Down routine to last from midnight until 6am because I noticed that I’d stay up until midnight or wake up early just to use certain apps, leading to doom scrolling on YouTube, Farcaster, Instagram, etc. Extending the routine helped curb this habit.

The apps and sites I choose to block include all social media, news outlets, streaming services, and entertainment apps. Most are blocked by default in Opal, but I’ve added a few extras (like Warpcast—the memes are strong, but my desire to control my attention is stronger!). Overall, it’s a pretty standard, minimal customization for me, and it’s been highly effective.

After extending my routines, I noticed the app became even more impactful. Since my apps are periodically disabled, I’ve adjusted my daily routine to fit this structure. I now know my apps will be accessible at 9am, so I set aside time to respond to notifications then. Similarly, I check in at 1pm, knowing the apps will be re-enabled.

During this period, I also downloaded the OS X app. While it’s useful, I need to be more disciplined with it. Unlike on iOS, OS X doesn’t prevent force-quitting apps, and Opal doesn’t have a headless daemon (to my knowledge), allowing me to quit the app and browse as usual. I’ve taken additional steps to protect my focus, including:

• Editing the /etc/hosts file on my Mac to permanently reroute social media sites to 127.0.0.1

• Installing the Unhook YouTube extension, which removes recommended content, shorts, and distractions, displaying only my subscribed channels. This has revived my interest in content I intentionally subscribed to and has been refreshing.

These measures have been effective for maintaining my focus. I occasionally browse YouTube Shorts on my phone during downtime but don’t find this too disruptive.

Overall, the adjustment period highlighted both my determination to bypass the restrictions I set for myself and the power of technology to keep me aligned with my goals. During this time, my screen time dropped from 5 hours to about 40 minutes per day, and my average daily phone pickups were around 140. Today, my average pick ups are around 90 pick ups.

The main hurdle has been retraining my mind to focus. Before Opal, I was used to consuming content in rapid 10-15 second bursts, which made reading books difficult. I had to rebuild my reading habit, and now I enjoy books consistently. Instead of scattered interactions and countless articles, my evenings are spent reading or spending quality time with others, helping me be more present.

The OS X app remains a challenge, as I frequently bypass it to access social media and YouTube. This is an ongoing issue, and I hope a stronger solution is introduced soon. I deal with it by reducing my laptop screen time and being stricter about what I am on my laptop. As a programmer and technical centric person, this is hard, but I make it work. Simple things like closing the lid while cooking, or not taking it to bed with me. I recently removed my monitors, so now I work from just the one screen of my laptop. This has had a positive impact and I now focus on one thing at a time, rather than multiple monitors with multiple activities. This is all creating room for me to do deep work.

Another hurdle has been managing contacts. Initially, I blocked WhatsApp, but since some of my contacts are only reachable there, I found myself constantly disabling the app to use it, which was frustrating. I now leave WhatsApp accessible and have removed myself from group chats I don’t actively engage in.

With iOS 18.0, Apple released advanced screen time controls, including downtime periods, focus modes, app limits, and web restrictions. While I considered whether to cancel my Opal subscription, I decided to continue using it as it simplifies screen time management, saving me time. iOS 18.0’s suite is robust and offers great features, but I haven’t dedicated the time to fully configure it. Opal abstracts that, and does a good job at it. The average monthly cost is not too high, and I am comfortable paying that.

I highly recommend screen time limiters. Before using Opal, I believed I needed constant connectivity. I thought, “I don’t use Instagram or Warpcast that much,” but in reality, I was using them frequently and without structure. Since using screen time blockers, I’ve adopted the mantra, “Sunlight before screen light,” and it has had a positive impact.

As an independent remote worker, having a structured routine is essential, and these apps help provide that. My time on social apps is now more intentional. When my social apps are accessible, I engage thoughtfully, often by commenting, something I never did before.

In the evenings, I now enjoy activities like cooking, watching a movie, or reading. Falling asleep is easier, my mind is more at ease, and my thoughts feel less scattered. I’ve also noticed a contrast when spending time with people who don’t use screen time controls; phone use is so normalized that we sometimes forget to truly connect with those around us. I hope for a future where we spend less time on devices and more time connecting in person.

So, to conclude: screen time control apps are worth it! I’ll continue using Opal for the foreseeable future—give it a try and see what happens!

Invader is an artist from Paris who places mosaic murals in different cities around the world. There are various websites that document this artwork. One of the websites, invader-spotter.art provides invaluable data about his work. There are articles, journal posts, and thousands of pages with pictures, status' and names for each Invader piece. Invader Spotter has a research database, where Invaders can be searched for based on their traits. There is also detailed documentation, with pictures, about the last reported status, and numbering of Invaders in different cities. It is a behemoth of information, offering data about almost any Invader.

I have been lucky enough to be able to travel to a few of the cities containing murals. For each city I have been to, I have researched Invader locations. This involves Google, Instagram, The Web Archive, Flickr and adhoc websites. It's fascinating to look back at the history of the Space Invaders website (https://www.space-invaders.com/home/) using the time machine. For a while, When a city was invaded, a map of each Invader location was released. These maps were added to space-invaders.com - this is all viewable on the web archive! The maps are the first bit of data I look for when looking for Invaders in a new city. There are also SlideShares which offer insight into Invaders in cities. A collective, or individual, creates these and sometimes links to google MyMaps with the latest locations.

This is great, and has allowed me to create a MyMaps for Invaders world wide, which can be found here:

https://www.google.com/maps/d/u/1/edit?mid=1qE51pu_PH7xYWxQg0SsY02NeGAu050Q&ll=39.0988217957248%2C-30.863683216200005&z=4

https://web.archive.org

space invaders > the story of an invasion

space invaders invade the world >>> the story of an invasion by an anonymous artist called space invader.

https://www.slideshare.net

Space Invader Geneva

Space Invader Geneva - Download as a PDF or view online for free

There are blog posts, Instagram, and Flickr pictures of Invaders. Once all of this information has been collected, it becomes possible to build up a plan for a day of discovering Invaders.

All of this has got me thinking. There is so much information out there about Invaders, but it's all over the place! To find all the Invaders in in one city, in an efficient way, you need to visit many websites, and collect all of the information. Alongside this, there are events, and articles, and YouTube videos (exit through the gift shop for example) which are so interesting and contribute to the artwork out there for Invader. For example, one of the Invaders in Geneva was really hard to find because it blended in with a bridge (yep looking at you GNV_28).

The intention here is not to centralise all information about Invaders. Having information spread out all over the internet is self-documenting in itself. It also adds to the excitement of hunting Invaders, as it requires first hunting the web for locations and status updates. The intention here is to build a system that pulls together all the documentation that is currently in different places around the web and provides pointers and an aggregated layer of said data.

What if there was one place to go to get all Invader content? What if people could be rewarded for positively contributing images, status', locations? What if there was an incentive structure, on a decentralised database, that anyone could contribute to, and be rewarded for doing so?

I would like to propose a system that allows for the documentation of Invaders. I believe that a blockchain and smart contract is a good approach for building this system due to the decentralised, and open nature of data on there and the financial requirement for contributing data.

This system should:

• Allow people to retroactively upload images, descriptions and locations of street art, using data that is already available online

  • There is so much information online already, but it is spread all over the place. For example, many Invaders can be found on Instagram and Flickr simply by searching the Invader number. For Flickr, it's the same. Googling PA_35, for example, reveals a link to Flickr straight away.

  • The goal here is to allow people to research and collate this information online, then upload it to the platform and document it. This would be helpful for others, who would like to find everything out about a particular Invader piece.

• Reward positive, constructive contributions, and continued contribution

  • Positive, constructive contributions are fundamental to a platforms longevity and success. Recently, I ran a competition on Farcaster (ref: https://warpcast.com/~/channel/invaders), which encouraged people to upload their own images of Invaders in return for DEGEN tips. This was a success, and as a result many community members uploaded high quality and interesting images. To this point, farmers can get involved. This is a constant battle for the Degen team, who recently disqualified many Airdrop farmers from their Season 2 Airdrop (ref: https://degen.tips).

  • The Ethereum network utilises two standards for NFTs today. ERC-721 and ERC-1155. By using this, people who contribute get ownership of their contribution. The ERC-721 itself will have a few unique properties. The first being an array of images. When a user uploads for the first time this array would just be the first image. As a subsequent contributor, I can decide whether I want to create a discovery, or contribute to an existing discovery. The latter would cost less to do, however the former would allow for ownership of a discovery.

  • This system works for a few reasons...

    • People can display that they found a certain number of Invaders on any platform that allows for displaying NFTs

    • These are stored in the contract, and can be sorted through based on data related to them (status, city, name, number etc)

    • Once created, they are, to a certain extent immutable. Perhaps you are on a hunt, and you discover an Invader. If you document it immediately, then the exact time you discovered that Invader is preserved on the blockchain as you create the NFT at the time of discovery.

    • The opportunity for Token Gating can be introduced, requiring people who have over 1,000 discoveries and so on. This is tightly couple to the community aspect, and ensuring that the content uploaded is of high quality and is a positive contribution. I outline a voting system below.

  • There is also an opportunity for a points system utilising ERC-20 tokens. However, this needs thorough thought. Vitalik discussed community tokens being used to fund causes. There is potential here. The tokenomics would probably shape around the way the platform is adopted by the community. (ref: Vitalik article https://bitcoinmagazine.com/markets/markets-institutions-currencies-new-method-social-incentivization-1389412608)

• Allow for voting, reporting and discussion about contributions

  • By allowing for voting, and contribution discussion, the system can be self-governing. This may require some hand-holding in during the first few cycles of the release, just like many decentralised platforms. However, once traction is gained, the system will govern itself. This can be seed on Reddit, Farcaster, and many other platforms which have moderator roles for communities.

  • There should be a system for recording votes. This should be tied to a public wallet address, and record the sentiment of the vote. These votes should be publicly accessibly. A vote is used here to describe a few things:

    • Status of an Invader (OK, Damaged, Destroyed)

    • The intent of a contribution (negative/positive)

  • When a vote is changed, there is a record of this. The blockchain can be used to discover all votes placed by wallets, or all votes placed on a particular Invader, or the vote history of a specific city.

• Allow for the storing of metadata related to a certain Invader or city

  • As described before, there is a wealth of information available including maps, journals, and articles. These are all on servers distributed far and wide through the world wide web. There should be a way for a user to upload a metadata item and link it to an Invader, specific city, or keep it completely loose and have it appear in a contributions list.

  • This data should be stored in the contract and accessible to clients through read functions. Ideally, if there items should utilise IPFS. The goal here is to have pointers to all information in one, accessible place.

Of course far more needs to be spec'd out, however I am in the phase of building now. I have so far achieved two things:

• Deployed a contract to the DEGEN Layer 3 Chain (explorer: https://explorer.degen.tips/) and experienced the incredibly low gas cost to do so

• Created a basic contract, and collaborating with a friend to introduce a lot of the features discussed here. It continues to be worked on and, once a testnet version is released, I will share a progress update

I can be found on Farcaster as @henry or in the Invaders channel /invaders

Farcaster

henry on Warpcast

my first DEGEN contract deployment 0.00001411856 DEGEN gas costs...and able to use an existing solidity smart contract this changes the game https://explorer.degen.tips/tx/0x3a9dda99a2f4bc84643137f755db47ecfca0b8eb71c748a884b21942ee014406 https://zora.co/collect/base:0x8eb3fa90d7b9bed575f90c804346f8d269886a85/1

I enjoy using SAFE, a product offered by Gnosis, for creating and managing multisigs. This blog post offers a step by step guide for creating and interacting with a multisig crated via SAFE on the Base L2 Network.

There are a few concepts we need to tackle first in order to get set up.

• Make sure you have a wallet provider set up on your desktop wallet client with a Base RPC. The official base docs have a great set up guide which you can read here and an explainer on what an RPC is can be found here

• Bridge some Layer 1 Ethereum to Base. You can do this on the official base bridge: https://bridge.base.org/

• Switch to the Base RPC by selecting the new network within your wallet provider. Everything we do in this guide will be done on the base network

Okay now we can get started. Head to app.safe.global. On the main interface you will see the option to connect your wallet - go ahead and connect the wallet you wish to use to set up your SAFE.

After connecting, click Create Account. From here you will see the Create New Safe Account interface. On the left hand side, you can name your safe. "This name is stored locally and will never be shared with us or any third parties". (quote from app.safe.global)

On the right hand side, you see 'Your Safe Account preview'. The preview has a couple of important things. Make sure that the wallet is the wallet you would like to use to set up the safe, and the network is Base. If you don't see either of these, you may need to switch connected wallet, or select Base as your RPC URL in your wallet provider. If you have not done this yet, then see the Setup section of this post for information on how to set up your RPC URL.

Once you are happy with all the information here click the next button. The following page is information about your safe signers. The setup on this page is important and if misconfigured you may not be able to access assets in your multisig.

Before doing anything, check that the Wallet, Network and Name in the 'Your Safe Account preview' section are what you expect. They should match the content entered in the previous screen.

Now, on to the signers. A safe uses the concept of Transactions that multiple signers all sign in order to execute. As explained before, a multisig wallet works by requiring two or more private keys to execute a transaction. Whereas in a self-custody wallet you can just send an asset directly, in a multisig all signers must approve the transaction before the asset can be sent. I have included a diagram that demonstrates this below.

Hopefully this diagram conveys how multisig wallets improve overall security. It's important to note that any signer can reject a transaction. Doing so, will require a gas fee, as it cancels the transaction onchain. There is a great guide on this:

https://help.safe.global

Cancel Transactions | Safe{Wallet} Help Center and Support.

In case you change your mind and want to not execute a transaction after all, there is a way to cancel pending transactions. However, the only way to effectively prevent already (partially) signed transaction to not be executed in the future, an "empty" transaction has to be executed at the same nonce.

Now, find the public addresses of the wallets you wish to use as a signer. A good practice here is to use one hot wallet (i.e. Rainbow, MetaMask, etc) and one cold wallet (i.e. Trezor, Ledger). Next, we can configure the threshold.

Now, let's set up the signer threshold. For this safe, we are setting it up to be 2 out of 3 signers. This means in order to perform any actions on assets in my safe, 2 signatures will be needed. One from a hot wallet, and another from a cold wallet (as the 3 signers added are a hot wallet and 2 ledgers).

Some fun information on this page - you can create an address book for wallets you regularly send to! And the address book is stored only on your local storage, and not shared with third parties.

Okay great, on the next screen we will review all of our setup information before proceeding to interact with the SAFE smart contract and create our multisig.

Once you have review all the information here and it looks accurate, hit Create. This will trigger a transaction and allow you to actually deploy your SAFE smart contract. Because this is Base, the gas fees are extremely reasonable. A great benefit of doing this on an L2!

Congratulations! you just created your first multisig safe! Now before we transfer any valuable assets into it, let's check everything works with the signers we sent to it. To do this, we are going to send a small amount of Base ETH into the SAFE, then send it out again. An important concept to note here:

• Gas is never paid from your SAFE. Instead, it is paid by the final signer that you use. Good practice here may be to keep one of your signers topped up with gas fee ETH, and use that as your final signer every time

Alright alright alright let's get started. First we need to locate the public address of your safe. This is the address we will be sending the Base ETH to. We can get this from the top left of your SAFE.

Click this to copy the address to your clipboard. Now, head to a wallet you have access to send transactions on, and we are going to send a small amount of ETH from it, to the address we just copied. Note: the address will be prefixed with 'base' so you may need to remove this before sending. Here's the error that MetaMask will give you if this is the case:

You can change this in your safe settings.

Okay great, now that you have some assets in your SAFE, we can send them out again. To do this, select New Transaction. Once you have done this, you will be presented with the following screen. As you can see, the SAFE UI has a transaction builder. This is great when sending multiple assets. The SAFE documentation is detailed and great at explaining these concepts.

https://help.safe.global

Transaction Builder | Safe{Wallet} Help Center and Support.

Create a batched transaction with the Transaction Builder Safe App.

Select Send Tokens, and input an address you have access to (or your frens address if you're feeling generous!)

Once you have happy with all the information here, hit Next. You will be presented with the Confirm transaction screen. Some important things to check:

• to(address): clicking the external link icon will take you to Basescan so you can verify the address you are sending assets to

• value(uint256): the WEI value of your transaction (if sending ether). You can convert this to ETH using a converter.

http://eth-converter.com

Ethereum Unit Converter

After constantly using a calculator to convert between Ether units i thought it would neat to implement the EthereumJS-Units library and BigNumber library on a website for everyone to use. There is a simple and full converter page, since the majority wouldn't bother for any other units than Ether, Gwei and Wei.

Once you are happy with this, you can sign to create the transaction. Note that the Transaction status has 'Confirmed (0 of 2)' meaning that your 2 signers are being recognised, and are required to execute this action onchain. Let's sign this transaction.

Great! Now we can see that our Transactions tab has 1 pending transaction.

Now, let's head over to our Transactions tab and see what's waiting for us there.

We can see in the screenshot above that there is one pending transaction for a send. It has one confirmation, and needs to be signed by one other signer. So let's connect the second wallet we added as a signer, and execute this transaction. We can do this by selecting the wallet from our wallet provider and connecting it to the SAFE website. If you connect with the correct wallet, you will see a popup telling you there is one transaction that needs to be confirmed.

We can see that the interface has changed and there is now a 'Confirm' button.

Clicking confirm will take us again to the confirm transaction screen, only this time it will look slightly different. This is because we are executing the transaction. Before, we signed the transaction and created it. This time, we are applying it to the blockchain. This is why we need to pay gas. This is the final step before the assets leave your multisig. Check everything here!

Just as with signing, we can double check the transaction details by selecting the accordion.

Once you are happy with all the information in this screen, click Execute. This will trigger an onchain transaction requiring gas. You will see a similar transaction screen to before, just like when we were setting up the safe!

If you wish, you can view the Transaction after completion.

And that's it! Congratulations, you just set up a multisig using SAFE on the Base Layer 2 chain!

The same concept we just tackled for sending ETH is mirrored for sending NFTs and ERC-20s. There are fantastic guides on the official SAFE docs which go in depth on how to do this

https://help.safe.global

Send funds | Safe{Wallet} Help Center and Support.

How to send ETH and ERC-20 tokens from your Safe.

https://help.safe.global

How to manage CryptoPunks with Safe | Safe{Wallet} Help Center and Support.

How to store and transfer your punk.

https://help.safe.global

Supported asset types | Safe{Wallet} Help Center and Support.

This article provides an overview of the different types of assets that can be managed with your Safe.

Thank you for reading. If any concepts are unclear, or you have any feedback please reach out! Would love to write more educational articles on different concepts within web3.

Have a nice day!

Upon my arrival in Paris, a couple of weeks prior to the the show, I couldn't help but feel enamoured by the city. Paris, perhaps, stands as one of my cherished European destinations. The amalgamation of exquisite cuisine, pedestrian-friendly streets, and fine wine renders the city remarkably cozy and delightful to me. This early arrival afforded me the luxury of unwinding, recuperating, and acquainting myself with the city's charms before the unfolding of the show – an ideal prelude indeed.

Bright Moments embarked on its journey in Venice Beach, pioneering the minting of CryptoVenetians for the local community. The only caveat? Personal attendance was required to mint one. Subsequently, the team headed to New York, delving into generative art. It was during the New York show that Tyler Hobbs released "Incomplete Control", and Grailers DAO was started. Following this, the art circus headed to Berlin. The Berlin chapter, NFT Art Berlin, showcased the works of 10 artists, each commanding Kraftwerk for their exhibition. It was in Berlin that I joined the team. Departing from Berlin, we headed to London. Six artists unveiled their works, each hosting individual exhibitions in Soho Square over a 1 month period. The minting experience for Londoners utilised a tiny screen, accompanied by generative piano melodies. Next: Mexico, NFT Art CDMX. Eleven artists presented their works, each allocated their own room for displaying their pieces in the heart of Mexico City. Following CDMX, Tokyo, with 30 artists releasing work some in Asakura House and some in a skyscraper – Digital Garage. Following Tokyo, Buenos Aires, where 10 artists exhibited their works within a palace in Buenos Aires, followed by a collectors' expedition to Patagonia.

The rationale behind my detailed description of each city stems from my firm belief that without their unique characteristics, participants, and settings, the experience in Paris would not have been as exceptional. Each preceding city involved a lot of standing, with minimal communal seating. Berlin unfolded within the vast confines of a power plant, while London spanned across two storefront locations. Mexico City boasted a sprawling venue featuring a restaurant, yet the art display areas primarily involved standing.

So, over the last couple of years a community of people has formed in each city. Everyone has been gathering in one place, to celebrate generative, and more recently, AI art. Paris was an art cafe. The reveal experience involved sitting at a table with the artist, and revealing your piece sat with them. This made the experience extremely intimate. Being able to actually sit down and enjoy your piece, ask questions about it, and enjoy it with friends and the artist, made a huge difference to the enjoyment. Each piece holds a story. The cafe extends beyond simply enjoying the piece with the artist. Throughout the 4 days, the seating was used for various different things. On day 1, a couple of people arrived at midday, ordered a coffee, and sit in the cafe surrounded by art, chatting away. As the evening drew in, people used the chairs to relax and chat. On day 2, the same happened. On day 3, there was an unofficial Farcaster meetup held in the cafe. People gathered at the far end of the cafe, by the huge reveal screen, and chatted about all things Farcaster. On day 4, the same with an unofficial Grailers meetup. On all days, as the evenings drew in, people sat in the chairs chatting away and just spending time. Every time I went into the cafe I was grabbed by someone who I hadn't seen yet and was able to say hello to old and new faces! It was perfect.

I firmly believe that this experience wouldn't have been as transformative had it been held in an earlier Bright Moments city. The preceding events laid the groundwork for fostering camaraderie and familiarity, imbuing the Parisian cafe setting with an unparalleled sense of intimacy, personal connection, and enjoyment.

The opportunity to sit down with artists and engage in discussions about their work served as a catalyst for this immersive experience. Spectators were welcomed to observe, contributing to the creation of a genuine community ambiance centred around the artists and their creations. I vividly recall a moment when I entered the venue to find my friends gathered around Luke Shannon's table. Joining them, we quickly begun to discuss his work and play with the algorithm. It was truly special.

To conclude, Paris earns a perfect score of 10/10 for its rich cultural tapestry, sense of community, and vibrant atmosphere. This brief reflection encapsulates why I deem the cafe a resounding success. Thank you for taking the time to read!

So, this morning I woke up and checked the Farcaster feed. There were a lot of posts in frames and some of the posts had embeds with buttons...what?! The first frame on my feed was by @w1nt3r and it was just a click away to see how high the counter would go. At this point the concept of frames totally clicked and I was hooked. Pardon the pun.

The first thing to appreciate is how straightforward the onboarding process for building a frame is. The documentation is well written, and a near step-by-step guide with example repository can be found in the notion here: https://warpcast.notion.site/Farcaster-Frames-4bd47fe97dc74a42a48d3a234636d8c5

The notion also provides a template git repository for a poll frame application. It's written in NextJS and can be hooked straight up to Vercel. All that needs to be configured is the HOST env variable and the Vercel KV Redis database. Coincidentally, before today I did not know about the Vercel KV db. It looks pretty powerful and useful. Definitely need to look into it further. But that's for another day, and another post.

The GitHub also provides template code for reading Farcaster IDs. So I got to work ripping apart and understanding the template GitHub. The first thing I wanted was to get rid of the 'poll' element. I just wanted users to see a counter, similar to @w1nt3rs embed. Everything in the template is in TypeScript so it was super easy to see the types being expected everywhere. Refactoring the code so it would just be a title and a counter took a couple of hours, but meant I got into all the open graph features, and data fetching / storage features. The main refactoring took place in the actions.tsx class. In here, I created an object that just stored the name of the 'hit' that shows. Here's the updated code:

export async function saveNumHits(numhits: NumHits, formData: FormData) {
  let newNumHits = {
    ...numhits,
    created_at: Date.now(),
    title: formData.get("title") as string,
  };

  await kv.hset(`num_hits:${numhits.id}`, newNumHits);
  await kv.zadd("num_hits_by_date", {
    score: Number(numhits.created_at),
    member: newNumHits.id,
  });

  revalidatePath("/num_hits");
  redirect(`/num_hits/${numhits.id}`);
}

export async function giveHit(numHits: NumHits) {
  await kv.hincrby(`num_hits:${numHits.id}`, `numHits`, 1);

  revalidatePath(`/num_hits/${numHits.id}`);
  redirect(`/num_hits/${numHits.id}?results=true`);
}

export async function redirectToNumHits() {
  redirect("/num_hits");
}

I retained the created_at logic and the title logic. This proved pretty helpful as I was able to create and view all previous test runs in one view, and see previous mistakes.

Finally, I checked out the /api/vote endpoint and made some changes. The main change was removing all the optionality that would be there for a poll. Instead, it would just be tracking integers and Farcaster IDs. The key logic I wanted to keep: If someone had already taken their free vibe, they could not take another free vibe. The code for that can be seen below:

      const fid = validatedMessage?.data?.fid || 0;
      const alreadyVoted = await kv.get(`num_hits:${numHitsId}:votes:${fid}`);
      voted = voted || !!alreadyVoted;

      if (!results && !voted) {
        let multi = kv.multi();
        multi.hincrby(`num_hits:${numHitsId}`, `numHits`, 1);
        multi.set(`num_hits:${numHitsId}:votes:${fid}`, true);
        await multi.exec();
      }

Above, the code assumes validateMessage is successfully storing the FID of the person interacting. Then it checks the @vercel/kv store for the key of that particular number of hits frame, and the FID of the person interacting. If the caster has already voted, then it does not apply the hit. If the caster has not already voted, then the hit is applied and the key/value pair for that caster is updated to true. Nice! Working on this bit was interesting to learn about - two new concepts for me...retrieving and using an FID (more on that later) and using @vercel/kv to store data.

Finally, I was ready to do some UI work. The template app generates dynamic HTML to generate the OpenGraph hero. So, I chose a free font, chose the greenest background possible, found a purple colour palette I liked and set everything up.

I toyed around with a few ideas - gm was the first, then the thumbnail would say something along the lines of "times gm'd". Eventually I landed on free vibe. Pretty happy with the choice.

Alright grand! I posted this on frames, then started looking at what other people had been working on. Man, people are using frames for really interesting things. @df compiled a list of all the frames as of a certain time. It can be found here: https://warpcast.com/df/0x90ef6189

Here are some of my favourites...

The onceupon.gg frame stood out to me for potential use cases. Recently, we released a profile page on the Bright Moments portal. The main purpose of the profile page is to track progress to certain goals within the BM ecosystem. A Frame is a perfect place to plonk this data! There can be a default image for no citizens, then use the FID of the person interacting to check which Crypto Citizens they are holding, using the API we built for the profile page.

These are the technical challenges I faced:

• Resolving an FID to the connected wallet address. New to this, don't be too mean!

• Retrieving potentially large numbers of citizen images. Some wallets multiple hundreds of CryptoCitizens.

• Displaying all citizen images in a way that makes it clear and doesn't look...awful

Okay! Great challenges to have. The first challenge I solved was the resolving an FID to a wallet address. This was the easiest by far and I used searchcaster.xyz to achieve my goal. Searchcaster provides a very simple to use API for finding information about FIDs. Here is the API request for retrieving account information for the account with FID 1: https://searchcaster.xyz/api/profiles?fid=1

{
    "body": {
        "id": 1,
        "address": "0x8773442740C17C9d0F0B87022c722F9a136206eD",
        "username": "farcaster",
        "displayName": "Farcaster",
        "bio": "A sufficiently decentralized social network. farcaster.xyz",
        "followers": 1435,
        "following": 2,
        "avatarUrl": "https://i.imgur.com/I2rEbPF.png",
        "isVerifiedAvatar": false,
        "registeredAt": 1628882889000
    },
    "connectedAddress": "0x86924c37a93734e8611eb081238928a9d18a63c0",
    "connectedAddresses": [
        "0x86924c37a93734e8611eb081238928a9d18a63c0"
    ]
}

Okay, next challenge - image data. Originally, I wanted all citizens to appear and take up the entire tile. This went great for wallets with fewer than 50 citizens. However, 50 and above, the requests just intermittently timed out and sometimes the Open Graph image didn't load. Not ideal. Clearly, displaying all citizens was not a feasible option.

I changed tact and decided to display just full sets. Similar to the full set tracker on the Bright Moments profile which makes uncollected backgrounds slightly higher in opacity

This worked a treat and sort of killed two birds with one stone. I was able to parse large amounts of data, taking just one citizen with each background, and display everything well. I tried with some of the higher collectors - people with 600+ citizens, and it worked a charm. Farcaster have provided a great tool for testing frames. It can be found here: https://warpcast.com/~/developers/embeds

A drawback I found in testing with this tool is that it didn't consider my currently connected FID. I may have been misusing the tool, but it took me a good few minutes to track that one down. One nuance here, is that the /api/image endpoint, which generates the final OpenGraph SVG, can not read the state of the request. Instead, the /api/vote endpoint is posted to by the frame itself. The code makes this pretty clear...

  const fcMetadata: Record<string, string> = {
    "fc:frame": "vNext",
    "fc:frame:post_url": `${process.env["HOST"]}/api/generate`,
    "fc:frame:image": `${process.env["HOST"]}/api/image`,
    "fx:frame:button1": "View Citizens",
  };

I admit, I did initially try to fetch the FID in the image route. It didn't work. But glad I tried. I did some tidy up, and made sure the default image state looked polished and reflected how the profile looks on Bright Moments. I ensured the default states worked for wallets with many citizens, no citizens and a few citizens - success! The final result is something that I am proud of.

The image is generated on this URL: https://frame.brightmoments.io/api/image?fid=<FID>

I hope you have fun playing around with this, I definitely did while building it! Overall, a productive Saturday spent learning about a new technical concept and producing something useful & fun. Have a nice weekend!

I signed up for the basic tier Midjourney subscription - $10 per month, allowing me to use 3.3 hours of GPU generation per month. I then hopped into the Midjourney discord and used the newbies channel to generate an output with a prompt:

2024 arriving, walking in through a door and kicking the door down, creating dust and explosions in the process

All of this took about 10 minutes to do, and I had a set of 4 images to choose from.

I then chose to upscale image 2 (by selecting U2), then asking for variants on the upscaled version until I received an image I wanted to use.

All of this took about 20 minutes. My subscription plan allows for stealth mode generations via direct messages, which is something I will for sure use in the future. I was quite nervous at the idea of generating images in the public channel! Plus, it was difficult to track my generation as the feed was full of all the public generated images.

Once I had generated an image I liked, I created several compressed variants for the collection thumbnail using compresspng and generated one upscaled (4x) version for the main collection image. I then saved them, and moved to the Zora create flow.

I am pretty excited by both the Base and Zora chains. They offer really affordable contract deployment gas fees. I like that they are bridge-based too, and both chains offer simple bridging options. Zora have built the bridging into their website. It's straightforward to use, understand and access.

The next step is actually adding the Zora network to your wallet. Again, Zora has this covered. I always advise to double check every on-chain action you make even if you are using a throwaway or hot wallet. There is a support article here about how to add the Zora RPC to your wallet client. In a few clicks, I was up and running and had the Zora network on my wallet client. Next step - minting. Zora has a '+' on their website, which is used for creating on the Zora platform. It takes you straight to the create page.

I think this page is designed really well. As a user, I was provided only with the information needed to mint a token. In a few clicks and keyboard taps, I had created my collection, set the price, and uploaded my imagination. The choice of words here is interesting too, making the creation process a bit more personable and exciting. Adding the mint duration up front is thoughtful as well, and provides a good amount of optionality for the mint duration of the collection.

There are a few advanced options as well. I didn't use these, but can see how they would be helpful in the future. When I release major art projects I like to use wallet splits so I can donate primary revenue to relevant / important causes and charities.

I filled out all of the fields required for my mint, and hit create. Zora have their own chain explorer which is great. The deployment cost for this new collection was 0.00004539 ETH. Reasonable! Here's the deploy transaction and here is the token contract

Zora have made an accessible, fun-to-use token deployment flow.

Using Midjourney and Zora, I was able to create a fun open edition mint for the start of the New Year in about 1 hour, and not pay an exorbitant amount of ETH to deploy. This was a great exercise, and something I may use in the future.

The AI art landscape is exciting. The boundaries of computed imagination are being pushed every day. Artists are developing their own styles, using computers as a tool for doing so. Computers are becoming adept at creating imagined works by using words and sentences from humans. All of this combines to create fun, accessible opportunities for artists to share computed art with the world.

It feels like AI generated imagery is emerging from a previously technologically heavy concept to an accessible art medium that everybody can enjoy. This compliments how far web3 has come in the last few years. Both Midjourney and Zora are demonstrating how technology can be accessible and usable by people who have little-to-no technical experience in order to share messages with the world by providing understandable on-ramps without an over saturation of information.

Thanks for reading!

For the last few years, I've made a pilgrimage to Old Trafford on Boxing Day to catch our home fixture. Last year we played freshly promoted Nottingham Forest. An easy win, 3-0. This year we faced Aston Villa. Aston Villa are having a pretty insane season. Their history in the Premier League is varied, having finished mostly between 15th and 20th, interspersed with a couple of years of finishing between 5th and 7th. This year, they are sitting in 3rd place and having a pretty great season, it is fair to say! They managed to bag Emiliano Martínez, the Argentine goal keeper who won the world cup, and he's doing a pretty great job for them. I had a mixed bag of emotions - excitement to see my team play, and the nerves of facing the team that are currently 3rd in the league, knowing they would be 2nd only on goal difference if they won. Here's how the evening went.

It started with a drive through the moors, down the notorious Snake Pass between Sheffield and Manchester. Shortly before making the trip, I found out it was once voted the third most treacherous drive in the world. Interesting. There were three of us in the car, and on the way we were eagerly discussing the game. This drive reflected my feelings about the game. Pretty scary winding down from the moors, feeling pretty sick from the constant change of direction that the turns provided, and excited when the city of Manchester came into view. Onward!

We arrived at the stadium about 3 hours before the game started. I love arriving early, as there are only a few crowds outside the stadium, and it's easy to walk around and see things like the Munich Air Disaster Clock. We headed down the Sir Matt Busby Way to a fella who has been selling the teams merch for longer than I can remember. He has the first stand you come to as you head away from Old Trafford. I don't know his name but he always has a smile on his face and a crowd around the stand. I decided to grab a David Beckham scarf, owing to the documentary released this year, and him being one of my favourite United players, and a Sir Bobby Charlton scarf, as a homage to one of the greatest and most dedicated Manchester United players of all time, who unfortunately passed away this year. Here's a touching video of the silence observed, and reef being laid, shortly after his passing was made public. Here's a pretty accurate comment that sums up Sir Bobby.

He survived Munich. He saw close friends die at Munich. He brought Manchester United back to the top after Munich. Then, he won European Cup for United. He won the World Cup for England, too. Dignified, respected, legendary. He’s England’s finest. RIP Sir Bobby Charlton.

As 8pm crept closer, we joined the pilgrimage from the Cricket Ground to Old Trafford and the magic feeling started to creep in. The pubs started to empty out, and the chants began. Outside the stadium was a mixture of a low hum, and excitedly nervous atmosphere. Nobody really knew what to expect. The last couple of games United played at home, we got a spanking. Both ending 3-0, one to Manchester City and another to Bournemouth. This game could go any way - score predictions were all over the place, I was pretty set on it being 2-1 to Aston Villa.

Queues outside the main stalls were forming. There were a lot of grim looking faces, but the Manchester spirit shone through and people were starting chants left right and centre. As I got up to the security and was being searched, I held my hand out over the railing and had it promptly high fived by a passing supporter. I turned around and the cheeky chappy gave me a smile. Which in turn caused me to smile - smiles are infectious, hey?!

We hustled in to the stadium without a hitch and headed up the concrete staircase. I love this part of the evening. From inside the stand, you can't hear the sounds of Old Trafford (unless there's a goal, or a chant). So it's possible to stand at the bar and be completely unaware that you're stood in the largest capacity stadium in the Premier League. We got into the foyer, and after a short toilet break, headed straight up to our seats. Again, I love this part. As we headed up, the sounds of the stadium were slowly increasing in volume. At the top of the stairs, there's a small door, about the size of a fire exit. No bigger than my bedroom door. This is the part of the evening that feels like stepping through the wardrobe to Narnia. Like transcending from one dimension to another dimension. I cannot describe the feeling of climbing that staircase, with the sounds of excited Manchester United supporters slowly getting louder. Stepping through that doorway, suddenly 74,000 fans are on the other side, every single person facing the pitch, and every single person chanting. Incredible. Like nothing else on earth. We were all wide-eyed, quiet, and stunned. This happens every time. It never gets old.

The game begins, and Aston Villa are off to a pretty good start. They were all over United. They were working the ball well, showing energy, teamwork and spirit. All the while United were looking pretty lethargic and tired....Uniteds forward, Rashford was left in heaps of space and no one ever played him the ball, resulting in him looking increasingly annoyed. About twenty minutes passed, and Villa are awarded a free kick. Quite far out, doesn't look too threatening. The mood in the stadium went quiet, and Villa lined up to take the free kick. McGinn lined up to take it, the United penalty box was filled with light blue...McGinn took a couple steps back, ran up to the ball and punted it over the wall. The ball curved round, hit the ground, made a bounce, there was a bit of a scuffle and.... 1-0. Dead silence from United fans. We all sat with jaws dropped. But, we are Manchester United! A one goal deficit? A walk in park, not a problem for the reds. We're used to being down a goal and coming back. 90 minutes is a long, long time!

5 minutes after play was resumed, Villa were awarded a corner. I got a sinking feeling in my stomach...Villa looked quite strong on their set pieces. Again, the stadium went quiet as Villa stepped up to take their corner kick. Players gathered in the penalty area again, McGinn stepped up to take the corner. He took a few steps back, then punted the ball over the players in the box. It hit a player on the outside who punted it square back in, Dendoncker got a back heel to it...2-0. Dead silence. Dead, dead silence. 26 minutes in and we're 2-0 down. It's fair to say at this point, we were all a bit upset. We all sat in our seats, looking at the pitch with glum faces. The Villa fans were having an absolute party - two goals at Old Trafford! They all looked like they'd had a few too many red bulls and were in a club. Their chants filled the stadium. This was a pretty heinous feeling.

At this point, I'd like to introduce the concept of a 'Manchester United Classic'. We have a habit of making a comeback. This is why we're never too worried about losing. We always come back. Below are a couple of my favourite comebacks in Manchester United history.

So, we're sat there in Old Trafford. Manchester United are 2-0 down, the stadium is full of gleeful Aston Villa fans chanting away. Seriously, the couple of thousand supporters were filling the stadium with their chants. It was cute. What do we do in the light of this? We start singing - first 'Glory Glory Man United' and then 'United United United'. The whole stadium, singing in unison. Pretty special.

Half time comes around, we head inside to get some hot chocolates and chocolate buttons, as it's getting quite cold. We took our time, the highlights were on the TV's and we were huddled around chatting about the first half. There was a bit of consolation going on, but the mood was definitely not dampened. People were cracking jokes and chatting, and spirits were quite high. It got close to the second half and people started heading back to their seats. We hung back and decided to wait in the warmth.

The second half started, and given the drop in the crowds I decided to head to the toilet. Not 30 seconds passed and I heard a massive roar from the crowd. What on earth...I ran out front, dumped my hot chocolate in the bin and made a mad scramble for the door back into the stadium, to discover we'd scored a goal! Not sure what Eric Ten Haag (ETH) did at half time but suddenly we were on form, attacking left right and centre and pushing Villa back. I reckon it's because we were shooting towards the Stretford End at this point, where the die hard United supporters stand. This, combined with the magic ETH performed during half time, meant we were in for an incredible second half. The goal was checked by the VAR and eventually disallowed, but it was enough to lift all of our spirits. We were back on.

Suddenly, we had our swagger. We were attacking, we had form, we had style, we were communicating. Players were creating opportunities left right and centre. The left and right wings were a flurry of activity with players receiving the ball and running it forward. The centre field was cracking apart Villa's defence. It was a dream to watch. Around the 58th minute, Bruno Fernandes made a short run with the ball. At the same time, Rashford made a run forward too. Fernandes played him the ball, and Rashford ran into the box. He had Højlund and Garnacho running into space to the right of him. The Villa defence were focused on Højlund and both running to him. Garnacho was left with heaps of space, running to the far side of the goal. Rashford looked to the right, and released the ball. Højlund didn't quite make the touch, and the ball rolled out to Garnacho, who was there and ready to receive. He punted it, aimed at the goal, and took his shot! Goal! Manchester United 1, Aston Villa 2. Højlund knew exactly what to do and while Garnacho celebrated, he grabbed the ball and ran back to the centre of the pitch. The stadium exploded with cheers of joy and celebration! We were on!

Clearly, Garnacho gained a lot of adrenaline from this. Suddenly, his energy levels peaked and he was creating opportunities, and closing them. The true energy and spirit of this 19-year-old Argentine man shone through. The 70th minute closed in and Garnacho won the ball in the centre field. Again, he found himself in bags of space and quickly ran the ball forward. The stadium was going nuts. Complete and utter ecstasy. He made a decent run and found Fernandes on the wing, passed the ball to him and ran in to the box. Fernandes ran forward, and made a cross. Villa deflected the cross. Straight into the line of Garnacho, who was again in heaps of space. He took his time, and punted the ball towards the goal. It hit a Villa defender, and bounced right around the outside of Martínez. Another goal! We were level! At this point, my scarf had obscured my face due to me jumping up and down, screaming and I was deafened by the screaming echoing around the stadium. In just 12 minutes we had gone from being 2-0 down to drawing level. Even if the game ended then, I would remember this for the rest of my life. Stood in Old Trafford for a Manchester United classic comeback. Wow.

But, with 70 minutes of play, there were still 20 minutes left. And we knew it. Oh we knew it. Every Manchester United chant under the sun was being belted out, and every single person in the stadium was stood singing, dancing, and throwing their arms in the air. Just like the Quidditch World Cup in Harry Potter.

I was vibing. I was riding the best wave in the set. I was driving along the smoothest road ever built. In my gut, I had a feeling we would get another. Everything was just perfect - the fans, the score, the mood inside the stadium, the energy levels of the players on the field. We were going to get another goal. I could feel it.

Villa tried another attack, working the ball around the box, finding their strikers in space and creating opportunities. Around the 82nd minute, they managed to find their forward in space, who punted the ball towards the United goal, only to be met with Jonny Evans' left leg and a deflection, placing the ball back into play.

Queue up the 85th minute. United won a corner. A jubilant Stretford End was there to lift up the taker; chanting, cheering and clapping. Fernandes stepped up to take it, and crossed the ball in to the box. There was a scuffle, and the ball met a Villa player's foot, deflected right into the path of Højlund. He had bags of space, enough to watch the ball coming towards him. He lined up, he lifted is leg, he saw the open goal to the right of the Martinez and punted the ball right into the back corner of the net. Everyone. Lost. Their. Mind. 3-2 to United. Højlund's first Premier League goal.

I have to explain why this moment was so special, and why I have written so extensively about it. I have been a Manchester United fan for my entire life now. I am not in the UK that much nowadays, so enjoy going to the Boxing Day fixtures when I can. This means I watch highlights on YouTube, and watch historic videos about the club - the building of Old Trafford, the Busby Babes, the Class of '92, the journey to the treble in 1999, Rooneys debut hat trick, the loss of the Premier league in 2012 to City, the 2023 FA cup final against City. Manchester United are a comeback club. As a supporter, this means that the game is never truly over until the final whistle is blown. To be in Old Trafford, during a game where we were 2-0 down and to then come back and win 3-2, is a moment that every supporter lives for. During the drive back, we all had grins from ear to ear. We went to bed with grins from ear to ear, we woke up the next day with grins from ear to ear.

In the words of Gary Neville...

This is Manchester United. This is Manchester United. This is Manchester United!

Thank you for reading, and Happy New Year!