Meeting 1: Kickoff (Monday 2 November 2026)
Chipp Walters (ALTUIT) and Mark Pratt (Museum For A Rainy Day). About two hours. Outline taken from the meeting recording; it follows the order of the conversation. Pre-meeting document: Website Design Review and Redesign.
Housekeeping
-
Payment: Patreon to be replaced by PayPal
- Patreon takes a big cut and holds money for over a month
-
The website design review was sent on Discord
- AI-generated, partly corrected by hand
- Mark asked to review the executive summary, customer value proposition and at-a-glance section in detail
Project goals
Working together
- Mark is a generalist, not a designer; serious only about photography
- Wants a cohesive identity and design language across app and website
- Wants it to feel professional
- Open to bringing in other help if needed
The museum
-
Started as a place to show Mark’s own photography
- He doesn’t post photos online; avoids feeding Instagram and AI
- A museum or gallery context feels different
-
Extends to other diligent, unseen artists
- Example: a Berlin painter with a house full of paintings, shown only five or six times, in cafés
-
Goal: permanent exhibitions for at least 1,200 artists across many devices
- Best on Meta Quest or Vision Pro; decent on iPad and iPhone
-
Long term: a serious digital museum that includes 3D artists
- Blender competition entrants and ArtStation game-asset artists have no such outlet
- Should be serious, but cool and fun
Revenue and GROW
- Revenue is secondary for now; costs kept very low
-
Future special exhibitions
- Talks with a curator of contemporary Chinese art (ex-Sotheby’s, now at the Confucius Institute Nuremberg)
-
GROW, the sister site: tools for curators and artists, some free, some
paid
- A platform to propose work and have it reviewed
- Output mainly to Museum For A Rainy Day, possibly other digital museums
Chipp’s summary of the need
- Help with branding and website design
- Customers: visitors, 3D and traditional artists, and Mark’s own photography
Meeting record and questions page
- Chipp posts a synopsis on a web page after each meeting (this site)
-
The same site holds questions for Mark
- Fields and buttons, with a Copy JSON button at the top right
- Mark copies the result and sends it on Discord
Logo critique
Logotype
- The pixelated type reads as youth, gaming or rave
-
The only color is four green dots, which look like a mistake
- Visitors never get to hear the logo explained
- The brand isn’t recognizable anywhere yet
- The name is strong: it communicates the idea and sparks curiosity
- The umbrella mark (with green raindrops) can work; the logotype is the question
- Chipp suggested considering a change; Mark’s choice
Mark’s reasons for the logo
- Trademark applications already filed
-
Pixelation signals gaming technology and the digital nature
- The name alone may not say “digital museum”
- Didn’t want to get stuck on the logo early
- Partly agrees; the logotype can evolve later
Criticism style and priorities
- Mark welcomes blunt, constructive criticism
- Agrees that no designer has ever touched the website
- Prefers to prioritize a cohesive design language over logo work
The logo at small sizes
- Mark shared a Photoshop view of the logo on Discord
- 72 DPI is too low for such a complex logo
- Idea: use the logo mark with the name set beside it
- The font and logo look fine large, like a title card; the site design made them too small
- The umbrella alone at 512 × 512 is used as the icon and works small
-
Serious kerning issues in “for a rainy day”
- The audience is artists with an aesthetic sensibility, and the logo is the first thing they see
- Chipp will rework it and suggest other ideas
Color palette
- The current green is close to British racing green; Chipp doesn’t like it
- Chipp asked for a mood board
Mood board approach
- Tools: PureRef, Photoshop, Google Slides
-
Sources: Behance, website design galleries
- Paste images straight into Discord
- Mark wants to build a mood board app in GROW
- Chipp’s approach: look at it from the designer, marketer, curator and artist viewpoints
-
Personal likes are welcome even if they won’t be copied
- Example: a favorite camera, a sports car
- Semantic questions: if the museum were an animal, a food, a celebrity spokesperson
- Aim: avoid a design that suits the designer but not the client
Mark’s look-and-feel wishes
- Not in love with the Tate or MoMA sites
- Not too technology-driven; the logotype is the nod to tech
- Licensed logo font; D-DIN (open source) is used for museum signage
- Sophisticated, but not overly museum-like or institutional
-
More accessible
- Curation isn’t based on followers, sales or exhibition counts
- Looks for artists who practice regularly in their field
- Cameras as a model would feel too technical
- It must be clear that it’s a digital museum
-
Colors chosen to complement and break up layouts
- Avoid a “Windows tile” look across the artist panels
Chipp’s color guidance
- Large areas of color will overwhelm the art
- Hue, saturation, value and color schemes as the vocabulary
- Suggests whites and grays, maybe muted colors
-
Option: a clean, monochromatic, playful interior
- The logo is the playful part
- A colorful front image that rarely changes, or art that matches the theme
The story is the artists
-
Journalists and curators urge Mark to tell his own story
- Including a curator in Canada and the Chinese contemporary art expert
- Mark: the story is the artists, not him
-
Raise little-known artists to hero status
- Including artists known in China but not in the US or Europe
- Social campaigns draw other artists to download and visit
- Featured artists rotate
- Monthly MetaHuman videos presenting the artists
Presenter videos and the uncanny valley
- Chipp has seen voice cloning paired with a stylized rendered presenter, which avoids the uncanny valley
- MetaHumans can feel off; cloned voices can sound stilted
- Mark: a Blackmagic Design contact thought cloned characters were real
-
Mark accepts some uncanny valley for high production value
- The presenter matters less; the artists must feel real and look like heroes
Mood images and AI tools
- Chipp tried generating images live; the results were poor
- Midjourney is good for concept work
-
Mark likes the “a place for curiosity” image but keeps coming back to
black, which he dislikes
- Problem: it works with 2D images, not 3D
- ChatGPT is best for marketing copy and headlines, and good for refining imagery
- Chipp will build a website of ideas
Logo direction decision
- Mark agrees with the logo and type criticism
- Proposes using the umbrella alone as the logo, with the rest in regular type
- Chipp agrees
Umbrella changes
- Chipp wants to add a green dot on the far right
- Mark: a new trademark variation adds filing cost
Trademark discussion
- Chipp asked why trademark at all
-
Mark’s reasons
- T-shirts, tote bags, a museum gift shop
- Talks with a Chinese manufacturer about logo cases bundled with the apps
-
Chipp: money not well spent
- He has trademarked only once, when investors demanded it
- The TM mark may be usable without registration; worth looking into
- Dated public use also protects the image
- Mark: the money is already spent; avoid the sunk-cost fallacy
- Mark is open to logo changes; keep the pixelation and evolve gradually
- Chipp: keep the umbrella, add a dot, maybe add color
-
A black-and-white version is needed; Mark has one
- The umbrella works on white and on black
Mood board assignment
- The mood board can come from Midjourney, ChatGPT or anywhere
- Mark posted the black-and-white logo and the postcard design on Discord
Design tools
- Mark uses Affinity Designer
- Chipp is building his own Photoshop-style editor with free AI features, and his own vector editor based on Inkscape (path effects work like Blender modifiers)
-
Mark: any tool is fine as long as it’s accessible
- He switched to Blackmagic and dislikes Adobe’s subscription lock-in
Fonts and logo files
- Mark to send D-DIN, the logo font and the latest vectors
-
The logo was redrawn by a graphic-designer friend after the original
files were lost
- The original was pixelated and feathered; there is no perfect version
- Chipp will do the mood board himself and show only what he likes
-
Fundamental challenge: 2D art for now, 3D later
- Mark loves robot images but has no 3D robot
-
Chipp focuses on look, feel and color themes before headlines and type
- The robot idea could become a 3D picture frame holding 2D art
- Both fonts were sent on Discord during the meeting
Chipp’s own brand as an example
- chippwalters.com: teal and red-orange, one favored font
- Abstract tiles rather than literal product pictures
-
The same brand on Patreon, Superhive, Gumroad and YouTube thumbnails
- Not identical everywhere, but continuous
- Mark noticed the change; wants the same continuity between experiences
- Goal: logo, color theme, typeface
-
Color is the strongest design element
- A Volkswagen painted Ferrari red
-
Typography matters hugely; the logo font is too distinctive for reading
- Mark: it was never meant for reading; D-DIN is for text
The app and frames
- The app drops visitors straight into a 3D environment
-
Three frame types, modeled in Blender by an environment artist
- Current choices: basic wood, basic black, none (canvas look)
- Chipp offered to model frames
-
Mark: frames are a choice for artists
- Currently a pull-down; artists can’t see the frames
- Frames differ between Unreal, the website and GROW
Frames revisited
- Mark worries ornate frames distract from the art
-
Chipp: framing is an art in itself
- Examples: an ornate frame on his Joker and Harley Quinn piece; a signed Peter Max lithograph
-
Mark: simple frames keep the Unreal footprint small
- Target 1,200 artists in one app; it may be split into parts
- Advice from a former Epic VP: load levels in and out like chapters
-
Chipp: built the Alamo AR app for San Antonio
- A three-year project; the most accurate model of the 1836 Alamo
- Built in Unity; Unreal’s LOD tools make frames no problem
- Chipp pasted frame examples into Discord
Platforms and team
- Two-person team
- Live: iPhone, iPad, Android phones and tablets, Meta Quest
- macOS in TestFlight
- PC in Microsoft review
- Vision Pro due in November
- Compromises were made; Mark hates the current navigation and UI
Scope
- Not meant as scope creep
-
The design language must cover the website and the app
- Avoid a great website followed by a disappointing app
Marketing message and headlines
- Mark had already fed the landing-page document into the redesign
-
A headline should grab attention without click-bait
- “A place for curiosity”: good, intriguing
- “Everyday extraordinary”: a platitude that fits any site
- The headline should fit only this museum
-
The subheadline gives evidence that backs the headline
- “Objects, ideas and stories from a curious life”: weak
- Needed: a headline, a subheadline and some copy
- Chipp read the early copy and still couldn’t tell what the museum is about
- Mark: the current version is far better
Mark’s copy process
- Fed a landing-page workflow into Codex
- Layered a leading corporate-communications strategist’s methodology on top
- Being on many devices can sound like a gimmick; the real aim is access
Two audiences
- Chipp: the site appeals to visitors and artists at once
- Mark: two experiences, Museum For A Rainy Day and GROW
- Mark has done a competitive analysis of museum apps
Chicken and egg
- Needs a compelling visitor experience and a simple artist upload platform
-
Minimum of 24 works per artist
- Filters for practicing, serious artists
- Gives visitors a real feel for the artist
- Other museum apps are hard to update; they need consultants or agencies
- Meta Quest: in two weeks about 10,000 people saw the title card; roughly 85–100 downloads
Marketing advice
- Developers prefer building to marketing
- Chipp’s rule of thumb: 30% development, 70% marketing
-
The chicken-and-egg problem is solvable with smart marketing
- Partner with local galleries; bring headsets to shows
- Use AI to help market the artists
- Collaboration videos like Chipp’s with Superhive reach both audiences
Mark’s marketing plans
- His sixth venture; knows “the indifference of the universe to your great idea”
-
Not worried about marketing
- A tight Vision Pro community; Unreal contacts
- Technical and press angles
- Needs a tight website and a coherent UI first
- Then he switches roles to marketing and PR
- Bots research artists; a similar press list
Outreacher
-
Mark’s internal tool, tracking artist meetups
- Testing in Berlin every couple of months; a future standard format
-
Authoring and tracking for posts and messages, LinkedIn and beyond
- An orange dot marks a finished version
- An artist-finding tool: delete removes, approve queues, promote adds to the CRM
- Chipp: Mark already knows the solution
Website imagery critique
- The top homepage image is poor; it looks like basic 3D
- Mark: that is the current app experience
-
Chipp: make it look like a beautiful real museum
- Could be generated with Blender and AI from real-museum reference photos
-
Mark: Anthony Kimura built a modular, performant environment
- Chipp: Anthony’s strength is modeling, not environment art
- Mark: the work is iterative; open to fixing the app first
Cognitive visualization
- Chipp shared a cognitive-visualization link
- Start with exciting images, not the final design
- There is one chance to impress
- Poorly framed, poorly matted images don’t work
- Chipp shared matted-frame examples; framing matters
Order of work
- Mark will follow Chipp’s lead on the order
-
Chipp: branding first
- Replace the homepage image with a generated one
- The For Artists page has too many colors
- People may want GROW more than the museum
GROW
- A future GROW Studio: an optional subscription
- Many design ideas for it
-
Not a social network; no pressure to post
- Share work in progress, collaborate creatively
-
Curator tools to build exhibitions
- Deploy to Museum For A Rainy Day or elsewhere
-
A permanent exhibition is unprecedented and less gatekept
- The test: practicing, continuous output, consistent style
- No favorite child between GROW and the museum
Next steps
- Meet next week, same time and channel
-
Chipp works on branding and design concepts
- Logo, color theme, typography, design
- No messaging yet; maybe sample messaging
- Concepts are ideas, not finished designs
Typography
- Chipp likes Mark’s high x-height, condensed fonts
- A bit informal; may try Futura or something more formal
- Needs research
Artist photos
-
Chipp dislikes prominent artist photos
- Galleries show a small photo off to the side
-
Mark’s reasons
- Meetup visitors thought they were seeing AI art
- “Ego boo”: the artists’ satisfaction at being shown
- A reward for submitting 24 curated works; promotion was promised
- Standard fields and image instructions are coming
- Chipp: a smaller face photo; don’t put the artist ahead of the art
- For Artists page: blue is the artwork area, pink is the artist area
App experience
- Chipp finds the 3D images unappealing
- Mark and Alex are open to changing the in-museum experience
Making money vs. worthwhile projects
- Mark is annoyed when people first ask how it makes money
- Chipp asked that last time; investors would need a clearer answer
- A leaf in the river vs. a laser beam; both are valuable
- Chipp is retired and values the journey
Mark’s view
- Thrilled that Chipp is excited
- The project should exist; it helps people be seen
-
Business may come with traction
- Museums earn only about 20–25% of revenue from tickets
- A possible investor fit: Caterina Fake (Flickr, Etsy, Kickstarter; a designer)
- Not about 10× or 100× returns
-
Meta Quest and Vision Pro lack non-gaming content
- Museum apps usually die after launch
- GROW keeps the app alive and evolving
- An alternative to Instagram doom-scrolling
Other art sites
- Renderosity, ArtStation, DeviantArt, BlenderArtists
- ArtStation is mainly a hiring portfolio for games and concept art
-
Mark at a UK gaming conference
- Few game artists make personal art
- One woman’s lace designs made in Substance Painter
-
DeviantArt is an upload site; the museum is curated
- Curation like a DJ mixing old and new
- Chipp once commissioned an ArtStation painter for an Alamo painting
- Curated sections and artist subscriptions exist elsewhere
Web vs. app delivery
- Chipp asked about viewing from a web page
-
Mark chose apps for low cost and long life
- The stores pay for distribution; one-time uploads
- No web-traffic costs
- Looking at Three.js and Cloudflare long term; storage is a concern
- Unreal targets include Apple TV, next after Vision Pro
- Chipp: walking a gallery gets old; people want to scroll
App navigation
- Chipp will download it from the Play Store on an Android tablet
- First launch plays a MetaHuman video explaining the controls
- Gaming thumb controls, tap an image to jump, or swipe
- Switch galleries from the top menu
Data and scale
-
All images are stored in the app
- Performance focus; reduced galleries
- Smart compression of power-of-two images
- Chipp: the issue is simultaneous hits, not data size
- Mark runs the Metro Publisher CMS; sites get hammered by bot traffic
- Cloudflare now makes a web component possible, but costs grow
- Alex is experimenting: an endless beanstalk, an endless image frame
- Challenge: navigating 1,200 artists
Chipp’s web suggestion
- Walk before running; don’t architect for 100,000 images yet
-
A quick Three.js web page to excite people
- “If you think this is cool, wait till you see it on Meta Quest”
- Mark agrees it’s possible now, but is committed to Unreal
-
Meta Quest: about 10,000 unique people saw the title card free;
about 85 downloads
- The store suggests apps; it has little non-gaming content
- Vision Pro next; outreach to Vision Pro YouTubers
- On the web it would be one of a million
- Chipp is focused on conversion; maybe a landing page later
Unreal art and lighting
- Chipp will generate concepts with gallery treatments
- No strong Unreal environment artist on the team
-
Alex: full-stack developer and partner in Mark’s company, learned Unreal
- Unreal’s MCP integration has improved
-
Anthony Kimura: met at Unreal Fest Stockholm
- Twinmotion work; good modeling and texturing, not environments
- A successful Unreal plugin
- His brief was performance, with few advanced features
- Mark thinks a lighting artist is needed
Lighting is an art skill
- Website images: the lighting, textures and camera angles are poor
-
A VFX artist’s video compares human vs. AI work from $10 to $10,000
- The AI output was weaker than the expert’s
- Tools don’t help a novice; skill matters
- MCP control doesn’t mean knowing how to light
- A reference-image approach could help
-
Mark interned six months at a large Washington, D.C. photo studio
- Knows light modifiers, but isn’t a lighting artist
- Mark is strong in cloud and networking and built his Unreal contacts from scratch
- Goal: expand the network and bring in the right person
Mark’s thanks
- Grateful for the feedback at any budget
- A cohesive design language shows a serious endeavor
Gallery screenshot critique
- Bad leading lines, no focal point
- Bad perspective: verticals should stay vertical
- Aspect ratios differ: Android 16:10, iPad 4:3
-
The bottom image shows off the roof beams
- Gallery ceilings should disappear, painted black
- Mark: the images are meant to show a 3D space
- Chipp: a wireframe overlay on the right third would show it’s 3D
Work sequence and aspirational concepts
- Mark: many construction sites; work through them in sequence
- Chipp: start with the website
-
Cognitive visualization first: aspirational images before any polygons
- Generate with ChatGPT and Midjourney
- Then solve the technology
- Design thinking: the Steve Jobs and the smartphone example
- Next stage goal: something aspirational
- Mark is fully on board, at whatever pace