Atlas of the work of Ryan Kamp

Plate I. The measure of things

Ryan Kamp

One plate, drawn at every size from a galaxy to a ribosome.

Ryan Kamp is a graduate student in computer science at the University of Cincinnati. His research interests are scientific visualization, computer graphics, digital media, software engineering, and information retrieval. This atlas follows the first of them across twenty-nine powers of ten.

The plate is one continuous zoom. It takes the same journey as Powers of ten in Splashery, his Gaussian splat playground: from the Milky Way down to a single ribosome, by way of a garden bed beside the Smithsonian Castle. Here the journey is cut in line. The planets and the Moon stand where they are at this minute, the Earth is turned as it is turned now, each star sits at its cataloged distance, and the scale bar is true at every step.

Turn the wheel over the plate, drag it, slide the rule, or press the arrow keys.

The planets, out to Neptune

The plate is about 84 AU across. Light crosses it in 11 h 40 min.

The stops, largest first

  1. The Milky Way, about 100,000 light-years across. No one can draw it from outside: the spiral is a diagram, set at the galaxy's true size and tilt, with the Sun 26,700 light-years from the center.
  2. The naked-eye stars, each where it really is. The ring is 1,600 light-years from the Sun.
  3. The Sun's neighbors. The ring is 65 light-years out; the bright ones are named.
  4. One light-year: the Sun alone.
  5. The planets, out to Neptune. The orbits are about 60 AU (9 billion km) across, and every planet stands where it is today.
  6. Mercury, Venus, the Earth and Mars. Mars's orbit is about 3 AU (456 million km) across.
  7. The Earth and the Moon, about 384,400 km apart on average, with the Moon's path for the coming month.
  8. The Earth, about 12,700 km across, its coasts turned as they are at this minute. The hatched half is night.
  9. The Chesapeake Bay, about 320 km long, and the Mid-Atlantic coast.
  10. Washington, D.C., on the Potomac: a square whose sides are about 16 km, less the corner across the river.
  11. The National Mall: 3.6 km from the Lincoln Memorial to the Capitol, center to center.
  12. The Enid A. Haupt Garden beside the Smithsonian Castle, about 170 m long and 100 m deep, with its parterre, its paths and its trees.
  13. A garden bed with a young full-moon maple, about 2 m long, drawn as contour lines of its height.
  14. One maple leaf, about 9 cm across, its edge and its veins traced from a photograph.
  15. Cells inside a leaf. The picture is 200 µm wide, and each small ring is a chloroplast.
  16. A chloroplast, where a leaf turns light into food, about 3.8 µm long in this trace, with the membranes inside it.
  17. A ribosome, the machine that builds proteins, about 22 nm across: the outline of its cryo-EM map, with its chains of RNA and protein inside.

Down to the Earth, every plate looks from the north of the ecliptic. Closing in on Washington, the view swings round to look straight down on the garden, north up. From the city inward, every outline is the true shape of the thing at its true size: the city from OpenStreetMap, and the rest traced from the capture, the pictures and the cryo-EM map that Splashery shows at each scale. The leaf, the cells, the chloroplast and the ribosome each come from a different plant or cell, as they do in Splashery.

Plate II.

Splashery

2026 to present

An interactive Gaussian splat playground that runs entirely in the browser. Pick a toy (a real captured object, or one generated from a seed), spin it, poke it, blow on it, splash paint on it, drop it, watch it fall apart and come back, make your own, and share it as a link, an embed, a GIF or a video.

Play Code Tinkerer's Manual

Fig. 1

The night sky, for any date, time and place

Splashery puts you under the stars: every star to magnitude 6 from the HYG database, the constellation lines, the Sun, the Moon and the planets, turned to the date, the hour and the place you choose.

This chart is the sky over Cincinnati, drawn from the same catalog with the same formulas. The zenith is at the center and the horizon is the rim. Drag the chart sideways, or slide the hours.

Cincinnati, Ohio: 39.10° N, 84.51° W.

The Night sky toy

Fig. 2

Real landscapes from public data

The Grand Canyon that floods from the river up, the Mariana Trench under an ocean that drains away, and Mount St. Helens before and after May 18, 1980, each built from public elevation data.

This map is Mount St. Helens from the same two grids, 11 km on a side, with a contour every 100 m. Change the date and the summit goes: the highest point in the grid falls from 2,945 m to 2,533 m, and the crater opens to the north.

Highest point in the grid: 2,945 m.

Mount St. Helens The Grand Canyon The Mariana Trench

Fig. 3

Powers of ten

One continuous zoom from a galaxy down to a single molecular machine, through real pictures and data at every scale: about 98,000 stars where they really are, the planets on their orbits, aerial pictures of Washington at every half power of ten, a maple leaf, the cells of a leaf under the microscope, and a ribosome from its cryo-EM map. Plate I follows its stops.

It is Splashery itself, running in a frame: about 5 MB to begin with, and more pictures as you zoom. Or open the Powers of ten toy on its own page.

Also on the shelves

From atoms to weather

Climate records drawn from data: Mauna Loa CO2 as a spiral and GISTEMP temperature anomalies. A ripple tank and a light bench. A molecule viewer that fetches entries from the Protein Data Bank. A volume viewer for CT and MRI scans. Captured toys that are real scans of up to 450,000 splats, and effects that run on the GPU for every splat, every frame.

It is static files and ES modules only: no build step, no framework, no server. Ryan defined the project's vision and its toy categories, refined the interaction design through hands-on testing, and contributed mathematical recipes.

Open Splashery with its labs shelves

Plates III to VI.

Charted in detail

builds checks a person decides revises stop

Plate III.

grooph

2026 to present

Visual loop graphs for coding-agent workflows. You draw who builds, who checks, where a person decides and when the work stops. grooph checks that every loop names a stop, then compiles the graph into a prompt package for a separate coding-agent harness. It runs no agent and calls no model, and graphs stay on your device.

Open Docs Code

anchor text URL radiant

Plate IV.

Link Meteor

A Chrome extension that collects links with their anchor text and URL in separate fields, for reviewing sources and exporting a research collection. No account, ads, telemetry or backend.

Website Code

1 10 100 1,000 104 105 106 about 345,000 words

Plate V.

English OpenList

December 2025 to present

A free, open, continuously maintained list of valid English words, about 345,000 of them, with the provenance of every word (which sources attested it, and when), auditable correction records, and a browser explorer.

Explore Code Dataset

sources swept on a schedule; repeats struck out

Plate VI.

resmon

April 2026 to present

A local-first literature-monitoring desktop application. It runs scheduled searches across open-access scholarly sources, removes duplicates, and compiles reading reports, with optional summaries.

Code

Index of further plates

  1. reword_nerd

    A local-first browser workbench that turns a reviewed document, or a set of images, into a portable workbook of rewriting prompts. Everything stays reviewable in the browser, the download is a separate, explicit step, and the application contacts no model provider itself.

    Open Write-up Code

  2. Ars Magna

    An anagram finder. Give it any text and it returns every way those exact letters can be re-partitioned into real English words, by a deterministic search.

    Try Code

  3. aqrobat

    A local text and emoji QR lab: QR artwork built from letters, symbols or emoji.

    Make one Code

  4. Amordle

    A free word game: any word length from 2 to 35 letters, three difficulties, a daily puzzle, and ranked multiplayer with an Elo rating.

    Play Code Write-up

  5. The Speed-Accuracy Trade-Off in Perceptual Decision-Making

    Ryan's undergraduate research capstone, April 2025: evidence accumulation modeled as a diffusion process, with dynamic programming and value iteration, in a simulated obstacle-avoidance scenario.

    Code Video

Legend.

Research interests

  • Scientific visualization
  • Computer graphics
  • Digital media
  • Software engineering
  • Information retrieval

MS in Computer Science, concentration in Artificial Intelligence, University of Cincinnati, August 2025 to expected May 2027. Thesis advisor: Dr. Yizong Cheng.

BA in Mathematics and BS in Neuroscience, University of Cincinnati, January 2020 to May 2025. Capstone advisor: Dr. Hyunjoong Kim.