Early Access Testing — dadeda.design
Welcome — and thank you. You're seeing the v0.1.0-preview release. The site is functional and most things should work; the goal of this round is to surface what surprises, frustrates, or feels good before the apex domain cutover.
Staging URL: https://website.kathryn-89d.workers.dev/
Plan to spend ~10–15 minutes. The guided tasks below are not a script — they're prompts to help you encounter the major surfaces. If something pulls your attention sideways, follow it; that's exactly what I want to hear about.
Guided tasks (~10 minutes)
For each: did you complete it? How did it feel — fast / normal / slow / couldn't? Anything to flag?
Discovery (See Work). Find work I've done on AI. Expand the card and read about it. Find a demo to watch.
Discovery (Get to Know). Find an article I've written. Open it (it'll link out to the publishing site).
Reading. Open one expanded card and scroll through its full body content. Use the navigation inside the card.
Sharing. Share the site with someone. Any path counts: contact > share, browser share menu, copy-paste, etc.
Contact. Email me about a hypothetical project (no need to actually send unless you want to). The path you take matters more than the message.
Architecture. Find documentation about how the site is built.
After the tasks: 5 minutes of open exploration. Click around. Try things that don't make sense. Try the site on your phone if you can.
Optional — WebMCP / agent testing
If you have a WebMCP-capable Chrome extension (e.g., WebMCP Inspector 3.0.0 or WebMCP - Model Context Tool Inspector) and the following pre-requisite Chrome flags enabled, the site exposes 10 tools.
⚠️ Prerequisites: Go to chrome://flags and enable the "WebMCP for testing" flag in Chrome 146.0.7672.0 or higher.
Try these prompts via your agent:
- "List the clusters and cards on this site."
- "Find cards tagged with 'data-viz' and summarize them."
- "Open the Do Websites Need Pages card and summarize its body."
- "Get Kathryn's resume URL."
- "Share this site with me as a brief intro."
Tools registered: getSiteMap, getCard, searchCards, focusCard, setPrivacyPreferences, getResume, sendEmail, openLinkedInProfile, openGitHubProfile, shareProfile. (Also callable via the browser console: await window.dadeda.callTool('getSiteMap').)
If you don't have a WebMCP-capable extension, skip this section — there's no fallback you need.
Submitting feedback
Three paths, easiest first:
A. The form at /testing. Below this doc on that page is an interactive form. Fill the fields, click Submit via GitHub (opens a prefilled issue in a new tab; you click submit on GitHub to send) or Send via email (opens your mail client with everything in the body). Universal — works whether or not you have a GitHub account.
B. The GitHub issue template directly. https://github.com/Data-Design-Dimension/website/issues/new?template=tester-feedback.yml. Same fields, filled in on GitHub.
C. Plain email. kathryn@dadeda.design with the same fields. Please include the specific browser and device you tested on — be precise, the exact model matters for perf / CPU / GPU diagnostics (e.g., "Chrome 145 / MacBook Pro 16-inch 2019 / macOS 14.5", "Safari / iPhone 14 Pro / iOS 18", "Firefox 130 / Dell XPS 13 9310 / Windows 11"). No other structure required — just write what you saw.
Either way: screenshots help a lot, especially for visual issues. Drag-drop them into the GitHub issue or attach to the email — neither prefilled URL can carry binary files, so add them after submission.
What I'm NOT looking for in this round
- Polish-level word smithing on case-study text. The bodies are first drafts and will iterate.
- Missing case-study content for other works. Older projects were archived and I'll be adding more anonymized examples from NDA-protected agency work post-launch. For now, my goal is a slim balanced set of projects across AI / engineering and dataviz.
What I AM looking for
- Did you find what you were looking for? Each task should resolve in seconds, not minutes.
- Did the site feel responsive? Cursor lag, expand jank, scroll weirdness — name it.
- Did anything surprise you? Visual artifacts, unexpected behaviors, "wait, what just happened?" moments.
- Did anything feel really good? Equally important — those are the threads to pull on.
- Mobile. If you tried on a phone, even just for a minute: report.
Outreach copy I send to testers
Reusable for whatever channel I'm using to invite people:
Hi. I'm rolling out an early-access preview of dadeda.design, my personal professional website, and would love your eyes on it. Visit https://website.kathryn-89d.workers.dev/. There's a short list of guided tasks to follow (~10 min) plus open exploration, with a feedback form at the bottom of the page: https://website.kathryn-89d.workers.dev/testing. The form submits via GitHub if you have an account, or by email if you don't (please include the specific browser and device you tested on — model matters for perf diagnostics). Anything that surprises, frustrates, or feels good will help me. Thanks ✌️