Eyra internal. Remove this page and the banner link before sharing with D3I.
Notes on this draft
What the example site does differently from datadonation.eu today, why, and what still needs checking before it goes to D3I.
The goal
Researchers should understand how data donation works in practice and get a study running without confusion about what is what and who can help. The example is written from D3I's perspective ("we" is D3I) so that D3I can picture it on their own site. It is not about making Eyra look better.
What D3I's site already does well
- The Getting started page separates researchers in the Netherlands from researchers elsewhere, and sends the second group to Eyra's service and pricing page. The example keeps that split and builds on it.
- It says the service runs on Next, that Eyra manages Next, and that the software can be self-hosted.
- The study design, privacy and completed-projects material is useful to everyone and is left untouched.
- D3I's data donation task is actively maintained (v3.0.0, last updated 28 October 2026). The "outdated" problem concerns older per-study repositories, not D3I's main codebase. The example makes that distinction on purpose.
What the example changes, and why
| Change | On datadonation.eu today | Why it matters |
|---|---|---|
| Support scope stated up frontStart here, Who helps you | "Dutch researchers can use it for free with support from the D3I project." The same contact address appears on every page, with no statement that it is for Dutch institutions only. | Researchers abroad assume the support is for them, find out late that it isn't, and then come to Eyra. |
| Two codebases, two maintainersWhat is what, Who helps you | Feldspar is not mentioned. Nothing says the data donation task differs from Eyra's template. | People take a D3I flow application, get stuck, and ask Eyra for support that Eyra cannot give because the code was refactored. |
| One name per thingWhat is what | "Port", "the data donation software service", "script", and (on GitHub) "data donation task" are used for overlapping things. Eyra says "flow application". Next is not explained. | Researchers can't tell which part they are looking at, or search for help with the right term. |
| Realistic expectationsYour flow application | "Standard scripts for the most common platforms, with no coding required to use them." "Drop the wheel into your data donation study like any other script." | Download file formats change constantly. Without Python skills and testing, most people will not get a flow application working. |
| Dates and owners on every flow applicationYour flow application | The script selector shows no versions or dates and is labelled "in development". Completed projects link to study repositories without saying whether they still work. | A researcher can't judge how much work a starting point will be. |
| A route for researchers abroad that ends in working, supported softwareStart here | Getting started sends them to Eyra for the platform, then "Next step: pick a platform" sends everyone to the D3I script selector. | The current path leads non-Dutch researchers onto D3I's unsupported-for-them codebase by default. |
| Eyra's role namedWhat is what | Footer: "supported by Eyra". Data donation page: "we developed the software Port". | Researchers don't know Eyra exists as a place to get help until they are already stuck. |
Statements the example makes on D3I's behalf
These are written as D3I's own words but are Eyra's understanding. D3I needs to confirm or correct each one.
- D3I supports only researchers at Dutch institutions, and the reason is the ODISSEI funding. Their site implies the first part and does not state the reason.
- Dutch researchers contact D3I first for platform problems, and D3I involves Eyra where needed.
- The script selector produces output built on the data donation task. The exact steps to get that output into a study on Next are not described on their site, so the example leaves them out.
- D3I "can advise" Dutch researchers on other teams' flow applications if they are built on the data donation task.
- Nobody supports self-hosted Next installations.
- "Port" is the name for the software as a whole, and equals the data donation service on Next.
- Which repositories in the list are built on the data donation task versus Feldspar. The grouping follows who built them; older D3I repositories (2023, 2024) may predate the refactoring.
Things to settle on Eyra's side first
- The example proposes "flow application" as the shared term, with "script" for the Python inside it. D3I uses "script" throughout. Worth deciding how hard to push on this.
- eyra.co/projects/data-donation appeared to contain placeholder text in the pricing section when it was read for this draft ("Lorem ipsum" tier descriptions, a "€XXX00" price, an unexplained "D3i workflow" heading). The example links to that page as the destination for researchers abroad, so it should be checked.
- The example says Eyra offers paid support, review and development on Feldspar-based flow applications to anyone, including Dutch researchers. Confirm that is the offer.
- No prices are quoted anywhere in the example.
Suggested framing for the conversation
- Start from the shared problem: researchers arriving at either organisation with a flow application nobody can support.
- The refactoring is D3I's right and serves their users. The ask is only that the site says so.
- Offer the support table and the what-is-what picture as pieces D3I could adopt separately, without restructuring their site.