Field note
What should I know before using the studio?
Answers about local processing, supported formats, source-row handling, exports, reviews, editions, pricing status, and product limitations.
Does my file leave this device?
No. The shipped content policy lets the studio reach only this site, the billing entitlement check and Cloudflare's cookie-free page-view count, and the studio has no upload path, no account and no sync. An imported file is parsed and profiled in browser memory, the rows are released when profiling finishes, and only the field names and bounded aggregates are kept in the project saved on your device.
Does the studio write definitions automatically?
No. An import supplies field names and, from a JSON Schema, titles, descriptions, required flags and enumerations. Every definition, unit, universe, missing-value meaning and owner in a folio was typed by a person, and the studio will not draft, complete or reword one for you.
Can the packet certify my dataset?
No. The packet is a formatted account of what you wrote, plus a manifest and a SHA-256 list. The checksums show whether a file changed after export; they say nothing about whether a definition is true, whether the data matches it, or whether a repository, a regulator or a reviewer would accept it.
Worked example from the sample project
Most questions about this product have the same shape: does it decide something for me, and the answer is almost always no. It profiles what it can observe, formats what you author, and lints for the kinds of gap a rule can actually see. Meaning stays with the people who know the dataset.
The sample makes that concrete. The studio can report that trunk_cm held only numeric values in the rows it read. It cannot report that trunk_cm is a circumference rather than a diameter, that it was measured at a particular height, or that centimetres is the right unit. Every one of those came from a person typing them into the field folio, and all three matter more to a reader than the observed type does.
What this does not do
- It does not upload your file, and it discards source rows after profiling rather than storing them.
- It does not certify a dataset, a definition, a privacy posture or a repository submission.
- It has no account, no collaboration and no live sync; handoff is by file.
Do this in the studio
The fastest answer to most of these questions is fifteen minutes with the sample project.
- Open the studio and load the Neighbourhood tree inventory sample.
- Read the Field folios station, then run the deterministic review.
- Export a handoff packet and open the reader page with the studio closed.
Open the studio at the Overview station