Can npm download statistics be segmented by country?

No. The measurement layer aggregates worldwide installs and carries no geographic partition. Provenance arrives implicitly through maintainer usernames, declared repositories and keywords, so anything geographic worth building is derived downstream rather than delivered pre-cut. The limitation runs across the whole category - PyPI and crates.io likewise have no country column - which keeps cross-language studies apples-to-apples.

Why do some packages show zero downloads?

Because they were never installed. Of roughly 4.3 million package documents, about 2.5-3 million carry download records at all; unpublished experiments and abandoned scaffolding read as zeros rather than errors. Treating that absence as a state instead of a fault is the single biggest pipeline win this corpus hands you.

What is the finest granularity of npm download data?

Per package per day. Per-version breakdowns cover only the trailing seven days - enough to spot a bad-release spike, not enough to rebuild version-level adoption curves - while publication timestamps per version run back to the registry's 2010 inception, so lifecycle questions resolve off the records even where counting does not.

Datadory notebook

npm Registry Download Statistics: Measured Adoption, Delivered

1,744 datasets. Pick your catch. Every guide here is built on what the catalog can actually prove.

1,744 datasets. Pick your catch.

How do npm, PyPI and crates.io adoption numbers compare?

The three flagship language registries measure adoption three different ways, and the differences decide which question each one can answer:

One caveat travels across all three: each registry meters its own installs, so absolute numbers never compare across ecosystems. Trajectories within one registry do.

Pick up where this leaves off

Every one of these ships with sample rows before you commit to anything.

Systems Software Global - the whole JavaScript and Node.js ecosystem in one…

Get a sample

npm Registry Download Statistics: Measured Adoption, Delivered

How do npm, PyPI and crates.io adoption numbers compare?

Pick up where this leaves off

npm Public Registry API

deps.dev - Open Source Insights API

GitHub Octoverse and GitHub REST/GraphQL API

PyPI - Python Package Index API and Stats

crates.io - Rust Package Registry API

The sample tray

Start here — the deepest field dictionaries in the catalog

Wyoming Statewide Time-Stamped Oil and Gas Activity, 1900-2020

BTS T-100 Data Tables Query Interface

Open Payments (Physician Payments Sunshine Act)

USAJOBS API Data

Steel Plates Faults (UCI)

AviationStack Flight & Airport API