Skip to content

03-problem-statement.md

Latest commit

 

History

History
48 lines (36 loc) · 2.33 KB

03-problem-statement.md

File metadata and controls

48 lines (36 loc) · 2.33 KB
Raw

03 — Problem Statement

Non-normative.

Problem

Organizations can now calculate their digital carbon footprint with high fidelity, but there is no standardized, discoverable, machine-readable way to publish it at the origin. The calculation layer (Salesforce Net Zero Cloud, Microsoft Sustainability Manager, Watershed, Persefoni, Sweep, Climatiq, Kepler, GSF Impact Framework) is rich and fragmented; the disclosure layer is a PDF or a bespoke authenticated API. Consumers — clients, load balancers, aggregators, regulators — have nowhere uniform to look.

Consequences

  • No discovery. There is no robots.txt-style convention for carbon; every integration is bespoke (N producers × M consumers).
  • Rebound risk. The naive alternative — per-request HTTP carbon headers — increases the footprint it reports and opens DoS/fingerprinting vectors.
  • Unverifiable claims. Without a machine-readable methodology link and attestation, published numbers cannot be checked, inviting greenwashing.
  • Wasted calculation. Audit-grade numbers already exist inside enterprise suites but never reach a public, programmatic surface.

Who has the problem

Actor Pain
Web-server / platform operator Wants to disclose a footprint without buying an enterprise suite or inventing a format.
Enterprise sustainability lead Has the numbers in NZC/MS/Watershed; cannot expose them in a standard, automated way.
Aggregator / regulator Must collect comparable data across many origins; faces N bespoke formats.
Carbon-aware client / scheduler Needs a programmatic read of an origin's intensity to make routing/procurement decisions.

Goal

A single, standardized HTTP endpoint — GET /.well-known/sustainability-data — that returns a small, validated JSON document of energy and carbon metrics, fed by whatever source the operator already has, and safe to serve at scale.

Success looks like

  1. Any operator can publish a schema-valid document in minutes.
  2. Existing enterprise data (NZC/MS/Watershed) and live telemetry (Kepler/Climatiq) can be projected to the endpoint automatically.
  3. A consumer can GET one path across many origins and parse uniform fields.
  4. The endpoint cannot become a DoS or fingerprinting vector (safeguards built in).

See 04-requirements.md.