Skip to main content
Version: v2 beta

Positioning against existing frameworks

This method retains its foundations, rules, models, hardware and use-case structure. PCRs specify assessment requirements for a defined boundary; D4B uses them to improve its framework and make its choices explicit, without claiming overall compliance.1

Framework map​

Digital environmental assessment includes several complementary approaches2:

FrameworkCoverageStrengthsBoundaries and limitations
ADEME / ARCEP product category rulesFrench digital servicesOfficial framework, multiple indicators, three tiersLimited granularity for some equipment; French context
BoaviztaHardware components and equipmentDetailed decomposition, open sourcePartial use-case coverage; service scope needs additional definition 3
GreenITAwareness and practical use casesAccessible explanations and digital sufficiencyDocument types and methods vary by publication
Green Software Foundation SCISoftware carbon intensityExplicit boundaries and functional unitCarbon only; hardware manufacturing included through allocation
PAIA, MITConsumer electronicsParametric product modelingProduct-centered rather than service-centered
imec NetZeroSemiconductors and componentsManufacturing process detailSpecialized; applicability depends on access and available inventories 4
Cloud Carbon FootprintCloud infrastructureComponent energy coefficients, open sourceCarbon-centered and cloud-focused 5
UBA Green Cloud ComputingCloud and datacentersVirtual-resource allocation, multiple indicatorsGerman and datacenter context

This map describes complementary uses, not a precision ranking or certification of compliance. The dated source review identifies recent contributions including SCI for AI and component models.

What is adopted​

FrameworkContributions adopted in this methodology
ADEME / ARCEPLCA framing based on ISO 14040/14044. A multicriteria perspective, with nine D4B indicators and an explicit mapping to the 2026 PCR. Three tiers: user devices, networks and computing centres. Equipment or system approaches according to control over the scope.
BoaviztaMaterials → components → equipment decomposition and component-to-equipment modelling. Distinction between explanatory models and generic factors. Transparent data and open source.
UBA Green Cloud ComputingVirtual-resource allocation using vCPU, vDisk and vNIC. Supporting infrastructure, including cooling and PUE, accounted for through documented allocation. Multiple indicators: GWP, ADP, cumulative energy demand and water.
Green Software Foundation (SCI)Explicit boundaries and a unit of analysis related to the function delivered. Connection between usage, infrastructure and impact.
GreenITPractical use cases and user journeys. Sufficiency as a reduction strategy. Explanations accessible beyond LCA specialists.

What is adapted​

TopicRelationship with the PCRsD4B choice retained
BoundaryWhole application and explicit study periodPage, session or request calculations, to be aggregated
IndicatorsSixteen EF categories in the 2026 parent PCRNine indicators, with coverage and gaps stated
NetworkFixed and variable contributions in Internet access/application PCRsPer-GB example; difference documented
CloudService-unit factors and specific allocation rulesResource estimates where appropriate, without automatic equivalence
QualitySix criteria in the parent PCRSeparate D4B grid with an explicit mapping
AIApplication time boundary differs from model lifecycleTraining, media and limitations retained

These differences are detailed on the relevant pages. Compatible documentation additions can be incorporated directly. Changes to formulas, factors or scores require a separate analysis of their effects on results.67

Distinctive organization​

  1. A modular architecture for discovery, scoping and implementation.
  2. One central assumptions and defaults register.
  3. An explicit link to the versioned canonical digital4better/data repository.8
  4. A hardware chain from materials to components, equipment, infrastructure and services.
  5. Visible allocation, double-counting, robustness and time conventions.

Assumptions and limitations​

  • This is an editorial and methodological positioning, not an exhaustive normative comparison.
  • Convergences and differences should be revisited as frameworks evolve.
  • Specialized frameworks such as PAIA and imec are only partially reflected here.

Footnotes​

  1. Green Software Foundation SCI Online documentation, unpinned version. ↩

  2. Reference corpus: Pathfinder Framework, Green Cloud Computing (UBA, 2021) and PAIA; Boavizta and SCI documentation. Versions and boundaries are detailed in Sources and corpus. ↩

  3. Boavizta documentation Online documentation, unpinned version. ↩

  4. imec SSTS Online documentation, unpinned version. ↩

  5. Cloud Carbon Footprint Online documentation, unpinned version. ↩

  6. ADEME, RCP services numériques, version 2, juin 2026, p. 13–25. ↩

  7. ADEME, RCP des applicatifs numériques, juin 2026, p. 13–16, 24–35, 54. ↩

  8. Digital4Better, digital4better/data.Evolving catalogue; pin a commit for reproducible factors. ↩