{"schema_version":"0.1","map_id":"paper-62-map","publication_id":62,"publication_anchor":"paper-62","slug":"paper-62","canonical_path":"/knowledge/papers/paper-62/","machine_path":"/knowledge/papers/paper-62.json","root_node_id":"paper-62","stage":"mapped_draft","contribution_type_vocabulary_version":"0.1","contribution_types":["scheme"],"title":"In-App Cryptographically-Enforced Selective Access Control for Microsoft Office and Similar Platforms","short_title":"In-App Selective Access Control","year":2022,"venue":"6th International Symposium on Cyber Security, Cryptology, and Machine Learning (CSCML)","publication_status":"Published","topic":"privacy-identity","labels":["Applied"],"authors":["Karim Eldefrawy","Tancrede Lepoint","Laura Tam"],"keywords":["selective access control","ciphertext-policy attribute-based encryption","hybrid encryption","Microsoft Office add-in","fine-grained document protection","enterprise deployment"],"research_question":"Can fine-grained access policies be enforced cryptographically inside familiar Office documents, while preserving the document format and keeping encryption and decryption responsive enough for ordinary use?","central_answer":"The paper implements an Excel add-in that encrypts selected cell ranges locally under random symmetric keys and protects those keys with FAME ciphertext-policy attribute-based encryption. The same document remains usable by recipients with different attributes, and prototype measurements report sub-second to low-second workflows for the tested policies and spreadsheet sizes, subject to key-management, platform, and deployment assumptions.","curation":{"drafted_at":"2026-07-11","drafted_by":[{"actor_type":"ai","name":"OpenAI Codex","role":"full-text extraction, version comparison, evidence mapping, and initial assessment"}],"method":"Source-grounded review of both checked-in PDFs: the 16-page published paper and the distinct 26-page extended version. Both title pages and a substantive evaluation page were rendered and visually inspected. Technical details below use the extended version where it adds deployment and platform analysis, while publication identity and the published abstract remain tied to the proceedings version.","source_scope":"full_source_audit","approval":{"status":"pending","note":"AI-authored source map awaiting author verification. In particular, the version relationship, prototype-to-production boundary, and security interpretations should be checked before approval."}},"sources":[{"id":"source-paper-62-published","type":"author_hosted_copy","title":"In-App Cryptographically-Enforced Selective Access Control for Microsoft Office and Similar Platforms","url":"/pubs/2022/cscml2022-abe-msoffice.pdf","provenance_category":"author","media_type":"application/pdf","sha256":"2e7ff82af4619bd136419bc6efe8e1d42f6dd441ab4181896008c60e44bd9b00","page_count":16,"version_note":"Published CSCML proceedings version."},{"id":"source-paper-62-extended","type":"author_hosted_copy","title":"In-App Cryptographically-Enforced Selective Access Control for Microsoft Office and Similar Platforms (Extended Version)","url":"/pubs/2022/cscml2022-abe-msoffice-extended-version.pdf","provenance_category":"author","media_type":"application/pdf","sha256":"182274d9bbabd5049e614f795a9da4dc274d885ec9b0f377fdbafa5d19b16239","page_count":26,"version_note":"Distinct extended version with additional enterprise-deployment, alternative-encryption, and platform-extension material."},{"id":"source-paper-62-official","type":"official_publication_record","title":"Springer CSCML 2022 chapter record","url":"https://doi.org/10.1007/978-3-031-07689-3_32","provenance_category":"official"},{"id":"source-paper-62-openalex","type":"citation_index_snapshot","title":"OpenAlex record W4285178185","url":"https://api.openalex.org/works/W4285178185","accessed_at":"2026-07-11"}],"source_anchors":[{"id":"anchor-paper-62-problem","source_id":"source-paper-62-extended","label":"Motivation, contribution, and prototype boundary","locator":"Abstract and Section 1, PDF pages 1-3","url":"/pubs/2022/cscml2022-abe-msoffice-extended-version.pdf#page=1"},{"id":"anchor-paper-62-setting","source_id":"source-paper-62-extended","label":"Key authority, hosting, encryption/decryption services, and recipients","locator":"Sections 3.1-3.2, PDF pages 6-7","url":"/pubs/2022/cscml2022-abe-msoffice-extended-version.pdf#page=6"},{"id":"anchor-paper-62-workflows","source_id":"source-paper-62-extended","label":"Spreadsheet creation, hybrid encryption, storage, and decryption workflows","locator":"Sections 3.3-3.5, PDF pages 7-11","url":"/pubs/2022/cscml2022-abe-msoffice-extended-version.pdf#page=7"},{"id":"anchor-paper-62-policies","source_id":"source-paper-62-extended","label":"Policy language, numerical attributes, parser modification, and bit-length caveat","locator":"Section 3.6, PDF pages 11-13","url":"/pubs/2022/cscml2022-abe-msoffice-extended-version.pdf#page=11"},{"id":"anchor-paper-62-evaluation","source_id":"source-paper-62-extended","label":"FAME/Charm environment and encryption, decryption, and size measurements","locator":"Section 4 and Tables 1-3, PDF pages 13-16","url":"/pubs/2022/cscml2022-abe-msoffice-extended-version.pdf#page=13"},{"id":"anchor-paper-62-alternative","source_id":"source-paper-62-extended","label":"Standardized multi-receiver alternative, informal security reasoning, and limits","locator":"Section 5, PDF pages 16-20","url":"/pubs/2022/cscml2022-abe-msoffice-extended-version.pdf#page=16"},{"id":"anchor-paper-62-deployment","source_id":"source-paper-62-extended","label":"Enterprise building blocks and cloud, hybrid, and on-premises deployment options","locator":"Section 6, PDF pages 20-22","url":"/pubs/2022/cscml2022-abe-msoffice-extended-version.pdf#page=20"},{"id":"anchor-paper-62-platforms","source_id":"source-paper-62-extended","label":"Extension to other platforms and client-side-execution boundary","locator":"Sections 7-8, PDF pages 22-24","url":"/pubs/2022/cscml2022-abe-msoffice-extended-version.pdf#page=22"},{"id":"anchor-paper-62-published-version","source_id":"source-paper-62-published","label":"Published version identity and abstract","locator":"Title page and abstract, PDF page 1","url":"/pubs/2022/cscml2022-abe-msoffice.pdf#page=1"},{"id":"anchor-paper-62-publication","source_id":"source-paper-62-official","label":"Official peer-reviewed publication record","locator":"CSCML 2022, DOI 10.1007/978-3-031-07689-3_32","url":"https://doi.org/10.1007/978-3-031-07689-3_32"},{"id":"anchor-paper-62-citations","source_id":"source-paper-62-openalex","label":"Dated citation-count snapshot","locator":"OpenAlex cited_by_count = 0, queried by DOI on 2026-07-11","url":"https://api.openalex.org/works/W4285178185"}],"nodes":[{"id":"paper-62","kind":"paper","parent_id":null,"order":1,"epistemic_status":"published","title":"In-app selective access control","summary":"A working Office add-in design for encrypting selected document regions under attribute policies, evaluated as an Excel prototype and extended with enterprise-deployment analysis.","source_anchor_ids":["anchor-paper-62-problem","anchor-paper-62-published-version"]},{"id":"paper-62-question","kind":"question","parent_id":"paper-62","order":1,"epistemic_status":"research_question","title":"Research question","summary":"Can users selectively share one intact Office document while cryptography, rather than a trusted document server, determines which protected regions each recipient can read?","source_anchor_ids":["anchor-paper-62-problem"]},{"id":"paper-62-answer","kind":"contribution","parent_id":"paper-62","order":2,"epistemic_status":"implemented_and_measured","title":"Central answer","summary":"Integrate a policy UI and hybrid encryption into an Office JavaScript add-in, place encrypted cell data in the document's custom XML, and release a data key only when a recipient's CP-ABE attributes satisfy the policy.","source_anchor_ids":["anchor-paper-62-workflows","anchor-paper-62-evaluation"]},{"id":"paper-62-scope","kind":"scope","parent_id":"paper-62","order":3,"epistemic_status":"explicitly_scoped","title":"System and trust boundary","summary":"The prototype protects selected spreadsheet content while relying on a key-issuing authority, Office's add-in and document APIs, cryptographic implementations, and secure handling of recipient keys.","source_anchor_ids":["anchor-paper-62-setting","anchor-paper-62-deployment"]},{"id":"paper-62-scope-authority","kind":"assumption","parent_id":"paper-62-scope","order":1,"epistemic_status":"trusted_component","title":"Key authority and identity integration","summary":"An authority alone holds the CP-ABE master secret and issues attribute keys; an enterprise deployment must connect that lifecycle to IAM, private-key storage, revocation, and logging infrastructure.","source_anchor_ids":["anchor-paper-62-setting","anchor-paper-62-deployment"]},{"id":"paper-62-scheme","kind":"scheme","parent_id":"paper-62","order":4,"epistemic_status":"implemented","title":"Selective-encryption scheme","summary":"The design combines local authenticated symmetric encryption for cell contents with FAME CP-ABE encapsulation of a random content key and an embedded policy describing eligible attributes.","source_anchor_ids":["anchor-paper-62-workflows","anchor-paper-62-evaluation"]},{"id":"paper-62-scheme-document","kind":"component","parent_id":"paper-62-scheme","order":1,"epistemic_status":"implemented","title":"Document-preserving storage","summary":"Ciphertexts and metadata are stored in a custom XML part, protected cells are cleared or shown as unavailable, and formulas remain recoverable because the workflow encrypts underlying cell inputs rather than only rendered text.","source_anchor_ids":["anchor-paper-62-workflows"]},{"id":"paper-62-scheme-policy","kind":"component","parent_id":"paper-62-scheme","order":2,"epistemic_status":"implemented_with_caveats","title":"Policy representation","summary":"The UI supports AND/OR combinations over string and numerical predicates; numerical comparisons are compiled into bit attributes, increasing policy size and requiring a fixed bit length chosen consistently by issuers and encryptors.","source_anchor_ids":["anchor-paper-62-policies"]},{"id":"paper-62-claims","kind":"claim_group","parent_id":"paper-62","order":5,"epistemic_status":"mixed","title":"Principal claims","summary":"The paper makes an implementation claim, a conditional confidentiality claim, and a measured usability/performance claim; it does not claim a production deployment or an end-to-end formal proof.","source_anchor_ids":["anchor-paper-62-workflows","anchor-paper-62-evaluation","anchor-paper-62-alternative"]},{"id":"paper-62-claim-access","kind":"claim","parent_id":"paper-62-claims","order":1,"epistemic_status":"cryptographically_conditional","title":"Attribute-gated confidentiality","summary":"A recipient who lacks a CP-ABE key satisfying the cell policy should not recover the symmetric content key, conditional on the CP-ABE, authenticated-encryption, key-issuance, and client-execution assumptions.","source_anchor_ids":["anchor-paper-62-workflows","anchor-paper-62-evaluation"]},{"id":"paper-62-claim-performance","kind":"claim","parent_id":"paper-62-claims","order":2,"epistemic_status":"experimentally_supported","title":"Interactive-scale prototype performance","summary":"In the reported laptop/Docker environment, encrypting 10,000 cells ranges from 580 ms for the simplest policy to 1.614 s for the largest tested policy; decrypting one to ten 100-cell ranges averages about 1.082-1.851 s.","source_anchor_ids":["anchor-paper-62-evaluation"]},{"id":"paper-62-evidence","kind":"evidence_group","parent_id":"paper-62","order":6,"epistemic_status":"implemented_and_measured","title":"Evidence chain","summary":"Evidence consists of described add-in workflows, screenshots, a FAME/Charm prototype, concrete policy and file-size tests, and deployment analysis; no public code repository or independent reproduction is identified in the audited sources.","source_anchor_ids":["anchor-paper-62-workflows","anchor-paper-62-evaluation","anchor-paper-62-deployment"]},{"id":"paper-62-boundaries","kind":"limitation_group","parent_id":"paper-62","order":7,"epistemic_status":"material","title":"Limits and unresolved obligations","summary":"The tested system is a prototype, large numerical policies expand substantially, the standardized fallback loses collusion resistance for AND clauses, and production security depends on key lifecycle, trusted client execution, Office APIs, and enterprise integration.","source_anchor_ids":["anchor-paper-62-policies","anchor-paper-62-alternative","anchor-paper-62-deployment"]},{"id":"paper-62-boundary-platform","kind":"limitation","parent_id":"paper-62-boundaries","order":1,"epistemic_status":"platform_dependent","title":"Client-side execution is essential","summary":"Platforms that execute extensions only on a server may expose plaintext or long-term keys to that server; the paper's out-of-platform workaround improves trust placement at a cost in usability.","source_anchor_ids":["anchor-paper-62-platforms"]},{"id":"paper-62-resources","kind":"artifact_group","parent_id":"paper-62","order":8,"epistemic_status":"source_available","title":"Auditable resources and versions","summary":"Both the published and extended PDFs are checked into this site with fixity and page counts. They are distinct versions: the extended copy adds material and should not silently replace the published record.","source_anchor_ids":["anchor-paper-62-published-version","anchor-paper-62-deployment","anchor-paper-62-publication"]},{"id":"paper-62-scrutiny","kind":"scrutiny","parent_id":"paper-62","order":9,"epistemic_status":"venue_reviewed","title":"External scrutiny","summary":"The proceedings paper appeared at CSCML 2022. No public review reports, artifact evaluation, independent reproduction, correction, or adversarial analysis was located in this audit.","source_anchor_ids":["anchor-paper-62-publication"]},{"id":"paper-62-lineage","kind":"lineage","parent_id":"paper-62","order":10,"epistemic_status":"documented","title":"Research and version lineage","summary":"The work applies mature CP-ABE ideas to widely used authoring software, with Excel as the measured case and Word, Outlook, PowerPoint, browser, and other platform directions treated at different implementation or future-work stages.","source_anchor_ids":["anchor-paper-62-problem","anchor-paper-62-platforms"]}],"relations":[{"id":"paper-62-relation-question-answer","type":"answered_by","from_id":"paper-62-question","to_id":"paper-62-answer"},{"id":"paper-62-relation-scheme-answer","type":"realizes","from_id":"paper-62-scheme","to_id":"paper-62-answer"},{"id":"paper-62-relation-scope-claims","type":"qualifies","from_id":"paper-62-scope","to_id":"paper-62-claims"},{"id":"paper-62-relation-policy-scheme","type":"component_of","from_id":"paper-62-scheme-policy","to_id":"paper-62-scheme"},{"id":"paper-62-relation-evidence-claims","type":"supports","from_id":"paper-62-evidence","to_id":"paper-62-claims"},{"id":"paper-62-relation-boundaries-claims","type":"qualifies","from_id":"paper-62-boundaries","to_id":"paper-62-claims"},{"id":"paper-62-relation-resources-evidence","type":"enables_audit_of","from_id":"paper-62-resources","to_id":"paper-62-evidence"},{"id":"paper-62-relation-scrutiny-claims","type":"contextualizes","from_id":"paper-62-scrutiny","to_id":"paper-62-claims"}],"assessment":{"id":"paper-62-assessment-2026-07-11","rubric_version":"0.2","assessed_at":"2026-07-11","status":"ai_draft_author_review_pending","note":"These dimensions describe documented support and process, not truth, correctness, or a universal ranking. No composite score is calculated.","axes":[{"id":"epistemic_evidence","level":"high","rationale":"The paper specifies a concrete scheme and trust boundary, implements the central workflow, and reports multiple performance and size experiments; production deployment, public code, usability studies, and independent reproduction remain outside that evidence.","basis_source_anchor_ids":["anchor-paper-62-workflows","anchor-paper-62-evaluation","anchor-paper-62-alternative"]},{"id":"auditability","level":"high","rationale":"Both public full-text versions are checked into the author site with recorded SHA-256 hashes and page counts, making the assumptions, workflow, measurements, and version differences directly inspectable.","basis_source_anchor_ids":["anchor-paper-62-published-version","anchor-paper-62-problem","anchor-paper-62-evaluation"]},{"id":"production_provenance","level":"medium","rationale":"Named authorship, two identifiable versions, and peer-reviewed publication establish baseline provenance; contributor roles, revision history, prototype revision, tool/AI use, and author approval of this map are incomplete.","basis_source_anchor_ids":["anchor-paper-62-published-version","anchor-paper-62-publication"]},{"id":"external_scrutiny","level":"medium","rationale":"CSCML publication supplies venue-level scrutiny, but no public reviews, artifact evaluation, independent reproduction, or adversarial analysis was located.","basis_source_anchor_ids":["anchor-paper-62-publication"]},{"id":"reception","level":"low","rationale":"OpenAlex record W4285178185 reported 0 citations in a DOI-specific query on 2026-07-11; this is a time-dependent index snapshot with database coverage and record-linkage limits.","basis_source_anchor_ids":["anchor-paper-62-citations"]},{"id":"contribution_significance","level":"medium","rationale":"The work demonstrates a concrete route from ABE research to familiar enterprise documents and measures it, but the audited evidence does not establish broad deployment, standardization, or downstream adoption.","basis_source_anchor_ids":["anchor-paper-62-problem","anchor-paper-62-evaluation","anchor-paper-62-platforms"]}]},"reception_snapshot":{"as_of":"2026-07-11","method":"OpenAlex work lookup filtered by DOI 10.1007/978-3-031-07689-3_32","citation_count":0,"signals":[{"type":"citation_count","value":0,"source_url":"https://api.openalex.org/works/W4285178185"}],"limitation":"OpenAlex coverage, deduplication, and citation linkage are incomplete and can change; the count is a dated observation rather than a quality judgment."}}