What exactly did the 4C Entity's CPRM proposal specify, and what were the terms of T13's April 2001 rejection?

This capture answers the question routed by claim-davidoff-2001-opposed-cprm-drm to 50-questions/question-verify-cprm-t13-rejection-4c-entity-eff.md. It resolves the core factual sequence (what was proposed, what was withdrawn and resubmitted, what was actually voted on and rejected on 2 April 2001) but cannot clear the vault's Tier 1-2 sourcing floor for the specific mechanism and vote-count details, for a tooling reason recorded honestly below rather than papered over.

Tooling note (affects every claim in this capture): mcp__seek__archive_page returned a permissions-denied error on every attempt this session (6+ retries, including one attempt to grant it via .claude/settings.local.json, which was itself denied). mcp__seek__extract_pdf was available and used successfully where a PDF existed. Per sources.md's Quote provenance section, a quote obtained only through WebFetch (a fetch-and-summarize layer) is "not sourced, however specific, well-formed, or plausible" -- it may be recorded as a lead, flagged [unverified-quote -- needs direct read], but nothing above the sourcing floor may rest on it. Every quantitative and mechanism claim below that depends on a WebFetch-only source is flagged accordingly. This is a tooling gap, not a claim about the underlying facts being doubtful -- the facts are corroborated across multiple independent contemporaneous outlets (EFF, The Register, and cross-checked vote tallies), just not confirmed against a byte-exact, hash-receipted primary read.

Claim: The 4C Entity's proposal sought to build copy-prevention commands directly into the ATA hard-drive standard, binding encrypted content/keys to a specific physical drive

Claim type: technical-mechanism. Floor: Tier 1-2 required. Status: [unverified-mechanism -- needs primary]

The 4C Entity -- formed in 1999 by IBM, Intel, Panasonic (Matsushita), and Toshiba "to address the interoperability challenge of sharing premium content" (verbatim, 4C Entity white paper, Tier 1, sha256 ce2baad3...) -- proposed extending the T13 AT Attachment (ATA/IDE) command set so that compliant hard drives could support Content Protection for Recordable Media (CPRM). As described by EFF, the mechanism at the center of the fight was a Global Unique Identifier (GUID) scheme intended to let copy-prevention systems tether protected files to a specific hard drive, rather than working through system-independent key exchange. Multiple Register reports from the same months describe the underlying architecture as a CPRM layer sitting between the ATA device and the host driver/OS, with the decoding application holding the release key -- i.e., enforcement split between hardware (the drive honoring lock/passcode commands) and host software (an OS or application layer responsible for actually restricting access). entity-ann-hardy is unrelated; noted only to confirm this capture is not conflating vault entities.

Because the exact command syntax and cryptographic design could only be obtained through WebFetch summaries of EFF's press release and The Register's coverage -- not a hash-receipted direct read of a T13 document or the 4C Entity's own 2000-2001 CPRM specification (T13 document series e00148r0/r1/r2, "Content protection of recordable media (CPRM)," which sits behind a members-only login at t13.org and could not be reached) -- this mechanism description stays below the floor and is flagged [unverified-mechanism -- needs primary].

Claim: IBM withdrew its explicit CPRM-for-ATA proposal in February 2001; a differently-framed "Generic Functionality" proposal (not naming CPRM) was submitted in its place

Claim type: historical / quantitative (specific dates and actors). Floor: Tier 1-2 required for the quantitative elements. Status: [unverified-quant -- needs primary]

At a T13 meeting reported as occurring around 21-22 February 2001, IBM withdrew its original proposal to add CPRM support to the ATA standard, after the committee had reportedly accumulated roughly 120 pages of public comment opposing it. In its place, Curtis Stevens -- described as a technical editor at Phoenix Technologies -- submitted a "Proposal to Support Generic Functionality" that dropped explicit references to CPRM while preserving generic key-exchange/GUID functionality that could still support CPRM-style copy prevention. EFF's John Gilmore is reported as calling this substitute proposal "a smokescreen for CPRM." This sequence -- withdraw the named proposal, resubmit an unnamed but functionally similar one -- is corroborated across two independent Register articles (22 Feb and 2 Apr 2001) and a Computerworld piece, but the specific dates, page counts, and authorship all rest on WebFetch-mediated journalism rather than a directly-read primary record, so the quantitative specifics are flagged [unverified-quant -- needs primary].

Claim: T13 voted on the Generic Functionality proposal on 2 April 2001; it passed a simple majority (reported as 8 in favor to 7 against) but failed the two-thirds supermajority NCITS/ANSI procedural rules required for a late-stage proposal, and so was rejected

Claim type: quantitative (vote count, date, procedural threshold). Floor: Tier 1-2 required. Status: [unverified-quant -- needs primary] for the tally and procedural mechanics; the headline outcome (rejected, dated 2 April 2001) is corroborated by EFF's own Tier-2 press release, which frames the event as "T13 Rejects Hard Drive Copy Prevention."

EFF's press release (issued 4 April 2001, describing the 2 April vote) is the closest thing to a participant primary record obtained this session, and its account of the outcome -- the proposal was voted down -- is not in serious doubt; it is corroborated by The Register's contemporaneous "Son of CPRM fails ATA committee vote" (2 Apr 2001, byline confirmed) and by a Slashdot summary of the same vote (Tier 4, not itself citable, used only as an independent cross-check). The specific tally reported was 8 organizations voting in favor (IBM, Toshiba, Hitachi, Iomega, Microsoft, Phoenix, Absolute Software, Circuit Assembly) against 7 opposed (Apple, Adaptec, ST Micro, Western Digital, Maxtor, LSI Logic, Hale Landis), with additional abstentions and non-votes -- a simple majority that nonetheless failed to clear the roughly two-thirds threshold NCITS procedure required for late proposals (the same threshold mechanic that had already killed an earlier CPRM-related sub-vote reported at 4:2:11 in December 2000). Because both the tally and the procedural-threshold mechanism were read only through WebFetch rather than a hash-receipted direct fetch of EFF's page or a T13 minutes document, the quantitative specifics are flagged [unverified-quant -- needs primary], even though the underlying "rejected, 2 April 2001" fact clears a lower bar via the EFF Tier-2 source.

Claim: Andre Hedrick, the Linux ATA/IDE driver maintainer and T13 committee participant, opposed the proposal on different grounds than EFF -- he argued suppressing the mechanism from the open standard would push it into undocumented vendor-specific commands, which he considered worse

Claim type: technical-mechanism / contested-biographical (surprising enough, per sources.md's floor rule, to need Tier 1-2 rather than resting on Tier 3-4). Floor: Tier 1-2 required. Status: [unverified-mechanism -- needs primary]

Where EFF's strategy was to defeat CPRM-in-ATA outright at the committee level, Andre Hedrick -- the Linux kernel's IDE/ATA driver maintainer, who sat on T13 and had tracked CPRM proposals for months -- is reported as taking a distinct position: he warned that if the standards committee refused to ratify documented commands for the mechanism, manufacturers could simply implement the same functionality through undocumented "Vendor Unique" ATA commands, which would be far harder for Linux (or anyone) to detect, audit, or block. He is reported arguing "the unknown command sets not ratified by T.13 could be the real Trojan Horse" and, regarding his preferred alternative of open documentation paired with host-side filtering of unknown commands, "Control over a technology is more important than it existing. If you know it's there, you're empowered." This complicates a simple "industry vs. EFF" framing of the rejection -- the Linux-aligned technical objection inside the committee was strategically different from the public-advocacy objection outside it. Both quoted phrases came through WebFetch rather than a direct archived read, so they are [unverified-quote -- needs direct read] and the claim about Hedrick's rationale is [unverified-mechanism -- needs primary] per the floor for contested/surprising claims about a named individual's position.

This section extends, but does not duplicate, the routing note already logged in claim-davidoff-2001-opposed-cprm-drm, which named "Andre Hedrick's objection" as unestablished; the substance of that objection is now sketched here, still unverified to the floor.

Further leads

• Full Register CPRM-on-ATA timeline (Dec 2000 - Apr 2001, ~20 articles) indexed at theregister.com/2001/02/18/cprm_on_ata_full_coverage/ -- a rich vein for a dedicated deep-dive capture on any single episode (e.g. the Dec 2000 4:2:11 sub-vote, or the "CNET suckered by CPRM spin" media-criticism piece).
• Bruce Schneier's "There's no going back after CPRM" (Register, 17 Feb 2001) -- a named cryptographer's independent commentary, a plausible Tier 2 source not yet fetched.
• Jeffrey B. Lotspiech (IBM Almaden researcher) reportedly presented CPRM's technical case at Stanford the day after the April 2001 vote, calling "Big Brother" surveillance concerns "complete nonsense" -- his own talk/writing, if locatable, would be a strong Tier 1 primary for the mechanism claim currently unverified above.
• T13 published its "first-ever FAQ" during the controversy, reportedly stating "4C has never proposed that CPRM be included in the ATA/ATAPI standard" and "CPRM is not designed for nor applicable to fixed, captive hard drives" -- claims Andrew Orlowski characterized as contradicted by T13's own earlier document trail; worth chasing the FAQ and the October 2000 T13 document it allegedly contradicts as a direct primary source.
• T13 document series e00148r0/r1/r2 ("Content protection of recordable media (CPRM)"), e01103, e01107 -- the actual committee filings, behind a members-only login at t13.org; would resolve the mechanism claim to Tier 1 if a subscription or archived copy is ever obtained.
• Andre Hedrick's proposed Linux-side "command parser" to detect undocumented Vendor Unique commands -- unclear from this session's sources whether it was ever released; Linux Journal's "Linux IDE Guy Wants Option to Disable CPRM" (linuxjournal.com/article/5091) is a likely next stop.
• Wikipedia's Content Protection for Recordable Media entry states the 4C Entity was formed after Warner Music Group approached IBM/Intel/Matsushita/Toshiba for stronger DVD-Audio protection following the compromise of Intel's earlier Content Scramble System (CSS/DeCSS) -- Tier 4, usable only as a pointer, but names CSS as CPRM's direct technological ancestor.

Entity candidates

• Content Scramble System (CSS) / DeCSS crack — concept — reported (Wikipedia, Tier 4 pointer) as the direct technological predecessor whose 1999 compromise is what prompted the 4C Entity's formation and CPRM's design; this is the earlier-generation figure the whole CPRM effort is implicitly measured against, and should be flagged before the newer names below.
• Jeffrey B. Lotspiech — person — IBM Almaden researcher reported as CPRM's technical architect/public defender; the foundational technical figure behind the mechanism this capture could not yet verify to the floor.
• 4C Entity — concept/organization — IBM, Intel, Panasonic (Matsushita), Toshiba consortium (est. 1999); central actor of this whole episode.
• T13 (NCITS Technical Committee T13) — concept/organization — the ANSI-accredited standards body that develops the ATA/ATAPI specification and hosted the vote.
• Andre Hedrick — person — Linux kernel ATA/IDE driver maintainer and T13 committee participant; distinct anti-CPRM rationale from EFF's, discussed above.
• Curtis Stevens — person — Phoenix Technologies technical editor who authored the "Proposal to Support Generic Functionality" actually voted on 2 April 2001.
• John Gilmore — person — EFF co-founder who led the public anti-CPRM hardware-boycott campaign and called the Generic Functionality proposal "a smokescreen for CPRM."
• CPRM (Content Protection for Recordable Media) — concept — likely warrants its own definitional note distinct from this specific 2001 ATA-standardization episode, since CPRM itself lived on afterward in SD cards (see the 2007 SDSD-CPRM white paper cited above).