US 6995870

I'll verify the current status of US 6,995,870 against live sources before summarizing.

The CAFC docket search returned nothing on point. Let me try more targeted queries for the claim set and any 2026 appellate activity.

Let me try to retrieve the actual claim text and check for any 2025–2026 appellate activity.

US 6,995,870 — verified findings

Important scoping note: every search I ran was keyed to the literal string 6995870 / 6,995,870. Where results below mention other RAH patents (e.g., 7,761, 8,416,444, 7,729,008, 7,312,897, 8,760,704), I say so explicitly and do not attribute their content to the '870 patent.

1. Bibliographic data (as listed on the patent record)

Source: https://patents.google.com/patent/US6995870/en (the authoritative text supplied to me). The chain of divisionals is confirmed by the front page of the related continuation US 8,760,704 (https://patentimages.storage.googleapis.com/55/bf/c8/d19b5ef2ed6bad/US8760704.pdf), which recites: "…continuation of application No. 10/040,664, filed on Jan. 7, 2002, now Pat. No. 6,995,870, which is a division of application No. 09/229,002… which is a division of application No. 08/606,883, filed on Feb. 26, 1996, now Pat. No. 6,043,909."

Two discrepancies worth flagging rather than resolving:

• The Unified Patents patent page for US-6995870-B2 lists the priority date as 1996-02-25, whereas the Google Patents record and the '704 front page give 1996-02-26. This is likely a time-zone/derivation artifact, but I am not auto-correcting either value.
• The "adjusted expiration 2017-05-07 / Expired – Fee Related" entry is Google's own legal-status assumption ("not a legal conclusion"), and I could not independently confirm the terminal date. Treat it as unverified.

2. Abstract

The '870 abstract, as reflected in the patent text and its co-specification family member US 8,760,704:

"The system provides for controlling color reproduction of input color image data in a network having nodes (or sites). The system distributes the input color image data from one of the nodes to other nodes, and provides data structures in the network. The system has means for providing color calibration data at each node characterizing output colors (colorants) of the rendering device of the node, and means for producing at each node, responsive to the color calibration data of the rendering device of the node, information for transforming the input color image data into output color image data at the rendering device of the node. The rendering device of each node renders a color reproduction responsive to the output color image data, wherein colors displayed in the reproduction at the rendering device of each node appear substantially the same within the output colors attainable by the rendering devices."

(Quoted from the family text at https://patentimages.storage.googleapis.com/55/bf/c8/d19b5ef2ed6bad/US8760704.pdf; the '870 specification contains the same "Summary of the Invention" language, e.g. '870 at 8:27–34.)

3. Independent claims — what I can and cannot verify

Caveat, stated plainly: the patent text supplied to me is truncated inside the description and does not include the claims section, and none of my searches returned the verbatim claim set of the '870 patent. I therefore cannot give you a reliable claim-by-claim walkthrough of each independent claim, and I will not guess at claim numbers or wording. What I can ground is the following:

(a) Claim terms actually construed for the '870 patent. In the MDL proceeding, the district court's claim construction order in RAH Color Technologies LLC v. Dalim Software GmbH et al., N.D. Cal. Case No. 3:18-cv-03277-SI, Document 120 (filed 01/22/2021), stated:

"The parties have identified three terms requiring claim construction that are in the '870 and '444 patents: (1) 'sites' ('870 and '444 Patents); (2) 'appear substantially the same' ('870 Patent); and (3) 'interactive conference' ('444 Patent)."

Source: https://storage.courtlistener.com/recap/gov.uscourts.cand.[327437](/patent/327437)/gov.uscourts.cand.327437.120.0.pdf

This confirms that the asserted '870 claims (i) are directed to multiple "sites" and (ii) require that reproduced colors "appear substantially the same," limited by the "output colors attainable by the rendering devices." The same order records RAH's expert testimony that "appear substantially the same" is measured through the patent's verification procedures and confidence intervals ('870 at 42:41–47, 42:53–62) — i.e., a tolerance-based, not exact-match, standard.

(b) Plain-language gist of the claimed subject matter, tied to the specification. On the record I retrieved, the independent claim coverage of the '870 patent appears to be a networked/multi-site color reproduction control system: a computer at a site with memory storing color reproduction information (calibration/transformation data, tonal transfer curves, color transformations, gamut data), communicating over a network interface with at least one other site so that the shared information is used by a color input or output device at that other site. I could not verify how many independent claims the '870 patent has, nor which of them are system vs. method vs. user-interface claims, because the claim text was not retrievable in this session. Any statement I made about specific claim numbers would be fabrication.

Related-family context (explicitly not the '870 patent's claims): prosecution-paper excerpts surfaced in PTAB filings show claims 55–77 and 72–89 style language in the later continuation applications (e.g., claim 72, "A system for controlling color reproduction at multiple sites comprising: a computer at a site; memory storing information and programs… a network interface enabling communication of said information by said computer with at least one other site…"). Those excerpts come from a different application in the family and must not be read as the '870 claim set.

4. Litigation and CAFC docket check (including 2026)

• District court: the Google Patents record for the '870 patent links a large family of US cases (2010–2020), including N.D. Cal. 3:18-md-02874 (the coordinated MDL No. 2874), N.D. Cal. 3:18-cv-07464 / 07465 / 03277-adjacent filings, D. Del./E.D. Pa./W.D.N.Y./E.D. Mich./S.D. Ill./N.D. Ill./E.D. Va. and E.D. Tex. cases. Accused-party alignment discussed in the PTAB record includes Adobe, Xerox, EFI, Heidelberger, Dalim, Kodak, and Quad/Graphics (see IPR final written decision excerpts at https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1523435](/patent/1523435)/...).
• PTAB: IPR2019-00627 (US 7,729,008), IPR2019-00628 (US 8,416,444), IPR2019-00629 (US 7,312,897), IPR2020-00646 (US 7,791,761) — these are related RAH patents, not the '870 patent.
• CAFC: the only Federal Circuit matters I could confirm involving this family are Adobe Inc. v. RAH Color Technologies LLC, Appeal Nos. 2021-1067 and 2021-1068 (appeal from the PTO; 2021-1068 filed 2020-10-20, terminated 2021-09-22; oral-argument audio posted at https://www.cafc.uscourts.gov/9-02-2021-2021-1067-adobe-inc-v-rah-color-technologies-llc-audio-uploaded/). I also note the Unified Patents entry showing RAH Color Tech LLC as assignee for several family members with priority cited as 1997-08-24 (https://portal.unifiedpatents.com/patents/patent/US-[5319437](/patent/5319437)-A).
• 2026 docket specifically: I found no 2026 CAFC docket, appeal, or other appellate proceeding involving US 6,995,870. My searches for "RAH Color Technologies" + 2025/2026 Federal Circuit activity returned nothing on point. Given the patent is listed as expired (adjusted expiration 2017-05-07), the absence of current appellate activity is consistent, but I want to be clear that this is a negative search result, not a confirmed all-clear — a dedicated PACER/CAFC docket query would be needed to rule out recent filings definitively.

5. Explicit uncertainties

1. Claim text of the '870 patent is unverified — I could not retrieve it, so no independent-claim summary is offered beyond the construed terms "sites" and "appear substantially the same."
2. Number and categorization of independent claims is unknown to me.
3. The "expired 2017-05-07" expiration date derives from Google's automated status, not from an authoritative USPTO term calculation.
4. The 1996-02-25 vs. 1996-02-26 priority-date discrepancy between Unified Patents and Google Patents/the '704 front page is unreconciled.
5. Assignee string varies across sources ("Individual" as original assignee; "RAH Color Tech LLC" as current assignee; a 2005-12-19 assignment to RAH Color Technologies, LLC and a 2013-05-14 merger record to RAH COLOR TECHNOLOGIES LLC).

I'll research litigation involving US patent 6995870. Let me start with the sources you mentioned.

I have substantial data. Let me gather more detail on specific cases, outcomes, and the PTAB/CAFC proceedings.

Let me check the Stanford NPE litigation database for this specific patent and look for case outcomes.

Litigation Involving U.S. Patent No. 6,995,870

Important caveat up front: I was able to retrieve a large volume of primary-source material (complaints, PTAB filings, MDL orders) and Unified Patents/Google Patents litigation metadata for this patent, but I hit a tool-use limit before I could independently confirm the final disposition of every case. The status entries below are therefore a mix of (a) verified facts and (b) reasonable inferences from docket metadata. Where I am uncertain, I say so explicitly.

The Patent and the Plaintiff

• Patent: U.S. Patent No. 6,995,870 B2, "System for Distributing and Controlling Color Reproduction at Multiple Sites." Priority date 1996‑02‑26; filed 2002‑01‑07; granted 2006‑02‑07. (One complaint in the record contains the typo "6,999,870," but the correct identifier is 6,995,870.)
• Owner/Plaintiff: RAH Color Technologies LLC (Alexandria, Virginia), formed 2005 by Dr. Richard A. Holub, the sole named inventor. RAH Color Technologies is the record assignee (assignment recorded 2005‑12‑19; later a "merger" assignment in 2013).
• Field: Color management — print servers, proofing devices, presses, color imaging software.

District Court Litigation

The '870 patent was asserted (or placed at issue) in a large family of suits filed by RAH Color Technologies, nearly all of which were consolidated into MDL No. 2874, In re: RAH Color Technologies LLC Patent Litigation, N.D. Cal. Case No. 3:18‑md‑02874‑SI (Judge Susan Illston). The MDL was later closed by an order of final judgment (per the JPML docket, "***THIS MDL IS CLOSED").

Note on the designations above: the case numbers, courts and filing dates in the table come directly from the patent's own litigation metadata (Unified Patents, via Google Patents — https://patents.google.com/patent/[US6995870](/patent/US6995870)/en) cross-checked against the underlying complaints and JPML/PTAB filings. Several metadata entries could not be resolved to a named defendant, and two entries show a court that conflicts with the underlying court document; I have flagged those rather than silently "correcting" them.

PTAB (IPR) Proceedings — Important Distinction

The Adobe‑filed inter partes reviews did not challenge the '870 patent itself. They challenged four related RAH patents:

• IPR2019‑00627 — U.S. 7,729,008
• IPR2019‑00628 — U.S. 8,416,444
• IPR2019‑00629 — U.S. 7,312,897
• IPR2019‑00646 — U.S. 7,791,761

Petitioner: Adobe Inc. (Perkins Coie LLP, lead counsel Christopher L. Kelley). Patent Owner: RAH Color Technologies LLC (Global IP Law Group). Filed 2019‑02‑01; consolidated oral hearing held 2020‑05‑28. The IPR2019‑00627 Final Written Decision rejected Adobe's challenge (claims 28–33, 36–38 of the '008 patent not shown unpatentable). I did not find any IPR petition directed to claim(s) of U.S. 6,995,870 itself. The '870 patent nevertheless appeared throughout these IPRs as a "related matter" because it was co-asserted in the district court litigation.

I also could not confirm any Federal Circuit (CAFC) appeal specifically involving the '870 patent from the materials I retrieved; I therefore cannot state one exists.

Outcome / Current Status

• MDL No. 2874 is closed; a final judgment order was entered by Judge Illston in the N.D. Cal. (JPML docket: "THIS MDL IS CLOSED").
• Settlement: The MDL orders show at least the Eastman Kodak case was stayed pending finalization of a settlement, and RAH's complaints state that RAH "has licensed the technology covered by its patents to five of the largest manufacturers of color imaging and printing products… and an additional manufacturer… [and] 13 major companies have entered into end‑user license agreements." The strong inference (which I could not independently verify for each defendant) is that most of the '870 suits resolved by license/settlement or voluntary dismissal rather than trial.
• Patent status: Per the patent record, U.S. 6,995,870 is Expired – Fee Related, with an adjusted expiration of 2017‑05‑07. That maintenance-lapse/expiration date is material to the value of any surviving claims and is consistent with the closure of the litigation campaign.
• Claim construction: In the Adobe case (3:18‑cv‑03277‑SI, order filed 2021‑01‑22), the court construed the terms "sites" ('870 and '444), "appear substantially the same" ('870), and "interactive conference" ('444).

Sources (grounding)

• Google Patents litigation metadata for the patent: https://patents.google.com/patent/US6995870/en
• Unified Patents litigation portal entries linked from the patent page (e.g., https://portal.unifiedpatents.com/litigation/Illinois%20Northern%20District%20Court/case/1%3A17-cv-00894)
• JPML / MDL No. 2874 dockets and Pretrial Orders (N.D. Cal. 3:18‑md‑02874‑SI)
• PTAB IPR petitions and Final Written Decision, Adobe Inc. v. RAH Color Technologies LLC (IPR2019‑00627/‑00628/‑00629/‑00646)
• Complaints: RAH v. R.R. Donnelley & Sons (1:17‑cv‑00894); RAH v. Quad/Graphics (2:18‑cv‑00087); RAH v. Xerox (1:17‑cv‑06813); RAH v. Kodak (6:19‑cv‑06112); RAH v. Ricoh (E.D. Pa.); Electronics For Imaging v. RAH (N.D. Cal. 3:18‑cv‑01612 and E.D. Va. 1:18‑cv‑01436)
• Stanford NPE Litigation Database, RAH Color Technologies LLC party page

If you need, I can try to pull the specific docket sheets (via CourtListener/RECAP or PACER) to pin down the final disposition and any settlement/license terms for each individual case — that is the one area where my retrieved evidence is thinner than I would like.

I'll start by verifying the structured data against what web sources show for this patent.

Let me run targeted searches to confirm whether the '870 patent itself ever attracted an IPR petition.

Let me check whether any IPR was ever filed against the '870 specifically, and look at the outcome of the related '444 IPR that was filed around the same time.

PTAB Proceedings Analysis — U.S. Patent No. 6,995,870 B2

"System for distributing and controlling color reproduction at multiple sites"
Patent Owner / Current Assignee: RAH Color Technologies LLC · Inventor: Richard A. Holub · Priority: 1996-02-26 · Granted: 2006-02-07

Proceedings overview

Total AIA trial proceedings on file against US 6,995,870: zero (0). Breakdown: 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denials. The USPTO Open Data Portal returns no IPR, PGR, or CBM proceeding for this patent, and my independent web search surfaced no evidence that any petition was ever filed against the '870 — no institution decision, no FWD, no termination, and no Rule 36 or merits appeal. The bottom-line defensive posture is therefore unusual and defendant-favorable in one specific sense: the '870's asserted claims (34, 37, 39, 41, 42, 43) have never been tested at the PTAB, so this is not a "hardened by IPR" patent — but it is also an expired patent (adjusted expiration 2017-05-07 per the patent's legal-status record), which caps any forward-looking infringement exposure at essentially nothing. The PTAB silence on the '870 is not for lack of assertion: it was asserted in six-plus district court cases, yet every IPR petition in the RAH campaign was aimed at four different patents in the family.

Caveat on the null result: the absence is sourced from the structured ODP block plus search corroboration. I did not have direct PTAB E2E docket access in this session. If you are relying on this for a filing, confirm against the PTAB E2E "Patent Number Search" for 6995870 at https://ptacts.uspto.gov/ptacts/.

For each proceeding

None to report. There are no {PROCEEDING_NUMBER} — {Petitioner} v. {Patent Owner} entries because no AIA trial was ever instituted, denied, or terminated as to the '870.

Why the null set is credible, not a data gap. The RAH Color Technologies MDL (In re RAH Color Technologies LLC Patent Litigation, N.D. Cal. MDL No. 2874, Case No. 3:18-md-02874-SI) produced a well-documented IPR cluster filed by Adobe Inc. on 2019-02-01, and the '870 is conspicuously absent from it. Petitioner Adobe's own Exhibit 1037 claim-mapping chart — reproduced in the public IPR record — lists the asserted-and-challenged claims per patent and shows the 6,995,870 row blank in the "Claims Against Adobe & Challenged in Adobe's IPRs" column, while the same row shows '870 claims 34, 39, 43 asserted against Xerox, 34/39/43 against EFI, 34/39/41/43 against Heidelberger, 34/39/41/42 against Dalim, and 34/39/41/42/43 against Kodak.

Related-patent PTAB activity (context only — NOT proceedings on the '870)

These do not invalidate, narrow, or estop anything as to the '870, but they define the litigation landscape you will be negotiating inside:

All four were consolidated for a single oral hearing on 2020-05-28. Sources: the '008 FWD public version at https://bannerwitcoff.com/wp-content/uploads/2020/09/PTAB-IPR2019-00627-124.pdf ; the '897 FWD at https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1523317](/patent/1523317)/download-documents?artifactId=dWQYzSy8yHtdm60fwWOQmZPRzytdgelbTrRIHzgiWlhj9E-y-x2wwRw ; docket index at https://www.docketalarm.com/search/?q=patent%3A%287791761%29 .

Strategic summary

What is CANCELED vs. SUSTAINED vs. UNTESTED as to the '870. Nothing is canceled and nothing is sustained — every claim of the '870 is untested before the PTAB. For accuracy and to avoid any risk of mis-citation in a brief: claims 32 and 33 that Adobe successfully killed in IPR2019-00629 are claims of U.S. 7,312,897, a different patent in the same family. Do not let anyone (including a plaintiff's demand letter) blur the RAH patents together. The '870 claims actually asserted in the MDL were 34, 37, 39, 41, 42, and 43 (claim 34 as the recurring independent method claim, with 37, 39, 41, 42, 43 as dependents asserted across Xerox, EFI, Heidelberger, Dalim, Kodak, and Microsoft). Those claims are alive as issued but have never been construed by the Board. In district court, Judge Susan Illston construed only two disputed terms at the 2021-01-22 claim construction order in RAH Color Techs. LLC v. Adobe Inc., N.D. Cal. No. 3:18-cv-03277-SI, Dkt. 120 — "sites" and "appear substantially the same" — and the order expressly notes "None of the PTAB decisions construed any of these claim terms." That order is available at https://storage.courtlistener.com/recap/gov.uscourts.cand.[327437](/patent/327437)/gov.uscourts.cand.327437.120.0.pdf .

Estoppel landscape. Section 315(e)(2) estoppel is claim- and proceeding-specific. Because no IPR ever reached a final written decision on the '870, no party is estopped as to any '870 claim, and the entire prior-art universe remains available as an invalidity defense in any '870 litigation. There is a second-order wrinkle worth knowing but not over-reading: RAH argued throughout the Adobe IPRs that Quad/Graphics, Dalim, Xerox, Heidelberger, and EFI were unnamed real parties-in-interest in Adobe's petitions (which, if accepted, would have imported Adobe's estoppel onto those MDL co-defendants). The Board rejected that argument both at institution and again in the '008 FWD — "Petitioner has sufficiently shown that Quad, Dalim, Xerox, Heidelberger, and EFI are not RPIs." So the estoppel did not spread, and the '870 remained fully open to challenge in the district court by every defendant who cared to raise it.

Pattern signals. (1) The same petitioner, Adobe, filed four IPRs on the same day (2019-02-01) against four RAH patents — and deliberately did not challenge the '870, even though the '870 was the most broadly asserted patent in the campaign. That is a strategic tell: the '870's asserted claims are largely method-of-use / user-interface / network-architecture claims, and Adobe's strongest art (Stone, Tajika, Beretta, Sugiura, Newman/Kuwamoto, the ICC v.3.x specs) mapped most cleanly onto the '008, '444, '761, and '897 output-device claims. (2) RAH litigated hard and appealed — it fought the RPI issue, moved for additional discovery under Garmin, filed motions to exclude, and IPR2019-00628 was appealed to the Federal Circuit as 2021-1067. (3) No defensive aggregator appears in the '870 chain. Unified Patents appears in the litigation-data links attached to this patent's Google Patents record in other contexts, but I found no Unified Patents (or RPX) petition on the '870 — the RAH challenge was defendant-driven (Adobe as lead, with MDL co-defendants Xerox, EFI, Heidelberger, Dalim, Kodak, Microsoft, and Agfa on the periphery).

The fact that matters most for a 2026 demand letter: the patent's legal-status record shows "Expired – Fee Related," adjusted expiration 2017-05-07. Twenty years from the 1996-02-26 earliest non-provisional filing date plus term adjustment lands exactly there. Treat that as Google's legal-status assumption rather than a legal conclusion (as Google itself disclaims), but verify it against the USPTO PatentCenter maintenance-fee and term-adjustment records before you rely on it. If correct, no post-2017-05-07 conduct is actionable, and § 286's six-year lookback (a suit filed today reaches back only to 2020-09-27) yields a zero recoverable-damages window for any newly filed case.

Recommended next steps

Because there is no PTAB activity on the '870, the "link to the FWD" step has nothing to link to — and that is itself the deliverable. Say it plainly to your client: no AIA trial petition has ever been filed against U.S. 6,995,870, so there is no PTAB cancellation, no claim-level FWD, and no estoppel narrowing the '870. Do not let a demand letter imply otherwise by gesturing at the Adobe IPRs; those hit the '008, '444, '761, and '897.

Concrete actions:

1. Verify the null before you rely on it. Run a PTAB E2E Patent Number search on 6995870 (https://ptacts.uspto.gov/ptacts/) and the USPTO PTAB decisions library (https://www.uspto.gov/patents/ptab/decisions). My determination rests on the ODP block in this prompt plus search corroboration; a docket-level confirmation takes ten minutes and closes the loop.

2. Lead with expiration and § 286, not with validity. If the 2017-05-07 adjusted expiration holds, the strongest motions in the case are a Rule 12 dispositive motion / summary judgment on no recoverable damages, and a laches-and-§ 286 framing on the pre-2017 conduct. Confirm the maintenance-fee lapse and PTA figures at USPTO PatentCenter (https://patentcenter.uspto.gov/).

3. If you must litigate validity, the art is wide open. No § 315(e)(2) estoppel constrains any '870 ground. The Adobe IPR record is a free R&D library: the Stone, Tajika, Beretta, Sugiura, Newman/Kuwamoto, ICC v.3.x, Scott-Taggart, and Herzog references — plus the Poynton, Munford, Green, and Vrhel expert declarations — were all developed and publicly filed in IPR2019-00627/-00628/-00629/-00646 and are reusable. Note the authenticity fights the Board actually resolved (Stone as Fed. R. Evid. 902(6) periodical prior art, Munford's library-authentication declarations, the ICC v.3.0/v.3.01 public-accessibility dispute, and Adobe's admission that an ICC v.3.0 appendix was downloaded from the Internet rather than a USPTO file history) — that history tells you exactly which authentication attacks will and won't land.

4. If an active PTAB trial somehow exists and is unindexed, the statutory clock is fixed: institution decision within 6 months of the petition's filing/accord date, FWD within 12 months of institution (35 U.S.C. §§ 314(b), 316(a)(11)), with the CAFC appeal following. Monitor PTAB E2E and the Federal Circuit docket (https://cafc.uscourts.gov/) and CourtListener (https://www.courtlistener.com/?q=RAH+Color+Technologies) for any late-filed petition or appeal.

5. Watch the family, not just this patent. The Adobe appeals and the surviving '008 and '761 claims are where RAH's remaining leverage lives. IPR2019-00628 ('444) went up to the Federal Circuit as 2021-1067 — pull that disposition, because it may bear on claim-construction positions (e.g., "interactive conference," "sites") that RAH could try to port over to the '870.

What I do not know with high confidence, stated explicitly: I could not independently retrieve the PTAB E2E docket page for 6995870 in this session, the full text of the IPR2019-00628 and IPR2019-00646 final written decisions (including which specific claims survived), or the disposition of CAFC Appeal No. 2021-1067. I have not asserted any claim-level outcome for those proceedings beyond what the sources quoted above state verbatim. Treat the '870 expiration date as a legal-status data field to be verified, not as a legal conclusion.

I'll research this patent's ownership chain before writing up the analysis.

Let me try to pin down the actual assignment records and correspondence details.

Ownership Analysis — US 6,995,870 B2

Patent: US 6,995,870 B2 — "System for distributing and controlling color reproduction at multiple sites"
Appl. No. 10/040,664 · filed 2002-01-07 · issued 2006-02-07 · priority 1996-02-26 (continuation of App. 08/606,883, which issued as US 6,043,909)
Legal status: Expired – Fee Related, expiration adjusted to 2017-05-07

Sourcing note (read first): The authoritative text provided for this patent (Google Patents, fetched 2026-09-27) exposes two post-issuance legal events but does not print reel/frame numbers or the recording correspondent. I attempted to retrieve the Assignment Center reel/frame images and was unable to complete that retrieval in this session. Rather than invent reel/frame identifiers, I have marked those fields UNVERIFIED below and flagged exactly what must be confirmed at the Assignment Center. Everything else is grounded in the patent text and the litigation/IPR record cited inline.

Inventors

Employment history (from his own testimony and RAH's complaints, e.g. RAH Color Techs. v. X-Rite, N.D. Ill./W.D. Mich. and Adobe v. RAH Color Techs. IPR2019-00627/00628/00629/00646, Ex. 2124/2238):

• Eikonix Corp. → acquired by Eastman Kodak, 1983–1990
• Agfa/Bayer (Compugraphics), 1990–1993
• SuperMac Technologies, 1993–1994

Unusual-pattern note: This is a single-inventor patent, so there is no "co-inventors departing within 12 months" pattern to evaluate. The relevant pattern here is the inverse: the inventor never assigned to an operating employer at all — he filed pro se/in his own name as an individual, and the patent-holding entity was created by him, not acquired by him. His commercialization vehicle, Imagicolor Corporation (formed 1998), was dissolved when investment did not materialize; he kept the patents personally and formed the licensing LLC in 2005. That "failed operating startup → inventor-owned patents → purpose-built licensing LLC" arc is the structural signature of an individual-inventor-started NPE, not a corporate fire-sale.

Original assignee

On the issued patent the assignee of record is "Individual" — i.e., Richard A. Holub personally. There was no corporate original assignee. Current assignee is RAH Color Technologies LLC (spelled "RAH Color Tech LLC" in the Google Patents metadata).

• Primary line of business of the original owner: none. Holub was a solo researcher developing a prototyped color-instrumentation + soft-proofing system; he rented lab/demo space from RIT Research Corp. and hired RIT students and contractors.
• Did they ship a product embodying the claims? No. Holub's own declaration (IPR2019-00646, Ex. 2121) and RAH's complaints (e.g. ¶¶18–22 of the N.D. Ill. complaints) state: "commercialization of the prototype did not come to fruition"; Imagicolor was "eventually dissolved." The technology was never commercialized by the owner.
• Status of the original owner/entity: Inventor alive and active (deposed Feb 2020 in the Adobe IPRs). The commercialization entity, Imagicolor Corporation, is dissolved. The patents are now held by an entity Holub himself controls.

Assignment timeline

The sources available to me document two post-issuance recorded events (these are the Google Patents "legal events," which mirror the same Assignment Center records):

1. 2005-12-19 (recorded; execution date not shown) — Reel/frame UNVERIFIED

• Conveyance: Assignment (Assignment of Assignor's Interest — "see document for details")
• Assignor: Richard A. Holub (individual)
• Assignee: RAH Color Technologies, LLC
• Correspondent: UNVERIFIED — not exposed in the retrieved records. (Separately, the recurring prosecution lawyer across the whole family is Martin LuKacher, later Kenneth Jay Lukacher (USPTO Customer No. 24902), attorney docket prefix "ML-0376…" — see Adobe IPR Ex. 2122 and EFS receipts. That is a prosecution-side recurrence, not a confirmed assignment correspondent; do not conflate the two without pulling the reassignment cover sheet.)
• Context: Transfer-to-asserter / licensing-entity formation. RAH Color Technologies LLC was formed in 2005 — per RAH's own pleadings, "as a vehicle for an on-going licensing program" — and this 2005-12-19 recording is the assignment of the patent from the inventor to that vehicle. Recording occurred in the same month the LLC's licensing program began.

2. 2013-05-14 (recorded) — Reel/frame UNVERIFIED

• Conveyance: Merger
• Assignor: RAH Color Technologies LLC
• Assignee: RAH Color Technologies LLC
• Correspondent: UNVERIFIED
• Context: Internal reorganization / merger, not an arm's-length transfer. Note the anomaly: the assignor and assignee carry the same name, which is consistent with a same-name merger/state conversion (or a corrective re-recording) rather than a sale to a third party. The conveyance type is recorded literally as "MERGER." This must be confirmed against the reel/frame image before being characterized with certainty.

No further recorded assignments appear after 2013-05-14. The chain therefore terminates at RAH Color Technologies LLC, the entity that has been the plaintiff in every infringement action.

If, on independent verification at the Assignment Center, additional links exist (e.g. a security interest or an unreleased lien), add them — but nothing in the retrieved record shows any transfer to a third-party aggregator, and there is no recorded release, license, or change-of-name entry.

Timeline diagram

timeline
    title Ownership of US 6995870
    1996 : Priority application filed by Holub
    1998 : Imagicolor Corporation formed and later dissolved
    2002 : Continuation filed as app 10 040 664
    2005 : RAH Color Technologies LLC formed
         : Assignment recorded to RAH LLC
    2006 : Patent US 6995870 issues
    2010 : First infringement suit vs Seiko Epson
    2013 : Same-name merger recorded for RAH LLC
    2017 : Patent expires
         : Suits vs RR Donnelley and Xerox
    2018 : Multidistrict litigation consolidated
    2019 : Adobe inter partes reviews filed

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT.
The patent moved from the individual inventor to RAH Color Technologies LLC, an entity formed in 2005 whose only stated purpose is licensing. Concrete evidence, not naming inference: (a) RAH's own complaints state the LLC was "formed as a vehicle for an on-going licensing program"; (b) Holub declares he is "the sole member of RAH Color Technologies LLC" (IPR2019-00646, Ex. 2121 ¶1); (c) the entity maintains a single stated office at 7012 Colgate Drive, Alexandria, VA 22307 (a residential address, not a plant or campus), per the X-Rite complaint ¶1; (d) no product embodying the claims was ever shipped. Mitigating nuance: the transferor was an individual inventor, not an operating company, so this is the "individual-inventor-started" variant rather than the classic corporate spin-out.

2. Known asserter in the chain — PRESENT.
RAH Color Technologies LLC is listed in the Stanford NPE Litigation Database as an asserter (categorized "Individual-inventor-started"), and this patent's case is tracked there (see the RAH-v.-Adobe MDL entry tied to the portfolio, https://npe.law.stanford.edu/patent/[7710560](/patent/7710560)). The Google Patents record itself carries ~20 Unified Patents litigation entries for this family. The '870 patent was asserted in, at minimum: RAH v. RR Donnelley (N.D. Ill. 1:17-cv-00894, filed 2017-02-02 — '870 expressly named); RAH v. Quad/Graphics (E.D. Wis. 2:18-cv-00087, filed 2018-01-17 — '870 expressly named); RAH v. Xerox (W.D.N.Y. 6:18-cv-06746); and the consolidated MDL In re RAH Color Technologies LLC Patent Litigation, 3:18-md-02874 (N.D. Cal.). It is not an Acacia/Marathon/IV-style mass aggregator, so this is a "patent-owner-as-asserting-entity" signal rather than a "known aggregator acquired it" signal.

3. Repeat correspondent across the chain — UNCLEAR / INSUFFICIENT DATA.
The assignment-cover-sheet correspondent is not exposed in the records I could retrieve, and I will not infer one. What is verifiable is a repeated prosecution lawyer across the family: Martin LuKacher (and later Kenneth Jay Lukacher, Customer No. 24902), docket prefix "ML-0376…," who filed/amended applications in this family (Adobe IPR2019-00628 Ex. 2122 ¶¶9–14; EFS receipts). Recurrence on assignment recordings specifically cannot be confirmed here.

4. Cascading transfers — NOT PRESENT.
Only two recorded events, 2005-12-19 and 2013-05-14, ~7.5 years apart. There is no chain of LLC-to-LLC hops within 24 months, and no cluster of assignees sharing an address or principal.

5. Pre-litigation transfer — NOT PRESENT.
The operative transfer to the asserting entity (2005-12-19) predates the earliest '870-based complaint by roughly twelve years (RR Donnelley, filed 2017-02-02), and the 2013-05-14 merger predates it by ~4 years. Neither link falls within the 6-month window. (There is a separate, earlier wave — RAH v. Seiko Epson, 6:10-cv-06710, filed 2010-12-21 — but the 2005 transfer likewise long predates it.)

6. Bankruptcy fire-sale — NOT PRESENT.
Imagicolor Corporation was dissolved (per RAH's own pleadings), but the patents were held by Holub personally, not by Imagicolor, so there was no estate sale or court-supervised patent sale. No Chapter 7/11 is implicated.

7. Privateering — NOT PRESENT.
No operating company transferred the patent to an NPE to assert on its behalf. The transferor is the inventor himself.

8. Defensive aggregator (anti-NPE) — NOT PRESENT.
The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates at RAH Color Technologies LLC, which is the asserting party. If anything, Unified/RPX appear here as adversaries (defensive trackers), not as owners.

Verdict

NPE — high confidence.

RAH Color Technologies LLC is a single-member, licensing-only entity that never shipped a product embodying the claims and that took this patent by assignment from its own sole inventor/sole member (recorded 2005-12-19), then asserted it against RR Donnelley (2017), Quad/Graphics (2018), Xerox (2018) and others consolidated in MDL 3:18-md-02874 — with the patents challenged in four Adobe IPRs (2019). That satisfies two strong signals: Signal 1 (transfer to a licensing-only shell with a single principal, residential office, and no products) and Signal 2 (recognized asserting entity / high-frequency plaintiff in a public NPE database). The 2013-05-14 same-name "merger" recording is an internal reorg, not a third-party sale, so it neither strengthens nor weakens the finding — but it should be confirmed against its reel/frame image.

Verification links:

• USPTO Assignment Center: https://assignmentcenter.uspto.gov/ (search patent number 6995870 to capture the exact reel/frame and correspondent fields I could not retrieve)
• Google Patents legal events (mirrors the two recorded events above): https://patents.google.com/patent/US6995870/en

Two open items for a follow-up run: (1) the reel/frame and correspondent-of-record on both recordings; (2) whether any security agreement or lien exists on the 2005 reel that was later released — if the correspondents on either recording match the LuKacher prosecution docket (ML-0376…), flag Signal 3 as present, as that would tie the recording firm to the same small circle of counsel running the assertion program.

I'll research the prior-art citations for US 6,995,870 specifically. Let me search authoritative sources.

Let me retrieve the actual front page and citations of the '870 patent itself.

Let me pull the actual front-page references list for the '870 patent from the USPTO/EPO original documents.

I hit my tool-use limit before I could complete a direct PatentCenter/Patent-Full-Text retrieval of the '870's own front page, so I'll give you exactly what I verified, what I could not, and what the evidence supports — and nothing more.

Prior-art analysis for US 6,995,870 (literal ID: 6995870)

A. Scope confirmation — and an important verification gap

Every query was keyed to the literal strings 6995870 / US6995870B2. Results returned the correct patent and its family, not similar numbers:

• US 6,995,870 B2, System for Distributing and Controlling Color Reproduction at Multiple Sites, sole inventor Richard A. Holub; appl. no. 10/040,664; filed 2002-01-07; issued 2006-02-07; earliest priority 1996-02-26. Confirmed against the Google Patents record and Espacenet (https://patents.google.com/patent/US6995870/en; https://be.espacenet.com/publicationDetails/originalDocument?...NR=[6995870B2](/patent/6995870B2)...).

Two things I could not obtain, stated plainly:

1. The '870's own "References Cited" front-page list was not retrievable in this session. The authoritative text you supplied is truncated in the description and omits both the claims and the References Cited section. My USPTO-linked retrievals (PTAB ptacts.uspto.gov) surfaced family prosecution documents but not the '870 front page itself.
2. Therefore the "U.S. Patent Documents" lists I did retrieve come from family-member continuations, not from the '870. Specifically US 8,760,704 (https://patentimages.storage.googleapis.com/55/bf/c8/d19b5ef2ed6bad/US8760704.pdf) and US 8,917,394 (https://patentimages.storage.googleapis.com/7c/77/74/18cc9876f58995/US8917394.pdf), both continuations of the 10/040,664 line. These lists are a proxy, and I flag them as such rather than representing them as the '870's own citations.

Consequence for your request: I can identify and characterize the prior art discussed in the '870 specification and the family citation lists, but I cannot certify these as the exact references cited on the '870 front page, and I cannot reliably map any reference to specific claim numbers — a point the earlier sections already established (the '870 claim text was unverified there too; that remains unreconciled and I do not resolve it by guessing).

B. Prior art expressly discussed within the '870 specification itself

These are the references the patent's own text identifies as prior art. They are the most defensible "citations" I can attribute to the '870 with confidence, because they appear in the authoritative text you supplied. All are printed publications/patents predating the 1996-02-26 priority date, so all qualify as prior art under pre-AIA 35 U.S.C. § 102(a)/(b) (the '870 is pre-AIA because its priority chain goes to 1996).

§ 102 bottom line for Section B: No single one of these references appears to disclose the whole of a networked, multi-site, Virtual-Proof-based color-reproduction control system as claimed. They are individually strong § 102(b) references against narrower limitations (color measurement instruments, printer calibration loops, remote proofing, profile formats, gamut operators), and are the natural § 103 combination set. The Vincent and Sodergard references are the two the '870 itself works hardest to distinguish.

C. U.S. Patent Documents on the front pages of family continuations (proxy list — NOT verified as the '870's own citations)

Retrieved from the '704 and '394 front pages. Dates are as printed on those documents. I present them because the Holub family carried a near-identical background-art list across issuances, but treat this as a proxy for the '870 list, not the list itself.

All of these predate 1996-02-26 and are facially available as § 102(b) art. I cannot tell you which of these, if any, was actually cited on the '870 front page — that is precisely the retrieval I could not complete.

D. The correct § 102 framework for this patent

• Pre-AIA § 102 governs (priority 1996-02-26; first-to-invent regime).
• Critical date for the claimed subject matter: 1996-02-26 (earliest priority), assuming the asserted claims are supported by the 08/606,883 disclosure. The later filing dates (1999-01-12 division; 2002-01-07 division) matter only if a claim lacks § 112 support in the 1996 application.
• Under § 102(b), any printed publication or U.S. patent published/issued before 1996-02-26 (more than one year before the 1996-02-26 date, counting the 1996 chain) is a statutory bar. All Section B references and all Section C patents qualify.
• Watch for anachronistic citations: the family front pages also list references dated 1997–1998 (e.g., Ohno et al. 1997/1998; ICC spec. 1998; Adobe Photoshop 5.0, 1998 — these appear in the related Holub patent US 8,279,236's reference list at https://patents.justia.com/patent/[8279236](/patent/8279236)). Those post-date 1996-02-26 and therefore are NOT § 102 prior art against the '870; they belong to later filings and must not be treated as anticipatory art for the '870.
• For § 102 anticipation, a single reference must disclose every claimed limitation. Given the multi-site / network-interface / shared-Virtual-Proof architecture, no single reference in Sections B or C appears to do so for the broad independent claims. The realistic § 102 candidates are against narrower dependent claims (individual calibration, measurement, or gamut-mapping limitations).

E. Why I am not giving you a claim-by-claim § 102 map

Your request asks which claim(s) each reference potentially anticipates. I am deliberately not inventing claim numbers, for two grounded reasons:

1. The '870 claim text was not retrievable in this session (and was flagged unverified in the earlier sections). Without verbatim claims, any "Reference X anticipates Claim Y" statement would be fabrication.
2. The only '870 claim terms I can ground are those construed in the MDL (N.D. Cal. 3:18-cv-03277-SI, Doc. 120, filed 2021-01-22): "sites" and "appear substantially the same" (constrained by "output colors attainable by the rendering devices") — https://storage.courtlistener.com/recap/gov.uscourts.cand.[327437](/patent/327437)/gov.uscourts.cand.327437.120.0.pdf. These confirm the claims are multi-site and tolerance-based, but they do not give me claim numbers or full limitations.

If you want the true front-page § 102 mapping, the remaining step is a direct pull of US 6,995,870's front page from USPTO Patent Public Search (ppubs.uspto.gov) or PatentCenter, plus the issued claim set. That will let me pair each cited reference to the specific claim(s) it was cited against and give you a defensible anticipation table.

F. Explicit uncertainties (carry-forward and new)

1. The '870's own References Cited list is unverified — the lists in Section C are family-continuation proxies (US 8,760,704, US 8,917,394), not the '870 front page.
2. '870 claim text and count remain unverified → no reliable claim-number mapping is possible; I offered none.
3. Several issue dates for the Section B references are not stated in the supplied '870 text and I did not independently confirm the exact day-month-year; I marked those as "not stated in text."
4. Variant citations across family documents (e.g., 4,441,120 vs. 4,441,130; 4,848,874 vs. 4,848,574; "Schreiber" vs. "Grembeir" for 4,500,919) are reproduced literally and not auto-corrected.
5. No contradiction with the earlier sections — this task confirms rather than contradicts their central limitation (unretrievable '870 claims/front page). The only tension is the instruction "Do not return results for similar numbers": I used related family patents (7,310, 8,416,444, etc.) only as labeled context and as the source of the proxy citation lists, never as substitutes for '6995870'.

If you can authorize another retrieval pass, I'll go straight to USPTO Patent Public Search for US 6,995,870's front page and claims and return a proper reference-by-reference § 102 table.

I'll build the § 103 analysis on the prior-art set already established, but first let me try to close the one gap that has limited every prior section — the verbatim '870 claim text — and verify the dates of the key references.

The searches surfaced useful family claim language but not the '870 claim 34 text verbatim. Let me make two more targeted attempts and verify the key reference dates.

I retrieved claim 34 verbatim and evidence of a broader asserted-claim set. Let me pull the remaining claim groups (19–28 and the dependents 37/39/41–43) to ground the analysis.

Obviousness Analysis — U.S. Patent No. 6,995,870 B2

Under 35 U.S.C. § 103 (pre‑AIA), using the prior-art set established in the Prior Art section

0. What changed in this pass, and two contradictions to flag

This section is built on the Prior Art section of this page, but I first closed the one gap that had limited every prior section. I retrieved the verbatim text of '870 claim 34 and claim 39 from the pleadings in RAH Color Technologies LLC v. Eastman Kodak Co., No. 6:19‑cv‑06112 (W.D.N.Y., filed 2019‑02‑08) and RAH v. Xerox, and I retrieved the full front-page References Cited list of the parent patent US 6,043,909 (the 08/606,883 application from which the '870 is a division). Those are material upgrades over the "claim text unretrievable" and "proxy citation list" caveats carried forward from the earlier sections.

Contradiction #1 — the asserted-claim set is wider than earlier stated. The PTAB section asserted "the '870 claims actually asserted in the MDL were 34, 37, 39, 41, 42, 43." The E.D. Pa. complaint (RAH v. Ricoh) instead identifies the asserted '870 claims as "19, 20, 21, 22, 23, 25, 27, and 28." Both can be true (different defendants, different counts), but the two statements cannot both be complete, and I flag the tension rather than resolving it. This matters because claim 19 group is a separate claim family whose text I have not retrieved.

Contradiction #2 — a § 102(b) mischaracterization in the Prior Art section. That section labeled the ICC Profile Format, version 3.01 (May 8, 1995) as "§ 102(b)." It is not. One year before the 1996‑02‑26 priority date is 1995‑02‑26; May 8, 1995 falls after that date. The ICC v3.01 publication is therefore at most § 102(a) art (or § 102(e) via family members' filing dates), not a statutory bar. The same date-sensitivity applies to the TAGA 1995 Sodergård paper, Garini (Jan 1995), Hoyt (Apr 1995), Holub IS&T (May 1995), Newman (Jul 1995), Fukushima (Nov 1995), and the Komori/Blessing/imagexpo items. I correct this because it changes how a challenger must plead the ICC reference.

Everything below is grounded in the literature identified in the Prior Art section, in the '870/'909 specification text, and in the retrieved claim language.

1. The claims to be analyzed

Claim 34 (independent; verbatim, from the Kodak and Xerox complaints):

"A method for providing control to a user for processing color images comprising the steps of:
[A] providing an interface operable at a computer through which the user is able to select a plurality of sites having one or more color input or output devices;
[B] communicating between said sites through a network interface at said sites; and
[C] providing information for transforming input color image data into output color image data for the color input or output devices at said plurality of sites such that colors produced by the color devices appear substantially the same within colors attainable by each of the devices, wherein said information for transforming comprises information relating the color gamuts of different ones of said color devices to each other and user preferences for color reproduction for at least one of the color devices."

Claim 39 (dependent; verbatim): "The method according to claim 34 wherein said user preferences for color reproduction include at least one aspect of the utilization of one or more neutral colorants."

Claims 37, 41, 42, 43 and the 19–28 group: text not retrieved in this session. From the asserted-claim pattern and the parallel parent claims, claims 41–43 almost certainly recite the forward-model / forward-model-table / gamut-descriptor / prototype-transformation-table sequence and "gamut configuration data" + "neutral output color function" (cf. the analogous parent claims 41 and 43 of US 6,043,909, retrieved verbatim). I do not assert their text.

Construed terms (from the MDL order, N.D. Cal. 3:18‑cv‑03277‑SI, Dkt. 120, 2021‑01‑22): "sites," and "appear substantially the same" — the latter constrained by "output colors attainable by the rendering devices" and measured against the patent's own verification procedures and confidence intervals ('870 at 42:41‑62). Critically, the [C] limitation is a conjunctive requirement: the transformation information must include both (i) inter-device gamut-relating information and (ii) user color-reproduction preferences. That conjunction is the only place the claim is potentially novel over the art, so it is where the § 103 analysis must concentrate.

2. Legal framework

• Pre-AIA § 103 governs (priority 1996‑02‑26; first-to-invent). The AIA §§ 102/103 do not apply.
• Graham v. John Deere, 383 U.S. 1 (1966) factors: scope/content of the prior art; differences between the prior art and the claims; level of ordinary skill; secondary considerations.
• KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) governs the motivation analysis and applies retroactively to pre-AIA patents: a combination is obvious if there is "an articulated reasoning with a rational underpinning" — including the rationales of (a) combining prior-art elements by known methods to yield predictable results, (b) substituting one known element for another to obtain a predictable result, (c) using a known technique to improve a similar device in the same way, and (d) a design need or market pressure with a finite number of identified, predictable solutions.
• Level of ordinary skill (proposed, based on the field and the '870's own disclosures): a bachelor's degree in electrical engineering, imaging science, or printing technology (or equivalent), plus 2–4 years of experience in color management, prepress, or press instrumentation, with working familiarity with CIE colorimetry (CIE 15.2), device profiles (ICC/MCT), polynomial colorant-to-color models, and closed-loop press control. This is a relatively high level of skill, which cuts toward obviousness of combinations of well-known building blocks.

3. Prior-art qualification (all pre‑1996‑02‑26)

The front page of the parent US 6,043,909 lists the following, all of which predate the critical date. The '870 and '909 share a specification, so this list is the best available proxy for the '870's own References Cited (a point I still mark as not literally verified for the '870 front page):

Take-away: every element of claim 34 exists somewhere in this set, and the two elements of the [C] conjunction are separately and expressly disclosed by Gordon et al. 1987 (inter-device gamut relating) and by the ICC/MCT profile formats plus ICC rendering-intent user selection (user color-reproduction preferences).

4. Grounds of obviousness

Ground 1 — The primary combination: Sodergård (1991) + Holub et al. (1988) + Gordon et al. (1987) + ICC v3.01 (1995)/MCT (1993)

Why a PHOSITA would have combined them (KSR rationales):

1. Same field, same problem. Every reference addresses one facet of the single acknowledged problem: printed color "varies within a press run and from site to site because there have been only limited means of coordinating control of product appearance among sites" (the '870/'909 Background — an admission of the design need). Sodergård's own paper is the teleproofing response to that need; Holub 1988 is the colorimetric-calibration response; ICC/MCT is the portable-transform response. Combining them is combining complementary solutions to a known, unitary problem.
2. Explicit pointer in the art. Vincent states the instrument is usable "to insur[e] consistent color across a network" and that it forms "part of a feedback network." That is an express invitation to network the calibration instruments — i.e., to build exactly the [B]/[C] architecture.
3. Predictable result from known techniques. Applying a known color transform (ICC/MCT profile), a known gamut-relating operator (Gordon 1987), and a known network link (Sodergård/ISDN) in series yields nothing more than the sum of the parts — a remote device rendering the same appearance within its own gamut. There is no unpredictable interaction.
4. Finite, identified solutions. By 1995 the industry had converged on device-independent color + profiles + gamut mapping as the architecture (ICC was formed precisely to standardize it). "[A] finite number of identified, predictable solutions" (KSR) existed, favoring obviousness.

Ground 2 — The "network color calibration" combination: Vincent + Chan/Sobol + Bonino + Gordon et al.

Vincent supplies the networked colorimeter and the network-consistency object; Chan and Sobol supply the closed-loop calibration architecture (render a calibration form on the device, measure it, compare to expected, generate correction coefficients) that maps onto the '870's produce-then-verify structure; Bonino supplies inter-device gamut harmonization ("all devices operating within the same gamut"); Gordon supplies explicit gamut operators. Motivation: Vincent expressly aims at network-consistent color; Chan/Sobol expressly aim at automated device calibration; combining a network of Vincent-type instruments with Chan/Sobol-type closed-loop calibration is the predictable way to keep a multi-site network in agreement.

Ground 3 — The "multi-site press control" combination: Kipphan + Sodergård (1993/1995) + ICC/MCT + the press-control references (Heidelberger CPC 23; Komura)

Kipphan supplies CIELAB error → ink-density correction to the press control panel; Sodergård supplies on-line measurement of the image area guided by the digital page description; the ICC/MCT supplies the portable transform. Motivation: Kipphan's admitted weakness (off-line, color-bars-only, operator must sample proactively, no calibration self-check) creates the specific design incentive to move to on-line image-area measurement (Sodergård) and to a shared transform format — i.e., to combine.

Ground 4 — The conjunction of limitation [C]

The hardest limitation is that the transmitted transformation information must include both gamut-relating information and user preferences for the device. The art discloses the conjunction in a single, easily-combined location: the ICC profile stores, in one transmissible data structure, (i) the device's color transform and its associated gamut information (via the PCS relationship and the profile's gamut description) and (ii) a user-selectable rendering intent that governs how out-of-gamut and near-gamut colors are reproduced — which is, on its face, "a user preference for color reproduction for at least one of the color devices." Add Gordon's gamut operator and Bonino's harmonization for the express "relating the color gamuts of different ones of said color devices," and the conjunction is met. Motivation: the entire purpose of a profile is to be shared between devices, which is precisely why it is transmissible over Sodergård's network.

5. Dependent claims

Claim 39 ("user preferences… include at least one aspect of the utilization of one or more neutral colorants") is, if anything, more clearly obvious than claim 34. The 1985 Schwartz/Holub/Gilbert TAGA paper is titled Measurements of Gray Component Reduction in neutrals and saturated colors and describes "a method for finding colorimetrically equivalent ('indistinguishable') 4‑colorant solutions… that used varying amounts of black." Its entire subject matter is the trade-off between neutral (black) and non-neutral colorants, i.e., "the utilization of one or more neutral colorants." Boll 1994 independently discloses GCR flexibility across seven inks. Hardy & Wurzburg 1948 and Neugebauer 2,790,844 are earlier 3‑vs‑4‑colorant equivalents. A PHOSITA would combine Schwartz's black-utilization teaching with the claim-34 system simply because black utilization is the standard, user-exposed control for making a four-colorant device reproduce a target — a substitution of a known element (GCR/UCR as the "user preference") for a predictable result.

Claims 41–43 (analogous to the parent's forward-model / gamut-descriptor / "gamut configuration data" and "neutral output color function" claims): should the retrieved text confirm this scope, the mapping is: forward model ← Pobboravsky 1962 (polynomial colorant→CIE models) + Holub 1988; gamut descriptor ← Gordon 1987 + Neugebauer 1956; inversion ← Schwartz 1985 + Rose 5,200,816 (neural nets) + standard Numerical Recipes (Press et al. 1986, cited on the front page); gamut configuration data/neutral function ← Gordon 1987 + Schwartz 1985. I state this as conditional and do not assert it as established.

Claims 19–28 (the Ricoh group, text unverified): cannot be analyzed responsibly; see § 7.

6. RAH's likely rebuttals, and secondary considerations

Rebuttals that carry some force:

1. No single reference teaches the whole. True. Sodergård teleproofing transmits image data, not transformation data, and the patent expressly distinguishes it: it "lacks any provision for calibration or verification of the devices at either end of a link and also lacks the data structures needed to support remote proofing and negotiation of color approval." Vincent "does not discuss how distributed calibration would be implemented." Bonino is photometric, not colorimetric, and assumes common primaries. These are the strongest non-obviousness arguments and they are real — a challenger must prove the combination, not a single anticipatory reference.
2. Possible teaching away. The specification's disparagement of Sodergård ("lacks… the data structures") could be argued as teaching away from the image-data-teleproofing approach. Counter: disparagement of a reference's omission is a statement of the problem to be solved (a motivation), not a teaching away from the claimed solution, and under KSR a mere alternative disclosure is not teaching away absent criticism that would discourage the solution.
3. The "user preferences" element is thin in the art. ICC v3.01's rendering-intent/mode field and its profile-editor ecosystem are the best evidence, but I did not verify the exact v3.01 field set, so this limitation is the weakest link in Ground 1/4.

Secondary considerations: RAH's complaints state the technology was licensed to "nine of the largest manufacturers and service providers of color imaging and printing products," "two additional manufacturers," and that "13 major companies have entered into end-user license agreements," and that industry players "recognized the contributions" — arguments of industry recognition / licensing as a nexus-bearing secondary consideration. A challenger should expect these to be pressed, and should test the nexus (licenses taken to settle litigation carry little weight) and the timing (post-invention licensing under litigation pressure). Against these, note the complete absence of any PTAB or court validity test on the '870 itself (no estoppel, no FWD, no construction of claim 34's novelty-limiting conjunction beyond "sites"/"appear substantially the same"), so there is no adverse validity record either way.

7. Explicit uncertainties and clear statements of what I do NOT know

1. Claims 37, 41, 42, 43 and 19–28 of the '870 are unretrieved. I analyzed claim 34 (verbatim) and claim 39 (verbatim) only. Any statement I made about the content of 41–43 is expressly conditional on analogy to the parent and is flagged as such; I did not invent their text.
2. The '870's own front-page References Cited list is still not literally verified. Section 3 uses the parent US 6,043,909 front page (retrieved verbatim) as the proxy. Given the division relationship and shared specification this is a strong proxy, but it is a proxy.
3. § 102 classification corrected. Implicit in § 0: the Prior Art section's labeling of ICC v3.01 (May 8, 1995) as § 102(b) is wrong — it is § 102(a) at most. Other 1995-dated items are similarly date-sensitive and I flagged them.
4. Exact § 102(e) filing dates for Newman 5,432,906 and Fukushima 5,465,111 were not confirmed; I listed them as § 102(e) on the assumption their filings predate 1996‑02‑26, which must be checked.
5. Disclosures of Naka 5,109,275, Shalit 5,345,315, Shigeta 5,394,204, Newman 5,432,906, and Fukushima 5,465,111 were not independently verified. I used them only as listed art, not as mapped references.
6. I did not retrieve any PTAB or court obviousness holding on the '870 — there is none. This is a prospective analysis, not a review of an existing decision.
7. Practical note: because the patent's legal-status record shows "Expired – Fee Related," adjusted expiration 2017‑05‑07, a § 103 invalidity attack today is a defensive/declaratory tool with no damages at stake (and § 286 reaches back only to 2020‑09‑27). If the goal is a validity opinion rather than a defensive one, the expiration question should be confirmed at USPTO PatentCenter before relying on it — Google's status entry is an express "assumption," not a legal conclusion.

Bottom line: On the art as established, claim 39 is plainly obvious over Schwartz/Holub 1985 combined with any of the claim-34 systems. Claim 34 is obvious on the combination of Sodergård 1991 (multi-site color communication over a network) + Vincent 5,272,518 (networked colorimeter aimed at network-wide consistency) + Gordon/Holub/Poe 1987 (explicit inter-device gamut operators) + Bonino 5,309,257 (inter-device gamut harmonization to the "commonly achievable" attainable colors) + ICC v3.01/MCT (transmissible color profiles carrying gamut information and a user-selectable rendering preference for a device) — with motivation supplied by the field's admitted, unitary problem (site-to-site color variation), the express network-consistency object in Vincent, and the finite, converged ICC architecture. The strongest non-obviousness argument remaining for RAH is the conjunctive "[C]" limitation and the specific transmission of transformation-and-preference data (rather than image data) — a narrow gap that the ICC/MCT profile-sharing disclosure is the single most effective reference for closing.