US 4719615

I'll search for this specific patent number across USPTO and CAFC sources.

Now let me search for CAFC 2026 dockets and any litigation involving this patent.

US patent 4,719,615 — Analyst Summary

Bibliographic data (per USPTO / Google Patents)

Source: https://patents.google.com/patent/US4719615/en

Abstract (verbatim)

"An erasable optical data storage medium including a substrate and a dual layer of a first material and a second material on the substrate. The dual layer is susceptible to expansion and contraction, to writing data thermally, to erasing data thermally and mechanically, and to reading data optically, the first material and the second material, respectively, being bonded together and remaining in the same physical state upon the expansion and relaxation of the dual layer. Methods and apparatus are also disclosed for recording data bits on the medium."

Note a scope point worth flagging: the abstract and specification describe a substrate plus a dual layer, and the specification also describes methods and two apparatus embodiments (FIGS. 8 and 9). But the granted claims of this patent are directed only to the medium itself. The method and apparatus subject matter was pursued in the divisional (US 4,852,075). This is a common source of confusion when reading the '615 abstract in isolation.

Technology in brief

A deformable "bump" (topographical) optical recording medium, not a phase‑change medium. Bottom "expansion" layer (material 18A) is an elastic, highly expansive, low‑Tg elastomer; top "retention" layer (material 20A) is a higher‑Tg glassy polymer. A first laser wavelength (e.g., Krypton, 647 nm) passes through the transparent top layer and is absorbed in the bottom layer, which expands; the heated top layer is above Tg, pliable, and is pushed up into a bump. When it cools below Tg it freezes the bump and holds the expanded bottom layer in tension. A second wavelength (e.g., Argon, 488 nm) absorbed in the top layer heats it above Tg, releasing the tension so the elastic bottom layer pulls the surface flat — a fast, low‑power erase. Neither material changes physical state. The patent emphasizes: identical appearance/refractive index matching for optical coupling, dye/pigment "optical tuning" to selectable wavelengths, no delamination from the substrate (avoiding "creep"), operation on non‑smooth surfaces, and erase‑then‑write within one bit‑cell time. Working example: Epon 828 epoxy + Savinyl RLS Scarlet dye as the retention layer; Dow Corning 734 silicone rubber + carbon black as the expansion layer; 2.5–5 µm bumps; ~2:1–3:1 read contrast.

Plain-language overview of the independent claims

Claim 1 (medium, structural): An optical recording medium with (a) a first material layer and (b) a second material layer next to it, where (c) the two layers together can form an optically detectable deformation by thermal expansion of the first material, and this happens without liquefaction or gasification of either material when the first layer is locally heated with light of a first preselected wavelength; and (d) the two materials are bonded strongly enough that the second layer holds the first material in its stretched, expanded state after the light stops and the first material cools — thereby retaining the deformation — with the second layer being strong and rigid enough to do that holding. In short: a two‑layer, non‑ablative, non‑bubble bump‑forming medium where the stiff top layer mechanically locks in the expanded bottom layer.

Claim 4 (medium, defined by thermal properties): An optical recording medium with (a) a first material layer having glass transition temperature "Tg1"; (b) a second material layer bonded to the first, with glass transition temperature "Tg2" above ambient and above Tg1; (c) the two layers forming a dual‑layer recording medium; (d) the first material having a high coefficient of thermal expansion; (e) the second material permitting localized thermal expansion of the first when the second is above Tg2, and retaining the first bonded in its expanded configuration when the second cools below Tg2; and (f) the dual layer forming and retaining an optically detectable deformation by localized thermal expansion of the first material — again without liquefaction or gasification — when locally heated with light of a first preselected wavelength.

Claim 19 (medium, polymer‑specific): Essentially claim 4 recast in polymer terms — (a) a first polymer layer, (b) a second polymer layer bonded to it, (c) the first layer with a high coefficient of thermal expansion and Tg1, adapted to produce a deformation by localized thermal expansion on heating; (d) the second layer with Tg2 above ambient and above Tg1, permitting the first layer's localized expansion when heated above Tg2 and retaining the first layer in expanded configuration when cooled below Tg2; and (e) the two layers cooperating to form and retain an optically detectable deformation when locally heated with light of a first preselected wavelength.

Representative dependent claims: 2 (erase by locally heating the second material with a second wavelength, without liquefaction/gasification); 3/14 (second material substantially transmits the first wavelength); 5 (contraction of first material when second is heated above Tg2 by a second wavelength); 6–7 and 12–13 (dye‑based light‑sensitive absorbers in the second and first materials respectively); 8–9 (elastic/relaxation‑force recovery); 10 (both materials are polymers); 11 (both have relatively low thermal conductivity); 15 (the materials are not metals or metal oxides — a direct swipe at Cornet's U.S. 4,371,954); 16–18 (first material elastic with high yield strain; second material thermoplastic/thermoset with low yield strain; second material glassy at ambient and rubbery above Tg2); 20–21 (dependents of claim 19).

Prior art cited

U.S. 3,689,768 (Sato); 3,787,873 (Fuji); 4,264,986 (Willis); 4,268,575 and 4,282,534 (Fuji); 4,278,734 (Ohta et al.); 4,285,056 and 4,320,489 (RCA); 4,336,545 (Kodak); 4,360,895 and 4,371,954 (Cornet, Thomson‑CSF); 4,398,203 (Thomson‑CSF); JP 57‑60546; JP 58‑62096; JP 58‑94145. The specification discusses Ohta, Willis and Cornet as the three principal references it improves upon.

Litigation / CAFC 2026 docket check — no results found

I searched for the patent number in combination with Federal Circuit and 2026 docket terminology and found no CAFC 2026 docket, appeal, or petition referencing US 4,719,615. My searches surfaced only patent‑document and bibliographic hits (Google Patents, Justia, PatentLeaderboard, a Korean family member KR930006351B1 citing it, and an academic bibliography listing "Optical Data, Inc., US 4719615, 1988").

Two explicit caveats:

1. I cannot confirm the absence of a 2026 CAFC docket with authority — my search tooling is a general web search, not a live PACER/CM‑ECF or CAFC docket query, so a very recent or sealed filing would not necessarily appear. If a specific case number is known, that should be checked directly.
2. A 2026 appeal on this patent would be extraordinary on its face. The patent expired (anticipated expiration 2005‑01‑12) and is listed as "Expired – Lifetime." Any 2026 appellate activity would most plausibly be a lingering damages/§285 fee issue from pre‑expiration infringement, not an ongoing infringement appeal. I found no evidence of even that, so treat any such report as unverified.

If the intent behind the CAFC query was to find appellate history for the family rather than this exact number, note that Optical Data's related patents (e.g., 4,852,075; 4,780,867; 4,901,304; 4,825,430) were broadly licensed and litigated in the late 1980s–1990s optical‑media era — but I have not verified specific case names or dockets and will not guess at them.

Confidence

• Title, inventors, assignee, application/filing/issue/priority dates, abstract, claim text, and claim count: high — taken directly from the USPTO‑sourced full text and corroborated by Google Patents, Justia, and PubChem/EPO family data.
• Legal status "Expired – Lifetime" with 2005‑01‑12 anticipated expiration: high as a recorded status, but it is explicitly an assumption in the source, not a legal conclusion.
• Absence of CAFC 2026 activity: low‑to‑moderate — negative search result only, with the limitations described above.

I'll search for litigation involving US patent 4719615 across patent litigation databases and court sources.

The initial searches returned noise about unrelated "961" patents (e.g., 6,594,961, 9,693,961). Let me narrow the search to the specific patent and its assignee.

Let me search dedicated litigation databases directly.

Findings on U.S. Patent No. 4,719,615

Result: No litigation identified

Based on my searches, I could not identify any known litigation specifically involving U.S. Patent No. 4,719,615 ("Erasable optical data storage medium," inventors Feyrer, Gordon, and Skiens; assignee Optical Data Inc. of Beaverton, Oregon; granted January 12, 1988; expired January 2005).

What I searched and what I found

Important caveat on search noise

A large fraction of the results returned were false positives for other patents whose numbers happen to end in "961," such as:

• U.S. 6,594,961 — Richard Leines decking patent; litigation Leines v. Homeland Vinyl Products (E.D. Cal., No. 2:18‑cv‑00969), a royalty/license dispute.
• U.S. 9,693,961 — Purdue Pharma; PGR and Hatch‑Waxman litigation (D. Mass.).
• U.S. 6,xxx,961-type patents in the CAO Lighting v. OSRAM Sylvania and CAO Lighting v. Technical Consumer Products matters.

None of these involve U.S. 4,719,615, and I have excluded them rather than treating them as related.

What the record does show about U.S. 4,719,615

The only substantive references I located to the '615 patent outside its own file wrapper are:

1. Optical Data Inc.'s THOR-CD relationship with Tandy Corp. — Reported in Tech Monitor ("Optical Data Involved in Development of Tandy's THOR-CD Technology"). Optical Data announced it had a patented dye-polymer optical disk coating forming laser-induced bumps, and stated that Philips & DuPont Optical Ltd. had already taken a license to the technology. This is a licensing/commercial relationship, not litigation.

2. Academic/survey citation — the patent is cited in a survey of optical storage patents as Optical Data, Inc., US 4719615, 1988 (with its sibling US 4852075, 1989). This is bibliographic, not litigation.

3. Patent family — the '615 patent's sibling is U.S. 4,852,075 (divisional, "Erasable optical data storage medium and method and apparatus for recording data on the medium"). If any enforcement occurred, it may have been asserted under the 075 patent or under the wider Optical Data portfolio (e.g., U.S. 4,780,867; 4,825,430; 4,896,314; 4,901,304), rather than under '615 alone.

Honest limitations

• I was unable to query PACER and the Unified Patents case list directly (the searches against those domains did not return usable records and I hit the tool-call limit). My conclusion rests on general web and court-document searches, not an exhaustive docket sweep.
• Because '615 expired in January 2005 and its priority dates to 1983, any enforcement would necessarily predate 2005 — and pre‑2000s district court dockets are less completely digitized, so a null result is weaker evidence than it would be for a modern patent.
• I am not asserting that no suit ever existed; I am stating that no litigation is known or documented from the sources available to me.

Recommendation for a definitive answer

To confirm with authority, the following should be checked directly (I could not complete these):

1. PACER / CourtListener RECAP — search docket text for "4,719,615" and for assignee "Optical Data, Inc." (D. Or. was the likely home venue, given the Beaverton, OR assignee).
2. Unified Patents litigation portal — https://portal.unifiedpatents.com/litigation (I was unable to retrieve its case list).
3. USPTO Patent Assignment / file wrapper for recorded licenses or security interests (e.g., the Philips & DuPont license referenced above).
4. Derwent / Lex Machina / Darts-IP historical U.S. litigation records.

Bottom line: As of this search, no litigation involving U.S. Patent No. 4,719,615 can be identified. If you need a litigation-grade negative, direct PACER/Unified Patents lookups by a human searcher are required, and I would flag any assertion of "no litigation" as unverified until those dockets are checked.

I'll start by verifying the structured ODP block against public sources, then search for any proceeding traces.

Let me run more targeted searches — including the family member and the patent's litigation history.

Let me check CourtListener/dockets and reexamination history for this specific patent.

Proceedings overview

Total AIA trial proceedings on US 4,719,615: 0. The USPTO Open Data Portal (ODP) structured block returns no IPR, PGR, or CBM for this patent, and my independent web searches surfaced no petition, institution decision, Final Written Decision, or appeal naming US 4,719,615 — so the breakdown by status is trivially "0 active / 0 invalidated / 0 sustained / 0 settled / 0 institution-denied." The bottom line for a defendant is not "the patent is hardened" and not "claims 1–21 are canceled" — it is that the patent is a 1983-priority, 1988-grant artifact whose term expired on 2005-01-12, so it is unsuited to an AIA challenge and, more importantly, incapable of supporting prospective relief or meaningful past damages today.

Because there are no proceedings, the per-proceeding template is inapplicable. I set out below (a) the evidence for the null result, (b) why the null result is structural rather than accidental, and (c) the strategic and estoppel consequences.

---

No proceeding on file — verification and scope

• Type: N/A (no IPR / PGR / CBM docketed)
• Filed: N/A
• Status: No AIA trial proceeding exists in the ODP ingest (the canonical source for this task). No contrary proceeding surfaced in web searches of PTAB materials.
• Judge panel: N/A — the Board has never constituted a panel on this patent that I can find.
• Petition grounds: N/A
• Institution decision: N/A
• Final Written Decision: N/A — there is no FWD to link because none was ever issued.
• Settlement / termination: N/A
• Appeal: N/A — no PTAB decision existed to appeal, so there is no CAFC docket to report.

Search-quality caveats — do not conflate these false positives. Querying "4719615" / "the '471 patent" against PTAB and litigation corpora returns noise from different patents:

• The Janssen Biotech v. Celltrion line of cases (D. Mass. No. 1:15-cv-10698-MLW; Fed. Cir. appeal 2017-1120) repeatedly refers to "the '471 patent," but that is an antibody patent asserted by Janssen/NYU, with a 1990s priority chain (the '413 parent application) and an ex parte/reexamination double-patenting fight — it has nothing to do with Optical Data's erasable optical medium. The coincidence is the last three digits, not the patent.
• PTAB petitions discussing US 6,465,961 reexamination control nos. 95/002,324 and 95/000,680 likewise match on a different "4719615"-adjacent string and involve a plasma-lamp patent, not this one.
• The Justia and Unified Patents portal hits for US 4,719,615 and US 6,355,432 are citation relationships only — 4,719,615 appears in other patents' "cited-by" lists (it is cited as prior art in, e.g., US 6,916,519, and in the Affymetrix VLSIPS citation thicket). A citation is not a challenge.

I could not, with the sources available to me, confirm any post-grant Office proceeding (including pre-AIA ex parte or inter partes reexamination) against 4,719,615. Treat that as an unverified negative, not a proven one — the ODP block governs AIA trials only, so a 1990s–2000s reexamination file, if any, would live in the PatentCenter file wrapper rather than in the ODP trial data.

---

Strategic summary

Claim status: everything UNTESTED, nothing CANCELED, nothing SUSTAINED-by-a-tribunal. The full claim set — claims 1–21 as issued on 1988-01-12 — remains as granted; no claim has been canceled, narrowed by certificate, or confirmed in a contested post-grant proceeding. The nuance is that "untested" here is not a sign of strength. It is a function of timing: the AIA trial regime (IPR/CBM/PGR) began on 2012-09-16, and by then this patent's pre-URAA term — 17 years from the 1988-01-12 grant — had already run out on 2005-01-12, as reflected in the "Expired - Lifetime" status with an "anticipated expiration" of 2005-01-12 on the face of the record. PGR was never available (it reaches only patents with effective filing dates on or after 2013-03-16; this one claims 1983-08-22 through a 1986-03-04 CIP), and CBM — which could in principle reach pre-AIA business-method patents — sunset for new petitions on 2020-09-16 and would in any event require a financial-services covered business method. There was simply no window in which a rational challenger would pay to invalidate an expired medium patent.

Estoppel landscape: essentially a blank slate, but a blank slate you probably never need to use. Because no IPR was ever instituted, 35 U.S.C. § 315(e)(2) binds nobody — no petitioner, and therefore no privy of any petitioner, is estopped from raising any § 102 or § 103 ground in district court based on patents or printed publications. There is no IPR record, no FWD findings on claim construction or objective indicia, and no Board claim-construction ruling to exploit or defend against. If you find yourself on the receiving end of an assertion, every prior-art ground is technically available to you — the 1970s–80s art cited on the face of the patent (US 3,689,768 Sato; US 3,787,873 Fuji; US 4,264,986 Willis; US 4,278,734 Ohta; US 4,371,954 Cornet; US 4,398,203 and US 4,360,895 Thomson-CSF, etc.), plus the Cornet/Willis bump-forming theory the specification spends pages distinguishing. But the more efficient answer is statutory, not evidentiary: with the term ending 2005-01-12, § 286 caps recovery at infringements occurring no more than six years before the complaint, so any complaint filed after 2011-01-12 — and we are now at 2026-09-30 — recovers nothing on the last day of the term and, by extension, nothing at all. There is also no ongoing infringement to enjoin, because there is no enforceable term left.

Pattern signals: absent across the board. No petitioner filed multiple (or any) IPRs against this patent. The patent owner (Optical Data, Inc.) never had occasion to defend an AIA trial and therefore never pursued a PTAB appeal to the Federal Circuit on this patent. No defensive aggregator (Unified Patents, RPX, LOT, etc.) appears in the chain as a filer — the only Unified Patents hit is a citation on a wholly unrelated patent page. What the "Cited By" list does show is a dense web of Optical Data's own follow-on filings from 1986–1990 (US 4,780,867; US 4,825,430; US 4,901,304; US 4,896,314; US 4,856,992-adjacent Toshiba/Tandy work), i.e., the family was commercially real and actively prosecuted — but that activity is a prosecution and licensing story, not a post-grant trial story. The sibling case US 4,852,075 (filed 1986-04-04, claiming the 1983-08-22 priority) presents the same picture and is likewise not subject to any proceeding I could locate.

---

Recommended next steps

1. If a demand letter or complaint cites US 4,719,615, lead with expiration, not invalidity. The record shows "Expired - Lifetime," publication 1988-01-12, and anticipated expiration 2005-01-12 (see https://patents.google.com/patent/US4719615/en and https://patentcenter.uspto.gov/ for the file wrapper). There is no live term and no injunctive exposure; the § 286 six-year lookback forecloses damages on a 2026 filing. Respond with a term-expiration letter rather than a merits defense.
2. There is no FWD to quote or link, because there is no FWD. Any internal memo or client alert asserting that "claims 1–5 were canceled" or that "the patent survived two IPRs" would be fabricated; do not repeat such statements if you encounter them. The correct statement is: zero AIA proceedings, zero claim-level outcomes.
3. Do not treat the absence of IPRs as a validity signal. For a live patent, an empty PTAB docket is a mild signal that the claims are weak or unasserted; for a patent that expired before the AIA trial regime existed, it is purely a matter of arithmetic. Contrast the two explicitly in any risk memo.
4. If you need to confirm the null result yourself (belt-and-suspenders before relying on it), pull the complete file wrapper at https://patentcenter.uspto.gov/ (covers reexaminations, assignments, and maintenance-fee/expiration records) and search PTAB E2E at https://ptacts.uspto.gov/ptacts/ by patent number; for any litigation or CAFC history, query https://www.courtlistener.com/?q=%224%2C719%2C615%22 (my searches there returned no hits, but this is the right place to re-verify before you certify the absence to a client).
5. Budget zero on estoppel strategy. With no instituted IPR, no § 315(e)(2) estoppel exists, and there is no reason to buy into an IPR coalition — the patent cannot be asserted prospectively, so there is nothing to invalidate defensively.

---

Confidence and limits. I am confident in the ODP-sourced count of zero AIA trial proceedings and in the expiration date of 2005-01-12, both of which the authoritative record in this prompt supports. I am not confident as to whether any pre-AIA ex parte or inter partes reexamination ever occurred on this patent — my searches produced no evidence of one, but the sources available to me do not conclusively rule it out. That is the one factual gap worth closing from the file wrapper before advising a client that the patent has never been challenged in any forum.

I'll research the assignment record and ownership history before drafting.

US 4719615 — "Erasable optical data storage medium" — Ownership & Assignment Reconstruction

Verification caveat up front: my tooling could not open assignmentcenter.uspto.gov / assignment.uspto.gov directly this session (only web-search retrieval was available, and it did not return Assignment Center record pages for this patent). What follows is built on (a) the Google Patents legal‑events table for US4719615A, which mirrors the USPTO assignment feed, and (b) contemporaneous trade-press and EPO family documents. No reel/frame numbers and no correspondent names were retrievable, so none are stated. I flag every inference as such rather than inventing record metadata.

Inventors

Pattern notes:

• All three inventors are the same trio across US4719615 and its sibling US4852075, and Feyrer/Skiens recur across US4780867, US4901304 and US4912696 — a small in-house research group, not a rotating cast.
• All three executed confirmatory assignments to the company after the 1986-03-04 CIP filing (recorded April–August 1987, i.e., 13–17 months post-filing, in the run-up to the 1988-01-12 issue date). That is the normal "employee‑inventor assigns to employer" lag; it is not a departure/fire‑sale pattern, and I found no evidence any inventor left within 12 months of filing.

Original assignee

Optical Data, Inc., a corporation of the State of Oregon, 9400 S.W. [Beaverton-Hillsdale Hwy], Beaverton, Oregon 97005 (per the EP counterpart applicant block). Original assignee of record on the issued patent and, per the retrieved records, still the assignee of record today.

• Business: developer of dye‑polymer, bump‑forming erasable optical recording media — the two‑wavelength "expansion layer / retention layer" chemistry claimed here. It was a technology and IP house, not a disc manufacturer.
• Product embodying the claims: Did it ship a product? Unclear / probably not at scale. Contemporaneous coverage describes it as "a tiny firm" whose model was licensing. Tandy Corp. took a licence and announced the THOR‑CD recordable/erasable system in April 1988, publicly acknowledging the underlying technology came from Optical Data of Beaverton; Philips & DuPont Optical (pdO) also took a licence to evaluate the dye material. (Audio, July 1988; Electronics Australia, Aug. 1988; Hi‑Fi News, Nov. 1989.)
• Current status: No evidence that Optical Data, Inc. (Oregon) was acquired, merged, or filed for bankruptcy. I could not verify its ultimate corporate fate → unclear. Note also the important name collision: the Warren, New Jersey "Optical Data" profiled in NJ Biz (Aug. 1991) — videodisc science curricula, "Windows on Science," founder William Clark — is a different company and must not be merged into this chain.

Assignment timeline

Recorded reassignments (Google Patents legal‑events mirror; execution dates and reel/frame are not exposed in the retrieved record):

• Recorded 1987-04-13 — Reel not retrievable

  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST"
  • Assignor: Skiens, W. Eugene
  • Assignee: Optical Data, Inc.
  • Correspondent: not retrievable from the record (no attorney of record captured in the Google Patents mirror)
  • Context: Original employee‑inventor assignment to the operating company.

• Recorded 1987-05-15 — Reel not retrievable

  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST"
  • Assignor: Gordon, N. Ross
  • Assignee: Optical Data, Inc., a corp. of OR (recorder added the state of incorporation, suggesting a cleanup/confirmatory filing)
  • Correspondent: not retrievable
  • Context: Original employee‑inventor assignment.

• Recorded 1987-08-03 — Reel not retrievable

  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST"
  • Assignor: Feyrer, Clyde D.
  • Assignee: Optical Data, Inc.
  • Correspondent: not retrievable
  • Context: Original employee‑inventor assignment.

After 1987 there are no further recorded assignments. Google Patents lists "Current Assignee: Optical Data Inc." — consistent with no post‑issuance transfer of title ever having been recorded. The patent's legal status is Expired – Lifetime, with anticipated expiration 2005-01-12 (17 years from the 1988-01-12 issue date, pre‑URAA term).

Licenses (not assignments — do not confuse with a title transfer): Tandy Corp. (THOR‑CD, 1988) and Philips & DuPont Optical. You asked for recorded conveyances; a bare licence is not one, and I found no record of a licence being recorded against this patent. Mentioned only because it is the reason the patent never moved.

Sibling continuity: the division US4852075A and the CIP parent 06/525,813 (filed 1983-08-22, abandoned) sit in the same Optical Data family with the same three inventors — no separate title event for US4719615 flows from either.

Timeline diagram

timeline
    title Ownership of US 4719615
    1983 : Parent application filed
    1986 : CIP application filed
    1987 : Skiens assigns to Optical Data
         : Gordon assigns to Optical Data
         : Feyrer assigns to Optical Data
    1988 : Patent issued
         : Tandy licenses THOR CD technology
    1989 : Philips and DuPont Optical licenses
    2005 : Patent expired

NPE / troll-pattern signals

1. Shell-entity transfer — not present. No transfer to any LLC, and no entity with an "IP / Holdings / Ventures" suffix appears anywhere in the chain. The assignee of record has been Optical Data, Inc. (a state‑chartered Oregon corporation) since 1987.
2. Known asserter in the chain — not present. No Acacia, Marathon, IV, Wi‑LAN, Conversant/Mosaid, Vringo, Pendrell, Round Rock, Spangenberg vehicle, or any Unified Patents / RPX high‑frequency plaintiff appears as assignee or assignor. Caveat: this check is only meaningful because there is no chain at all after 1987.
3. Repeat correspondent across the chain — unclear / not assessable. No correspondent, attorney, or recording firm is exposed in the records retrieved, and the only attorney of record I could find in the family is John Norton Haigh Skerrett (H.N. & W.S. Skerrett, Charles House, Birmingham, UK) — the foreign agent on the EP counterpart EP0338776, not a US assignment correspondent. I explicitly decline to treat a single UK prosecution address as a US recording-correspondent finding.
4. Cascading transfers — not present. Zero post‑issuance title transfers over ~18 years of enforceability; the opposite of a <24-month LLC cascade.
5. Pre-litigation transfer — not present. No assignment within 6 months of any suit, because no suit naming this patent surfaced and no assignment occurred after 1987 at all.
6. Bankruptcy fire-sale — unclear, leaning not present. No Chapter 7/11 record for Optical Data, Inc. (Oregon) was found; I could not confirm the company's ultimate dissolution path either. Flagging as unverified rather than asserting.
7. Privateering — not present; direction is inverted. Here the small innovator licensed out to large operating manufacturers (Tandy; Philips & DuPont Optical), which is ordinary upstream technology licensing, not an operating company arming an NPE against competitors.
8. Defensive aggregator (anti-NPE) — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN; it terminates at the original assignee and then at expiry.

Verdict

Insufficient data — meaning: only the original inventor-to-employer assignments exist, and there is no recorded post-issuance assignment chain to analyse. The three records (recorded 1987-04-13, 1987-05-15 and 1987-08-03, all Skiens/Gordon/Feyrer → Optical Data, Inc.) are the entire recorded history; title never left the original assignee, so every NPE-structure test except "no records" is untestable, and the ones I can test (shell transfer, known asserter, cascading transfers, defensive aggregator) all come back negative.

The correct characterisation of this patent's owner is a small operating R&D company that monetised by licensing, not an NPE: Optical Data, Inc. licensed the bump‑forming dye‑polymer technology to Tandy (THOR‑CD) and to Philips & DuPont Optical, per April 1988–November 1989 trade press. The patent itself has been expired since 2005-01-12, so no assertion risk attaches today.

Verify at the USPTO Assignment Center: https://assignmentcenter.uspto.gov/ (search patent number 4719615; also https://assignment.uspto.gov/patent/index.html). If that search returns only the three 1987 inventor assignments with no correspondent or reel/frame detail beyond what is above, this analysis stands unchanged — and if it returns a later record I could not see, that record, not this document, should govern.

I'll search USPTO and related sources for this specific patent and its cited prior art.

I have the full list of references. Now let me research the specific cited references I don't yet have details on.

Prior-Art Analysis — US 4,719,615 ("Erasable optical data storage medium")

Building on the previously generated bibliographic, technology, and claim overviews. This section addresses only the reference set and its §102 implications; it does not repeat the claim summaries.

---

Scope and method

Patented subject matter being tested. All 21 claims are directed to the medium itself. Three independent claims — 1, 4, 19 — are product claims. The novelty-critical limitations that any §102 reference must disclose, singly, are:

The negative limitation "without liquefaction or gasification" is the single most important differentiator, because the two closest references (Willis and the Cornet family) both operate by melting or by a phase transformation. This is also flagged in the specification, which criticizes Willis's "melting of the bump" and Cornet's martensitic→parent transformation.

§102 date window. The '615 application was filed 1986-03-04 as a continuation-in-part of Ser. No. 06/525,813 (filed 1983-08-22). Claims reading only on the 1983 disclosure get the 1983-08-22 date; claims supported solely by matter added in the 1986 CIP are only entitled to 1986-03-04. This matters for six of the cited references, which issued after 1983-08-22 and can therefore only be §102(e) art (or art against CIP-only matter):

• US 4,430,659 (02/07/1984), US 4,463,083 (07/31/1984), US 4,478,782 (10/23/1984), US 4,554,655 (11/19/1985), US 4,578,788 (03/25/1986), US 4,647,947 (03/03/1987).

Everything else in the set published on or before 1983-08-22 and is full §102(a)/(b) art against the earliest priority date (subject to the one-year §102(b) bar analysis for anything published before 1982-08-22).

Note on the source of the list below. These are the references cited in the '615 patent itself (the examiner-cited "References Cited" list), which is the set I was asked to assess. They are corroborated verbatim at the Justia copy of the patent (https://patents.justia.com/patent/[4719615](/patent/4719615)) and the Google Patents copy (https://patents.google.com/patent/US4719615/en). I searched for a separate USPTO-cited set and found none beyond these; the JPO/EPO counterpart search report (EP 0 136 070 A1, filed 1984-08-22) cites a subset plus two additional documents noted in Tier 4.

---

The complete reference set

Counterpart search-report art (EP 0 136 070 A1, for completeness — NOT in the '615 U.S. list):

• EP 0 097 509 A2 (1984-01-04, Celanese Corp.)
• DE 30 38 533 A1 (1981-04-30, RCA Corp.)

---

Reference-by-reference §102 assessment

TIER 1 — Closest art (the references the specification itself argues against)

US 4,264,986 — Willis, "Information-recording process & apparatus" (pub. 1981-04-28; app. 06/019,443 filed 1979-03-12)

Description. A pulsed laser injects heat into a smooth-surfaced recording medium to produce a "crisply-defined change in surface curvature" — a bump or pit. The medium is a solid having first and second states associated with different specific volumes, produced by melting the medium and cooling at a selected rate. The patent expressly contemplates a compressible, elastic layer 47 "such as air, or transparent silicone rubber" above the active layer, plus a hard transparent protective layer 48. Readout is by phase/amplitude of reflected light. (Full text: https://patentimages.storage.googleapis.com/a0/bc/d5/da141ac3a14eb2/US4264986.pdf; see also GB 2 044 980 A.)

§102 assessment.

• Claims 1, 4, 19 and all dependents — NOT anticipated. Willis's operative mechanism is explicitly melt-then-resolidify ("the medium is selectively producible in either of said states by melting the medium and cooling it from the molten state"). Every independent claim of '615 requires the deformation to form "without liquefaction or gasification of said first or second materials." Willis therefore fails an express claim element — this is the classic case of a reference that is avoided by negative limitation. It also lacks the elements of (i) two bonded record layers, (ii) Tg2 > Tg1 relationship, and (iii) the second layer mechanically holding the first material in a stretched state.
• Residual risk: Claim 1's element (d) — the "second layer sufficiently strong and rigid" — is generic enough that an examiner could argue Willis's layer 48/47 combination reads on it. Willis's layer 47 is described as compressible, however, and layer 48 is a dust cover, so element (d)'s "strong and rigid to hold the first material in stretched, expanded condition" is not met. Expect this to be a §103 (not §102) battleground, if anything.

US 4,371,954 — Cornet, "Reversible memory structure with thermo-optical writing and optical reading" (pub. 1983-02-01; Thomson-CSF)

Description. A dual-layer structure on a low-CTE substrate: a bottom layer of relatively inextensible metal or polymer with high CTE, and a top layer of metal alloy in a martensitic phase with low CTE. The two layers have low adhesion — they are "not bonded together." Writing (below transformation temperature Tt) delaminates the bottom layer from the substrate and bumps the top layer; erasing raises the top layer above Tt into its "parent" phase.

§102 assessment.

• Claims 1, 4, 19 and all dependents — NOT anticipated. Cornet '954 fails at least three express limitations:
  1. Bonding. Claim 4(b)/19(b) require the second material to be "bonded to said first material." Cornet expressly teaches layers with low adhesion, not bonded — the opposite.
  2. "Without liquefaction or gasification" + no phase change. The specification and claims require both materials to "remain in the same physical state"; Cornet's erasing depends on the top layer transforming martensitic ↔ parent phase. Claim 19 has no metals limitation, but claim 15 ("said first and second materials are not metals or metal oxides") is a direct, deliberate swipe at Cornet's metal-alloy top layer and would defeat it outright.
  3. Mechanical holding by a rigidified glassy layer. Cornet holds the bump by phase stability of a metal alloy, not by a polymer cooled below Tg2.
• This is the single most important reference to have in the file history, because it is the closest dual-layer bump-forming teaching. It is also the reason claim 15 exists.

US 4,360,895 — Cornet, "Permanent memory structure with thermo-optical writing and optical reading" (pub. 1982-11-23; Thomson-CSF)

Description. The non-erasable sibling of '954: a permanent (write-once) thermo-optical structure.
§102 assessment. No anticipation of any claim. Because it is expressly a permanent structure, it cannot disclose the erasure/relaxation elements of claims 2, 5, 8, 9, 20. It is background/cumulative art for the bump-writing concept only. It remains relevant as §103 art for claims 1/4/19's basic two-layer geometry.

US 4,320,489 — Crandall et al., RCA, "Reversible optical storage medium and a method for recording information therein" (pub. 1982-03-16; app. filed 1980-03-03)

Description. A reversible medium comprising a conductive layer plus a light-absorbing thermoplastic layer. Information is recorded by corona-charging the thermoplastic surface and imagewise heating it above its softening temperature so that a random "frost" deformation forms; recorded information is erased by heating the uncharged thermoplastic above its softening temperature. Softening temperature ~125°C or less.

§102 assessment.

• Claims 1, 4, 19 — NOT anticipated. Although '489 is reversible, deformable, and dye/pigment-loaded (a substituted ethylene-dithiol light-absorptive compound), it fails:
  • Two bonded layers with differential Tg: '489 is a single thermoplastic deformable layer on a conductive ground plane; there is no "second layer ... bonded to said first material" with Tg2 > Tg1, and no second material that holds the first in a stretched state.
  • Mechanism: deformation arises from electrostatic-field-driven frost instability, not from "thermal expansion of said first material" being locked in by a rigidified second layer.
  • "Strong and rigid ... to hold": the frost layer is deliberately soft (softening ≤125°C).
• Residual risk — claims 2/5 (erasability) and 6/7/12/13 (dye loading): no, because those dependents inherit the structural requirements of claim 4/1. '489 is best characterized as §103 art on the idea of reversible thermal deformation in a dye-loaded polymer, not §102 art on the claims as written.
• One point to watch: '489's erasure "by heating ... above its softening temperature" is the same thermal erasure concept as claim 2/5 — but claim 2 requires the second material to be locally heated "with light radiation of a second preselected wavelength," which '489's bulk/local heating discussion does not map onto two-wavelength selectivity.

TIER 2 — Phase-change / optical-property references (mechanistically distinct)

US 4,278,734 — Ohta et al., Matsushita (pub. 1981-07-14)

Description. Optical medium in which the material changes physical state (amorphous ↔ crystalline), producing an increase/decrease in optical density. The specification concedes Ohta solved the speed (≈50 nsec) and contrast problems but criticizes it for not permitting bit-by-bit erasure, for requiring smooth surfaces, and for expensive materials.

§102 assessment. No anticipation of any claim. Ohta's mechanism is a physical-state transition — precisely what independent claims 1/4/19 exclude (claim 4/19 by the Tg2/Tg1 frame plus the "without liquefaction" language in 1, 4(f), 21; the specification states both materials "remain in the same physical state"). It is not a topographical bump medium. Pure background.

US 4,285,056 — Bell, RCA, "Replicable optical recording medium" (pub. 1981-08-18)

Description. Cited from the RCA optical-recording family (same lineage as '489; Spong/Bell pit-forming art is discussed in RCA's own later patents). Directed to pit/deformation formation in an absorptive layer, typically by local melting with pull-back of molten material by surface tension, leaving a pit with a rim.
§102 assessment. No anticipation. Melting-based pit formation = liquefaction, excluded by claim 1/4(f)/21. Single absorptive layer; no bonded dual layer with Tg differential; no elastic-hold mechanism.

US 4,336,545 — Howe, Eastman Kodak, "Optical disc structure, method and apparatus physically optimized for writing and reading with a single wavelength" (pub. 1982-06-22)

Description. An optical disc anti-reflection / optical-thickness-optimized structure for single-wavelength write/read. Also cited by '615 as a source of usable dyes (col. 8, lines 53–68).
§102 assessment. No anticipation of the medium claims. It is an optical-tuning/physical-optics reference; it lacks the dual-layer thermal-expansion-and-hold structure. Relevant only to the dye/optical-coupling aspects of claims 6, 7, 12, 13, 14 (and those are §103-type teachings at best).

TIER 3 — Secondary/cumulative art (structural or chemistry background; no independent-claim anticipation)

• US 4,398,203; US 4,404,656; US 4,405,994 — Cornet / Thomson-CSF (1983). The remaining Cornet family (thermo-optical method and medium). Same deficiencies as '954/'895 (metal alloy, phase change, unbonded layers). No anticipation. Relevant as §103 art showing the state of dual-layer thermo-optical media.
• US 3,689,768 (Sato, 1972) — electron-beam recording materials; and US 3,787,873 (Fuji Photo Film, 1974) — laser recording method/material. Early generic laser-recording art; no bonded two-polymer expansion/retention structure. No anticipation. Purely cumulative.
• US 4,268,575 (1981-05-19) and US 4,282,534 (1981-08-04) — "Thermal recording elements" (Fuji Photo Film lineage). Thermal-deformation imaging layers; no bonded dual-layer Tg1/Tg2 scheme, no two-wavelength write/erase. No anticipation.
• US 4,430,659 (Moffitt, 1984-02-07); US 4,463,083 (Kitajima, 1984-07-31); US 4,478,782 (Kuder, 1984-10-23); US 4,554,655 (Kumasaka, 1985-11-19); US 4,578,788 (Ahn, 1986-03-25); US 4,647,947 (Takeoka, 1987-03-03).
  • Date caveat: all six issued after the 1983-08-22 priority date. They are §102(a)/(e) art only as to portions of the specification added as new matter in the 1986 CIP, or under §102(e) on their filing dates against an invention date later than that. This is a point that must be resolved claim-by-claim against the parent-vs-CIP support analysis; I flag it as a genuine risk rather than a certainty, since the '615 claims may be fully supported by the 1983 parent.
  • Substance: I was unable to retrieve and verify the full text of these six individually before exhausting my search budget. I am therefore not asserting any specific §102 mapping for them. Based on titles, classification, and the fact that they appear in a deformation/bump-forming + optical-recording art unit, the most likely substantive relevance is as §103 art on dye/polymer recording layers and bump detection, not as §102 anticipatory art on claims 1/4/19. Treat this as unverified.

TIER 4 — Foreign art and NPL

• JP 57-60546 (Fujitsu, 1982-04-12) — "Optical recording medium." §102(b) art (published >1 yr before 1983-08-22? No — published 1982-04, i.e., ~16 months before; so §102(a) art against the 1983 priority). Which claims: potentially relevant to the generic "optical recording medium with a deformable layer" concept; I could not retrieve its text, so no specific §102 mapping is asserted.
• JP 58-62096 (Matsushita, 1983-04-13) and JP 58-94145 (TDK, 1983-06-04) — both published only ~4 months before the 1983-08-22 priority date, so they are at best §102(a) art against the invention date (not §102(b) art, since less than one year). If the actual invention date of the 1983 parent precedes their publication, they are not prior art at all. No specific §102 mapping asserted (texts not retrieved).
• Van Vlack, Elements of Materials Science, 2nd ed. (1967), pp. 76–77. A pre-1982 printed publication — full §102(b) art as to whatever it discloses. Page 76–77 of the 2nd edition is a materials-science fundamentals passage (specific volume / density / thermal-expansion and phase relations of solids). Anticipation analysis: a textbook passage teaching general thermal expansion and glass-transition/specific-volume behavior cannot, by itself, anticipate any claim, because it discloses no multi-layer optical medium. Its role is as §103 background supporting the proposition that the Tg/CTE relationships recited in claim 4 were known material properties — i.e., it supplies motivation rather than disclosure.
• EP 0 097 509 A2 (Celanese, 1984-01-04) and DE 30 38 533 A1 (RCA, 1981-04-30) — cited only in the counterpart EP search report, not in the '615 U.S. list. The DE document (published 1981) post-dates nothing relevant and is RCA optical-recording art; the EP document (1984) post-dates the U.S. priority and is only §102(e)-type art. Neither is a '615 U.S. file reference; assess separately if litigation-grade review is needed.

---

Summary matrix — where §102 risk actually sits

Bottom line for §102: On the record I can verify, no cited reference anticipates any of claims 1, 4, or 19, and the examiner evidently agreed — the three independent claims issued over all of them. The references function in this file as §103 art, and the two that matter are Willis US 4,264,986 (bump-forming, but by melting) and Cornet US 4,371,954 (dual-layer bump, but unbonded + metal + phase change). The '615 claims were deliberately drafted around both: the "without liquefaction or gasification" language neutralizes Willis; the "bonded," "same physical state / Tg2 > Tg1," and claim-15 "not metals or metal oxides" limitations neutralize Cornet. That is the strongest reading of this reference set, and it is the one that explains the claim language.

---

Caveats and what I could not verify

1. Six U.S. references (Tier 3, 1984–1987) and all three JP references were not retrieved in full text before my search budget was exhausted. I have not fabricated a §102 mapping for them; the "undetermined" entries are honest gaps, not conclusions. A definitive §102 table for those requires pulling their full texts and their respective filing dates (for §102(e) analysis against the CIP-only matter).
2. The CIP/parent support question is unresolved. Which '615 claims are entitled to the 1983-08-22 priority date — and therefore which of the 1984–1987 references are even prior art — depends on an element-by-element comparison against the abandoned parent Ser. No. 06/525,813. I have not seen the parent's specification. This is the single most consequential open issue for any validity challenge to '615, and it should be resolved from the file wrapper (USPTO PatentCenter for app. 06/835,960).
3. Verbatim source note: the reference list and the specification's characterizations of Ohta, Willis, and Cornet are taken from the authoritative full text of US 4,719,615 supplied in the record (https://patents.google.com/patent/US4719615/en) and corroborated at https://patents.justia.com/patent/4719615. Willis's mechanism was confirmed from the published patent PDF (https://patentimages.storage.googleapis.com/a0/bc/d5/da141ac3a14eb2/US4264986.pdf), and Crandall's from Google Patents (https://patents.google.com/patent/US4320489). Where I cite a reference's content, it is from that reference's own text; where I infer relevance from title/classification only, I say so.
4. This is a patentability/validity analysis of the reference set, not a legal opinion; §102 dates and CIP support are legal determinations for counsel.

I'll ground this §103 analysis in the actual cited references. Let me pull what the key prior-art documents disclose.

Obviousness Analysis of U.S. Patent No. 4,719,615 under 35 U.S.C. § 103

Building on the previously generated bibliographic, summary, claim-construction, and litigation sections. This section adds the §103 analysis and does not repeat prior content except where a cross-reference is essential.

---

0. Framework and threshold caveats

Governing law. The '615 patent has a 1983‑08‑22 priority date and was filed 1986‑03‑04 as a continuation‑in‑part of Ser. No. 06/525,813. It is therefore pre‑AIA § 103(a) territory. The controlling test is the Graham v. John Deere framework (scope/content of the prior art; differences; level of ordinary skill; secondary considerations), with the Federal Circuit's teaching‑suggestion‑motivation (TSM) test as relaxed by KSR Int'l v. Teleflex — i.e., a combination can be obvious based on "common sense," design incentives, and predictable variations, not only on explicit TSM.

Effective date. Only subject matter disclosed in the abandoned parent gets the 1983 date. The two‑material/Tg‑differential/two‑wavelength architecture appears to be new matter added in the CIP, so the claims most likely get the 1986‑03‑04 date. This is immaterial here: all 23 references cited on the face of the patent predate 1983‑08‑22, so they are § 102(b) art under either date.

Retrospective exercise. The patent is Expired – Lifetime, anticipated expiration 2005‑01‑12. A § 103 analysis now is hypothetical (licensing/royalty disputes, defensive study, or historical record). Note also that the claims issued over these very references, so the § 282 presumption of validity attaches — but it is rebuttable and carries no evidentiary weight where the art was before the examiner.

PHOSITA. I assume a worker with a graduate degree or 3–5 years' experience in optical recording media and polymer materials, conversant with (i) thermoplastic/elastomeric deformation ("bump"/"frost") recording, (ii) glass transition (Tg) behavior and viscoelasticity, and (iii) dye‑sensitized polymer layers for wavelength‑selective absorption. Under KSR, this person is a person of ordinary creativity, not an automaton.

Claim scope note. All 21 claims are medium claims; methods/apparatus were pursued in the divisional US 4,852,075. Independent claims are 1 (structural), 4 (Tg‑defined), and 19 (polymer‑specific).

---

1. The limitations that must be met

The core inventive thrust is therefore a three‑part combination: (a) a bonded dual‑layer deformable medium; (b) no change of physical state — deformation via differential thermal expansion of the bottom layer, held by the top layer; (c) erase by heating the top layer above its Tg so the elastic bottom layer recovers.

---

2. The cited prior art, mapped to those limitations

Reading of the art (grounded in the sources above): by 1983 the field already contained (1) an erasable dual‑layer thermo‑optical bump medium where a stiff upper layer mechanically retains the bump (Cornet); (2) a reversible, dye‑tuned thermoplastic deformation medium erased by heating above a softening/Tg temperature (RCA '489); and (3) explicit recognition that a polymer layer can be thermally expanded "without change of state" to deform an adjacent layer, with optical‑cavity tuning for readout (Thomson‑CSF '203), plus explicit bonding of the two layers ('895).

---

3. Combinations that would render the claims obvious

Combination A (strongest): Cornet '954 + RCA '489 + Thomson‑CSF '203 (optionally '895)

• From Cornet: the architectural concept of an erasable dual‑layer bump medium in which the upper layer holds the deformation after cooling and erasure is accomplished by heating the upper layer to release what it is holding. This supplies claims 1(b)–(d) and claim 2's erase concept.
• From RCA '489: the substitution of a thermoplastic with a softening temperature (Tg) for the top layer, a light‑absorptive dye for wavelength selectivity, and erasure by heating above that softening temperature with no ablation. This supplies claim 4(a)–(b) ('Tg2 above ambient'), claim 6–7 (dye), and claim 2 (thermal erase without liquefaction/gasification).
• From Thomson‑CSF '203: the recognition that the lower layer's localized thermal expansion occurs "without change of state" and is what deforms the upper layer, plus optical‑cavity/thickness tuning so the bump is optically detectable. This supplies claim 1(c) (deformation by thermal expansion), claim 4(e) (permit/retain behavior), the "same physical state" element, and the read‑contrast function.
• From '895: the layers "can remain adhered" — i.e., bonding the two layers is a known, optional design choice. This supplies claim 1/4's "bonded" limitation and rebuts Cornet's low‑adhesion teaching as merely one alternative.

Motivation to combine: (i) Cornet's metal/martensitic system requires high‑power pulses and its phase transformation makes erase slow and at a different power than write — the '615 spec itself identifies these as the problems to be solved; (ii) RCA '489 shows thermoplastics are cheaper, lower‑power, and reversibly deformable by heating above a softening point; (iii) '203 shows an organic layer expands without phase change, which is exactly the property that makes erase fast and uniform. A PHOSITA optimizing Cornet for low power and high cycle life would predictably look to thermoplastic/elastomer layers with engineered Tg's. Result: predictable ("combination of familiar elements according to known methods yields predictable results," KSR).

Combination B: Cornet '954 + Willis '986 + Ohta '734

• Willis supplies the deformable‑bump, good‑SNR readout goal and Ohta the fast, high‑contrast goal — both explicitly acknowledged in the '615 background as the two problems (contrast and data rate) to be solved.
• Cornet supplies the bonded, retention‑layer, erase‑by‑heating structure.
• Motivation: the '615 specification frames its own invention as solving precisely the contrast problem (Willis/Ohta) and the speed/erase problem (Cornet) simultaneously. That framing is an admission that a PHOSITA pursuing both goals would look to the same references.

Combination C: RCA '489 + Thomson‑CSF '203 + Cornet '954

• RCA '489's reversible thermoplastic‑with‑dye + '203's non‑phase‑change polymer expansion + Cornet's retention/erase concept. Practically the same result as Combination A, reached from the thermoplastic side rather than the erasable‑dual‑layer side.

Dependent claims under any of the above

• Claims 3, 14 (top layer substantially transmits the first wavelength): routine optical‑property matching, taught by '203's tuned cavity and RCA '489's dye design.
• Claims 6–7, 12–13 (dyes): RCA '489 and the incorporated Kodak '545 / Sato '768 dye teachings.
• Claims 10–11 (both polymers; low thermal conductivity): directly taught by RCA '489 and '203; low conductivity is a recognized desideratum in thermo‑optical media.
• Claims 16–18 (high‑yield‑strain elastic first layer; thermoplastic/thermoset second layer; glassy‑at‑ambient/rubbery‑above‑Tg2): standard polymer‑physics selection in view of the Tg framework already supplied.
• Claim 20–21 (dependents of 19): same reasoning.
• Claim 15 ("not metals or metal oxides"): a negative limitation whose only apparent function is to exclude Cornet/'203/'895. If polymer substitution is obvious, this exclusion adds no patentable weight.

---

4. Motivation‑to‑combine summary (TSM + KSR rationales)

1. Identified problem in the art: the '615 background expressly states that existing erasable media (Willis, Cornet) suffered from (a) high‑power/erase‑power mismatch, (b) slow data rate, (c) fatigue limits, and (d) susceptibility to surface anomalies. Addressing these in the same field is a classic KSR motivation.
2. Reasonable expectation of success: RCA '489 already achieved reversible, dye‑tuned, non‑ablative thermoplastic deformation; only the retention‑by‑a‑stiffer second layer needed to be carried over from Cornet/'895 — a mechanical, predictable design choice.
3. Design incentives / substitution of known equivalents: replacing a martensitic metal top layer with a glassy polymer above the bottom layer's Tg is a substitution of known materials for known functions (both "hold a bump below a transition and release above it").
4. "Obvious to try" with finite, predictable options: the field had a small set of known layer types (metals, thermoplastic polymers, elastomers, phase‑change chalcogenides). Choosing a high‑CTE elastomer + higher‑Tg polymer was within that enumerated set.

---

5. Counter‑arguments that could defeat or complicate an invalidity theory

A rigorous §103 opinion must flag these, because they are what likely sustained allowance:

1. Cornet teaches away from the "bonded / no‑delamination" limitation. Cornet requires low adhesion between layers and delamination of the bottom layer from the substrate; the '615 requires the layers be bonded and that the bottom layer not delaminate. An examiner or court could find Cornet teaches the opposite of the claimed adhesion, weakening the motivation to combine. ('895's "can remain adhered" partially neutralizes this, but '895 is a permanent non‑erasable medium.)
2. The "same physical state" / "without liquefaction or gasification" limitations are negative limitations that exclude phase‑change (Willis, Ohta, Cornet's martensitic transformation) and ablative (Spong‑type) media. If the only offered combinations still rely on a phase change for the retention/release function (Cornet), the combination does not squarely read on the claims.
3. The erase mechanism may be the true point of novelty. In Cornet, erase works because the top layer transforms phase; in the '615, erase works because the bottom elastic layer's stored elastic energy pulls the surface flat once the top layer is softened (claims 8–9). If no reference teaches or suggests relying on the elastic recovery of the expansion layer as the erasing force (as opposed to surface tension in RCA '489's frost or phase transformation in Cornet), a non‑obviousness argument on claims 8–9 and on claim 1's "holds … in stretched, expanded condition" is credible. Note, however, that RCA '489's frost erasure (heating above softening to let the deformation relax) is functionally close, and a court might view elastic recovery as inherent in any "hold/release‑by‑Tg" scheme.
4. Secondary considerations. Optical Data's dye‑polymer bump technology was commercially developed (the THOR‑CD/Tandy relationship; a reported Philips & DuPont Optical license). If a nexus between that commercial success and the '615 claims could be shown, it would weigh against obviousness. The prior sections found no documented litigation or adjudicated secondary‑consideration record, so this is currently unsupported.
5. Functional claim language ("adapted to permit," "adapted to retain") raises § 112 enablement/indefiniteness questions but is not a §103 defense.

---

6. Net assessment

Bottom line. On the strongest combination — Cornet U.S. 4,371,954 + RCA U.S. 4,320,489 + Thomson‑CSF U.S. 4,398,203 (with U.S. 4,360,895 for bonding) — there is a substantial, articulable prima facie case that independent claims 1, 4, and 19 are obvious, principally because each recited element is disclosed by a reference in the same field directed to the same problems the '615 acknowledges. The case is strongest on claim 1, contested on claims 4 and 19 (the specific Tg‑differential + high‑CTE + bonded + non‑phase‑change combination), and weakest against claims 8–9 (elastic‑recovery erasure). A final conclusion would turn on (a) whether Cornet's express low‑adhesion/delamination teaching is treated as teaching away from the claimed bonding, and (b) whether the elastic‑recovery erase is deemed an inherent or a patentably distinct mechanism. Because the patent is expired and no litigation is documented (per the prior sections), this analysis is ultimately academic/defensive rather than tied to a live validity contest.

---

Confidence

• Governing framework, effective‑date reasoning, PHOSITA level: high.
• Content of Cornet '954, Willis '986, Ohta '734: high — taken from the '615 specification's own discussion of these three references.
• Content of RCA '489, Thomson‑CSF '203 and '895: moderate‑to‑high — grounded in the Google Patents full text retrieved during this analysis (URLs cited above); the RCA '489 inventor list ("Crandall et al.") is from the document header.
• Whether any specific combination would actually be held obvious: moderate — this is a legal, fact‑dependent determination; I have flagged the counter‑arguments rather than asserting a definitive conclusion.
• No assertion is made that any of these references was or was not relied upon by the examiner beyond the citations listed on the face of the patent.