US 4204636

I'll search for authoritative information on this patent number, including any litigation or docket references.

Let me run one more targeted search on litigation/assignment records for this specific patent.

US Patent 4,204,636 — Analyst Summary

Bibliographic data

Uncertainty note: The assignee is rendered literally by Google Patents as "STANLEY HAYMAN AND Co" (both as original and current assignee). I cannot independently verify the corporate identity or any assignment history for that entity from the sources retrieved. No CAFC docket reference was found for this number.

Abstract (verbatim)

"Circuitry receives data information from a wand and decodes the same for duplicating a keyboard input to a display. The interface circuit includes a plurality of switches paralleling the keyboard input switches and generates autonomously a control signal indicative of a status of the display."

Plain-language overview

The patent addresses a practical era-specific problem: an OCR ("OCR WAND," trademark of Recognition Equipment, Inc.) wand that expects a "user ready" handshake from a host cash register, plugged into a point-of-sale terminal that has no circuitry to generate that handshake. The disclosed interface board sits between the wand's electronics (connector 16) and the terminal:

• Decoders (TI SN7442 and SN74154 in the described embodiment) split the six wand data bits into a 4-line and a 16-line selection, driving a relay coil matrix so that decoded ASCI character data becomes switch closures on wires 58.
• Those closures parallel the cash register keyboard's own switch lines (wires 24), so the terminal's existing processor treats wand input exactly as if a key had been pressed — no terminal modification needed.
• Two astable multivibrators (TI SN74123) with RC timing networks: one delays enabling of the decoder until the wand data stabilizes (eliminating glitches), and the other autonomously fabricates the "user ready" pulse on line 10 without any communication from the processor.
• Because the fabricated readiness signal is not a true feedback signal, the design deliberately relies on the human operator to notice the terminal's error beeper and rescan the label, rather than building a full handshaking loop.

Independent claims (claims 1, 3, 6); claims 2, 4, 5 are dependent

Claim 1 — apparatus in combination. The combination includes data reading means (encoded data output), display means, processing means, manual input means producing switch closure information, and means routing that switch closure information to the processing means. The interface apparatus is: (a) means for converting the encoded data output into data simulating the switch closure information, and (b) means connecting that simulated data to the routing means — such that the processor and display operate on data from either the manual input means or the data reader independently of the source of the display information.

Claim 3 — data entry system (improvement claim). In a system with keyboard means, display means accepting a predetermined data format, and processing means converting keyed data into that format, the improvement is (a) data reading means with encoded data output, plus (b) apparatus enabling the system to accept that encoded output for processing and display, including means for converting the encoded output into a format simulating the keyed data — so the system accepts both wand and keyboard data "without special modification."

Claim 6 — method claim. Steps: detecting a control signal that data is ready to be read from the wand; automatically generating a response simulating a user-ready condition; reading the data from the wand; converting the data to a preselected formatted signal; decoding that formatted signal as equivalent to corresponding keyboard entries (thereby simulating them); and processing the data the same as if keyboard-entered.

Dependent add-ons: claim 2 (and claim 4, tracking it) adds means for simulating the wand's expected first/ready signal independently of the processor, plus means for transmitting that simulated signal to the wand. Claim 5 specifies that the converting means comprises decoding means and relay means responsive to it.

Prior art cited on the face of the patent (6 references)

US 3,413,447 (Monarch Marking, 1968); US 3,735,350 (Lemelson, 1973); US 3,792,236 (Monarch Marking, 1974); US 3,826,900 (NCR, cordless scanning probe, 1974); US 3,963,901 (Intelcom, 1976); US 3,991,299 (Norand, 1976).

Discrepancy note (reported literally, not corrected): the patent body discusses additional/overlapping art not appearing on that six-item front-page list — e.g., Gott US 4,088,981, Cunningham US 3,812,347, Mak et al. US 3,238,501, Neff US 4,048,617, Dobras et al. US 3,792,236, Humbarger US 3,585,367, and Ferry US 920,940. Ferry US 920,940 as printed is a five-digit number appearing in a 1978-era specification citing 3-, 4-, and 7-digit art; I am flagging this rather than adjusting it.

Litigation / CAFC 2026 docket search

I searched for district-court and Federal Circuit activity referencing 4,204,636 / 4204636. No CAFC 2026 docket entry, appeal, IPR, or infringement suit citing this patent number was found. This is the expected result given the patent's status: filed 1978-10-13, it expired under the then-applicable 17-years-from-issue term on or about 1998-10-13 (Google Patents records this as "anticipated expiration"), so it has been off-term for over a quarter-century. Any 2026 court activity would be procedurally unusual; I found none, and I would not infer one from the absence of hits.

Sources

• Full text/PDF: https://patents.google.com/patent/US4204636/en and https://patentimages.storage.googleapis.com/32/49/44/5eb2848f5dce70/US4204636.pdf
• Justia: https://patents.justia.com/patent/4204636
• USPTO.report: https://uspto.report/patent/grant/4204636

I'll search for litigation involving US patent 4204636.

Let me check the specific litigation databases the user mentioned.

Let me try additional litigation-specific sources.

Litigation Search Report — U.S. Patent No. 4,204,636

Patent identified: U.S. Patent No. 4,204,636 — "Interface between data reading wand and cash register display"

• Inventor: Richard W. Hayman
• Assignee: Stanley Hayman and Co
• Application No. 05/950,967; filed October 13, 1978; granted May 27, 1980
• Legal status: Expired – Lifetime; anticipated expiration October 13, 1998
• Source: https://patents.google.com/patent/[US4204636](/patent/US4204636)/en

Bottom line

No litigation involving U.S. Patent No. 4,204,636 is known, and my searches returned none. I want to be explicit that this is a negative finding based on the sources I was able to reach, not a certified "no litigation exists" conclusion, because two things limit me here:

1. I could not directly query the sources you named. Repeated attempts to search portal.unifiedpatents.com/litigation/caselist, Docket Alarm, CourtListener, and PACER-indexed material did not return any usable, case-level result pages — the searches either returned unrelated documents or no results at all. I did not obtain a live return from the Unified Patents litigation caselist, and I have no PACER or CAFC docket access. So I cannot claim I have conclusively checked those databases.
2. No search result linked this patent number to any case. Searches for "4204636" combined with litigation, lawsuit, infringement, Docket Alarm, and CourtListener returned hits only for unrelated uses of that numeric string (e.g., a Lithuanian corporate-registry notice number, an Argentine trademark act number, an EPO/Lithuanian figure number, a Japanese patent JP 4204636 B2, and a CourtListener financial-disclosure investment ID). None of those concern U.S. Patent 4,204,636.

I also ran a search aimed at the assignee/inventor ("Stanley Hayman" in connection with a wand/cash-register interface patent), but that query was cut off before returning results, so I cannot report a result from it either way.

Case table

Because no litigation was located, there are no plaintiff/defendant/jurisdiction/case-number/filing-date/outcome rows to report. I will not populate the table with invented or approximated entries.

Why a null result is factually plausible here

This is not a surprising result, for several reasons grounded in the patent record itself:

• The patent has been expired for roughly 28 years. Google Patents records the anticipated expiration as October 13, 1998, and the status as "Expired – Lifetime." Any infringement action would have had to clear the courthouse steps before that date.
• The enforceable window was short and the technology narrow. Filed in October 1978 and granted in May 1980, the patent's commercial life coincided with early OCR-wand point-of-sale retrofits. The specification itself describes an interface board bolted between a Recognition Equipment "OCR WAND" and cash registers from Sweda, NCR, Data Terminal Systems, MKD, and TRW (see the Background section of the patent).
• The patent-era citation trail shows citations, not suits. Google Patents lists six examiner-cited references (US 3,413,447; 3,735,350; 3,792,236; 3,826,900; 3,963,901; 3,991,299) and a number of later citing documents (e.g., US 4,420,819; US 4,457,016; US 4,679,154; GB 2,259,597). Being cited by a later patent is a prior-art citation relationship, not evidence of litigation.
• Pre-1982 dockets are thinly digitized. Even if a suit had been filed in the early 1980s, PACER's electronic coverage for many districts generally begins around 1990 or later; earlier paper dockets are often not retrievable online. This is a real blind spot for a patent of this vintage.

Recommended next steps to close the gap (since I could not complete them)

To convert this into a defensible clearance/litigation check, someone with database access should:

1. Run a patent-number search on the Unified Patents litigation caselist (https://portal.unifiedpatents.com/litigation/caselist) using the literal string 4204636 and also 4,204,636 — the case list is built on patent-number matching and would surface any district-court or PTAB activity.
2. Query Docket Alarm / CourtListener (RECAP) for the patent number and for assignee "Stanley Hayman and Co."
3. Pull a USPTO Patent Assignment / PatentCenter "Litigation" or "Reexamination" tab for 4,204,636. Notably, there is no reexamination, reissue, or post-grant proceeding reflected in the Google Patents record — under the pre-AIA regime this patent would only have been exposed to ex parte reexamination (35 U.S.C. § 302), as inter partes reexamination and IPR did not exist during its lifetime.
4. Check pre-PACER paper dockets in likely districts (D.N.J., N.D. Tex., S.D.N.Y.) if a specific infringement concern arises.

Important caveat on the identifier: Per the strict rule you set, I have treated "4204636" literally as U.S. Patent No. 4,204,636 throughout and have not substituted any similar number. I want to flag that the Google Patents "Current Assignee" field reads "STANLEY HAYMAN AND Co" with an apparent spacing/formatting artifact; I have not auto-corrected it, and it should be verified against the USPTO assignment record before relying on it for a name-based litigation search.

I'll verify the structured "no proceedings" finding against live sources before writing up.

Let me run a couple more targeted checks before writing up.

Proceedings overview

Total AIA trial proceedings on U.S. Patent No. 4,204,636: 0. The structured "PTAB proceedings on file" block (USPTO Open Data Portal ingest) returns an empty set, and my independent web searches for the patent number, the title, and the inventor/assignee surfaced no IPR, PGR, CBM, or derivation proceeding — no institution decisions, no final written decisions, no settlements, no appeals. Breakdown by status: 0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution denials. There is therefore no PTAB-based defensive posture to report either way: no claim of this patent is helped by a Board holding, and none is killed by one. What the record actually gives a defendant is much simpler and much stronger than an IPR win — the patent expired on 1998-10-13, roughly fourteen years before the first AIA trial was ever filed (the Board's electronic trial-filing system went live 2012-09-16). There is no live claim to assert.

Proceedings detail

No proceedings to profile. I will not construct the ### {PROCEEDING_NUMBER} — {Petitioner} v. {Patent Owner} entries the template calls for, because doing so would require inventing proceeding numbers, and you directed me not to.

Why the null result is structural, not a gap in searching

Four independent reasons converge, and each is verifiable from the patent record already developed in the two prior sections:

1. Timing forecloses three of the four AIA vehicle types outright. AIA trials (IPR/PGR/CBM) became available only after the AIA's 2012-09-16 effective date. This patent's anticipated expiration was 1998-10-13 (Google Patents: "Expired – Lifetime"). PGR is unavailable as a matter of statute — it reaches only patents with an effective filing date on or after 2013-03-16. CBM review under AIA § 18 is unavailable both because the § 18 window closed on 2020-09-16 and because these claims (decoders, relay matrices, astable multivibrators for a wand-to-keyboard interface) are directed to hardware interconnection, not to "a financial product or service," and would in any event likely qualify for the "technological invention" exclusion.
2. IPR is legally available against an expired patent, but practically inert here. For the avoidance of doubt for any defendant: the Board does have authority to review expired patents. See Apple Inc. v. Gesture Tech. Partners, LLC, IPR2021-00922, Paper 10 (PTAB Nov. 29, 2021), and the Board's discussion in IPR2023-00590 (Paper ___, Patent 7,155,451 B1) rejecting the argument that "expiration removes the patent from the [Office's] jurisdiction." But the Office's current practice weighs heavily against instituting on long-expired patents: see the 2025 line of discretionary-denial decisions collected in the Director Review request excerpted at https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1557856](/patent/1557856)/ ("the challenged patent has been in force for approximately twenty years and is now expired, creating strong settled expectations"; SAP America, Inc. v. Valtrus Innovations Ltd., IPR2025-00415 et al., Paper 10 (PTAB July 10, 2025)). A patent that has been expired for ~28 years sits at the far end of that spectrum. Nobody petitioned, and nobody would have a reason to.
3. No pre-AIA-style administrative challenge either. The prior sections record no reexamination (ex parte or inter partes), no reissue, and no certificate of correction. Inter partes reexamination did not exist until 1999-11-29 — after this patent had already expired.
4. Citations are not proceedings. The 15–20 "Cited By" entries on the Google Patents page (US 4,420,819; US 4,457,016; US 4,679,154; GB 2,259,597; the SambaNova reissue/sibling-family entries; etc.) are prior-art citation relationships from later prosecution files. Being cited by a later patent is not evidence of any adversarial proceeding.

Strategic summary

Claim status: all six claims are UNTESTED — and all six are EXPIRED. No claim of 4,204,636 has ever been canceled, and none has ever been confirmed, because no tribunal has ever construed them. The only claim-level "disposition" of this patent is the calendar: filed 1978-10-13, issued 1980-05-27, expired on or about 1998-10-13 under the then-governing 17-years-from-issue term. For a defendant receiving a demand letter citing claims 1, 3, or 6, the correct response is not an IPR-style invalidity argument — it is that there is no enforceable right. Any sole remaining theory is a claim for damages accruing during the term (35 U.S.C. § 284), and the last possible day a cause of action could have accrued is 1998-10-13. The six-year limitations bar of 35 U.S.C. § 286 extinguished even that residual recovery around 2004-10-13. There is no theory of ongoing infringement of an expired patent, and no injunctive or exclusionary relief is available.

Estoppel landscape: empty, in both directions. Because no IPR/PGR/CBM was ever instituted, 35 U.S.C. § 315(e)(2) estoppel attaches to no one. No petitioner or privy is barred from raising any ground; conversely, no defendant can borrow a Board holding to knock out a claim. The practical consequence is asymmetric in the defendant's favor only because the patent is dead: the full universe of prior art remains available for a defense under §§ 102/103/112 if this patent were somehow revived into a live dispute — including the six examiner-cited references (US 3,413,447; US 3,735,350; US 3,792,236; US 3,826,900; US 3,963,901; US 3,991,299) and the additional art the patent body itself discusses but that never appears on the front page (Gott US 4,088,981; Cunningham US 3,812,347; Mak et al. US 3,238,501; Neff US 4,048,617; Dobras et al. US 3,792,236; Humbarger US 3,585,367; Ferry US 920,940). That body-text/front-page mismatch, flagged in the prior summary, is itself a § 102/§ 103 lead if anyone ever needed one. Ferry US 920,940 is reported literally as printed; I have not corrected the five-digit number.

Pattern signals: none of the tell-tales are present. There is no serial petitioner (no petitioner at all), no PTAB-appeal track record for the patent owner, and no defensive aggregator (Unified Patents or similar) in the chain. The Google Patents assignee field still reads "STANLEY HAYMAN AND Co" verbatim as original and current assignee, with the apparent spacing artifact noted earlier; I have not auto-corrected it, and it should be confirmed against the USPTO assignment record before anyone runs a name-based search on it. The absence of IPRs is itself a signal here, but a weak one: well-asserted patents attract IPRs, and this one was never asserted in any proceeding I could locate. That is consistent with a short-lived, narrow, era-specific commercial product rather than a patent that a troll would find worth reviving.

Recommended next steps

1. Do not commission an IPR. There is nothing to review. The Office's own 2025 practice — collected in the Director Review request at https://ptacts.uspto.gov/ptacts/public-informations/petitions/1557856/ — treats long-expired patents as strong candidates for discretionary denial on "settled expectations" grounds, and this patent has been expired for ~28 years. An IPR petition would be an expensive denial.
2. Ground the defense in the term, not the merits. Letter response should cite: (a) expiration on 1998-10-13 (Google Patents legal status: "Expired – Lifetime," anticipated expiration 1998-10-13), and (b) 35 U.S.C. § 286's six-year bar, which cut off the last conceivable damages accrual around 2004-10-13.
3. If any opposing party claims otherwise, verify at the source. Search PTAB E2E by patent number and by party name at https://ptab.uspto.gov/ (non-registered users may search and review AIA review documents without an account), and cross-check the newer petitions portal at https://ptacts.uspto.gov/ptacts/ and the ODP API. Confirm on CourtListener (https://www.courtlistener.com/?q=%224204636%22) and in USPTO Patent Assignment / PatentCenter that no assignment, reissue, reexamination, or litigation certificate has altered the status.
4. Caveat on completeness: the structured block reports no proceedings "as of the most recent ingest." A filing in the last few days could theoretically post-date the ingest. My independent searches found nothing, so I do not expect one — but for a clean file, re-run the E2E patent-number search on the day you need it, and note that PTAB E2E lists a proceeding only once the multi-step petition submission completes, so a started-but-unfiled petition will not appear.

Bottom line for a defendant: the PTAB record on 4,204,636 is blank, and it is blank for a structural reason — the patent has been off-term since 1998-10-13, fifteen years before an AIA trial could have been filed against it. If a demand letter cites claims 1–6 of this patent, the correct disposition is "expired and time-barred," not "invalid at the Board."

I'll attempt to retrieve the actual assignment records from the USPTO Assignment Center and cross-reference sources.

Initial searches returned patent text but no assignment records. Let me query the assignment databases and legal-event sources more directly.

Let me try additional angles — the Google Patents legal events, the assignee entity, and inventor background.

I hit the tool-call ceiling before I could reach a case-level page in the USPTO Assignment Center. Below I report strictly what I could and could not verify, with no fabricated reel/frame entries.

Ownership Chain — U.S. Patent No. 4,204,636

Retrieval caveat up front. I attempted to query the USPTO Assignment Center (assignmentcenter.uspto.gov) and the legacy assignment search (assignment.uspto.gov/patent/index.html) for patent 4204636 via web search. Both are dynamic search interfaces that do not expose case-level records to general web crawling, and my searches returned no usable assignment record for this patent. I therefore have no reel/frame, no conveyance type, no correspondent, and no execution/recording dates to report. I flag this as a negative retrieval, not a certified "no assignments exist" conclusion.

What I can report from the authoritative Google Patents full text (fetched 2026-10-01) is that its legal-events dataset for US4204636 contains no ownership-transfer entries — only:

• 1978-10-13 — application filed by STANLEY HAYMAN AND Co; priority to US05/950,967
• 1980-05-27 — application granted; publication of US4204636A
• 1998-10-13 — anticipated expiration (status: Expired – Lifetime)

Disambiguation warning (important). The Google Patents record contains two tables that are easy to misread as ownership data but are not: the "Citations (6)" table (US 3,413,447; 3,735,350; 3,792,236; 3,826,900; 3,963,901; 3,991,299) and the "Cited By (15/20)" table (e.g., US 4,420,819; US 4,457,016; US 4,679,154; GB 2,259,597). Both are prior-art citation relationships, not assignments. No entity in those tables acquired this patent.

Inventors

• Richard W. Hayman — sole named inventor (per the Google Patents bibliographic record and the front-page inventor field).

Employer at time of filing: The patent names STANLEY HAYMAN AND Co as original assignee, and the shared surname raises a plausible inference that the inventor assigned to a company associated with him. I could not retrieve the assignment document, the execution date, or the assignor's stated capacity, so I cannot confirm whether the assignment was made by the inventor (employment/obligation-to-assign), by a prior owner, or was filed by the company as applicant. Do not treat the surname match as established fact — it is an inference, not a record.

Unusual-pattern check (departures within 12 months): Not determinable. With no assignment documents and no PEDS/ODP prosecution-history access retrieved, I cannot establish inventor departure timing. There is also only one inventor, so the multi-inventor "everybody left" pattern is structurally inapplicable.

Original assignee

• Name as recorded on the patent: "STANLEY HAYMAN AND Co" (rendered literally by Google Patents as both original and current assignee). Consistent with prior sections, I am not auto-correcting this string; it may reflect a real corporate name, a formatting artifact, or an abbreviation of a longer registered name. Verify against the USPTO assignment record and any state corporate registry before relying on it for name-based searching.
• Product embodying the claims: Presumed yes, on the face of the specification — the patent describes a physical interface circuit board mounted between a Recognition Equipment "OCR WAND" and third-party cash registers (Sweda, NCR, Data Terminal Systems, MKD, TRW are all named in the Background). That is consistent with a small integration/OEM vendor, not a major terminal manufacturer. I could not verify shipping volumes, model numbers, or commercialization.
• Primary line of business: Unclear. The named field ("interface between data reading wand and cash register display") and the era (filed 1978) suggest point-of-sale peripheral integration. Not independently verified.
• Current status: Unknown — not determinable from the sources retrieved. I found no evidence of operating status, acquisition, dissolution, or bankruptcy. I will not guess.

Assignment timeline

No recorded assignment could be retrieved for this patent. Per the constraint against fabricating assignments, I am reporting a null result rather than approximating reel/frame entries:

• My queries against assignmentcenter.uspto.gov and assignment.uspto.gov returned no case-level record for 4204636.
• The Google Patents legal-events record for US4204636A contains no assignment, security-agreement, merger, change-of-name, license, or release entries — i.e., no reel/frame data at all.
• Best-supported reading: the record is consistent with the original assignee (Stanley Hayman and Co) having retained ownership through expiration in 1998, with no recorded post-issuance transfer. That is a finding of absence of transfer, not affirmative proof of it — a pre-digital paper chain, if any existed, would not necessarily appear in the online databases.

Because I cannot certify that the Assignment Center holds no records (as opposed to holding records I could not retrieve), I am proceeding with the remaining sections as an insufficient-data assessment rather than stopping outright.

Timeline diagram

Rendering only the ownership-relevant events that are actually documented. No transfers are asserted.

timeline
    title Ownership of US 4204636
    1978 : Filed by Stanley Hayman and Co
    1980 : Patent granted
    1998 : Anticipated expiration

NPE / troll-pattern signals

Each signal below is assessed against the available record. No signal is marked "present," because no transfer record exists on which to base one.

Note on temporal relevance: Even if a transfer had occurred, this patent expired 1998-10-13 under the pre-AIA 17-years-from-issue term. Any NPE assertion economics would have had to operate in the 1980s–1990s, and any such activity would predate most modern NPE databases' coverage of this vintage. The absence of a modern asserter trail is therefore expected and carries little diagnostic weight.

Verdict

Insufficient data.

Justification: The only ownership fact I can ground is the original assignee named at filing — "STANLEY HAYMAN AND Co" — recorded in the Google Patents legal events on 1978-10-13, with the patent granted 1980-05-27 and expiring 1998-10-13; no reel/frame, correspondent, or conveyance entry was retrievable from the USPTO Assignment Center or legacy assignment search, and no transfer, litigation, NPE-list, or aggregator linkage appears in any source reached. With zero recorded links in the chain, I cannot mark even one NPE signal "present," and I decline to infer a shell/asserter pattern from the assignee's name alone (which is explicitly against the constraints). If the Assignment Center in fact holds no records for 4204636, the correct reading is original-assignee retention through expiration, which is a non-NPE outcome by default — but I cannot yet certify that.

Verification link (for a human with database access to close the gap):
https://assignmentcenter.uspto.gov/ — enter patent number 4204636. Legacy interface: https://assignment.uspto.gov/patent/index.html.

Recommended follow-ups: (1) pull the recorded assignment (if any) directly in Assignment Center and capture reel/frame, conveyance, and correspondent of record; (2) check the print front page of the issued patent at https://patentimages.storage.googleapis.com/32/49/44/5eb2848f5dce70/US4204636.pdf for the "Assignee" and any assignment-reference field not surfaced in text extraction; (3) confirm the legal name and corporate status of "Stanley Hayman and Co" in state registries; (4) since the only cited/inventor-surname datum I located ("Hayman," 6,192,271 B1, dated 2/2001) is an unrelated later patent reference, do not conflate it with this chain.

I'll search for the cited references on USPTO/patent databases to verify and characterize each one.

Now let me search the remaining cited references.

Prior-Art Analysis — U.S. Patent No. 4,204,636

Scope note on the "USPTO database" search. I could not reach a native USPTO search interface (Patent Full-Text, PatentCenter, or the examiner's USPTO-format citation listing) from this session. What I was able to retrieve is the USPTO-derived front-page data as mirrored by Google Patents, Justia, uspto.report, and the individual reference documents themselves (PDFs hosted on patentimages.storage.googleapis.com, which are the official USPTO-issued prints). I therefore treat the front-page citation list below as the authoritative "cited references" set, and I flag explicitly where I could not verify a reference's content.

Two date/consistency flags before the analysis:

1. The task header states "Current Date: April 26, 2026," while the system/context date given for this session is 2026-10-01. Nothing in the analysis below turns on this, but I am flagging the inconsistency rather than silently picking one.
2. Minor citation-date discrepancy: Google Patents lists US 3,991,299 with a priority date of 1972-02-03; the Unified Patents record I retrieved shows 1972-02-02. Both agree on the 1976-11-09 issue date. Reported literally, not reconciled.

---

1. The formal citation set (6 references, examiner-cited on the face of the patent)

These are the "Patent Citations" listed on the printed patent / Google Patents record for US 4,204,636.

Statutory framework. US 4,204,636 filed 1978-10-13 with no earlier priority claim, so the pre-AIA § 102(b) critical date is 1977-10-13. All six references issued before that date (earliest-to-latest issue spread: 1968-11-26 to 1976-11-09), so each is available as a § 102(b) statutory bar, and each is also § 102(a)/(e) art. Pre-AIA § 103 governs obviousness because the application predates the AIA.

---

2. Reference-by-reference analysis

A threshold point on how § 102 is applied here: anticipation requires a single reference disclosing every element of the claim as arranged. The claims of 4,204,636 have two independent limbs — (i) converting wand data into switch-closure-simulating data routed to the terminal's existing keyboard-directing means, and (ii) autonomously simulating the reader's expected "user ready" handshake signal (claims 2, 4, and the corresponding step of claim 6). None of the six front-page references discloses both limbs. I therefore state, for each, which claims it potentially anticipates (as the task requests) and where the disclosure falls short, which is more properly § 103 territory.

2.1 US 3,991,299 — Norand Corp. (Chadima et al.), "Bar Code Scanner" — most relevant

• Full citation: Chadima et al., U.S. Pat. No. 3,991,299, "Bar Code Scanner," filed 1972-02-03, issued 1976-11-09, assigned to Norand Corporation.
• Description (per the patent body's own characterization of it): "a bar code reader bypasses a keyboard and generates signals equivalent to those converted from the keyboard for direct insertion into memory. An audio alarm is provided for the operator to retry transmission if the data is bad." A portable hand-held bar-code reader with an alarm-driven retry protocol.
• § 102 analysis:
  • Claim 3 — strongest candidate. Claim 3 requires (a) data reading means with encoded output and (b) "means for converting said encoded data output to a format simulating said keyed data," enabling the system to accept both keyboard and reader data "without special modification." If "signals equivalent to those converted from the keyboard" is construed to mean signals in the same format as keyboard-derived signals, US 3,991,299 maps onto claim 3 nearly element-for-element. This is the single most dangerous reference on the face list.
  • Claim 1 — likely not a clean anticipation. Claim 1 requires element (b) "means connecting said data simulating said switch closure information to said directing means" (the means that normally routes keyboard closures to the processor). The reference is described as bypassing the keyboard and inserting into memory directly, rather than paralleling the keyboard's routing path. That structural difference is a genuine anticipation gap.
  • Claim 6 — partial only. The reference's "audio alarm... to retry" does not disclose "automatically generating a response thereto simulating a condition wherein a user is ready to accept such data" fed back to the reader. The alarm is an operator-facing output, not a simulated handshake. Claim 6 is therefore not clearly anticipated; it is § 103 art when combined with a handshake-generating reference.
  • Claim 5 (decoding means + relay means): not verified in the reference; cannot assign.
• Honest caveat: my mapping rests on the instant patent's own summary of 3,991,299, not on the reference's claim text, which I did not retrieve in full. Before relying on this as an invalidity position, pull the printed claims of 3,991,299.

2.2 US 3,413,447 — Monarch Marking Systems, "Information-bearing label and reading method and apparatus therefor"

• Full citation: U.S. Pat. No. 3,413,447, filed 1963-02-25, issued 1968-11-26, Monarch Marking Systems, Inc.
• Description (verified from the document text): A label scanner reads machine-readable data (concentric-ring label format); the derived signals are stored in relay circuits; a unit fits over the cash register keyboard and carries a plurality of pneumatically operated keys, each positioned above a corresponding register key, with a separate control valve and valve solenoid per key. The relay circuits energize the valve solenoids, air operates the pneumatic key actuators, and the register "rings up" the amount read from the label. The pneumatic keys can also be depressed manually.
• § 102 analysis:
  • Claim 3 — moderate candidate, with a claim-construction hook. The reference discloses a data reader plus apparatus enabling a keyboard terminal to accept reader-derived data. But its conversion is mechanical actuation of the very keys of the terminal, i.e., it reproduces key actuation rather than simulating keyed data in an electrical signal format. Whether that meets "format simulating said keyed data" depends entirely on construction of "format"; a narrow reading (electrical switch-closure-equivalence) escapes anticipation.
  • Claim 1 — likely not anticipated. Claim 1 expressly recites "manual input means providing switch closure information" and conversion into data "simulating said switch closure information." The Monarch device does not generate electrical switch-closure-simulating data; it depresses what is arguably the mechanical equivalent of the switch itself. Different apparatus.
  • Claim 6 — not anticipated. No detection of a reader "data ready" signal and no simulated user-ready response.
• Body-citation discrepancy to flag: the specification attributes 3,413,447 to a "LeMers," whereas the bibliographic record shows the assignee as Monarch Marking Systems, Inc. I have not auto-corrected either, per the literal-identifier rule. The inventor name in the body text should be verified against the issued patent.
• Net: more useful as § 103 art (wand→keyboard-entry equivalence) than as § 102 anticipation.

2.3 US 3,792,236 — Monarch Marking Systems (Dobras et al.), "Record reading system"

• Full citation: Dobras et al., U.S. Pat. No. 3,792,236, "Record reading system," filed 1973-03-26, issued 1974-02-12, Monarch Marking Systems, Inc.
• Description (verified from the document text): Reading of width-modulated bar-coded records with a hand-held light pen; free-running clock gated into counters; comparator-based relative-width decoding; automatic transition between a scanning mode (searching for a start code) and a reading mode (translating/using a message); detection of an error automatically returns the system from read mode to scan mode; read-out to a utilization device such as a lamp display or data processor.
• Body-citation characterization to flag: the instant patent states that "Dobras et al. U.S. Pat. No. 3,792,236 includes handshaking signals circuitry associated with a bar code reader." My retrieved text of 3,792,236 confirms the scan/read mode transitions and error-triggered mode change, but I did not independently confirm a reader-side "handshake" in the sense used by 4,204,636. If the reference's mode-change signaling is what the body calls "handshaking," it is architecturally different from the claimed interface-generated simulated user-ready signal.
• § 102 analysis: The front-page citation of 3,792,236 maps most directly onto claims 2 and 4 (means for simulating the reader's expected first/ready signal independently of the processor, and transmitting it to the reader) and onto the "detecting a control signal indicating that data is ready to be read from the wand reader" step of claim 6 — if and only if the handshake disclosure is verified in the reference. The error→scan-mode logic alone does not read on those claims. Flag as: potentially anticipates claims 2/4 only if the handshake disclosure is confirmed; otherwise § 103 art.

2.4 US 3,826,900 — NCR (Moellering), "Cordless scanning probe"

• Full citation: Moellering, U.S. Pat. No. 3,826,900, "Cordless scanning probe," filed 1972-10-13, issued 1974-07-30, NCR (National Cash Register Co.).
• Description (verified from the document text and claims): A hand-held optical probe scribed across color-coded labels; the code is converted to electrical signals, frequency-modulated (two FM channels), and transmitted by radio to a local receiver; logic means converts the received waves to digital signals; processor converts digital to decimal; output to a display and a registering business machine. The stated purpose is to eliminate the fiber-optic tether to the register.
• § 102 analysis:
  • Claim 1 — not anticipated. Claim 1 requires manual input means providing switch closure information and a directing means routing it to the processor. US 3,826,900 has no keyboard/switch-closure limb and no conversion into switch-closure-simulating data. It routes reader data over an independent RF channel to a dedicated processor/display — the structural opposite of the 4,204,636 approach.
  • Claim 3 — not anticipated. No "converting... to a format simulating said keyed data."
  • Claim 6 — not anticipated. No simulated user-ready response; the probe is the transmitter, the register the receiver.
  • Net: this is the clearest § 103 reference on the list (wand + processor + conversion + display, using a wireless link instead of a keyboard path), not a § 102 anticipatory reference for these claims.

2.5 US 3,963,901 — Intelcom Industries Incorporated, "Data processing system"

• Full citation: U.S. Pat. No. 3,963,901, filed 1974-08-26, issued 1976-06-15, Intelcom Industries Incorporated.
• Description: Not verified. My content search did not return the disclosure or claims of this reference, and my attempts to retrieve it were cut off. I will not characterize it from inference.
• § 102 analysis: Cannot responsibly assign a claim mapping. Intelcom's title ("Data processing system") indicates a data-handling system of the mid-1970s vintage, but whether it discloses wands, keyboard simulation, or handshake generation is unverified. This is a recognized gap; it should be retrieved from the USPTO full-text database before any invalidity or clearance conclusion is drawn. Do not treat the absence of analysis here as a negative finding about its relevance.

2.6 US 3,735,350 — Lemelson, "Code scanning system"

• Full citation: Jerome H. Lemelson, U.S. Pat. No. 3,735,350, "Code scanning system," filed 1966-02-10, issued 1973-05-22.
• Description: Only partially verified. uspto.report confirms the grant record for "Code Scanning System" at US 3,735,350. I did not retrieve its claims; my Lemelson-specific searches surfaced other Lemelson patents (e.g., US 3,804,978, US 4,979,029) rather than this one.
• § 102 analysis: As a general "code scanning system" reference, it is at most § 103 art for the generic step of scanning a code and decoding it to a utilization device. It does not, on the record I could verify, disclose converting scanned data into keyboard switch-closure-simulating signals or simulating a reader handshake. Claim mapping: not assignable pending retrieval of the claims.

---

3. References cited in the body of 4,204,636 but absent from the front-page list

The specification discusses art that does not appear among the six examiner citations. Because these are literally "patent citations for 4204636" in the document, they are listed here — with the caveat that I verified content only through the citing patent's own summaries, not through the primary documents.

Cross-reference note (consistent with the earlier sections, not a contradiction): Dobras et al. US 3,792,236 appears both on the front-page list and in the body discussion; that duplication is normal and was already noted in the prior summary. The prior summary's observation that the body cites a larger set than the six front-page references is confirmed above.

---

4. Ranking — most relevant prior art for US 4,204,636

1. US 3,991,299 (Norand / Chadima et al.) — keyboard bypass + "signals equivalent to those converted from the keyboard" + alarm-driven retry. Closest single reference; best § 102 candidate (claim 3, possibly claim 1) and core § 103 art for claims 1–6.
2. US 3,413,447 (Monarch Marking) — wand output automatically operating the cash register keyboard keys. Closest to the functional result of claims 1/3; mechanically rather than electrically implemented. § 103 art; § 102 arguable only under a broad construction of "simulating."
3. US 3,792,236 (Dobras et al. / Monarch Marking) — the body asserts handshake-signal circuitry; verified scan/read mode switching and error-triggered mode change. Potential § 102 art against claims 2/4 only if the handshake disclosure is verified.
4. US 4,088,981 (Gott) (body-only; verify date) — "encoded similarly to... keyboards" is the most literal near-miss on claim 3's "format simulating said keyed data," but its post-1977-10-13 timing likely limits it to § 102(a)/(e).
5. US 3,826,900 (NCR / Moellering) — wand + processor + display, but over an independent RF channel with no keyboard limb. § 103 art.
6. US 3,963,901 (Intelcom) — unverified content; ranking provisional and could move materially once retrieved.
7. US 3,735,350 (Lemelson) — generic code scanning system; § 103 art for the scanning/decoding step.
8. Body-cited remainder (Cunningham US 3,812,347; Mak US 3,238,501; Neff US 4,048,617; Humbarger US 3,585,367; Ferry US 920,940) — § 103 background art; none reaches the two-limb combination.

Overall § 102 conclusion. On the verified record, no single one of the six front-page references discloses every element of any of claims 1–6. The claims' novelty resides in the combination of (i) converting wand data into switch-closure-simulating data routed through the terminal's existing keyboard-directing path, and (ii) autonomously fabricating the reader's expected "user ready" handshake without terminal feedback. US 3,991,299 comes closest on limb (i) but omits limb (ii) and (per the instant patent's own characterization) bypasses rather than parallels the keyboard; US 3,792,236 is the only front-page reference the body ties to "handshaking," but that disclosure is unverified. The realistic invalidity theory on this art is therefore pre-AIA § 103 (e.g., US 3,991,299 in view of US 3,792,236 or Cunningham US 3,812,347), not clean § 102 anticipation.

---

5. Verification gaps and items to close

• US 3,963,901 (Intelcom) — full text and claims not retrieved; cannot map to claims.
• US 3,735,350 (Lemelson) — full text and claims not retrieved; cannot map to claims.
• US 4,088,981 (Gott) — exact issue date unverified; date determines whether it is § 102(a)/(e) or § 102(b) art.
• US 3,991,299 (Norand) — claim-level mapping is based on the instant patent's summary, not the reference's own claims.
• Instructor-name discrepancy — body calls US 3,413,447 "LeMers"; bibliographic record shows Monarch Marking Systems, Inc. Reported literally, not corrected.
• Priority-date discrepancy — US 3,991,299 listed as 1972-02-03 (Google) vs. 1972-02-02 (Unified Patents).
• Fetch limitation — the "USPTO database" queries in this session were served by USPTO-derived mirrors (Google Patents/Justia/uspto.report) and the USPTO-hosted PDF prints, not by a live USPTO full-text search endpoint. Reference contents for items 2 and 5 above should be confirmed against the printed USPTO documents.

Sources:

• https://patents.google.com/patent/US4204636/en (front-page citation list and body text)
• https://uspto.report/patent/grant/4204636
• https://patents.justia.com/patent/4204636
• https://patents.google.com/patent/US3413447
• https://patents.google.com/patent/[US3826900A](/patent/US3826900A)/de ; https://trade.patenthub.cn/cpc/patent-[177710](/patent/177710)-US3826900A-…html ; https://uspto.report/patent/grant/3,826,900
• https://patentimages.storage.googleapis.com/88/28/ce/ae5b8105b3644e/US3792236.pdf
• https://uspto.report/patent/grant/3735350
• https://portal.unifiedpatents.com/patents/patent/US-[6866196](/patent/6866196)-B1 (confirming US 3,991,299, Norand Corp., "Bar Code Scanner," priority 1972-02-02)

I'll ground this analysis by verifying the prior-art references cited in the patent body, then build the §103 combinations.

Obviousness Analysis — U.S. Patent No. 4,204,636 under 35 U.S.C. § 103

Building on the earlier bibliographic, claim-text, and litigation sections. This section does not repeat those; it takes the six independent/dependent claim constructions and the prior-art listing as given and develops the § 103 case.

---

0. Scope, sources, and flags

Date discrepancy (flagged, not resolved): the system context for this analysis states today's date as 2026-10-01; the task instruction states April 26, 2026. The earlier generated sections used "2026" generically. This does not affect the legal analysis (the patent expired 1998-10-13), but I am recording it per the standing rule about contradictions.

"Prior Art section of this page" — what it actually contains. The Google Patents page for US4204636 supplies prior art in two places that do not match each other:

1. Front-page cited references (6): US 3,413,447; US 3,735,350 (Lemelson); US 3,792,236; US 3,826,900 (NCR); US 3,963,901 (Intelcom); US 3,991,299 (Norand/Chadima). Source: https://patents.google.com/patent/US4204636/en
2. References argued in the Background/Detailed Description but absent from that front-page list: Gott US 4,088,981; Cunningham US 3,812,347; Mak et al. US 3,238,501; Neff US 4,048,617; "LeMers" US 3,413,447; Dobras et al. US 3,792,236; Humbarger US 3,585,367; Ferry US 920,940; Chadima et al. US 3,991,299.

I use both sets, because § 103 is judged against "the prior art" available, and the applicant's own Background admissions are usable as prior-art evidence of what was known. Two identifier anomalies are preserved literally and not corrected: "LeMers U.S. Pat. No. 3,413,447" (the front page assigns 3,413,447 to Monarch Marking Systems Inc.) and "Ferry U.S. Pat. No. 920,940," printed as a five-digit number.

Governing statute: Because application 05/950,967 was filed 1978-10-13 (pre-AIA), pre-AIA 35 U.S.C. § 103(a) governs. The obviousness inquiry is made as of the effective filing date (1978-10-13) from the viewpoint of a person of ordinary skill in the art ("PHOSITA"), under Graham v. John Deere Co., 383 U.S. 1 (1966). KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), controls the motivation analysis: teaching/suggestion/guidance is a sufficient but not exclusive route; predictable variation and combination of known elements are also sufficient.

---

1. Level of ordinary skill in the art (PHOSITA)

Drawing from the references themselves (OCR wands, POS terminals, TTL logic):

A technician or engineer with a bachelor's degree (or equivalent) in electrical engineering plus roughly 2–3 years' experience, or a skilled logic designer with ~5 years' experience, familiar with (a) 1970s TTL MSI building blocks (BCD-to-decimal decoders such as the SN7442/SN74154 and retriggerable one-shots such as the SN74123, all named in the specification), (b) matrix/relay interfacing between machine outputs and keyboard-contact inputs, and (c) asynchronous start/stop data-transfer handshaking ("data ready"/"user ready") as practiced in peripheral interfacing.

This is a low-to-moderate skill level. That matters: under KSR, routine engineering choices within this skill set weigh against patentability.

---

2. The limitations that do the work

---

3. What each reference teaches (as verified)

Uncertainty note: I confirmed the content of the first six rows above from live sources. The final row is characterized only from the patent's own Background text and the front-page table; I did not retrieve those documents before my search budget was exhausted, and I do not assert their contents beyond what is quoted in the patent.

---

4. The § 103 case, claim by claim

4.1 Claims 1 and 3 — Gott, alone or in view of Chadima

Gott alone is close to a § 102 anticipation of the substance of claim 1. Gott discloses the full preamble set (data reading means with encoded output; display; processing/computer; keyboard producing coded input; means directing keyboard input to the computer for conversion to display format) and, critically, element (a) — "the recognition unit generates signals identical to those generated by the keyboard" — plus element (b) — those signals go "to the same computer in the same way." The final "independently of the source of said information for display" clause is the stated consequence of that arrangement, not a separate structural feature.

The one genuine gap: Gott's recognition unit re-encodes the character and drives the same input path; the 4204636 claims (as illustrated by decoders 48/50 plus relay matrix 56 and relay contacts on lines 58) generate physical switch-closure equivalents of keyboard contacts. If that distinction is given patentable weight, Chadima US 3,991,299 supplies it: data transferred "directly to memory and hence to the visual display," with scanner and keyboard "used interchangeably." LeMers/Monarch US 3,341,447→3,413,447 supplies it even more literally by paralleling actual cash-register keys with relay-driven actuators.

Motivation to combine (KSR-compliant): All three references are in the same field (label/wand reading into a keyboard-driven data terminal); each independently addresses the same problem named in 4204636's Background — the time and error cost of manual keying. A PHOSITA seeking to let a wand drive a register whose keyboard lines are simple contact closures would predictably select the contact-paralleling technique long used to retrofit machines (LeMers/Monarch) over designing a special terminal input port. That is a "combination of familiar elements according to known methods … likely to be successful" (KSR), i.e., an improvement in one of the prior art's own known techniques.

Claim 3 is a pure improvement claim directed to the same subject matter with "without special modification" appended. That phrase is a statement of result/advantage, not structure, and carries no independent patentable weight once the converting means are shown.

4.2 Claim 5 — decoding means + relay means

LeMers/Monarch US 3,413,447 discloses relay circuits that store scanned data and energize solenoids/actuators corresponding to keys. Combining that with a binary-to-one-of-N decoder (the SN7442/SN74154 chips the specification itself names, and which are described as "available" off-the-shelf) is the textbook KSR case: an enumerated prior-art element (relays responsive to stored decoded data) plus a known, off-the-shelf component substituted for a general-purpose element to achieve a predictable result. Claim 5's incremental contribution over claims 1/3 is a routine design choice of component.

4.3 Claims 2 and 4 — the fabricated "user ready" signal (the crux)

This is where the patent's actual novelty lies, and therefore where the § 103 analysis is most contestable. Two sub-limitations must be met:

• (i) means for simulating a first signal independently of the processing means — i.e., a local timing generator that fabricates the pulse the wand expects; and
• (ii) means for transmitting the simulated signal to the reader.

Prior-art supply of (i):

• Cunningham US 3,812,347 discloses a timing module 818 in an interface between an optical reader and a host, strobed by a sync detector, providing clock pulses — i.e., an interface-located timing generator satisfying a reader's timing expectations.
• Mak US 3,238,501 teaches generating "a bit related clock pulse" derived from the reader itself.
• Chadima US 3,991,299 controls its encoder/control circuit by timing signals from a timing generator; the patent's own text notes that standard 555-timer oscillators and one-shots "may be utilized," and that the astable devices are stock TI SN74123 chips.
• The 4204636 specification itself concedes that "other timing devices may be used … standard oscillators using 555 timers, available on integrated chips from National Semiconductor … Other one-shots or crystal oscillators may similarly be used."

That last concession is significant: it frames the timing-generator element of claims 2/4 as a selection among known, catalogued circuit options to perform a known function (delaying a strobe), which is the paradigm of predictable variation.

Prior-art supply of the design rationale for (ii) — relying on the operator instead of feedback:

• Chadima US 3,991,299 already teaches the exact human-in-the-loop error-recovery philosophy the 4204636 specification relies on: an audio alarm telling the operator to scan again, with automatic clearing on rescan.
• Cunningham US 3,812,347 teaches the same via read/re-read indicator lamps.
• Hence a PHOSITA facing a handshake-demanding wand and a register with no handshake output had a known, low-cost option: fabricate the expected strobe locally and let the existing alarm drive rescan behavior.

Motivation to combine: the problem is narrowly identified and the solution is a substitution of one known element (processor-generated ready pulse) by another known element (local one-shot/astable pulse) that performs the same function in the same way for the same purpose. Claims 2/4 expressly claim the simulation "independently of said processing means," which is precisely the substitution. KSR makes this combination-in-place-of-known-elements prima facie obvious absent evidence of unexpected results.

Weakest link for the challenger: the deliberateness of the open loop. Claims 2/4 (and claim 6's "automatically generating a response … simulating a condition wherein a user is ready") arguably capture a counterintuitive design decision — affirmatively lying to the peripheral about readiness rather than arranging genuine feedback. If an examiner or court accepted the applicant's framing that the change removes a feedback loop from what had been a feedback-based handshake protocol, the challenger would need the Cunningham/Mak/Chadima timing-generator teachings to do more work, or would need to show that open-loop peripheral strobes were themselves routine. On the record I have, I can state the combination but cannot assert with high confidence that it defeats claims 2/4 — this is the one place where the patent has a colorable non-obviousness argument.

4.4 Claim 6 — the method claim

Every step maps:

Claim 6 is therefore obvious over Gott in view of Cunningham, with Chadima as a cumulative secondary reference for the interchangeable-keyboard and operator-alarm teachings. Because claim 6 is method (steps), it avoids the § 112 ¶ 6 "means" construction latitude that claims 1–5 enjoy; the challenger has to show only that each step was known or obvious to perform, which the table above supports.

---

5. Motivation-to-combine summary (articulated per KSR)

1. Same field, same problem. All principal references are 1960s–70s label/wand reading into keyboard-driven data terminals; all address the same Background problem (manual keying is "time consuming and error prone"). KSR confirms field-of-endeavor identity supports combination.
2. Predictable result. Paralleling keyboard contacts with relay/decoder outputs was the standard retrofit technique (LeMers/Monarch; NCR's cordless probe context). No new principle of operation is required.
3. Known, catalogued components. The specification names SN7442, SN74154, and SN74123 as off-the-shelf TI parts and 555 timers as alternatives — the KSR "marketplace" scenario: a design need with only a finite number of identified, predictable solutions.
4. Design incentive. Register manufacturers cited in the Background (Sweda Model 800 vs. Models 600/800/35; NCR; DTS; MKD; TRW) created a mixed installed base of wand-capable and wand-incapable terminals — an explicit commercial incentive to build an adapter that needs no terminal modification.
5. Common ownership/disparate art is irrelevant here — all references are third-party or applicant-admitted.

---

6. Rebuttal positions the patentee could raise — and my assessment

---

7. Secondary considerations

Nothing in the patent record, the earlier litigation search, or the citation history supplies a Graham secondary consideration. Concretely:

• No commercial-success, long-felt-need, or failure-of-others evidence is presented in the specification or found in the record.
• No reexamination, reissue, or post-grant proceeding appears on the face of the record. Pre-AIA, only ex parte reexamination (35 U.S.C. § 302) was available during the patent's life; inter partes review and IPR did not exist.
• The 15–20 later "cited by" documents (e.g., US 4,420,819; US 4,457,016; US 4,679,154; GB 2,259,597) are prior-art citation relationships, not evidence of commercial success or of skepticism.
• The patent expired 1998-10-13 ("Expired – Lifetime"), so obviousness is now of historical/practice interest rather than of live validity consequence.

---

8. Bottom line

Candid confidence statement. Claims 1, 3, 5, and 6 I would expect to be held obvious, with claim 5 being the most mechanical and claim 1 being, in substance, anticipated by Gott. Claims 2 and 4 present a genuine contest: the limitation that the interface fabricates the "user ready" signal with no feedback from the terminal is a deliberate design choice, and I can assemble a motivation to reach it but cannot assert with high confidence that the assembled combination renders it obvious without additional evidence on the routine nature of open-loop peripheral strobes circa 1978. I have not located any reexamination or court decision that actually adjudicated obviousness for this patent, and nothing in the record supplies secondary-considerations rebuttal evidence either way.

---

9. Sources

• Patent full text: https://patents.google.com/patent/US4204636/en ; PDF: https://patentimages.storage.googleapis.com/32/49/44/5eb2848f5dce70/US4204636.pdf
• Gott US 4,088,981: https://patents.google.com/patent/US4088981
• Chadima et al. US 3,991,299: https://patents.google.com/patent/US3991299 ; PDF: https://patentimages.storage.googleapis.com/6c/9c/8b/b69ac8724af480/US3991299.pdf
• "LeMers" US 3,413,447: https://patents.google.com/patent/US3413447
• Dobras et al. US 3,792,236: https://patentimages.storage.googleapis.com/88/28/ce/ae5b8105b3644e/US3792236.pdf
• Cunningham US 3,812,347: https://uspto.report/patent/grant/3812347
• Mak et al. US 3,238,501: https://patents.google.com/patent/US3238501A
• Graham v. John Deere Co., 383 U.S. 1 (1966); KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007); pre-AIA 35 U.S.C. § 103(a).

Not retrieved (search budget exhausted), characterized only from the patent's own text: US 3,585,367 (Humbarger), US 4,048,617 (Neff), US 3,735,350 (Lemelson), US 3,826,900 (NCR), US 3,963,901 (Intelcom), Ferry US 920,940. Identifier anomalies preserved literally, uncorrected: "LeMers" vs. the front page's Monarch Marking assignee for 3,413,447; "Ferry U.S. Pat. No. 920,940" as printed with five digits; "STANLEY HAYMAN AND Co" spacing artifact.