US 5228118

I'll search for litigation and docket information on this specific patent number, then summarize.

Let me run a couple more targeted searches to confirm litigation status.

US Patent 5,228,118 — Summary

Important caveat up front: I searched for CAFC 2026 docket activity and USPTO litigation records for this number and found no infringement suits, PTAB proceedings, or Federal Circuit appeals involving US 5,228,118. This is consistent with its status: the patent's term expired 2012‑06‑25 ("Anticipated expiration," legal status Expired – Lifetime), so it is not a plausible candidate for a 2026 CAFC docket. I cannot rule out an unpublished/aggregator‑omitted filing, but I have no authoritative basis to assert any litigation. The Searches returned only unrelated 2026 CAFC news (Samsung v. Oura smart‑ring matters) — nothing touching this patent.

Bibliographic data

Noted discrepancy: Google Patents lists the priority date as 1991‑06‑29, while the Unified Patents portal entry shows 1991‑06‑28 for US‑5228118‑A. I have not resolved this; the Google Patents figure is the one most consistently reflected in the family data.

Abstract (as issued)

A printing system including a printer having at least one interpreter for interpreting print data for performing a printing operation, and a data processor connected to the printing device and having a plurality of printer drivers through which the print data are sent to the printer device. The data processor sends to the printer an inquiry signal asking the printer to send back to the data processor interpreter‑identification data representative of the interpreter or interpreters. The data processor selects one of the printer drivers compatible with the interpreter or one of the interpreters, on the basis of the interpreter‑identification data. The printer sends the interpreter‑identification data to the data processor, in response to the inquiry signal.

Independent claim 1 — plain language

A printing system with (a) a printer holding at least one language interpreter (e.g., a page/control‑language emulation) and (b) a computer connected to it over a bidirectional interface, the computer holding multiple printer drivers. The computer must include two functional elements:

1. Inquiring means — sends an inquiry signal to the printer asking the printer to return "interpreter‑identification data" describing which language interpreter(s) it has; and
2. Printer driver selecting means — picks, from the computer's several drivers, one that is compatible with one of the printer's interpreters, based on the returned data.

The printer must include ID data output means that respond to the inquiry by sending the interpreter‑identification data back. In short: the computer asks the printer "what languages do you speak?", the printer answers, and the computer auto‑selects the matching driver — eliminating the operator's manual driver‑selection step.

Independent claim 9 — plain language

Same core idea, but inverted and network‑scaled: there are multiple printers, each with at least one language interpreter, and at least one computer connected to all of them over a bidirectional interface, holding at least one printer driver. The computer must include:

1. Inquiring means — sends the inquiry to the multiple printers, asking each to return its interpreter‑identification data; and
2. Printer selecting means — uses the returned data to select the one printer whose interpreter matches a driver on the computer.

Each printer must include ID data output means that respond to the inquiry from the requesting computer by sending its interpreter‑identification data back to that computer. Rather than choosing a driver to match a printer, the computer chooses a printer to match its driver.

Dependent-claim highlights (context for the independents)

• cl. 2–3 (dep. 1): multiple interpreters → detect multiple compatible drivers, then pick per a predetermined priority order.
• cl. 4 (dep. 1): timeout — alarm if interpreter‑ID data is not received within a preset time.
• cl. 5 (dep. 1): alarm if no driver is compatible with any interpreter.
• cl. 6 (dep. 1): printer stores the interpreter‑ID data in memory and reads it out on inquiry.
• cl. 7 (dep. 1): computer has first memory for the drivers and second memory for the selected driver.
• cl. 8 (dep. 1): an automatic printer driver selection mode gate — the inquiry/selection only runs when that mode is enabled (with a default driver chosen otherwise, per the spec).
• cl. 10–15 (dep. 9): pre‑selected driver read from memory; inquiry may ask for only the compatible interpreter(s) or for all interpreters; multiple compatible printers resolved by priority; a second inquiry checks whether printers are unoccupied (busy status), selecting a free one, with priority order breaking ties.
• cl. 16 (dep. 9): alarm when no connected printer has a compatible interpreter.
• cl. 17–18 (dep. 9): printer‑side memory for the ID data; computer‑side memory for the driver(s) and for the selected printer.
• cl. 19 (dep. 9): return printer‑identification data along with interpreter‑ID data.
• cl. 20 (dep. 9): inquiry broadcast to the printers.
• cl. 21–22 (dep. 9): inquiry scope options — announce only the compatible interpreter vs. all interpreters, with the computer then matching.

Technical context from the specification (grounding)

• The embodiment pairs a laser printer LP (video controller 11 with first ROM 14 storing "LANGUAGE INTERPRETER A" and "LANGUAGE INTERPRETER B" plus the ID DATA OUTPUT program) with a personal computer PC (hard disk storing "PRINTER DRIVER C" compatible with Interpreter A and "PRINTER DRIVER D" compatible with Interpreter B).
• Interpreter A is for ordinary office‑printer text/document languages; Interpreter B is for a plotter/CAD language.
• FIG. 6 is the PRINTER DRIVER SELECTION routine (steps S10–S22, including the timer‑based timeout S12–S14 and alarm S15); FIG. 8 is the PRINTER SELECTION routine (steps S30–S41, including the broadcasting inquiry S31, compatibility test S33, and busy/unoccupied second inquiry S36–S38).
• The stated problem addressed: manual driver/printer selection was "cumbersome and time‑consuming," worsened when interpreters/drivers were revised or added, producing "a relatively long overall non‑printing time" and "relatively low overall printing efficiency."

Uncertainty notes

• Litigation: no CAFC/PTAB/district‑court activity located for this patent; the null result is expected given the 2012 expiration, but I am flagging it as "no evidence found" rather than a verified negative.
• Priority date: 1991‑06‑29 (Google/JP family) vs. 1991‑06‑28 (Unified Patents) — unresolved.
• Assignee naming: "Brother Kogyo Kabushiki Kaisha" (assignment record) and "Brother Industries, Ltd." (Google Patents current‑assignee field) refer to the same entity; the assignment filing date shown is 1992‑06‑25.

Sources: https://patents.google.com/patent/US5228118/en ; https://uspto.report/patent/grant/5,228,118 ; https://www.sumobrain.com/patents/us/Printing-system-wherein-one-printer/5228118.html ; https://portal.unifiedpatents.com/patents/patent/US-[10140073](/patent/10140073)-B2 (citing‑list entry for US‑5228118‑A).

Summary of Findings — U.S. Patent No. 5,228,118

Result: No litigation involving U.S. Patent No. 5,228,118 was identified in the searches I was able to run.

I was unable to locate any district court complaint, ITC action, PTAB proceeding, or Federal Circuit appeal in which U.S. 5,228,118 appears as an asserted patent. I therefore cannot provide plaintiff(s), defendant(s), jurisdiction, case number, filing date, or outcome for any case, because I found no case to report.

What I actually searched

• General web searches for "US patent 5228118 litigation," "5228118" patent lawsuit defendant, "5,228,118" patent infringement complaint, and "5,228,118" OR "5228118" patent case docket.
• Searches aimed at the Unified Patents litigation portal, plus attempts to reach docket-style sources.

The only substantive hits returned were patent documents citing 5,228,118 as prior art (e.g., U.S. 5,687,301 and U.S. 8,102,556 on Justia), which are citations during prosecution — not litigation. I did not obtain a confirmed hit on a litigation docket page (Unified Patents caselist, PACER, CourtListener/RECAP, or the Federal Circuit) for this patent number.

Why the absence of litigation is plausible

The patent's own bibliographic record (Google Patents, fetched 2026-09-17) shows:

• Assignee: Brother Kogyo Kabushiki Kaisha (Brother Industries, Ltd.)
• Priority date: 1991-06-29; Filed: 1992-06-25; Granted: 1993-07-13
• Status: "Expired – Lifetime"; anticipated expiration: 2012-06-25

A patent that expired roughly 14 years ago — and that was held by an operating company (Brother) rather than a monetization entity — would be an unlikely candidate for recent enforcement actions. Any assertion would most likely have occurred in the 1990s–2000s, a period for which free docket coverage is sparse.

Important caveats (please verify directly)

• I was not able to complete direct lookups on Unified Patents' caselist, PACER, Docket Navigator, or CourtListener in this session. Those paywalled/JS-driven databases are the authoritative sources for a definitive "no litigation" answer.
• My finding is not a legal conclusion that the patent has never been litigated; it means I found no evidence of litigation within the searches performed.
• If you need a legally defensible negative, run a docket search by patent number (5,228,118 and also 5228118) in:
  • PACER / CourtListener RECAP (all district courts)
  • Unified Patents Portal → Litigation (search by patent number)
  • Docket Navigator or Lex Machina (patent-level case lookup)
  • USITC EDIS (for any Section 337 investigations)
  • CAFC docket search (for any appeal referencing the patent)

If any of those searches return a case, I can analyze the complaint, claim-construction history, and disposition — but based on the evidence currently available to me, there is nothing to report for US 5,228,118.

Proceedings overview

Zero AIA trial proceedings. The count is 0 — no active, none with claims invalidated, none with claims sustained, none settled, none with institution denied. The USPTO Open Data Portal's structured "PTAB proceedings on file" block for US 5228118 returns no AIA trial records, and targeted web searches for IPR / PGR / CBM petitions against the patent (including combination searches on "5228118" and "5,228,118" with IPR, PTAB, and Brother) surfaced nothing but prior-art citation trails. The bottom line for a defendant: there is no PTAB record to work with — and there is also no live patent to defend against. US 5228118 reached its anticipated expiration on 2012-06-25 (20 years from the 1992-06-25 filing date) and carries a Google Patents legal status of "Expired - Lifetime" (Google Patents). The AIA trial regime only became available on 2012-09-16 for IPR/PGR and CBM — i.e., after this patent's term ended — so the absence of a PTAB file is structural, not evidence of a hardened patent.

No proceedings to report

There is no IPR20XX-XXXXX, PGR20XX-XXXXX, or CBM20XX-XXXXX number to enter here. I will not invent one. What follows is the verified negative finding and the surrounding context, framed in the requested template's terms.

(no proceeding) — n/a v. Brother Kogyo Kabushiki Kaisha

• Type: Not applicable — no Inter Partes Review, Post-Grant Review, or Covered Business Method review was ever filed or instituted against US 5228118.
• Filed: n/a. Note the structural bar: this patent's priority is 1991-06-29 (JP 185309/1991, granted as JP2661416B2), its US filing date is 1992-06-25, and it issued 1993-07-13 — well before IPR/PGR/CBM existed. The AIA trial mechanisms opened on 2012-09-16; the patent's 20-year term lapsed 2012-06-25, roughly three months earlier. Any CBM review would in any event have been unavailable on subject-matter grounds: the patent is directed to printer-driver/interpreter negotiation, not a financial product or service.
• Status: No PTAB activity on file (USPTO ODP: no AIA trials; confirmed by web search, which produced no petition, institution decision, FWD, or termination).
• Judge panel: None — no panel ever convened.
• Petition grounds: None. The only prior-art references of record are the two examiner-cited US patents: US4926347 (Citizen Watch, "Printer with emulation card installation reporting feature," 1990-05-15) and US4992957 (Seiko Epson, "Printer and method for utilizing character codes and control codes within a printer," 1991-02-12). These are prosecution citations, not petition grounds, and I have no record of any petition building on them.
• Institution decision: None issued.
• Final Written Decision: None issued. No claim of US 5228118 has ever been canceled, confirmed, or otherwise adjudicated at the Board. All 22 claims stand in their as-granted form.
• Settlement / termination: Not applicable.
• Appeal: No FWD, therefore no PTAB-to-Federal Circuit appeal. I found no CAFC docket or CourtListener entry involving US 5228118.
• Defensive value: Limited but decisive in one respect — the patent cannot be asserted at all today because its term expired 2012-06-25. If a demand letter or complaint cites US 5228118, the expiration date is the complete answer; you do not need a PTAB record, and you could not get one now for prospective relief.

Strategic summary

Claim status: UNTESTED across the board. Claims 1–8 (the single-printer/single-data-processor aspect, claim 1 independent) and claims 9–22 (the multi-printer network aspect, claim 9 independent) have never been through an AIA trial. There are no canceled claims, no sustained claims, and no claim-level Board construction to borrow. If you were litigating this patent, you would be building validity positions from scratch, with no Board precedent to lean on and no petitioner-side work product in the public record.

Estoppel landscape: § 315(e)(2) is a non-issue. Because no petitioner ever obtained an FWD on any claim, no party (and no privy) is estopped under 35 U.S.C. § 315(e)(2) from raising any § 102/§ 103 ground in district court. Conversely, there is no petitioner-side ground that has been "used up," and no IPR-tested art to distinguish. The practical consequence runs in your favor only because the patent is expired — there is nothing left to assert. If the patent were still in force, this would be the worst posture for a defendant: untested claims plus a full menu of available invalidity grounds (and no benefit from a prior petitioner's effort).

Pattern signals: none. No serial petititioner, because there was no first petitioner. No PTAB appeals by the patent owner, because there were no adverse decisions to appeal. No defensive aggregator (Unified Patents or similar) in the chain — Unified's portal lists US-5228118-A only as a citation/landscape entry attached to unrelated assertions, which is a prior-art indexing artifact, not a challenge. The patent's post-issuance footprint is essentially one of citation: it appears in the "References Cited" sections of later printer-driver and print-system patents (e.g., US5687301, US6614545, US7202961, US8456667), which is typical of a 1993 foundational-ish reference and is not evidence of enforcement activity. The same family also produced EP0521673B1, DE69226736T2, and JP2661416B2; I did not locate any EPO opposition or foreign revocation record, but I did not verify that to the standard I would want, so treat it as open.

Recommended next steps

• Lead with expiration, not with PTAB. US 5228118 is "Expired - Lifetime," anticipated expiration 2012-06-25 (Google Patents record). Any infringement theory premised on post-expiration conduct fails as a matter of law, and pre-expiration damages would face a six-year § 286 limitations bar that has long since run.
• If you nonetheless need a paper record, pull the prosecution history and the file wrapper via USPTO Patent Public Search / PatentCenter for application 07/904,491 to confirm no reissue, reexamination, or certificate of correction altered the claims. That is where a "later-life" validity proceeding would hide if one existed; I found none.
• Do not represent to a court or adversary that claims were canceled or confirmed. They were not. The correct, defensible statement is: no AIA trial proceeding has ever been instituted against US 5228118; the patent expired 2012-06-25; no claim has been adjudicated by the Board.
• If the patent term were restored or a continuation existed, the analysis would flip immediately: with zero PTAB history you would want to run your own pre-filing prior-art search against the two references of record (US4926347, US4992957) and the broader 1980s–1991 printer/emulation art before filing, since you would get no estoppel benefit from anyone else's earlier work.

Confidence and caveats. I am highly confident in the negative finding on PTAB activity: it rests on the authoritative ODP block in this prompt plus multiple web searches that returned no petition, no institution decision, and no FWD. I am highly confident in the expiration date (2012-06-25) and the "Expired - Lifetime" status. I have not verified district court litigation history for this patent; my searches did not surface any enforcement campaign, but absence of search results is weaker evidence there than on the PTAB side, so I am not asserting "never litigated." Any proceeding number, panel, ground, or FWD disposition would need to be invented to fill the template, and I have declined to do that.

I'll start by checking whether any recorded assignment data exists beyond what's in the authoritative patent text, then cross-reference.

No assignment data surfaced yet. Let me probe the assignment database directly and check for any litigation involving this patent.

Research caveat (read first)

The USPTO Assignment Center search application is a session-based interface that is not crawlable by general web search; my queries against assignmentcenter.uspto.gov and assignment.uspto.gov returned only USPTO's own announcements about the system (and the 2025-10-20 migration to the new Assignment Search app), not records for this patent. I therefore relied on (a) the authoritative Google Patents legal-events record reproduced in the source text for US 5228118, and (b) targeted searches for litigation and assignment records. I could not retrieve a reel/frame number for any assignment, and I will not invent one. Where a field is unknown, I say so.

---

Inventors

Unusual-pattern check: none found. There is one inventor, and the record shows the inventor's interest assigned to his employer rather than away from it. No evidence of inventors departing the original assignee within 12 months of filing, and no assignment evidence of a portfolio fire-sale. The multiple-record "cited by" list (254 citing documents, per the Google Patents page) shows the disclosure was absorbed into the industry mainstream by operating companies — HP, Canon, Xerox, Microsoft, Lexmark, Seiko Epson, Hitachi, Sharp, Samsung, Minolta, Oki, Ricoh, Toshiba — which is a citation pattern, not an ownership pattern.

---

Original assignee

Brother Kogyo Kabushiki Kaisha (trading as Brother Industries, Ltd.), Nagoya, Japan — named as original assignee on the issued patent and as the assignee in the sole recorded reassignment event on the Google Patents legal-events record.

• Primary line of business: operating manufacturer — printers, multifunction devices, sewing machines, machine tools, and related consumer/business electronics. Publicly listed on the Tokyo Stock Exchange (securities code 6448).
• Product embodying the claims: Yes, plausibly. The spec (FIGS. 3–6) describes a laser printer LP whose control ROM stores LANGUAGE INTERPRETER A (office/personal-printer control language) and LANGUAGE INTERPRETER B (CAD/plotter control language), plus a host PC holding PRINTER DRIVER C and PRINTER DRIVER D, with the host interrogating the printer and auto-selecting the matching driver. This maps directly onto Brother's HL-series laser printers shipping emulation/multiple-interpreter modes driven by vendor printer drivers.
• Current status: operating. No bankruptcy, dissolution, or acquisition event found. The patent itself is expired: Google Patents records 2012-06-25 Anticipated expiration and legal status Expired – Lifetime (20-year term from the 1992-06-25 US filing date).

---

Assignment timeline

There is exactly one recorded assignment in the chain, and it is the original inventor-to-employer conveyance — there are no post-issuance assignments.

• 1992-06-25 (executed) / recorded 1992-06-25 — Reel not retrievable (Assignment Center search not accessible to me; Google Patents does not expose the reel/frame for this event)
  • Conveyance: Assignment — recorded under the legend "ASSIGNMENT OF ASSIGNORS INTEREST"
  • Assignor: Ichiro Sasaki
  • Assignee: Brother Kogyo Kabushiki Kaisha (Brother Industries, Ltd.)
  • Correspondent: not available — no correspondent of record was exposed in any source I could reach. I am not inferring one.
  • Context: ordinary pre-issuance inventor-to-employer assignment for a corporation-filed application (the JPO priority was filed by the same applicant one day earlier).

No further assignments. The only other event on the legal-events record is a non-assignment docket entry: 2012-06-25 Anticipated expiration. There is no assignment to an IP holding company, no security agreement, no merger, no change of name, and no release.

Foreign family members (relevant only to confirm the US chain was not a fragment): EP0521673B1, DE69226736T2, and JP2661416B2 are country-status siblings of the same 1991-06-29 priority. No assignment records for these were surfaced either. Their existence supports the reading that Brother retained the family intact rather than breaking it up for sale.

Litigation check: No infringement action naming US 5,228,118 was identified in searches. The "Cited By" and "Families Citing" tables on the Google Patents page are citation relationships, not assignments or litigation, and should not be read as an assertion chain. Note that the "Families Citing this family" entry US6348971B2 (Seiko Epson, 1997-06-20 priority) is a patent that cites this family — again a citation, not a transfer.

---

Timeline diagram

timeline
    title Ownership of US 5228118
    1991 : Priority application filed in Japan
    1992 : US application filed
         : Inventor assigns interest to Brother Kogyo
    1993 : Patent issued as US 5228118
    2012 : Patent expires by operation of term
         : No assignment or litigation on record

---

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The only recorded transfer runs from the inventor to the operating manufacturer Brother Kogyo Kabushiki Kaisha (1992-06-25). There is no subsequent movement to a name bearing "IP / Patents / Licensing / Holdings / Ventures," no registered-agent address, and no single-purpose LLC anywhere in the record.

2. Known asserter in the chain — not present. Brother Kogyo Kabushiki Kaisha is a publicly traded operating manufacturer, not an entity appearing on any NPE roster. No Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, or Document Generation Corp entity appears in this chain.

3. Repeat correspondent across the chain — not present (unable to evaluate). No correspondent of record was surfaced for the single assignment entry. With one transfer in the chain, the recurrence test (the signal is recurrence, per your instruction) cannot be satisfied even in principle.

4. Cascading transfers — not present. No consecutive chained assignments exist; the chain has length one.

5. Pre-litigation transfer — not present. No infringement suit naming US 5,228,118 was identified, so there is no suit date against which a 6-month pre-litigation transfer window could be measured.

6. Bankruptcy fire-sale — not present. No Chapter 7/11 proceeding was found for Brother Kogyo Kabushiki Kaisha; the company is an operating concern. No Kodak/Nortel/Polaroid-style sale appears in the record.

7. Privateering — not present. Nothing in the record, and nothing in the 254-document citing set, indicates Brother funded an NPE to assert this patent against competitors. The citation footprint (HP, Canon, Xerox, Microsoft, Lexmark, Epson, Hitachi, Sharp, Samsung) is consistent with organic industry citation, not with a managed assertion campaign.

8. Defensive aggregator (anti-NPE) — not present as an event, but the outcome is the same. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. However, the patent is expired as of 2012-06-25 (Google Patents legal status "Expired – Lifetime"), so the asset is commercially neutralized regardless — expiration, not acquisition, is what ended any assertion potential here.

---

Verdict

Insufficient data — and, on the record available, affirmatively not an NPE chain.

The chain consists of a single inventor-to-employer assignment executed and recorded 1992-06-25 by Ichiro Sasaki to Brother Kogyo Kabushiki Kaisha; there is no recorded post-issuance transfer of any kind, and the only other legal event is the 2012-06-25 anticipated expiration. Because I could not access the Assignment Center record directly, I cannot supply a reel/frame for even that one entry, and I decline to fabricate one — hence "insufficient data" rather than "defensive / non-asserting," which would require a documented terminal assignment to an aggregator. The absence of any transfer out of a publicly traded operating manufacturer whose products embody the claims is itself the substantive finding: this is a plain operating-company-owned patent that died of old age.

Verify at: https://assignmentcenter.uspto.gov/ (search by patent number 5,228,118; the older mirror at https://assignment.uspto.gov/patent/index.html surfaces the same underlying records). If the Assignment Center returns a reel/frame for the 1992-06-25 conveyance, that would fill the one gap above; I would expect it to show a single entry with assignor SASAKI, ICHIRO and assignee BROTHER KOGYO KABUSHIKI KAISHA, with no later frames.

I'll build on the prior sections by focusing strictly on the prior-art citations of US 5,228,118. Let me verify the cited references directly.

Let me pull the family-level prior-art citations and the European counterpart's citation list, plus verify the JP references.

Prior-Art Analysis — US 5,228,118

Building on the earlier sections (bibliography, claims digest, litigation null result). This section does not repeat them; it addresses only the cited prior art and its §102/§103 exposure. Note on dates: your task header says April 26, 2026, while the fetched record and prior sections use 2026‑09‑17. I treat the substantive content as unchanged; flagging the inconsistency rather than silently reconciling it.

---

1. Scope and method caveat

The authoritative citation set comes from the Google Patents record (fetched 2026‑09‑17), which lists:

• "Patent Citations (2)" — the two references on the face of the patent, i.e., the examiner‑cited art reviewed during prosecution of US 07/904,491.
• "Family Cites Families (4)" — references cited against the family members (JP 2661416 B2, EP 0521673 B1), i.e., from the JP/EP search reports rather than the US face.

Two important analytic limits, stated up front:

1. I have full claim text and abstract/description excerpts for the two US references but not for the two JP references (JPH01230122A, JPH02178725A). My searches for them returned no results. I therefore cannot map their disclosures element‑by‑element to the claims, and I say so explicitly rather than guessing.
2. Anticipation (§102) requires a single reference disclosing every element of a claim, arranged as claimed. Because the two independent claims here contain a data‑processing‑device‑side inquiry + selection architecture, and every cited reference is a printer‑side architecture, my conclusion below is that no cited reference anticipates any of claims 1–22. The references are better characterized as §103 art and as the admitted background. Details per reference follow.

Forward citations are not prior art. The 116/254 "Cited By" entries are all published 1994+ (earliest substantive ones, e.g., US 5,392,419 HP filed 1992‑01‑24; US 5,402,527 Xerox priority 1993‑04‑23), i.e., after the 1991‑06‑29 priority date. They cannot be §102 art.

---

2. Master table of the six family-level citations

Priority application JP 3‑185309 (JP 18530991A → JP 2661416 B2) is the priority document, not prior art.

All six predate both the 1991‑06‑29 priority date and the 1992‑06‑25 filing date, so all are §102(a)/(b)‑eligible on their faces (subject to the JP documents' actual disclosure).

---

3. Reference-by-reference analysis

A. US 4,926,347 A — Citizen Watch (1990‑05‑15)

URL: https://patents.google.com/patent/US4926347

Disclosure: A printer with an IC‑card terminal accepting interchangeable emulation IC cards (each storing a different maker's control program — i.e., an interpreter). In a maintenance mode, the printer's CPU reads identification data at a predetermined card address, compares it against names stored in printer ROM, and prints the card name and version. Stated object: let the printer identify which emulation card is installed "even if the printer is not connected with a host computer." The specification contains the key admission that "only a host computer can discriminate a type of emulation IC card currently mounted in a printer" (prior‑art drawback the invention addresses).

§102 mapping — claims 1–22: Does not anticipate any claim.

• It teaches printer‑side identification of an interpreter/emulation module and printing that identification for an operator.
• Claim 1 requires inquiring means in the data processing device sending an inquiry signal, and printer driver selecting means selecting from a plurality of drivers. US 4,926,347 has no data processing device at all in the identification path — expressly the opposite (identification is done offline, without the host).
• It also lacks ID data output means that transmits interpreter‑ID data to a data processing device in response to an inquiry; it prints locally instead.
• Claims 6 and 17 (printer memory storing the interpreter‑ID data, read out on inquiry) are the only claims for which this reference is even topically relevant — but the read‑out is to a printer mechanism, not to a host over a bidirectional interface, so it still fails.

Net: best cited as the closest art on the printer‑side half of claim 1 (identifying what interpreter is installed) and as motivation for §103 combination with a host‑side driver table. Not anticipatory.

---

B. US 4,992,957 A — Seiko Epson (1991‑02‑12)

URLs: https://patents.google.com/patent/US4992957 ; EP family member EP 0332779 A2 (priority 1988‑03‑16, published 1989‑09‑19)

Disclosure (high confidence — full text reviewed): A single printer connected to a plurality of host computers, storing a plurality of print control units (23–25 / 3–4), each with its own character‑code and control‑code set (i.e., multiple interpreters). A character‑code/control‑code group judging unit analyzes the received data — comparing it to each control unit's syntax, assigning weighted probabilities, and taking ratios (H₁ vs. H₂) — and automatically selects which print control unit processes the data. The EP counterpart, published 1989‑09‑19, is the same disclosure via a different route.

§102 mapping — claims 1–22: Does not anticipate any claim, but it is the single most relevant reference.

• It discloses the printer‑side element of claim 1 ("a printing device which has at least one language interpreter") and even the multi‑interpreter case of claim 2.
• It lacks every data‑processing‑device‑side element of claim 1: no plurality of printer drivers, no inquiring means, no interpreter‑identification data returned over a bidirectional interface, and no printer driver selecting means.
• The selection mechanism is fundamentally different from the claims: US 4,992,957 resolves the language from the content of the print data itself (statistical/probability matching), whereas claim 1 resolves it from an identity reply solicited by an inquiry signal before driver selection.
• Same failure for claim 9: no printer selecting means at the host, no inquiry, no broadcast.

Why this matters: US 5,228,118's own Background § states, as known in the art, "one of the language interpreters of the printer is automatically selected depending upon the printing control language in which the print data received from a specific one of the data processing devices are described." US 4,992,957 is effectively that admitted art. It therefore cannot supply novelty; the patent's contribution must rest on the host‑side inquiry/response + driver selection, which this reference does not touch.

Net: best §103 primary reference (arguably combined with US 4,926,347's printer‑side ID reporting, or with ordinary host driver tables) against claims 1, 2, 9. Not anticipatory.

---

C. US 5,025,398 A — Storage Technology (1991‑06‑18)

URL: https://patents.google.com/patent/US5025398

Disclosure: A "data stream independent printer" — a non‑impact, all‑points‑addressable printer wrapped in conversion apparatus that emulates a line/band printer to the host. The printer stores a plurality of character‑set representations in a disk memory and uses host‑transmitted universal character set buffer (UCSB) entries as pointers to retrieve the corresponding set of printer control data, transparently converting line‑oriented control information into page‑oriented print activation signals.

§102 mapping — claims 1–22: Does not anticipate any claim.

• It involves a printer holding multiple character sets / emulations and selecting based on data received from the host — superficially near claim 1's printer side.
• But there is no inquiry signal from the host asking the printer what it has, and no return of interpreter‑identification data. The information flow is one‑way, host‑to‑printer (the host tells the printer which character set to use; the printer does not report its interpreters).
• Critically, there is no host‑side printer‑driver selecting means: the reference is directed at eliminating host reprogramming, i.e., at avoiding any host driver change, not at automatically choosing among host drivers. It teaches away from the claim‑1 architecture.
• Its 1991‑06‑18 issue date is 11 days before the JP priority date (1991‑06‑29) — a tight but valid §102(a)/(b) date; also §102(e)‑eligible as of its 1990‑05‑21 filing. Worth noting only as a dating curiosity; it is not anticipatory.

Net: §103‑relevant for the "printer stores multiple language/character‑set definitions and picks one based on host data" concept. Not anticipatory.

---

D. US 5,010,514 A — IBM (1991‑04‑23)

Disclosure (title/summary level): "Structured fields at a data stream boundary for delimiting files" — directed to delimiting file boundaries within a mixed data stream via structured fields (a print‑data‑stream architecture reference).

§102 mapping — claims 1–22: Does not anticipate any claim. It addresses data‑stream formatting/delimiting, not interpreter identification, inquiry/response, or driver/printer selection. It lacks every element of both independent claims. Its relevance is contextual (print‑data‑stream conventions of the era).

---

E. JPH01230122 A — Toshiba (priority 1988‑03‑10; published 1989‑09‑13)

F. JPH02178725 A — Canon (priority 1988‑12‑29; published 1990‑07‑11)

Status: I could not retrieve the disclosures. My targeted searches for these two JP publications returned no results, and they are not reproduced in the Google Patents record beyond the citation table. I will not fabricate their content or claim mapping.

What can be said reliably: both are Japanese "printing device"/"printing controller" publications with valid prior‑art dates (1989‑09‑13 and 1990‑07‑11), and they were cited by the JP/EP examiners against the family. Because Japanese printer‑controller art of that era routinely concerned multi‑emulation printers and host/printer command handling, they plausibly bear on the printer‑side elements (claim 1's "at least one language interpreter"; claims 6/17 memory‑readout; claim 9's "ID data output means"). But I cannot assert claim‑level §102 anticipation without their text, and my working hypothesis — consistent with the two US references — is that they too are printer‑side and lack the host‑side inquiry/driver‑selection architecture.

Recommendation: obtain the JP full texts (JPO/J‑PlatPat or the EP 0521673 B1 search report) before treating these as anything more than secondary §103 art.

---

4. Overall §102 / §103 assessment

§102 (anticipation): No cited reference anticipates any of claims 1–22.

• Independents claim 1 and claim 9 both require, in the data processing device, (i) inquiring means issuing an inquiry signal and (ii) driver‑selecting means (cl. 1) or printer‑selecting means (cl. 9) operating on returned interpreter‑identification data. None of A–F discloses an inquiry soliciting an interpreter identity, nor a bidirectional exchange in which the printer answers with its interpreter identity, nor a host that picks a driver/printer on that basis. Every dependent claim (2–8, 10–22) inherits those missing elements.
• The printed/returned‑data distinction is decisive for A: it prints the emulation ID for a human; it does not transmit it to a host in response to an inquiry.
• B is the strongest reference but is squarely the analysis‑based, printer‑side selection the patent itself admits as prior art.

The genuinely distinguishing feature of US 5,228,118 — and the reason none of the cited art anticipates — is the combination of: bidirectional interface + host‑originated inquiry + printer‑returned interpreter‑ID data + host‑side auto‑selection of the matching printer driver (cl. 1), or of the matching printer (cl. 9).

§103 (obviousness) is where the citing art actually bites. The most defensible combination an examiner could raise is B (US 4,992,957) as primary + A (US 4,926,347) as secondary, on the theory that a printer that (per B) already knows which of several interpreters it holds, and (per A) can already report its installed emulation identity, would obviously be queried by a host to auto‑select a matching driver. Note the counter‑argument, which I flag rather than resolve: A expressly frames offline printer‑side reporting as an advantage over host‑based discrimination, so A arguably teaches away from the host‑query architecture of claim 1. The §103 conclusion therefore turns on how the prior‑art admissions in US 5,228,118's own Background are weighed.

Claim‑level exposure summary:

---

5. Uncertainty notes (explicit, not fabricated)

• JP references E and F: disclosures not retrieved; no claim‑level mapping asserted. This is the principal gap in this analysis.
• EP 0521673 search report: I could not surface the EP search report's own citation list in this session; the four "Family Cites Families" entries are the family‑level art I can substantiate, and the EP examiner may have cited additional documents I have not seen.
• Dates: I use the Google Patents dates (priority/publication as tabled). The earlier section's noted 1991‑06‑29 vs. 1991‑06‑28 priority‑date discrepancy and the Uniform Patents variance remain unresolved and do not change the §102 eligibility conclusions above (all references predate even the earlier of the two candidate dates, except that US 5,025,398's 1991‑06‑18 issue is only 10–11 days prior).
• No §102 conclusion here is a legal opinion; it is a technical element‑by‑element comparison based on the text available.

Sources: https://patents.google.com/patent/US5228118/en ; https://patents.google.com/patent/US4926347 ; https://www.sumobrain.com/patents/us/Printer-with-emulation-card-installation/[4926347](/patent/4926347).html ; https://patents.google.com/patent/US4992957 ; https://www.sumobrain.com/patents/us/Printer-method-utilizing-character-codes/[4992957](/patent/4992957).html ; https://uspto.report/patent/grant/4992957 ; https://portal.unifiedpatents.com/patents/patent/EP-0332779-A2 ; https://patents.google.com/patent/US5025398 ; https://uspto.report/patent/grant/[5025398](/patent/5025398)

I'll ground this analysis in the actual prior-art references listed on the patent's face and family record. Let me pull details on each.

Let me pull the remaining prior-art-of-record references and one highly relevant non-record reference.

Obviousness Analysis — U.S. Patent No. 5,228,118 under 35 U.S.C. § 103

Patent under analysis: US 5,228,118 A, "Printing system wherein one of printer drivers through which print data is sent from a data processor to a printer is selected based on interpreters available on the printer" (Sasaki; Brother Kogyo Kabushiki Kaisha)
Effective filing date: 1992-06-25 (US 07/904,491); foreign priority: 1991-06-29 (JP 3-185309)
Source of record: https://patents.google.com/patent/US5228118/en

Scope note. This is a technical obviousness assessment, not a legal opinion or a validity conclusion. Because the application was filed 1992-06-25, pre-AIA § 103(a) governs (the first-inventor-to-file provisions of the AIA apply only to applications filed on/after 2013-03-16). Everything below builds on the Patent Summary and Litigation Summary already generated, and I do not repeat them. Two things in those prior sections matter here: (i) no litigation was found (so there is no claim-construction record to borrow), and (ii) the patent's 22 claims split into two independent families — claim 1 (driver-selection) and claim 9 (printer-selection).

---

1. Preliminary: which references actually qualify as prior art

The task directs me to the Prior Art section of the page. That section contains two distinct sets, and the distinction matters:

Verification caveat (stated explicitly rather than papered over): I confirmed the bibliographic data and publication dates for the two JP documents but was not able to retrieve their disclosures in this session (tool-step limit). I therefore do not build any combination on them; their titles and dates are noted only so the record is complete. If their texts are retrieved, they would be candidates for the "printer-side multiple-interpreter" role currently assigned to US 4,992,957.

One additional reference I am adding beyond the record section, flagged as such: US 5,165,014 (Hewlett-Packard / Vassar, filed 1990-09-12, issued 1992-11-17). Its issued patent post-dates the 1991-06-29 priority date, so it is not § 102(a)/(b) art; however, because its U.S. application was filed 1990-09-12 — before the 1991-06-29 priority date — it is available as prior art under pre-AIA § 102(e). (Its WO 92/04672 publication of 1992-03-19 and EP 0 548 251 publication of 1993-06-30 are not prior art by date.) I use it strictly as a motivation/secondary reference, not as the anticipatory core.

Negative caveat that avoids a common error: most entries in the "Cited By" list are not prior art to this patent, because they post-date the 1991-06-29 priority date and would at most be § 102(e) art if their own U.S. filings predated it. In particular, US 5,392,419 (HP, priority 1992-01-24), US 5,402,527 (Xerox, 1993-04-23), US 5,462,743, and US 5,511,149 (Brother, 1991-07-10) cannot be used against the claims. An analyst reaching for these will be reaching for the wrong corpus.

---

2. Level of ordinary skill in the art (POSITA)

A person of ordinary skill in the art as of June 1991 would have had a bachelor's degree in electrical engineering or computer science, or equivalent experience, plus approximately two to four years working with computer-to-printer interfaces. That person would be familiar with:

• escape-code / command-language printer control (e.g., PCL, PostScript, plotter/CAD languages) and printer emulation;
• the fact that a single printer may host multiple emulations/interpreters switchable by configuration or automatically (expressly admitted in the '118 Background, below);
• bidirectional serial/parallel and network device interfaces, and host-side printer drivers and configuration dialogs;
• data-communications protocol patterns such as capability interrogation / query-and-reply negotiation (this is the pattern the IBM reference below exploits).

This level is set at a "routine engineering" plane, not an inventor-genius plane, which is the plane against which KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), directs that predictable combinations of known elements be measured.

---

3. Critical foundation: applicant's own admissions (AAPA)

Before mapping references, note that the '118 specification concedes much of the claim structure. These admissions are the strongest single fact in the § 103 analysis because they remove elements from the "difference" side of the Graham inquiry:

1. Multiple printer drivers on the host is old. "A data or text processing device of modern vintage known as 'word processors', 'personal computers' or 'work stations' is provided with various printer drivers which are compatible with or applicable to respective language interpreters available on various types of printers…" ("Background," Discussion of the Prior Art).

2. A multi-interpreter printer that auto-selects an interpreter is old. "There is also known a printing system wherein a printer having a plurality of language interpreters is connected to different types of data processing devices… In this printing system, one of the language interpreters of the printer is automatically selected depending upon the printing control language in which the print data received from a specific one of the data processing devices are described." (Id.)

3. The only stated gap is host-side automatic driver selection. "However, each data processing device is not capable of automatically selecting one of the printer drivers through which the print data are sent to the printer." (Id.)

4. The motivation is stated in the patent itself. Manual selection is "cumbersome and time-consuming," worsened by revisions/additions of interpreters and drivers, producing "a relatively long overall non-printing time" and "relatively low overall printing efficiency."

Consequence: For § 103 purposes, element (a) of claim 1 (printer with ≥1 interpreter), element (b) (host with multiple drivers), and the printer-side automatic interpretation are admitted prior art. The only genuinely contested subject matter in independent claim 1 is the host-initiated inquiry ((c)), the host-side driver selection based on the printer's reply ((d)), and the printer-side reply means ((e)). And admission #3 is, in effect, a statement of the problem the invention set out to solve — which is itself a recognized motivation under § 103.

---

4. What each qualifying reference teaches

4.1 US 4,992,957 (Seiko Epson) — multi-interpreter printer that identifies/selects the right interpreter

https://patents.justia.com/patent/[4992957](/patent/4992957) | EP counterpart: http://data.epo.org/publication-server/rest/v1.2/patents/EP0332779NWA3/document.html

• A printer with an input data buffer (1, 21) and a plurality of print control units (3, 4; 23–25), each with "internal control codes and character codes specific to" itself — i.e., multiple language interpreters in one printer.
• A "character code/control code group judging unit (2, 22)" that "determines whether the input data corresponds to any one of the plurality of character code groups or control code groups" and then routes the job to the matching print control unit. Its stated objects include "a printer capable of automatically selecting and outputting character codes and control codes corresponding to a plurality of host units."
• The hardware figure (FIG. 5A–5C) shows a CPU 51, ROM 52, RAM 53, and an I/F unit 54 "connected to CPU 51 by bus 55 [that] receives the data from the host units," with an operation panel that can display the determined control code.
• It also teaches a timer/timeout: "If the next transfer of data is not performed within a predetermined time period, the last transferred data is considered to be the final bit of data."

What it supplies: printer-side plurality of interpreters (= "at least one language interpreter," and the multi-interpreter case of claim 2); the concept of matching a body of print data to one of several available interpreters; printer-side memory holding the interpreter programs; a host-facing interface.

What it does not supply: it matches interpreter → incoming data, not host driver → printer; and it does not have the host interrogating the printer.

4.2 US 4,926,347 (Citizen Watch) — printer whose emulation identity is present in the printer

https://patents.justia.com/patent/[4926347](/patent/4926347) | PDF: https://patentimages.storage.googleapis.com/80/fb/22/d8e89c62023b76/US4926347.pdf

• A printer with detachable emulation IC cards holding "control programs of different makers," because "code systems of ESC code vary depending on computer makers," and the cards let the printer work regardless of the host's maker. This is a language interpreter embodied in the printer.
• Crucially: "Contents of the emulation IC cards, e.g., information about the card and version names are stored at predetermined addresses thereof." The printer reads that data and prints indicia identifying the card in a maintenance mode.
• The reference itself states the deficiency and even the host-side capability: "However, only a host computer can discriminate a type of emulation IC card currently mounted in a printer, and hence inconvenience is caused. It is an object of the present invention to… provide a printer capable of identifying a type of IC card currently mounted therein even if the printer is not connected with a host computer."

What it supplies: (i) express recognition that a printer's emulation/interpreter identity is machine-readable identification data stored in printer-accessible memory (maps to claim 6 / 17 "memory means for storing said interpreter-identification data"); (ii) the design problem of what the printer is equipped with being knowable only with inconvenience; (iii) a two-mode architecture (maintenance mode vs. operating mode) suggestive of the claim-8 "automatic … mode" gate.

What it does not supply: the ID data goes to a printing mechanism for a human operator, not over an interface to the host. It is the opposite of the '118 architecture in signalling direction (self-report to paper vs. report to host).

4.3 US 5,010,514 (IBM) — the host-asks / device-answers capability-negotiation protocol

https://patents.google.com/patent/US5010514 | https://patents.justia.com/patent/[5010514](/patent/5010514)

This is, in my assessment, the single most important reference for the "inquiry" and "ID data output" limitations, because it discloses the protocol architecture almost verbatim:

• "Certain systems require prior knowledge of the receiver's functional capability before sending any information to it. This information is solicited by the sender device by means of a query structured field. The query structured field requests the receiver device to transmit information regarding its functional capability and the parameters associated with that function."
• "The host system generates and transmits a message seeking the identity of devices which require begin and end of file characters. The device responds with a novel format in which the structured field is coded in accordance with a unique pattern."
• Nested in a LAN with a print server servicing "a plurality of mainframe processors 3, 4 and/or personal computers 5, 6" and printers 9, 10, with "one or more disk drives 7, 8 and printers 9, 10 depending upon the volume of data."
• Selection among printers is done "in accordance with an established priority."
• The reference also discusses time-outs as a known mechanism ("Time-outs are used for a variety of purposes in both SNA and non-SNA communications").

What it supplies: a bidirectional host→device query and device→host reply; the reply conveying capability/identity data so the host can adapt; multiple devices on a network; priority-based device selection; print-server spooling. Its only substantive difference from the '118 "inquiring means"/"ID data output means" pair is the subject matter of the capability data (begin/end-of-file support rather than a language-interpreter list) — a difference of content, not of mechanism.

4.4 US 5,025,398 (Storage Technology) — printer-side conversion apparatus keyed to a host-supplied identifier

https://patents.google.com/patent/US5025398 | PDF: https://patentimages.storage.googleapis.com/ad/f7/d9/58f5ef7b37d0b5/US5025398.pdf

• A printer with conversion apparatus 111 that emulates a line printer to the host and stores "a plurality of character set definitions, overlays, copy modifications and setups on a disk drive memory."
• "In operation, the host processor transmits a band identification data to the printer system, which data is used to retrieve from the disk memory the form and format of the character set that is available to the host processor."
• The stated purpose is to make the printer work with host processors written for different printer classes — the same heterogeneous-host problem as the '118 patent.

What it supplies: a printer holding multiple selectable "personalities"/character sets in memory; a host-supplied identifier used to select the correct one printer-side; and printer-side memory of emulation data.

4.5 US 5,165,014 (HP/Vassar) — not of record; § 102(e) art; motivation reference

http://www.everypatent.com/comp/pat5165014.html | https://patents.google.com/patent/US5165014

• Directly addresses the same pain point and says so: "the personal computer user is currently required to make a necessary keyboard (keycap) or other equivalent selection in order to ensure that the correct electrical connections are made… The obvious disadvantage… not only involves an additional computer keyboard operation for the user, but a failure to execute this proper and necessary keyboard operation means that the improperly selected printer or command language conversion card will be totally inoperative."
• "The general purpose and principal object of the present invention is to totally eliminate the above keyboard and keycap printer language selection requirement."
• It discloses a network of a PC workstation and "a plurality of available printers operating in this same language," plus a print server that interconnects the PC to one of the plurality of printers based on the command language — i.e., selecting a printer/conversion card to match a language, in the same genus as '118 claim 9.
• Its own cited references include "IBM Technical Disclosure Bulletin, vol. 33 No. 1B, Jun. 1990, pp. 126–130, 'Printer Data Stream Method to Allow Dynamic Stream Switching'" — a June 1990 printed publication evidencing that dynamic switching among printer data streams was known >1 year before the '118 filing (§ 102(b)).

What it supplies: (i) the explicit articulation of the problem and motivation ("eliminate manual selection"); (ii) "matching" a host-side language to a printer-side language; (iii) selection among plural printers. Its difference from '118 is the sensing locus: it determines the language by sniffing the host's own outgoing data stream, not by asking the printer.

---

5. Claim 1 — element-by-element mapping

Independent claim 1 requires (a) printer with ≥1 language interpreter; (b) host with a plurality of drivers over a bidirectional interface; (c) host inquiring means sending an inquiry signal asking the printer to send back interpreter-identification data; (d) host driver-selecting means choosing a driver compatible with one of the printer's interpreters on the basis of the received data; (e) printer ID data output means responsive to the inquiry signal.

Assessment: Element (a) is admitted prior art and is taught by Epson (§4.1) and Citizen (§4.2). Element (b) is admitted prior art. The entire remaining delta is the pairing of (c)+(e) — host asks, device answers — with (d) — host then auto-selects its driver. Reference IBM supplies (c) and (e) as a general mechanism; HP/Vassar supplies (d) and the motivation; Epson and Citizen supply the printer-side hardware substrate and the recognition that the printer's language capability is knowable and mismatch is a problem.

---

6. The combinations, and why the POSITA would have made them

I set out three independent combination theories; any one is sufficient for a prima facie § 103 case against claim 1, and together they are mutually reinforcing.

Combination A (primary): US 4,992,957 + US 5,010,514 (+ US 5,165,014 for motivation)

Rationale / motivation:

1. Same field and same problem. All three concern making a computer drive a printer whose command language may or may not match. Epson's own object is "automatically selecting… character codes and control codes corresponding to a plurality of host units"; IBM's is adapting a host's data stream to a receiving device's capabilities; HP's is eliminating manual language matching. All three are in the printer-control-language art — the KSR "same field of endeavor" factor.
2. A known technique applied to a known structure. Capability interrogation (IBM) is a general data-communications technique expressly justified as answering "certain systems require prior knowledge of the receiver's functional capability before sending any information to it." Once a printer is known to host a set of selectable interpreters (Epson; Citizen), a POSITA has an obvious reason — and IBM gives an explicit template — to extend the query/reply to enumerate those interpreters rather than only EOF capability. Applying a known negotiation protocol to a new data field, with no change in mechanism, is the paradigm of a predictable combination under KSR.
3. The problem was expressly identified, which supplies the "reason to improve." The '118 Background itself concedes that the state of the art could auto-select a printer's interpreter but could not auto-select the host's driver, and the HP reference independently frames precisely that as the object of the invention: "totally eliminate the… keyboard and keycap printer language selection requirement." Where the art both perceives the deficiency and points to the general solution class (automatic matching), the improvement is obvious.
4. Reasonable expectation of success. Nothing in the combination is unpredictable: enumerate interpreters over an existing bidirectional link, compare names against a driver table, pick the match. IBM already does the enumeration-and-adapt step; Epson already does the table-matching step — only the side on which matching occurs changes.
5. Market/design pressure. Both Citizen and Epson explain the multi-vendor ESC-code problem (printers must work "regardless of a maker of a host computer"); the industry incentive to make a multi-emulation printer plug-and-play with existing drivers is the classic KSR "design incentive" factor.

Combination B (alternative, and arguably cleaner on (c)/(e)): US 4,926,347 + US 5,010,514

US 4,926,347 already puts machine-readable interpreter-identification data at a known address inside the printer ("information about the card and version names are stored at predetermined addresses thereof"), and expressly notes "only a host computer can discriminate a type of emulation IC card currently mounted in a printer." Combining that stored card-identity with IBM's query/reply protocol yields the (c)/(e) pair nearly by substitution: instead of printing the card ID for an operator (Citizen) or replying with EOF capability (IBM), reply with the interpreter ID. Adding US 5,165,014 (or the admitted prior art in the '118 Background) supplies the host-side driver-selection step, and supplies the motive (eliminating manual selection).

Combination C (multi-interpreter / priority variants): + US 5,025,398 and the June 1990 IBM TDB

For claims 2–3 (multiple interpreters → multiple compatible drivers → pick by priority), Epson's multi-unit printer supplies the multiple-interpreter case, and IBM's "in accordance with an established priority" plus the June 1990 TDB on dynamic stream switching supply priority-based selection and switching among streams.

Motivation statement (one paragraph suitable for an office action or IPR petition)

A person of ordinary skill in the art, faced with a printer containing multiple interpreters (Epson '957; Citizen '347) and a host already supplied with multiple printer drivers (admitted), and presented with the known technique of host-initiated capability interrogation over a bidirectional link (IBM '514) and the express desire to eliminate manual driver/language selection (HP '014; applicant's own Background), would have had both a reason and a reasonable expectation of success in having the host query the printer for its interpreter identity and automatically select the matching driver. The combination is a mere arrangement of old elements (a query, a reply, a lookup table, a stored ID) each performing its known function, yielding no more than the predictable result of removing a manual step — the very result the art was pursuing.

---

7. Dependent claims 2–8

Claims 2–8 add no element that is not either disclosed or a conventional implementation choice. My confidence is highest for 2, 3, 4 and 6; claims 5, 7 and 8 are "design choice" arguments that a POSITA could support but that are somewhat softer if a patentee argues specific implementation detail.

---

8. Independent claim 9 and dependents 10–22

Claim 9 inverts the architecture: multi-printer network, host queries the plurality of printers, each printer replies with its interpreter-identification data to the requesting host, and the host selects the printer whose interpreter matches a driver it holds.

Assessment: claim 9's core is well supported by HP '014 (printer-selection-to-match-language, on a network, over a print server) combined with IBM '514 (host-queries-device capability pattern). Claims 11–12/21–22 and 19 are alternative expressions or conventional data fields. Claims 14–15 (busy-status second inquiry) are the only sub-feature where I would want a supplemental reference not present in this record before asserting a strong § 103 position.

---

9. Secondary considerations (objective indicia)

I found no evidence in the record of secondary considerations favoring non-obviousness, and I flag that the absence is itself weakly probative only:

• No licensing program, no litigation (per the Litigation Summary) — no evidence the industry valued the claims as a barrier.
• No unexpected results / long-felt need evidence on the face of the patent; the asserted benefits ("improved printing efficiency," shorter non-printing time) are the predictable consequence of automating a manual step, not an unexpected one.
• Nexus caution: any commercial success of Brother multi-emulation printers would need a nexus to the claimed automatic driver/printer selection, not merely to the printers' general popularity. None is established.
• Possible contrary signal: the patent was granted over art that included exactly the two multi-interpreter references (Citizen, Epson), and the examiner (Arthur G. Evans) is the same primary examiner of record on US 5,025,398 and US 5,165,014 in the same art — suggesting he was aware of the relevant corpus. That cuts toward the allowance being a judgment that the host-side inquiry was the distinguishing feature (see § 10), not toward non-obviousness generally.

---

10. The strongest non-obviousness counterarguments (steelmanned)

I would be doing the analysis a disservice not to state where the patent is genuinely strongest, because these are the arguments a patentee (or an examiner defending allowance) would press:

1. Architectural inversion. The prior art splits cleanly into (i) printer-side detection — Epson '957 matches incoming data to an interpreter, and (ii) host-side detection — HP '014 sniffs the host's own outgoing stream. Neither discloses a host polling a printer for its interpreter inventory. Citizen '347 goes out of its way to make the report human-visible and printer-autonomous ("without connecting the printer to the host computer"). So the "pull" model of claim 1 — the host discovering the printer's interpreters by direct interrogation — is arguably absent from the art of record. If a court treated the direction of information flow as the inventive concept, this is the pivot point.
2. Epson's disclosure is the opposite mapping direction, so a combination argument must explain why inverting the matching side is not itself inventive. (Answer: the problem statement — admitted in the '118 Background — plus HP '014's express object of eliminating manual selection supplies the reason; the inversion is a predictable design choice, not a new principle.)
3. IBM '514's capability data is about begin/end-of-file delimiting, not language interpreters. A patentee would argue the reference teaches away from using the query/reply for language capability, or at least that extending it is not "the same field." (Answer: same protocol, same purpose — "prior knowledge of the receiver's functional capability" — and the difference is content, not mechanism; KSR permits combining references where the improvement is a predictable use of a known technique.)
4. The claim-8 mode gate and the claim-4 timeout/alarm combination give a fallback position if an independent claim falls: the specific interactive sequence (inquiry → timer → alarm → driver selection) could be argued as a particular, non-routine workflow. On the record I have, this is a weak fallback because each step is individually conventional and the combination yields expected results.

Where I would expect the case to be decided: on whether the art's failure to disclose interrogating the printer for its interpreter list is an "inventive" architectural choice or, under KSR, a predictable substitution of one known source of knowledge (the host's own stream; a printed card report; a data-stream sniff) for another (a direct query), in service of a goal (no manual selection) the art itself articulated.

---

11. Bottom line and confidence

• Claim 1 is, in my assessment, likely obvious over US 4,992,957 + US 5,010,514, motivated by US 5,165,014 (and by the applicant's own Background admissions). Combination B (US 4,926,347 + US 5,010,514 + US 5,165,014) is an equally viable alternative because Citizen already places machine-readable interpreter-identification data inside the printer and names the very deficiency the IBM reference's query/reply mechanism addresses. Confidence: moderate-to-high.
• Claims 2, 3, 4, 6, 10–13, 17–19, 21–22 are likely obvious as conventional implementations or alternative expressions of the same exchange. Confidence: moderate-to-high.
• Claims 14–15 (busy/unoccupied second inquiry with priority tie-break) are the weakest § 103 link on the present record, because the prior-art-of-record does not squarely disclose busy-status interrogation; a print-server/queue-dispatch reference beyond this record would be needed for a confident position. Confidence in the current record: low-to-moderate.
• No secondary considerations of non-obviousness are established on the evidence available.
• I could not verify the disclosures of JPH 01230122 A or JPH 02178725 A and have therefore excluded them from the combinations. If retrieved, they may strengthen (or complicate) the Epson-based "printer-side multiple-interpreter" strand of the analysis. I would also verify the June 1990 IBM TDB (Printer Data Stream Method to Allow Dynamic Stream Switching) directly before relying on it, since I have seen only its citation in US 5,165,014 and not its text.

Primary sources: https://patents.google.com/patent/US5228118/en · https://patents.justia.com/patent/4926347 · https://patents.justia.com/patent/4992957 · http://data.epo.org/publication-server/rest/v1.2/patents/EP0332779NWA3/document.html · https://patents.google.com/patent/US5010514 · https://patents.google.com/patent/US5025398 · http://www.everypatent.com/comp/pat5165014.html · https://uspto.report/patent/grant/[5165014](/patent/5165014)