US 4521677

I'll search for the specific patent number and check for any related 2026 CAFC docket activity.

Let me verify litigation status and confirm no Federal Circuit activity involving this patent.

US 4,521,677 — Search Results Summary

No‑auto‑correction note: I searched the literal string 4521677. Two unrelated registrations share that digit string — JP 4521677 (Toyoda Gosei-type automotive weatherstrip, app. 2007‑12‑01, pub. 2008‑10‑23) and EP 4521677 (a 2026 European publication appearing in a national-library gazette dated 17.06.2026). Neither is the US patent. Everything below refers only to US 4,521,677.

CAFC 2026 docket check — negative result

I found no Federal Circuit docket entry, appeal, or 2026 CAFC activity involving US 4,521,677. This is consistent with the bibliographic record: the patent's legal status is "Expired – Lifetime," with anticipated expiration 2003‑12‑02, and its listed assignee is an individual, so there is no live patent right to litigate in 2026.

Caveat on confidence: my search tools are general web retrieval, not a direct PACER/CAFC CM‑ECF or USPTO PatentCenter query. I cannot positively exclude an un‑indexed or differently‑formatted 2026 filing, but nothing surfaced. One incidental hit did appear in a Microsoft/HP IPR petition (Petitioner Exhibit 1002) where "4521677" appears inside a USPTO PLUS "closest related" search listing — that is a search‑result artifact, not an assertion of the patent, and I would not characterize it as litigation.

Bibliographic data

Sources: Google Patents US4521677A; Justia; uspto.report.

Uncertainty flags:

• Assignee. Both the original and current assignee are recorded as "Individual." I found no evidence of reassignment to NCR, IBM, or any retailer/EPL vendor. If a definitive chain of title matters, it must be confirmed via the USPTO Assignment database rather than third‑party aggregators.
• Sole cited reference. The only "Patent Citations" entry is a racket handle patent (US 4,180,264). That is almost certainly a Google Patents data artifact rather than a genuine §102/§103 reference, but I am reporting the record literally as it is displayed rather than interpreting it.
• Classification. Primary CPC clusters are G06K 7/10861 (coded labels on articles), G07G 1/145 (PLU management), G06F 3/147, G09G 2380/04 (electronic labels).

Abstract (as published)

A supermarket or like vending establishment equipped with a computer having a bar code reader at a checkout counter or terminal. The various shelves for the various items are provided with respective bar codes equipped with individual electronic displays linked to the computer so that these displays can display data at least in part coordinated with the bar code of the respective items stored there above so that the individual items need not be provided with price labels or the like.

Plain-language overview of the independent claim

Claim 1 is the only independent claim; claims 2–9 all depend from it (directly or via another dependent claim).

Claim 1 — plain language. This is drafted as a Jepson‑style improvement claim. It presumes a conventional supermarket product‑handling system that already has (a) a computer, (b) a checkout terminal linked to that computer, and (c) a bar code reader at the checkout terminal that identifies a purchased item. The improvement it claims is: a separate electronic display for each of several shelf locations holding items that carry bar codes, where each display is linked to the computer and shows indicia that are at least in part assigned to (i.e., correspond to) the bar codes of the items stored at that shelf section. In short: electronic shelf labels driven by the store computer and keyed to the products' bar codes, so items need not be price‑stickered.

Literal-text note: claim 1 recites displays "at each of a plurality of shelving rotations." Read literally that phrase is nonsensical; in context it plainly denotes shelf sections/locations, and I am flagging it as it appears rather than silently correcting it.

Dependent claims (compact map)

• 2 (dep. 1) — Input means for the computer, including a bar code reader, so scanning an item sets the display.
• 3 (dep. 1) — Patron checking terminal with its own bar code reader and a further electronic display that shows the same indicia as the shelf display when a shopper scans the item.
• 4 (dep. 2) — Same patron checking terminal feature as claim 3, but carried on the "input‑means" branch.
• 5 (dep. 1) — Wireless information‑transmitting link between display and computer.
• 6 (dep. 1) — Cable as the information‑transmitting link.
• 7 (dep. 5) — Patron checking terminal, layered onto the wireless embodiment.
• 8 (dep. 1) — Optical fiber information‑transmitting tube between displays and computer.
• 9 (dep. 6) — Patron checking terminal, layered onto the cabled embodiment.

The drafting strategy is clear: claim 1 captures the core electronic‑shelf‑label concept, claims 2–3 add the two distinguishing subsystems (bar‑code input; shopper self‑verification kiosk), and claims 5/6/8 pre‑empt the three transmission media (wireless, cable, optical fiber) described in the specification with reference to FIGS. 1–3.

Historical significance

US 4,521,677 is an early (1983 priority) articulation of what the industry now calls the Electronic Shelf Label (ESL) system combined with price verification. It sits in the "Similar Documents" lineage alongside US 4,139,149 (Display system), US 4,084,742 (Price marking system for automated check‑out) and US 4,833,308 (Checkout counter product promotion system), and it is cited as an [X] or [A] reference in numerous later EPO search reports — e.g. EP 0 299 355 A3, EP 0 348 579 A1, EP 0 961 253 A3, and the PCT search report for WO 92/09061 (Clares‑Regisbrook, claims 11–12) — which is the clearest objective evidence of its prior‑art weight in the EPL space.

If you'd like, I can next build a claim‑chart skeleton for claim 1 or trace the forward‑citation lineage into NCR/Kayser Ventures/Eldat artwork.

I'll research litigation involving US patent 4521677 across patent litigation tracking sites.

Let me search litigation-specific databases.

Bottom line

I found no evidence of any litigation involving U.S. Patent No. 4,521,677. No case caption, docket number, or judicial outcome naming this patent surfaced in any of the sources I could reach. I am reporting this as a negative finding, not as a positive one — see the caveats below, which matter a great deal for a patent of this vintage.

---

What I searched

Per your strict rule, I did not auto-correct or substitute identifiers: JP 4521677 (automotive weather strip), EP 4521677 (WLAN retransmission), Argentine Act 4521677 (JUJO CERÁMICA), and Brazilian record 4521677 (MARIA DUDU BEBÊ) are all distinct instruments that merely share the digit string, and none is the U.S. patent at issue. Likewise I did not treat the "Cited By" citations as litigation.

---

Patent identity (for the record)

• Patent: US 4,521,677 — "Product control system for supermarkets and the like"
• Inventor: Herbert S. Sarwin (Short Hills, NJ)
• Assignee: Individual (no corporate assignee of record)
• Filed: December 2, 1983 (App. No. 06/557,626)
• Granted: June 4, 1985
• Status: Expired – Lifetime; anticipated expiration December 2, 2003
• Claim structure: 9 claims; claim 1 is the only independent claim (a shelf-mounted electronic display at each shelving location, linked to the store computer, displaying indicia assigned to the bar codes of items stored there); claims 2–9 are dependents adding an input bar-code reader, a patron checking terminal with bar-code reader and display, wireless link, cable link, and fiber-optic link.
• Prosecution oddity worth flagging: the sole "cited by examiner / patent citation" on the record is US 4,180,264 (Acro, Inc., "Racket handle and method of making same") — plainly facially unrelated to supermarket product control. I am reporting it literally as it appears; it appears to be a data artifact rather than meaningful prior art.
• Legal events of note: a 1992 delayed-maintenance-fee petition (granted), a lapse for failure to pay the maintenance fee effective June 6, 1993, and acceptance of delayed payment October 26, 1993.

---

Why the negative finding should be treated cautiously

1. The enforceability window was 1985–2003. Any infringement action on this patent had to be filed on or before roughly December 2, 2003 (plus the six-year § 286 damages look-back for older conduct). A suit filed in, say, 1994 would sit almost entirely outside the digitally indexed dockets that Unified Patents, Docket Alarm, and RECAP cover.
2. Pre-2000 district court coverage is materially incomplete. PACER's earliest electronic dockets for most district courts begin in the 1990s, and coverage varies by court; paper-docket-era cases (1985–1990) are frequently absent from every free aggregator. A null result from a web search therefore is not a clean bill of health for the 1980s and early 1990s.
3. CAFC review would be indexed, but only for decided appeals. A 1980s–1990s appeal decided by the Federal Circuit would normally produce a published or unpublished opinion. Nothing naming U.S. Patent 4,521,677 surfaced. That is weak negative evidence of appellate activity, but I cannot rule out a per curiam affirmance without opinion.
4. I could not reach some sources at all. The tool returned empty result sets for several targeted queries (including the Unified Patents caselist query). An empty result set is the absence of a search hit, not a verified statement that the source contains no record. I have not verified any source end-to-end.

---

What I will not do

I will not supply plaintiff names, defendant names, jurisdictions, case numbers, filing dates, or outcomes. I have no evidentiary basis for any of those fields, and inventing them — even plausibly, even for a well-known art area — would be fabrication. There is no case number to give you.

I will also flag the temptation explicitly: this patent sits at the root of the electronic shelf label art (its "Cited By" list runs to Viscom, H.E. Butt Grocery, Diablo Research, Electronic Retailing Systems International, Display Edge Technology, Eldat, NCR, Kayser Ventures, and Information Planning & Management Service, among others). That dense citation network reflects the ESL patent thicket of the late 1980s and 1990s — and it is exactly the environment in which one would expect litigation — but expectation is not evidence. I found no case, and I am not going to infer one from the citation graph.

---

Recommended verification path

If you need a defensible clearance statement, the negative result above needs to be hardened by direct database queries I cannot run:

1. LexisNexis / Westlaw — full-text search "4,521,677" or "4521677" across all federal cases and the USPTO litigation releases; these have coverage back to the 1940s that free tools lack.
2. Lex Machina and Docket Navigator — patent-specific dockets; query by patent number, which searches the complaint and case-assignment fields directly rather than a web index.
3. Docket Alarm / Justia Dockets / CourtListener RECAP — for post-1994 district court coverage.
4. CAFC oral argument and opinion archives (cafc.uscourts.gov) — manually, by patent number.
5. Manual PACER civil case search in the districts where the inventor and likely defendants were located (D.N.J., E.D.N.Y., S.D.N.Y.) for the 1985–2003 window.
6. USPTO Patent Assignment records — confirm there was never an assignment to a litigating entity; the "Individual" assignee of record makes a later corporate enforcement campaign less likely but does not exclude a bare license or an exclusive-licensee suit under 35 U.S.C. § 281.

Summary in one line: No litigation involving U.S. Patent 4,521,677 is documented in any source I was able to reach; I cannot certify that none exists, and the paucity of indexed 1985–2003 docket data means the negative finding is strongest for the post-1994 period and weakest for the mid-1980s.

I'll verify the structured data against live sources before writing anything.

Confirming the negative and checking for any litigation/appeal history.

One more check on any appeal or enforcement history.

Proceedings overview

Zero AIA trial proceedings on file. The structured "PTAB proceedings on file" block returns an empty list, and independent checks of the PTAB public petition/decision records and web sources surface no IPR, PGR, or CBM naming U.S. Patent No. 4,521,677 — so the breakdown is 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denials. This is not a "hardened patent" signal and it is not a "troll got away with it" signal either — it is a dead-patent signal: US4521677 was filed 1983-12-02, granted 1985-06-04, and reached anticipated expiration on 2003-12-02, roughly nine years before AIA trials became available on 2012-09-16. A defendant facing a demand letter that cites this patent has a threshold, dispositive defense that has nothing to do with validity: the patent term ran out over two decades ago.

No proceedings to chart

There is no IPR/PGR/CBM number to report. I will not manufacture one. For completeness, here is what the canonical record does establish, and why the absence of AIA activity has a mundane explanation rather than a strategic one:

The 1993 lapse-and-revival episode (petition filed 1992-12-07, granted 1992-12-08, 8-year fee paid with surcharge 1993-07-16, lapse effective 1993-06-06, delayed-payment acceptance 1993-10-26) is the only contentious-looking event in the file history. It is a maintenance-fee administration matter, not an adversarial challenge, and it did not shorten the term — the patent is recorded as "Expired - Lifetime."

Structural reasons no AIA trial could exist here (this is the part that matters most to a defendant):

• IPR under 35 U.S.C. § 311 requires that the patent be available for inter partes review, and the Board's practice with expired patents assumes a live controversy over pre-expiration damages (§ 315(b) also bars a petitioner more than one year after service of an infringement complaint). With the patent expired since 2003-12-02 and no litigation on record, there is neither a petitioner with standing nor a § 315(b)-compliant window.
• PGR under § 321 is unavailable: the '677 claims have an effective filing date in 1983, far before the 2013-03-16 statutory cutoff for PGR-eligible patents.
• CBM (the transitional program, closed to new petitions on 2020-09-16) required the petitioner to have been sued for or charged with infringement of the patent. Nobody could be charged after 2003-12-02.

Strategic summary

Claim status: not "canceled," not "sustained" — untested because untestable. Claims 1–9 were never adjudicated by the Board or a court on this record. There is no FWD to cite, no claim-level disposition, and no estoppel to analyze. Under the template requested, the correct entries are: canceled = none; sustained = none; untested = all of claims 1–9 — and untested here is a function of expiration, not of the claims surviving scrutiny.

Estoppel landscape: not applicable. Section 315(e)(2) estoppel attaches only to a petitioner that obtained an FWD in an IPR it filed. No petitioner exists, so no party is estopped, and there is no "still-available prior-art ground" question to answer. Practically, the prior-art landscape is moot: an expired patent cannot support a § 271 infringement claim for any act occurring after 2003-12-02, and damages for earlier acts are time-barred by 35 U.S.C. § 286 (six-year lookback) many times over.

Pattern signals. No serial petitioner, no PTAB appeals by the patent owner, no defensive aggregator (Unified Patents or similar) in the chain — the assignee of record is simply "Individual." What the record does show is that the '677 is a heavily cited reference rather than a litigated asset: Google Patents lists it as cited by 84 (per the "Cited By" table) / and follow-on citing patents including Texas Instruments' US4654514 (Watson et al.), US4780599, US4766295, US4888709, US5198644, and later ESL-era filings into the 2000s. That citation pattern is the signature of a seminal 1985 disclosure that the electronic-shelf-label industry was built on top of — i.e., its practical significance today is as prior art against other patents, not as an assertion vehicle.

Recommended next steps

• If you are a defendant and a demand cites US4521677, stop analyzing validity and check the term. The patent expired 2003-12-02 (confirmed on the face of the Google Patents record: "Anticipated expiration 2003-12-02," status "Expired - Lifetime"). Any demand asserting present infringement is facially meritless; preserve the letter and consider whether the sender's conduct warrants a § 285 fee motion, an unfair-competition claim, or a referral. There is no FWD to quote — the dispositive citation is the expiration event on the patent's own face, not a Board disposition.
• Verify independently before relying on this memo. Run the patent number through PTAB E2E (https://ptacts.uspto.gov/ptacts/) and the USPTO Open Data Portal trial-proceedings endpoint, and confirm the maintenance-fee/expiration chain in USPTO Patent Center (https://patentcenter.uspto.gov/). For any litigation or appeal, search CourtListener (https://www.courtlistener.com/) and the Federal Circuit docket; I found none for this patent, and I have no high-confidence record of any enforcement campaign by Sarwin.
• Do not treat the absence of IPRs as evidence the claims are strong. The task framing that "well-asserted patents eventually attract IPRs" applies only while the patent is assertable. Here the silence is explained by a 2003 term expiry, and by the nine-year gap between expiry and the 2012-09-16 launch of AIA trials. The correct defensive posture is not "the patent survived attack" — it is "there is nothing left to attack, and nothing left to infringe."

Confidence note: I am highly confident there are no PTAB proceedings and no CAFC appeal on US4521677 — the structured ODP block is empty and multiple targeted web searches returned only citation and family data, never a trial number. I am not asserting that no infringement suit was ever filed at any point in 1985–2003; my searches did not surface a docket and I did not find one in CourtListener, but pre-2003 litigation records are incompletely digitized, so treat that specific negative as unverified.

I'll verify the ownership record with live sources before writing.

Let me check the USPTO assignment record directly and cross-reference any litigation.

Inventors

Herbert S. Sarwin — sole named inventor, listed in an individual capacity. The printed patent front page carries no "Assignee" line and no corporate inventorship designation; the only party of record on the face of the document is Sarwin himself.

• Employer at time of filing: not determinable from the record. US4521677 is a pro se–style individual filing (application 06/557,626, filed 1983‑12‑02). No SEC filer, no corporation, and no employment-obligation assignment appears in any indexed source I could reach. Google Patents records both Original Assignee: Individual and Current Assignee: Individual — the same single natural person on both ends of the chain, 40 years apart.
• Unusual patterns: none of the classic tells apply here (this is a pre-corporate-inventor-assignment problem, not a departure problem). There is no assignment out of Sarwin to any entity at any point, so there is no "all inventors left within 12 months" pattern to detect. The one notable artifact is the citation oddity: Google Patents lists exactly one cited reference, US4180264A "Racket handle and method of making same" (Acro, Inc., 1979‑12‑25) — an unrelated sporting-goods patent that cannot be a substantive anticipation reference for a supermarket shelf-display system. This is almost certainly a bibliographic indexing error in the ODP/Google data (the same single citation is mirrored in both Google's "Patent Citations (1)" and "Citations (1)" tables). I flag it rather than treat it as a real prior-art citation. The specification's actual prior-art discussion is narrative and refers to commercial hardware, not patents: the IBM 3660/3651 supermarket systems, the IBM 3666 Checkout Scanner, and Hewlett‑Packard HDSP‑6504 / HDSP‑6508 16‑segment alphanumeric displays.

Original assignee

None — the patent issued to Herbert S. Sarwin as an individual. There is no operating company, no holding company, and no institutional assignee on the issued patent.

• Products embodying the claims: unclear / likely none. The specification describes a system-level integration (shelf-mounted electronic display + store computer + UPC bar-code reader + patron verification terminal) and names only third-party IBM and Hewlett‑Packard components as the enabling hardware. There is no evidence Sarwin or any entity commercialized a product practicing claims 1–9.
• Primary line of business: n/a (individual inventor).
• Current status: The patent term ran to its natural end. Google Patents records "Expired – Lifetime" with an anticipated expiration of 2003‑12‑02 (20 years from the 1983‑12‑02 filing). Note the maintenance history shows a near-death experience: an FP Lapsed due to failure to pay maintenance fee event with an effective date of 1993‑06‑06, followed by a DP Notification of acceptance of delayed payment of maintenance fee on 1993‑10‑26. The owner paid late, with surcharge, and kept the patent alive to full term.

Assignment timeline

No recorded assignments exist for US 4,521,677. I found no assignment documents — no Assignment, Security Agreement, Merger, Change of Name, License, Release, or Correction conveyance — naming this patent, in either direction, at any time between the 1983 filing and today.

The evidence for this is consistent across sources:

Caveat on evidence quality: I could not load the Assignment Center's interactive result grid directly; my conclusion rests on the corroborating Google Patents / ODP-derived record plus the absence of any assignment reference in indexed search results. If you need a forensic-grade confirmation, run the query yourself at the Assignment Center with 4521677 as the patent number and also with 557626 as the application number. Per the reporting rules for this exercise, an empty assignment record is itself the finding: the original inventor-owner appears to have held the patent for its entire enforceable life, with no NPE, aggregator, or operating-company transfer at any point.

Note on report scope: the output specification directs the analyst to stop after this section when the Assignment Center shows no records. The remaining sections are included below because the ownership conclusion is the load-bearing part of the verdict, but they are necessarily driven by the absence of an assignment chain rather than by reel/frame citations.

Timeline diagram

timeline
    title Ownership of US 4521677
    1983 : Filed by sole inventor Herbert S Sarwin
    1985 : Patent issued to Sarwin individually
    1988 : Fourth year maintenance fee paid
    1992 : Late fee petition filed and granted
    1993 : Patent lapses then revived on delayed payment
    2003 : Patent reaches full term and expires

NPE / troll-pattern signals

Every signal below is assessed against the same underlying fact: there is no assignment chain to analyze. No reel/frame exists to cite, which is why most calls are "not present" rather than "unclear."

The one substantive observation worth carrying forward is not an NPE signal at all: US4521677 is the foundational prior art for the electronic shelf label industry. Its 84 (Google) / 114 (ODP) forward citations include H.E. Butt Grocery's US4766295, Viscom's US4888709, Diablo Research's US5198644, Electronic Retailing Systems' US5467474 / US5583487, and Kayser Ventures' US5537126 — i.e., nearly every early ESL patent had to design around Sarwin. That is why this patent shows up as an "X" reference in European oppositions and prosecution files rather than in litigation dockets. It is a citation workhorse, not an assertion vehicle.

Verdict

Insufficient data — more precisely, no assignment chain exists to evaluate. The patent issued in 1983 to a single individual, Herbert S. Sarwin, with Original Assignee: Individual and Current Assignee: Individual on the USPTO-derived record, and its legal-events history contains only fee events (FPAY 1988‑11‑21; FEPP petition 1992‑12‑07/08; FPAY + SULP 1993‑07‑16; FP lapse effective 1993‑06‑06; DP delayed-payment acceptance 1993‑10‑26) plus the 2003‑12‑02 anticipated expiration. There is no recorded assignment, no shell LLC, no asserter, no correspondent of record, and no litigation naming the patent as asserted. This is the textbook case where an empty assignment record is the finding: the original individual owner retained the patent to the end of its term.

For verification, run the patent number and the application number at the USPTO Assignment Center:
https://assignmentcenter.uspto.gov/ (search 4521677 and separately 557626) and cross-check https://assignment.uspto.gov/patent/index.html with the same two identifiers.

I'll search for the specific patent and its cited references.

Let me verify the cited reference list and check the analogous-field art in the same classification.

Prior-Art Analysis — US 4,521,677

1. The document itself (literal identification)

Source: https://patents.google.com/patent/US4521677/en and https://uspto.report/patent/grant/4521677

Caution on identifiers (per the strict "no auto-correction" rule): a search for the bare string "4521677" also returns a J-GLOBAL record for 自動車用ウェザーストリップ ("automotive weatherstrip"), where 4521677 is a Japanese publication number, not US 4,521,677. That record is not the patent you asked about and I have excluded it from the analysis.

---

2. The cited prior art — there is exactly one reference on the face of the patent

Both the Google Patents front page ("Patent Citations (1)") and Justia ("Referenced Cited") list a single patent document, marked * cited by examiner:

Sources: https://patents.google.com/patent/US4180264 ; PDF at https://patentimages.storage.googleapis.com/e8/ab/6a/3e676398ad6789/US4180264.pdf ; https://patents.justia.com/patent/4521677

§ 102 analysis of US 4,180,264

Timing qualifies; subject matter does not.

• § 102(a)/(b) timing: US 4,180,264 issued 1979-12-25, i.e. more than one year before the 1983-12-02 filing date of US 4,521,677. As a date, it is a § 102(b) statutory bar and would also be § 102(a) art.
• § 102 substance: it is non-analogous art in every respect. Its field of endeavour is molded plastic game-racket handles (Google's landscape tags it "Golf Clubs / Physical Education & Sports Medicine"); it contains no computer, no terminal, no reader of any kind, no record carrier, no display of data, and no vending or inventory function. Its only "reading" step is a human reading a mold drawing.

Anticipation conclusion under 35 U.S.C. § 102:

Assessment: US 4,180,264 anticipates none of claims 1–9 and is not even pertinent as § 103 background art (no reasonable fact-finder could combine a foam-molded tennis racket handle with a supermarket scanning system; there is no shared problem, field, or element). Its appearance on the front page is best characterised as an erroneous or clerical citation — the kind of "noise" citation that occurs when a front-page reference is mis-associated. I flag this because a naive citation-chain analysis would wrongly report a "prior art" for this patent. If you need the provenance (whether the examiner actually relied on it, or whether it is a front-page data error), the authoritative check is the PTO-892 "Notice of References Cited" and the examiner's search notes in the file wrapper via USPTO PatentCenter — I could not retrieve the IFW record in this session, so I am reporting the citation as listed by Google Patents/Justia/uspto.report rather than as confirmed by the examiner's own form.

Non-patent literature "citations": the specification refers to three IBM publications — IBM 3660 Supermarket Systems: Scanning System Introduction, the Key-Entry System Programmer's Guide, and the IBM Checkout Scanning Primer — and names commercial hardware (IBM 3666 Checkout Scanner, IBM models 3660/3651, Hewlett-Packard HDSP-6504/HDSP-6508 16-segment alphanumeric displays). These are cited in the body as enabling/alexandria references, not listed as prior art of record on the front page.

---

3. The genuinely most relevant prior art (uncited, but anticipatory in substance)

Because the only of-record citation is irrelevant, the real prior art must be searched in the field. Two pre-filing US patents in the same art (electronic retail price/shelf display; POS scanning) dominate:

(a) US 4,139,149 A — "Display system" (Crepeau et al.), granted 1979-02-13

https://patents.google.com/patent/US4139149 ; https://uspto.report/patent/grant/[4139149](/patent/4139149)

Brief description: a retail store system with a computer (10) having CPU 16 and a storage device 20 storing the coded prices of each item, plural point-of-sale terminals (12) each with a product code reader (14) ("an optical scanner or any other conventional device for reading product codes"), and plural electronic display units (24), each "associated with a type of item of merchandise available in the store and … located adjacent to such item so that the data displayed … can be seen by the customer when purchasing the item." The display units are wired in a series data-transmission line (22) to the CPU and show the price, and optionally "the name of the item, its size, or an identifying code number" — and the display may blink to promote a new or specially priced item. Critically, "the data used to charge the customer for items purchased at the POS terminal is the same data used and displayed by the display units."

§ 102(b) — this is the single most dangerous reference against claim 1. Element-by-element against claim 1's improvement clause:

• "a computer" → CPU 16 + memory 20 ✔
• "a checkout terminal linked to said computer" → POS terminals 12 on line 18 ✔
• "a bar code reader at said checkout terminal for identifying a purchased item" → product code reader 14 (optical scanner) ✔ (claim 1 says "bar code reader"; '149 says "product code," but an optical scanner reading a product code is the same structure — this is the main textual argument available to the patentee)
• "a respective electronic display at each of a plurality of shelving locations for respective items … each of said displays being linked to said computer" → display units 24 adjacent each item of merchandise, connected to CPU 16 ✔
• "for displaying indicia at least in part assigned to the bar codes of the items stored at the respective shelf sections" → display shows the price/name/code for its associated item, from the same master data ✔ (arguable: '149 associates display↔item by physical position in a series chain, not by assignment to a bar code; that is the only narrow point of distinction)

Conclusion: claim 1 is anticipated in substance by US 4,139,149 (or at minimum rendered obvious), notwithstanding that '149 was never cited against it. Claim 6 ("a cable forming an information transmitting link") reads directly on '149's series data-transmission line 22. Claims 2, 3, 4, 5, 7, 8 and 9 are not anticipated by '149: it has no bar-code input terminal for setting the displays (claim 2), no patron checking terminal/re-read display (3, 4, 7, 9), no wireless link — it affirmatively teaches hardwired series connection (5), and no optical fiber (8).

(b) US 4,084,742 A — "Price marking system for automated check-out of merchandize" (listed on Justia as Silverman), granted 1978-04-18

https://patents.google.com/patent/US4084742 ; https://uspto.report/patent/grant/[4084742](/patent/4084742)

Brief description: an automated checkout in which items carry machine-readable codes ("the so called bar code is now the most common type of code used for identifying processed foods"), read at a reading station by a laser scanner (source 7, oscillating mirror 9, reader 11); a computer 28 with a programmable memory 29 storing prices; the price is driven to the register printer, to a customer-visible visual display 21, and to a label printer/dispenser that re-marks the item at checkout.

§ 102(b): anticipate? No — it is checkout-side only; it has no shelf- or product-location display, no display-setting input terminal, and no patron checking terminal. Its "display" is the cashier's single customer display, not a per-shelf-section display. It is nonetheless the best background art for the preamble of claim 1 ("computer, a checkout terminal linked to said computer, a bar code reader at said checkout terminal for identifying a purchased item"), and it is the closest art on the no-individual-price-label motivation. Against claims 1–9 it is § 103 material at most.

(c) Other same-field pre-filing art worth pulling if you need a broader § 102/§ 103 picture

• US 4,071,740 (Gogulski), 1978-01-31 — cited in a later international search report against automated-checkout subject matter; POS/scanner environment.
• US 3,836,755 (Ehrat), 1974-09-17 — automated checkout, code reading/weighing environment.
• US 4,839,etc. and all other entries on the "Similar Documents" list (e.g. US 4,833,308 A; US 4,940,116 A) post-date the 1983-12-02 filing and are therefore not prior art against '677.

---

4. Documents that cite '677 (reverse citations — confirm its later importance)

'677 is heavily reverse-cited (Google shows 84 "Cited By" in one table and 114 in the other; the union includes the canonical electronic-shelf-label line: US 4,766,295 (H.E. Butt Grocery), WO 88/06773, US 4,888,709 / US 4,962,466 (Viscom), US 5,198,644 (Diablo Research), US 5,537,126 and US 5,736,967 (Kayser Ventures), US 5,690,etc. (Seattle Silicon), US 5,929,etc., up to US 11,715,115 and US 10,699,279 (Information Planning & Management Service)).

Important § 102(e)/(g) point: the earliest priority date among all of these citing publications is 1984-06-30 (US 4,780,599, Baus) and the earliest filing among them (US 4,707,785; US 4,654,514) is July–September 1984 — every one is after 1983-12-02. Therefore none of the 84/114 citing documents qualifies as prior art against '677 under § 102(e), (g), or otherwise. They are evidence of the patent's technological significance, not of anticipation.

Two EP search reports rate '677 as category [X] (particularly relevant; novelty-destroying alone) against later price-display systems — direct evidence of how third parties assessed its scope:

• EP 0 299 355 A3, "Price display system for sales areas and/or windows" (Zellweger Telecommunications AG): "[X] US 4521677 A 19850604 – SARWIN HERBERT S [US]" — http://data.epo.org/gpi/EP0299355A3-PRICE-DISPLAY-SYSTEM-FOR-SALES-AREAS-AND-OR-WINDOWS.html
• EP 0 961 253 A3, "Product activity data collection system" (NCR International): "[X] US 4521677 A 19850604 – SARWIN HERBERT S [US]" — http://data.epo.org/gpi/EP0961253A3-Product-activity-data-collection-system.html

---

5. Bottom line

1. Of-record prior art: exactly one reference — US 4,180,264 A (1977-04-25 / 1979-12-25, Acro, Inc. / Koenig, "Racket handle and method of making same").
2. It anticipates none of claims 1–9 and is non-analogous art of no § 103 value; its presence on the front page appears to be an erroneous/clerk-level citation rather than a substantive rejection reference. Verify against the PTO-892 in the PatentCenter file wrapper if provenance matters.
3. The prior art that actually matters for US 4,521,677 is uncited: US 4,139,149 A (Crepeau et al., 1979-02-13) is a substantive § 102(b) reference against claim 1 (and reads on claim 6), with only the "assignment to bar codes" vs. "series-position addressing" distinction as a possible escape; US 4,084,742 A (1978-04-18) is § 103 background against the claim 1 preamble.
4. Claims 2, 3, 4, 5, 7, 8 and 9 (bar-code input terminal for setting displays; patron checking terminal with re-read display; wireless link; optical-fiber tube) were not anticipated by anything of record, cited or uncited, that I could locate — those are the claims that carried the patent's novelty.
5. Scope of search / honest limitations: I worked from Google Patents, Justia, uspto.report and the EPO GPI data as of this session; I did not retrieve the USPTO PatentCenter IFW (full PTO-892, examiner search notes, or any foreign/NPL citations not printed on the face). The front-page citation set may therefore be incomplete, and any § 102(e)/§ 102(g) analysis of unpublished co-pending 1983 applications is outside what these databases expose.

I'll ground this analysis in the actual prior-art record shown on the page plus the "Similar Documents" art, then verify the key references.

Now let me verify the other pre-1983 art and dating so I don't mix post-filing references into a §103 combination.

Obviousness Analysis — US 4,521,677 under 35 U.S.C. § 103

1. Scope and ground rules for this analysis

Effective filing/priority date: 2 December 1983. There is no earlier priority claim in the record, so the § 102/§ 103 prior-art window closes on that date. This matters enormously here, because most of the references the page displays under "Cited By" and "Similar Documents" are post-1983 and therefore legally unavailable for an obviousness combination against '677.

The page's "Prior Art" section contains exactly three things:

1. One "Patent Citation" of record — US 4,180,264 (Acro, Inc., Racket handle and method of making same);
2. The "Similar Documents" list (25 items);
3. The "Cited By" list (forward citations — irrelevant to '677's own validity).

Of the 25 "Similar Documents," only two predate the critical date:

I want to flag this explicitly, because the "Similar Documents" table is a similarity ranking, not a prior-art ranking, and a careless chart would build a § 103 combination out of references that did not exist in 1983. Anything drawn from that list after 2 Dec 1983 can be used only as background evidence of the field — never as an obviousness reference.

Supplementary pre‑1983 references I located off-page (metadata verified only — I did not read full disclosures, so treat these as Medium/Low confidence leads, not sworn prior art):

The sole cited reference is a nullity. US 4,180,264 (Acro) discloses molding a foam-filled rubber grip onto a game-racket handle. I read its abstract and description in full during this analysis: there is no computer, no display, no bar code, and no retail subject matter of any kind. It cannot support a § 103 combination with anything, and it provides zero teaching toward claims 1–9. As flagged in the earlier summary, this is a Google Patents data artifact. I will not build any rejection on it.

---

2. Claim construction preliminaries (they drive the § 103 result)

Two construction issues must be resolved before the obviousness analysis, because both are outcome-determinative:

(a) "shelving rotations" (claim 1). Claim 1 recites "a respective electronic display at each of a plurality of shelving rotations for respective items having bar codes." Read literally, a "shelving rotation" is not a thing — it suggests a rotating shelf assembly (a rotisserie or carousel display), which is not what the specification describes. The spec describes ordinary supermarket shelves ("each shelf 10 carrying goods displayed for marketing"). Two coherent constructions follow:

• Construction A (broad, consistent with the spec): "shelf sections/locations" — i.e., the physical address on the shelf where a given product is stocked. This is the construction a court would almost certainly adopt under Phillips v. AWH (claim read in light of the specification, absurd result avoided).
• Construction B (literal): genuinely rotating shelves. Under this construction the claim is not enabled/written-described by the 1983 spec (§ 112(a)), and would be narrower — but useless as commercial coverage.

This analysis proceeds under Construction A. If Construction B were adopted, claim 1's validity problem shifts from § 103 to § 112 — a different but no less lethal posture. Either way the claim is weak.

(b) "indicia at least in part assigned to the bar codes." The phrase "at least in part" is deliberately broad and low-threshold. It reads on any displayed datum that the computer derives from, or correlates with, the product's bar code — price, unit price, name, size, code number.

(c) Claim 1 is a Jepson-style improvement claim. The preamble (computer + checkout terminal linked to the computer + bar code reader at the checkout terminal) is admitted prior art by the applicant's own choice of format. The whole § 103 battleground is therefore narrowed to the improvement: electronic shelf displays, one per shelf location, linked to the store computer, showing data assigned to the products' bar codes.

Note also the applicant's own admissions in the specification, which are § 102(b)/§ 103 evidence independent of any patent reference:

• The spec cites IBM 3660 Supermarket Systems: Scanning System Introduction, the Key-Entry System Programmer's Guide, and the IBM Checkout Scanning Primer — printed publications describing the computerized scanner checkout environment.
• The spec identifies the prospective display as an off-the-shelf part: "16 SEGMENT SOLID STATE, ALPHANUMERIC DISPLAY, HDSP-6504, HDSP-6508 by Hewlett Packard."
• The spec names the IBM 3666 Checkout Scanner and the IBM 3660/3661 computers as the ordinary equipment of the day.

Where an applicant characterizes his own display hardware and scanner as catalogue items and his own system architecture as conventional, the § 103 inquiry collapses to whether locating that display at the shelf and linking it to that computer was an obvious step. It plainly was.

---

3. The primary reference: US 4,139,149 ("Display system")

This is the case-dispositive reference. It is on the page's own "Similar Documents" list, it issued 13 February 1979 from a 30 August 1977 priority date, and its current listed assignee is NCR. Its disclosure, quoted from the record (Google Patents; uspto.report):

Every element of claim 1 is disclosed. The price shown on the shelf display is the price the CPU fetched from storage ("a magnetic disk or drum for storing coded prices of each item") by product code read at the scanner — which is precisely "indicia at least in part assigned to the bar codes of the items."

Single-reference § 103 rejection of claim 1

A single reference can render a claim obvious where the reference teaches all elements and the claim adds nothing of patentable weight. Here the only conceivable distinction is semantic: '677 says the display shows indicia "assigned to the bar codes," while '4,139,149 says the display shows "the price of its associated item" derived from the same data used at checkout. That is a difference in wording, not in structure, function, or result. Under KSR Int'l v. Teleflex (2007), such "a difference of wording" does not create patentability; and under Graham v. John Deere, the claim's scope, the art's scope, and the level of ordinary skill (a supermarket POS systems engineer circa 1983, well versed in the then‑mature IBM 3660 environment) leave nothing non‑obvious. Claim 1 is obvious over US 4,139,149 alone. Confidence: High.

Two-reference fallback combination

If for any reason a tribunal accepts a distinction keyed specifically to bar-code data (as opposed to POS key-entry data) driving the shelf display, then:

US 4,139,149 + US 4,084,742. US 4,084,742 (Alpert-type price-marking system, 1978) discloses a checkout scanner reading an item code, a computer whose memory stores prices, the computer fetching the price by code, and manual keyboard entry to alter prices and handle uncoded items: "Reader 11 emits item identifying signals to the computer 28. The computer obtains from the memory device 29 the price of the item … Keyboard 24 … may also be used for communicating with the computer to obtain sub‑totals or totals, to alter the prices of individual items."

Motivation to combine: both references are in the identical field (supermarket automated checkout), address the identical problem (keeping customer-facing price information synchronized with the centrally stored price that the scanner will charge), and the combination yields no more than the predictable sum of their parts — the shelf display of '149 driven by the code-keyed price table of '742. KSR rationales satisfied: same field of endeavor; same problem; predictable result; known technique improving a similar device in the same way.

---

4. Claim-by-claim obviousness

Claim 2 — "input means for said computer including at least one bar code reader … for correspondingly setting said displays"

Obvious over US 4,139,149 in view of US 4,084,742 (and the applicant's admitted art).

• '149 already discloses code readers at POS terminals feeding the CPU, with the same data driving the shelf displays — functionally, a scan sets the displayed data.
• '742 discloses scanner input plus keyboard entry to alter stored prices, i.e., the bar-code-reader-as-input-means concept.
• The applicant's own spec admits the IBM 3666 scanner reading the UPC as conventional input to the store computer, and admits that the input terminal's keyboard is the means of entering price/stock data.

Motivation: replacing or supplementing manual keyboard price entry with a scan of the very package whose bar code defines the product eliminates transcription errors and keystrokes — the classical KSR "known technique to improve similar devices in the same way." Obvious. Confidence: High.

Claims 3 and 4 — patron checking terminal with its own bar code reader + display showing the same indicia

This is the one limitation with no pre‑1983 reference squarely on all fours in the material I could reach — and I want to be candid about that.

• '149 already discloses that the shelf display is customer-facing: the display units are "located adjacent to such item so that the data displayed … can be seen by the customer," and each unit may show "the name of the item, its size, or an identifying code number." Customer price verification immediately at the goods is therefore taught.
• '742 teaches the scanner-to-price-lookup pipeline, and that a visual display device at checkout "is situated where it can readily be viewed by the customer."

The § 103 argument: given (i) a customer-visible shelf display, (ii) a scanner-driven price lookup, and (iii) the well-known retail practice of a fixed "price check" station at which a clerk scans an item to verify its price, placing a scanner and a display at a self-service station so the shopper can perform the verification herself is an obvious deployment of known components for their known purposes, motivated by the concrete benefits '677 itself recites: assuring the shopper "that, upon checkout, the checkout terminal will similarly respond," and reducing checkout friction.

Caveat (explicit): US 4,833,308 (Checkout counter product promotion system, 1986/1987/1989) is not available as § 103 art against '677 — it post-dates the critical date by years. It is evidence only that the marketplace moved in this direction afterward, which is not the test. On the material actually in the page's prior-art section, claims 3, 4, 7, 9 are obvious but on a weaker, more inference-dependent record than claims 1, 2, 5, 6, 8. Confidence: Medium.

Claims 5, 6, 8 — the transmission medium (wireless / cable / optical fiber)

These are the paradigm case of a claim that adds nothing patentable.

• Claim 6 (cable) is met outright by US 4,139,149: "the CPU 16 is also connected by a data transmission line 22 to plural display units 24." Obvious over '149 alone. Confidence: High.
• Claim 5 (wireless link) — the applicant's own spec treats the two media as interchangeable: "The alphanumeric display can be hard wired to the computer or connected by a wireless transmission system therewith … A cable connection to the computer can also be used." Choosing among known, functionally equivalent signal-transmission media is precisely the "predictable variation" that KSR holds unpatentable, and the claim recites no structure, protocol, frequency, modulation, or range. Where the inventor himself presents the medium as an arbitrary alternative, a person of ordinary skill would substitute it without the exercise of inventive faculty (In re Boesch-type reasoning).
• Claim 8 (optical fiber "information transmitting tube") — fiber-optic data transmission was established telecom technology well before 1983, and the '677 spec uses fiber as a light guide, not as a data channel ("a fiber optic unit 212 which is coupled with the light source 216 by the optical fibers 215"). The claim covers both; neither involves anything beyond known equivalents.

Motivation: routing data to a distributed array of shelf-mounted displays through an existing cable run, an RF link, or a fiber run are the three known options a skilled installer would consider; each is selected on the ordinary engineering criteria of cost, retrofit difficulty, and store layout. Obvious. Confidence: High.

Claims 7 and 9

Pure layering of the patron-checking-terminal feature (Medium confidence, per above) onto the wireless (claim 7) and cabled (claim 9) embodiments. They rise or fall with claims 3/4 and 5/6. Overall: Obvious, overall confidence Medium — bounded by the claim 3 record.

---

5. Why a person of ordinary skill would have combined these (the motivation analysis)

Secondary considerations (considered and rejected). There is no evidence in the record of unexpected results, skepticism, licensing, or commercial success with a nexus to the claimed subject matter. Any commercial success in electronic shelf labels accrued to NCR, Kayser Ventures, Eldat, Electronic Retailing Systems International and others years after this patent's 1983 filing (visible on the page's own forward-citation list) and is attributable to wireless-ESL and RF-battery developments — not to the '677 claims. Long-felt need is arguable but undercut by the fact that '149 had already addressed the shelf-display problem in 1979.

---

6. Confidence ledger and explicit uncertainties

What I did not do and will not do: I did not manufacture a reference, an inventor, a case caption, or a claim chart element. Two search calls in this session returned a step-limit error and produced no results; I have not substituted guesses for those gaps. If a definitive obviousness opinion is needed, the following must still be read in full: US 4,500,880; US 4,525,713; and the pre‑1983 EPO/PCT search-report citations for this family (e.g., FR 1,598,252, Compagnie Générale d'Automatisme, 1970, cited in WO 90/16051). I also could not verify a single pre‑1983 reference disclosing a customer-operated scanning/price-check station — that is the one genuine soft spot in this analysis.

---

7. Bottom line

US 4,521,677's claims are very likely invalid under § 103, and claim 1 is arguably invalid under § 102 as well.

• Claim 1 is fully disclosed by US 4,139,149 (1979) — computer, POS terminal, product code reader, and a per-item shelf-located electronic display linked to that computer and showing data "at least in part assigned to" the product code. Obvious on that reference alone; if a bar-code-specific reading is insisted upon, obvious over '149 + US 4,084,742 (1978).
• Claim 6 is met verbatim by '149's "data transmission line 22."
• Claims 2, 5, 8 are obvious as automation of manual price entry and as selection among known, equivalent transmission media — and the applicant's own specification concedes every hardware element as catalogue equipment.
• Claims 3, 4, 7, 9 are obvious, but on a materially thinner record; the patron self-scan verification station is where an infringement defendant would want the fight, and where the prior art needs to be developed beyond the material on this page.
• The sole "Patent Citation" of record, US 4,180,264 (Acro racket handle), supports nothing. It is a data artifact and should not appear in any defensibility chart.
• Everything attractive on the page's "Similar Documents" and "Cited By" tables post-dates the invention and is legally unusable — the instinct to reach for US 4,833,308, US 4,766,295, US 4,888,709 or US 5,198,644 when charting '677 must be resisted; those are descendants of this art, not its parents.

Since the patent expired 2 December 2003, none of this affects a live right; it matters only to retrospective infringement/remedies analysis, to the patent's weight as prior art against others (which is substantial — see its role as an X/A reference in EP 0 299 355, EP 0 348 579, EP 0 961 253 and WO 92/09061), and to the accuracy of the ESL art history.