US 5247670

I'll search for the patent and any related litigation dockets.

US 5,247,670 — "Network server"

No litigation found. My searches returned no CAFC 2026 dockets, district court suits, PTAB proceedings, or ITC actions involving 5,247,670. This is unsurprising: the patent reached its anticipated expiration on 2010-09-21 and its legal status is "Expired – Lifetime," so it is not a candidate for 2026 appellate activity. The only "5247670" hit in a litigation-adjacent database (CourtListener) was an unrelated financial-disclosure record ID, not a patent case. I cannot rule out a never-digitized historical suit, but there is no public record I can find.

Bibliographic data

Caveat on assignee: Fuji Xerox Co., Ltd. is the original assignee of record. The "Fujifilm Business Innovation Corp" entry reflects the 2021 corporate renaming of Fuji Xerox, not a separate arms-length transfer; Google expressly disclaims having performed an ownership analysis.

Abstract (as issued)

"A server in a network for performing services according to requests for service from clients on the network. The server includes a memory area for storing user commands, a memory area for storing a file program, an information exchange area, a temporary storage area, a service program area, and a control program area."

Plain-language overview of the independent claims

Claim 1 — A network server built by combining a conventional file server with an application-processing engine. Three required parts:

1. a file server that sends/receives and stores files on the network and analyzes the user's request;
2. service-program storage connected to that file server, holding multiple predefined service programs whose identifying data live in the file server; and
3. processing means that runs a selected service program against a selected stored file.

In essence: the server presents itself to the network as an ordinary file server, but a user's file/request is interpreted as a job trigger, and the appropriate application program is pulled from storage and executed on it.

Claim 6 — The same concept framed functionally without the word "file server": a server comprising (a) means to store request files received from the network, (b) means to store multiple predefined service programs available through the network, (c) means to process the request files by reading out and executing the requested service program, and (d) means to store the results of each processed request file returned by the processing means. This is the "receive request file → execute named program → store result" loop with no intermediate mail server.

Dependent claims at a glance

• Claim 2 — the processing means split into a service processing unit plus result-storage means.
• Claim 3 — adds a temporary information storage area buffering data between file server and service unit.
• Claim 4 — adds control means scheduling order/timing of services.
• Claim 5 — same scheduler, as applied to the claim-3 configuration.
• Claim 7 — scheduler added to the claim-6 server.

Technical gist

The patent's stated problem is that adding new services via a shared mail server (FIG. 6) or shared file server (FIG. 7) couples every node to the network and lets one faulty service program corrupt other users' mailboxes or files. The claimed solution makes each service server self-contained and "passive" — it owns its own filing function and storage, registers by name with a network control server, and exposes itself to clients through the ordinary file-server interface (a "file drawer" icon), so services can be added or removed without touching clients. The specification also describes a temporary storage area (34) between the file-server function and the service programs for re-scheduling and load buffering, a lower-layer multi-node variant (FIG. 4), and a "mutual mail box" mail-service variant.

Confidence / uncertainty

• Bibliographic data and claim text are from the patent document itself — high confidence.
• No CAFC 2026 or other litigation identified — moderate-to-high confidence, based on absence in searched sources; this is a negative finding from available public search results only.
• The current-assignee field is a Google Patents assumption and is likely a renaming artifact; treat it as unverified.

I'll search litigation databases for US patent 5247670.

That first result concerns a different patent (8,526,767), not 5,247,670. Let me search more specifically, including litigation databases.

Let me check additional litigation-specific sources and confirm the patent's expiration window.

Summary

No litigation involving U.S. Patent No. 5,247,670 was found.

I searched litigation-focused sources (Unified Patents litigation portal, RPX/insight docket documents, CourtListener, and general web indices of federal court complaints). I found zero cases — no district court complaints, no declaratory judgment actions, no ITC investigations, and no PTAB proceedings (IPR/PGR/CBM) — in which U.S. Patent No. 5,247,670 is identified as an asserted patent or patent-in-suit.

⚠️ Important false-positive note (do not conflate)

One search returned a highly active litigation record, but it concerns a different patent:

• The record involved U.S. Patent No. 8,526,767 ("the '767 patent"), asserted by Solas OLED against Samsung, litigated in the Eastern District of Texas with a parallel Samsung IPR petition (PTAB).
• That is 8,526,767, not 5,247,670. These are different patents in different technology areas and different decades. Per the instruction not to auto-correct or conflate identifiers, I am explicitly excluding that matter from the list below.

Patent identification (for the record)

Source: https://patents.google.com/patent/US5247670/en

Litigation list

No cases found. I am stating this explicitly rather than inferring from absence of a search hit alone.

Why the absence is plausible

1. The patent expired on 2010-09-21 (17-year term from the 1993 grant; the "Expired – Lifetime" status and anticipated-expiration date are recorded on the Google Patents page). Any infringement suit would therefore have had to be filed by roughly 2016 at the latest to reach pre-expiration damages (subject to the 35 U.S.C. § 286 six-year recovery bar), and most realistically before expiration.
2. Ownership sits with an operating company, not a monetization entity. The patent was assigned to Fuji Xerox Co., Ltd., and the current listed assignee is Fujifilm Business Innovation Corp — a large operating manufacturer. The Xerox–Fuji Xerox patent agreements (e.g., Sections B20/B21 enforcement rights, published on xerox.gcs-web.com) allocate enforcement rights between the parties; there is no public indication this 1988-priority server patent was ever asserted.
3. The 1988/1989 priority and 1993 grant predate the NPE/patent-assertion wave for this kind of software/network-server subject matter, and the claims (file-server-plus-service-program architecture) map to a limited set of products from that era.

Caveats and limitations

• My searches returned general results and no docket-level record naming 5,247,670 as a patent-in-suit. Some litigation databases (e.g., Unified Patents, Docket Navigator) require authenticated access; I could not query restricted dockets directly, so this is a negative finding based on available public search results, not a certification.
• CourtListener free-record coverage for older district court cases (1989–2010) is incomplete, so a very old suit could theoretically exist outside indexed sources.
• No IPR/PGR/CBM proceedings were found either (consistent with PTAB's post-2012 AIA window and the patent's pre-expiration timelines).

Bottom line: Based on the sources searched, there is no known litigation involving US 5,247,670. If you require a certified clearance, I recommend a paid Docket Navigator or Lex Machina query on 5,247,670 limited to U.S. district courts and the ITC.

I'll verify the structured data with targeted searches before drafting.

Let me run a few more targeted searches to confirm the absence.

PTAB Proceedings — US 5,247,670 ("Network server")

Proceedings overview

Zero AIA trial proceedings. The USPTO Open Data Portal structured block returns no AIA trial proceedings for US 5,247,670, and my independent searches (Google Patents/IPR cross-references, PTAB petition full-text, PTAB E2E docket strings, and Federal Circuit/CourtListener queries) surfaced none — no IPR, no PGR, no CBM, no derivation, no inter partes reexamination converted to trial. The defensive posture is therefore not "this patent is hardened by surviving IPRs"; it is the opposite: this patent is expired and has never been tested at the Board. For a defendant, that means the danger comes from the assertion itself, not from any PTAB record — and the assertion is very likely defective on its face (see "Defensive value" and Recommended next steps).

Why the absence is structural, not luck:

There are no proceeding numbers to report, and I will not invent any.

⚠️ False-positive warning: Searching "524" + IPR surfaces unrelated patents — US 8,412,524 (IPR2019-00496, word-correction UI) and US 7,058,524 (electrical power metering, Smart Meter Technologies v. Itron). Neither is US 5,247,670. Do not let an opponent's expert conflate these. Only the 5,247,670 number set (5,247,670 / JPH0776939B2 / GB2217067B / DE3908459C2) is this family.

Strategic summary

Claim status — all claim survived, none cancelled, all now unenforceable by expiry. US 5,247,670 has 7 issued claims. Claim 1 is the principal independent claim (a file server for sending/receiving files on the network that also stores service-program identifying data, plus processing means for executing a selected service program on a selected stored file). Claim 6 is the second independent claim (request-file storage + service-program storage + processing means + result storage). Claims 2–5 depend from claim 1; claim 7 depends from claim 6. No claim of this patent has ever been cancelled, narrowed, disclaimed, or held unpatentable — there is no FWD, no certificate of cancellation, and no reexamination history in the record. All 7 claims are formally "untested" at the PTAB, but that is an academic distinction: the patent's last day of enforceable term was 1993-09-21 + 17 years = 2010-09-21 (pre-URAA term, 17 years from grant; 20-years-from-filing would have been earlier, 2009-03-09, so the longer statutory term controls). Google Patents lists "Anticipated expiration 2010-09-21" and status "Expired – Lifetime," consistent with the maintenance-fee record (fees paid 1997-03-11, 2001-03-01, 2005-02-23 — i.e., through the 11.5-year window).

Estoppel landscape — § 315(e)(2) is irrelevant here. Because no IPR/PGR was ever instituted, no petitioner is estopped from anything. There is no § 315(e) bar, no § 325(e) bar, and no IPR-imposed narrowing. Conversely, a defendant gains nothing from someone else's estoppel and has no free ride from a prior FWD. If you need to invalidate the claims (e.g., to defeat a live threat or defeat a claim-construction-driven theory), the full universe of prior art under §§ 102/103 remains available in district court or a DJ action, unconstrained by IPR grounds — including the 15 references cited on the face of the patent (e.g., US 4,387,427 Intel; US 4,500,960 AT&T Bell; US 4,829,354 AT&T Bell; US 4,887,204 IBM; US 4,949,248 Caro) and the 11 "families citing" documents (e.g., US 5,151,895 Digital Equipment terminal server architecture; US 5,175,684 Trans-Link machine translation).

Pattern signals — none. No serial petitioner (there has never been a petitioner). No PTAB appeal history, therefore no Rule 36 or Kangxi-style CAFC mandate to look up. No defensive aggregator (Unified Patents, RPX, AST) is in the chain — none appears in any docket I could locate, which is expected given the patent's 2010 expiry. Chain of title: Yoshifumi Matsunaga → Fuji Xerox Co., Ltd. (assignment executed 1989-03-02, recorded 1989-03-09, REEL/FRAME 005053/0246); the current Google-listed assignee, Fujifilm Business Innovation Corp., is simply the 2021 corporate rename of Fuji Xerox. There is no portfolio-transfer or troll-acquisition signal in the structured record.

Recommended next steps

• Do not treat this as a PTAB question — treat it as a "is my demand letter even real" question. The patent expired 2010-09-21. Under § 286, the damages lookback is six years from suit; even if suit were filed today, the recoverable window closed on 2010-09-21, and any demand premised on post-2010 conduct has no statutory basis. Combined with § 271's requirement of an unexpired patent, an assertion of 5,247,670 against current products is groundless on its face.
• Verify the patent number in the demand letter character-for-character. Given the similarity of "5,247,670" to live, actively-litigated "524" patents — notably US 8,412,524 (IPR2019-00496, Apple v. Vlingo-line word-correction art) and US 7,058,524 (Smart Meter Technologies' electrical power metering patent, IPR2019-00682-line, Itron as petitioner) — a plausible explanation for a letter citing "the '524 patent" is a citation error by the sender. Demand letters are not uncommonly sent on mis-identified patents; confirm the number before spending a dollar on invalidity work.
• If a genuine assertion exists, the cleanest defenses are:
  1. Expiry / no enforceable right — patent term ended 2010-09-21 (17 years from the 1993-09-21 grant, pre-URAA). Cite the Google Patents legal-status entry ("Anticipated expiration 2010-09-21") and the maintenance-fee record.
  2. Damages cutoff — § 286 bars recovery for infringement more than six years before filing.
  3. No PTAB short-circuit available — because the patent is expired and pre-AIA, you cannot obtain a PGR or CBM, and IPR is a poor use of budget (see Sony). If validity must be adjudicated, it will be in district court or via a DJ action, where §§ 102/103/112 are all on the table.
• Where to verify the "no proceedings" finding yourself (the record is thin precisely because nothing happened):
  • PTAB E2E / PTAB Center: https://ptacts.uspto.gov/ptacts/
  • USPTO PTAB decisions search and API: https://www.uspto.gov/patents/ptab and https://developer.uspto.gov/ptab-api
  • Patent bibliographic and legal-status record: https://patents.google.com/patent/US5247670/en
  • CourtListener full-text (for any CAFC appeal, which would be the only PTAB-related record if one existed): https://www.courtlistener.com/?q=5247670
  • FreeLaw/Justia docket search for the patent number across all district courts, to confirm there is no parallel litigation history.

One honest caveat: I found no PTAB proceeding, but I also found no affirmative statement anywhere reading "no IPRs were filed against 5,247,670." My conclusion rests on (a) the ODP-structured list in the prompt returning empty, (b) zero hits in three targeted searches over PTAB petition full text and PTAB E2E URL patterns, and (c) the strong structural inference from the 2010-09-21 expiry versus the 2012-09-16 start of AIA trials. If you need a certifiable clearance statement for a litigation hold or an opinion letter, confirm directly against the PTAB E2E docket and the patent's file wrapper via PatentCenter rather than relying on search-engine absence.

I'll research the assignment record for US 5,247,670 and cross-reference the assignee history.

Let me verify the assignment record and the corporate name-change history.

Ownership Reconstruction — US 5,247,670 ("Network server")

Scope note / source limitation. My reel/frame and conveyance data below are drawn from the patent's own recorded legal-events table (the Google Patents mirror of the USPTO assignment record) plus corroborating web sources. I was unable in this session to independently open the Assignment Center detail page for this patent, so the correspondent-of-record field (the attorney/agent who filed each recording) could not be read and I will not guess at it. Treat the correspondent as a data gap to verify at the link in the Verdict, not as absent-by-finding. Everything else below is grounded in the record.

Inventors

Yoshifumi Matsunaga — sole named inventor on the face of US 5,247,670.

• Employer at filing: Fuji Xerox Co., Ltd. This is determinable two ways: (1) the sole recorded assignment (executed 1989-03-02, reel 005053/0246) conveys Matsunaga's entire interest to Fuji Xerox Co., Ltd., the classic inventor-to-employer instrument; and (2) Matsunaga reappears years later listed as "Matsunaga, Yoshifumi, c/o Fuji Xerox Co., Ltd." on a later Fuji Xerox-originated European filing (EP 0 740 258 A3, priority 1995-04-28), confirming sustained employment rather than a one-off engagement.
• No departure pattern. The 12-months-post-filing "inventor exodus" tell is not present: Matsunaga is still filing under the Fuji Xerox banner six years after this patent's 1988 priority date. There is no evidence of the inventor leaving to form a competitor or a shell licensing vehicle.

Original assignee

Fuji Xerox Co., Ltd. (Minato-ku, Tokyo, JP), assigned on the issued patent and on the underlying application US 07/321,134.

• Business line: a Tokyo-based developer/manufacturer of document and printing systems — multifunction peripherals (DocuCentre / ApeosPort lines), copiers, production printers, and document-management software (DocuWorks, launched 1996). Founded 1962 as a joint venture between Fuji Photo Film (later FUJIFILM) and Rank Xerox/Xerox.
• Product embodying the claims: yes. The specification itself names the commercial system the invention serves — the "8000 INS" (XINS) network, "sold by Fuji Xerox Corp." (see Background, reproduced in the patent text). This is an operating company shipping a networked server product, not a paper holder.
• Current status: operating, under a new name. Xerox exited the JV (its 25% stake sale was announced November 2019); Fuji Xerox became a wholly FUJIFILM-owned company and renamed itself FUJIFILM Business Innovation Corp. effective 2021-04-01. It remains a ~$9B, ~40,000-employee operating subsidiary of FUJIFILM Holdings. Not acquired, not dissolved, not in bankruptcy.
• Caveat (consistent with the prior section): Google Patents' "Fujifilm Business Innovation Corp" current-assignee entry is a renaming artifact, not an arms-length transfer — there is no recorded assignment evidencing a transfer to that name (see timeline).

Assignment timeline

The complete recorded chain is a single assignment. Chronologically:

• 1989-03-02 (executed) / recorded 1989-03-09 — Reel 005053 / Frame 0246
  • Conveyance: Assignment ("ASSIGNMENT OF ASSIGNORS INTEREST")
  • Assignor: MATSUNAGA, YOSHIFUMI
  • Assignee: FUJI XEROX CO., LTD. (Tokyo, JP)
  • Correspondent: not exposed in the sources I could access — verify at Assignment Center (see Verdict link). Because there is only one instrument, the "repeat correspondent across the chain" test cannot even be run here; a single appearance is not a finding in any event.
  • Context: initial inventor-to-employer assignment — standard corporate ownership capture, not an acquisition, fire-sale, reorg, or securitization.

Non-assignment legal events (for completeness — these are not conveyances):

This is itself a finding. Every maintenance fee was paid by the original assignee and the patent rode to its full statutory term without ever changing hands. There is no post-issuance assignment of any kind — no security interest, no merger recording, no change-of-name recording for the 2021 Fuji Xerox → FUJIFILM Business Innovation rename, and no transfer to any third party. The current owner is, and has always been, the original assigned operating company (now under its 2021 name).

Cross-reference check: no contradiction with the already-generated summary — that section likewise found the assignee field to be a renaming artifact and found no litigation. Both are reinforced here by the total absence of assignment activity.

Timeline diagram

timeline
    title Ownership of US 5247670
    1988 : Invented by Matsunaga at Fuji Xerox
         : JP priority filed 1988-03-16
    1989 : Assigned to Fuji Xerox
         : Reel 005053 frame 0246
         : US application filed 1989-03-09
    1993 : Patent issued
    1997 : Maintenance fee paid year 4
    2001 : Maintenance fee paid year 8
    2005 : Maintenance fee paid year 12
    2010 : Patent expires after full term
    2021 : Owner renamed Fujifilm Business Innovation

NPE / troll-pattern signals

Score: zero strong signals, zero weak signals evidenced.

Verdict

Insufficient data (only the original assignment).

This is the rubric's literal fit: the USPTO record for US 5,247,670 contains one instrument — the inventor-to-employer assignment executed 1989-03-02 and recorded 1989-03-09 at reel 005053/0246 — and nothing else. Being precise about what that affirmatively means: the absence of any post-issuance conveyance, together with three paid maintenance fees (1997, 2001, 2005) and full-term expiry on 2010-09-21, indicates continuous ownership by the original operating assignee (Fuji Xerox, now FUJIFILM Business Innovation) with no NPE involvement whatsoever. There is no shell entity, no asserter, no fire-sale, and no defensive aggregator — none of the eight troll-pattern signals is present on the evidence. I could not verify the correspondent-of-record field, which is the one open item and the only reason this is scored "insufficient" rather than an affirmative non-NPE call.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search property 5,247,670; the same records surface at https://assignment.uspto.gov/patent/index.html). Confirm the single entry, reel 005053/0246, and capture the correspondent name from the detail page to close the data gap.

I'll search for the patent and verify the cited prior-art references before analyzing them claim-by-claim.

Prior-Art Analysis — US 5,247,670 ("Network server")

Method / sources. I confirmed the patent record and its citation list against the patent document (authoritative, supplied) and cross-checked individual references against Google Patents, uspto.report, and Unified Patents mirrors via search. Note: I could not query the USPTO's PatFT/PatentCenter back-end directly through this tool; the bibliographic and citation data below therefore come from the patent document itself plus the mirror sources. No live search result contradicted the supplied patent text, so the patent text governs.

Important framing — the § 102 subsections that apply.
US 5,247,670 is a pre‑AIA patent: filed 1989‑03‑09, JP priority 1988‑03‑16. Therefore:

• § 102(b) one‑year bar date = 1988‑03‑09.
• § 102(a) is measured against the invention date (≈ the 1988‑03‑16 JP priority).
• § 102(e) applies to US patents granted on applications filed before the invention, i.e. effective filing before ≈ 1988‑03‑16.

This matters because several cited references published/issued after 1988‑03‑09 and are therefore only § 102(e) art (US patents) — and at least one cited reference is not § 102 art at all (see flags). I flag the subsection for each reference.

The single clearest prior-art ranking up front: the strongest § 102 candidate is US 4,780,821 (IBM); secondary relevance on the "file server" element comes from US 4,825,354 and US 4,887,204. No cited reference discloses, in a single document, all elements of claim 1 or claim 6 — specifically the combination of (i) a file server that stores files and (ii) an executable service‑program store whose programs are run by that server on a stored file with the result stored back. That missing "server executes the service program on the stored file" element is the recurring gap.

Cited references (the 15 examiner-cited U.S./foreign documents)

Per-reference detail

1. US 4,387,427 A — Intel — "Hardware scheduler/dispatcher for data processing system"

• Filed 1978‑12‑21; issued 1983‑06‑07. § 102(a)/(b) art.
• Description: Hardware/firmware scheduling and dispatching of tasks within a data-processing system; queues and selects work for execution.
• § 102 assessment: Discloses scheduling/dispatch of tasks. It does not disclose a network file server, service-program storage, or "identifying data stored in the file server." At most it is background for the scheduling means.
• Potential claims: Could be cited against claim 4/5/7 ("control means for scheduling an order and a time…"), but a single-processor hardware dispatcher is not the claimed network-service scheduler, so it does not anticipate those claims as a whole. No anticipation.

2. EP 0 077 008 A2 — Digital Equipment — "Interface apparatus in a data processing system"

• Filed 1981‑10‑05; published 1983‑04‑20. § 102(b) art.
• Generic interface apparatus; no network service model. No anticipation of any claim.

3. US 4,500,960 A — AT&T Bell Laboratories — "Geographically distributed multiprocessor time-shared communication processing system"

• Filed 1982‑06‑28; issued 1985‑02‑19. § 102(b) art.
• Distributed time-shared processing across geographically separated processors. Relates to distributed computing generally but not to a file-server that hosts and executes selectable service programs on stored files. No anticipation.

4. ⚠️ US 4,937,284 A — Neste Oy — "Modified polyolefin, method for making the same, and use thereof"

• Filed 1983‑02‑17; issued 1990‑06‑26.
• This is a polymer-chemistry patent with no relationship to computer networks. Listed in the patent's citation table, it cannot disclose any limitation of claims 1–7 and cannot anticipate anything.
• Two anomalies, stated literally per the no-auto-correct rule: (a) its subject matter is wholly unrelated to the '670 invention; (b) its issue date (1990‑06‑26) is after the '670 filing, so it could only ever be § 102(e) art (by its 1983 filing) — and even then it is irrelevant on the merits.
• Flagged as a likely citation-mapping error in the source record (e.g., a real network-server reference may have been mis-keyed to this number). I am not correcting the number, but the entry should be verified against the original examiner Form PTO‑892 / file wrapper before being relied upon.
• Potential claims: none.

5. US 4,769,772 A — Honeywell Bull — "Automated query optimization… distributed databases"

• Filed 1985‑02‑28; issued 1988‑09‑06. § 102(e) art (issue date post‑dates the 1988‑03‑09 bar).
• Query planning/optimization and materialization access planning across a distributed database.
• § 102 assessment: Distributed access planning, but no file-server + service-program-execution architecture. No anticipation of any claim.

6. US 4,835,685 A — Computer X — "Virtual single machine with message-like hardware interrupts and processor exceptions"

• Filed 1985‑05‑06; issued 1989‑05‑30. § 102(e) art.
• Virtual-machine-style resource presentation across nodes. Background only. No anticipation.

7. US 4,754,395 A — Computer X — "Network interface module with minimized data paths"

• Filed 1985‑05‑06; issued 1988‑06‑28. § 102(e) art.
• Hardware network-interface architecture. No service-execution model. No anticipation.

8. EP 0 205 946 A2 — IBM — "Flexible data transmission for message based protocols"

• Filed 1985‑06‑17; published 1986‑12‑30. § 102(b) art.
• Message-based protocol transmission. Background for the mail-service comparison (FIG. 6) but does not disclose the claimed server. No anticipation.

9. US 4,825,354 A — AT&T Bell Laboratories — "Method of file access in a distributed processing computer network"

• Filed 1985‑11‑12; issued 1989‑04‑25. § 102(e) art.
• File-access mechanisms (i-nodes, path names, processes, remote process identifiers) in a distributed network.
• § 102 assessment: Directly teaches the "file server for sending and receiving files to and from the communication network, for storing said files, …" element of claim 1 and the "means for storing request files received from said communication network" element of claim 6. It does not teach a store of selectable service programs or processing means that executes a selected service program on a selected stored file.
• Potential claims: relevant to the file-server element of claims 1 and 6, but does not anticipate either claim as a whole.

10. GB 2 187 009 A — Hitachi — "A knowledge-based system having a plurality of processors"

• Filed 1986‑02‑21; published 1987‑08‑26. § 102(b) art.
• Knowledge-based (inference) system spread over multiple processors; distributing processing/processing units.
• § 102 assessment: Relevant to distributing service execution across processors (the FIG. 4 lower-layer-server concept), but does not disclose the file-server-based request/service model of claims 1/6. No anticipation.

11. US 4,780,821 A — IBM — "Method for multiple programs management within a network having a server computer and a plurality of remote computers" ⭐ strongest candidate

• Filed 1986‑07‑29; issued 1988‑10‑25. § 102(e) art (filed before the invention date; issue after the bar date).
• Description (verified): A LAN with a server computer and remote/requestor nodes. The server includes storage devices for storing data files and application programs available for access on the network. A server "hypervisor" intercepts file/program requests from remote nodes, maintains a program matrix and a list of currently running programs and accessed files, compares a remote request against that matrix, and grants/denies access. A "node enabler" on the requestor node, licensed under the base product, entitles the requestor node to run the application program. License controls are enforced by recording a unique ID per signed-on remote computer.
• § 102 assessment vs. claim 1: This reference teaches (i) a server that stores files, (ii) a server that stores a plurality of application programs, and (iii) request analysis/management by the server. That maps strongly onto "file server for … storing said files, and for analyzing a user's request" and "means, connected to said file server, for storing a plurality of predetermined service programs." However, the application programs are executed at the remote node (the node enabler "entitles the requestor node to run the application program"), whereas claim 1 requires "processing means for executing a selected one of said plurality of predetermined service programs using a selected one of said files stored in said file server." The result is stored back to the file server in the '670; here the paradigm is distributed program licensing/management.
• Potential claims: Potentially citable against claims 1 and 6, but on the face of the disclosure it does not disclose the server-side execution-and-store-back element, so it does not fully anticipate. It is the best § 102 challenge and the best § 103 base reference among the cited set.

12. US 4,819,159 A — Tolerant Systems — "Distributed multiprocess transaction processing system and method"

• Filed 1986‑08‑29; issued 1989‑04‑04. § 102(e) art.
• Fault-tolerant, distributed transaction processing. Background for distributed processing; no file-server/service-program-store model. No anticipation.

13. US 4,887,204 A — IBM — "System and method for accessing remote files in a distributed networking environment"

• Filed 1987‑02‑13; issued 1989‑12‑12. § 102(e) art.
• "Statefull" distributed file server with client caching and synchronization modes (read-only / shared-read / read-write); server keeps state on which clients have files open.
• § 102 assessment: Strong teaching of the file-server element (send/receive/store files; server/client model; the reference explicitly defines "server" and "client"). It does not teach storing a plurality of service programs or executing them on stored files.
• Potential claims: relevant to the file-server element of claims 1 and 6; does not anticipate.

14. EP 0 278 317 A2 — IBM — "…using cached data at a local node after re-opening a file at a remote node…"

• Filed 1987‑02‑13; published 1988‑08‑17.
• Remote-file caching approach.
• § 102 timing flag: As a foreign (EP) publication, only § 102(a)/(b) can apply — not § 102(e) (pre‑AIA § 102(e) applies to US patents/applications). Its publication date (1988‑08‑17) is after both the 1988‑03‑09 § 102(b) bar and the ~1988‑03‑16 invention date. On these dates it is arguably not § 102 art at all for the '670 claims (unless a US counterpart application of the same family supplies an earlier § 102(e) date, or the examiner relied on a corresponding US filing). This should be confirmed against the file wrapper.
• Potential claims: none.

15. ⚠️ US 4,949,248 A — Caro — "System for shared remote access of multiple application programs executing in one or more computers"

• Filed 1988‑07‑15; issued 1990‑08‑14.
• Description (verified): LAN-based shared remote access/control of application programs (securities-trading example). Information services are represented as network nodes called "servers," each running an application program unmodified; clients request control and send keystrokes to the server to control its application program; display changes are broadcast to clients.
• § 102 timing flag: This reference's filing (1988‑07‑15) is after the '670's JP priority (1988‑03‑16). For § 102(e), the reference's effective date must precede the applicant's invention date; here it does not. Its issue date (1990‑08‑14) is also after the '670 filing, so § 102(a)/(b) do not apply. On its face, US 4,949,248 is not § 102 prior art to the '670 claims (it would matter only if the applicant were held to a later invention date than the JP priority).
• Technically: It does describe servers that run application programs and are accessible to remote clients — conceptually adjacent to claim 1 — but it lacks a general "file access request triggers execution of a selected stored service program on a selected stored file with the result stored back," and it is principally about remote control/display sharing of already-running programs.
• Potential claims: If it were art, it could be argued against claims 1/6; because of the date issue, it is not available as § 102 art without further proof of a later invention date.

Addendum — family-cited references (cited in the JP/GB/DE family files, not the US examiner set)

These appear under "Family Cites Families" and are relevant if the family members' prosecution is considered:

§ 102 note: CA 1,244,142 A (pub. 1988‑11‑01) and US 4,800,488 A (issued 1989‑01‑24) again post-date the 1988‑03‑09 bar; US 4,800,488 could be § 102(e) art by its 1985‑11‑12 filing. None of these four discloses the claimed server-executes-service-program combination.

Overall conclusions

1. No cited reference anticipates claim 1 or claim 6 as a whole. The recurring missing element is a file server that both stores the user's file and executes a selected, pre-stored service program on that file, storing the result back — the "automated-vending" server of the specification.
2. Best § 102 candidate: US 4,780,821 (IBM) — server-side storage of both data files and application programs plus request analysis — but execution occurs at the remote node, so it is an anticipation threat only under a broad reading and is more naturally a § 103 base reference.
3. File-server-element references: US 4,825,354 and US 4,887,204 teach the file server/client storage-and-access half of claims 1/6; neither teaches service-program storage/execution.
4. Scheduling-claim references: US 4,387,427 (and arguably US 4,819,159) touch scheduling/tasking, relevant only to dependent claims 4/5/7 and not anticipatory.
5. Two anomalies to flag (per the no-auto-correct rule, not corrected):

• US 4,937,284 A (Neste Oy) is a polymer-chemistry patent and is irrelevant — likely a citation-mapping error; verify against the original examiner citation.
• EP 0 278 317 A2 and US 4,949,248 A carry publication/filing dates that place them outside pre‑AIA § 102(a)/(b) and, for the EP document, outside § 102(e) as well; US 4,949,248's 1988‑07‑15 filing post-dates the 1988‑03‑16 priority and is therefore doubtful § 102(e) art.

1. Cross-reference check against the previously generated sections: the earlier "Patent summary" and "Litigation summary" are consistent with the patent text (7 claims, independent claims 1 and 6, priority 1988‑03‑16, expiry 2010‑09‑21, no litigation). No contradiction found. One nuance to restate: the effective § 102(b) bar is 1988‑03‑09 (one year before the 1989‑03‑09 US filing), not the 1988‑03‑16 JP priority — a distinction the citations analysis above depends on.

Caveats: These are technical-analyst assessments of § 102 exposure, not legal opinions of invalidity. Subsection attributions are date-based and should be verified against the file wrapper (Form PTO‑892 and any § 102(e)/§ 102(b) arguments of record). If you need certified prior-art status (e.g., to run an IPR-style invalidity chart), pull the original examiner citations from USPTO PatentCenter for US 07/321,134 and confirm the two flagged records.

I'll research the examiner-cited prior art references to ground the § 103 analysis in what each reference actually discloses.

Now let me examine the remaining cited references, particularly the scheduler and messaging references.

§ 103 Obviousness Analysis — US 5,247,670 ("Network server")

Built on the previously generated summary/litigation sections. Bibliographic facts, claim text, and the prior-art list come from the Google Patents page (patents.google.com/patent/US5247670/en); reference disclosures below are attributed to the sources I actually retrieved in this session, with unverified items flagged.

1. Method and legal frame

• Governing law: Pre-AIA § 103(a) (application filed 1989-03-09; priority 1988-03-16). The four Graham v. John Deere factors apply, with the KSR Int'l v. Teleflex gloss that a combination of familiar elements according to known methods is obvious when it yields no more than predictable results and a POSITA would have had an articulable reason to combine.
• Analytical caveat (important): These 15 references were considered by the examiner, and all 7 claims nonetheless issued. Any § 103 reconstruction below therefore carries hindsight risk. I flag where a ground is strong versus where it is a hindsight-flavored stretch.
• Evidentiary caveat: In this session I retrieved full text/abstracts for US 4,780,821; US 4,887,204; US 4,825,354; US 4,500,960; and US 4,387,427. For EP 0205946, US 4,819,159, US 4,769,772, GB 2,187,009, US 4,754,395, US 4,835,685, EP 0,077,008, US 4,800,488, CA 1,244,142, JP S61-150436, JP 2,574,784, US 4,937,284, and EP 0,278,317 I have title / assignee / date only. I do not describe their contents as verified facts.

2. Prior-art status of the cited references (pre-AIA framing)

A point worth stating plainly: nearly all of the operationally useful art (the two IBM references, the AT&T file-access reference, the Tolerant transaction reference) is § 102(e) "secret" prior art, effective only as of its US filing date and available for § 103. Because the applicant is Fuji Xerox and the primary references are IBM, the pre-AIA § 103(c) common-ownership safe harbor is unavailable.

3. Ground 1 — Claim 1 over US 4,780,821 in view of US 4,887,204

US 4,780,821 (IBM, Crossley) — retrieved content shows a LAN with a designated server computer whose storage devices 16 hold "data files and application programs which are to be available for access on the network"; a server-side hypervisor "inspects and responds to file/program requests and special data control blocks that are passed from a remote to the server system" and intercepts OS calls for file-sharing and record locking; a program matrix plus a list of currently running programs and accessed data files is compared against a remote request to decide whether to grant access; the server "maintains a file of all unique serial numbers which are authorized to use the system." (US4780821A, Unified Patents portal)

US 4,887,204 (IBM) — retrieved content shows a distributed services program installed on each of a plurality of nodes allowing processors to access data files distributed across nodes; a statefull server that "does keep information about who is using its files and how the files are being used"; client and server caches; and file synchronization modes in which, for a file open for read at more than one node with at least one write, "all read and write requests must go to the server node where the file resides." (US4887204, abstract at idiyas)

Element mapping for claim 1:

Why a POSITA would combine them: Both are IBM distributed-data-processing disclosures in the same LAN/server art, filed 18 months apart, and they address complementary halves of one known problem — how to let remote workstations use centrally held files and centrally held programs. Combining a statefull file-serving front end (4,887,204) with the server's program-availability and request-adjudication mechanism (4,780,821) is the aggregation of two known server functions onto one server, with predictable results. Under KSR, that is the paradigm case ("combination of familiar elements according to known methods").

4. Ground 2 — Claim 1 over US 4,825,354 in view of US 4,500,960 (or 4,780,821)

• US 4,825,354 (AT&T, Agrawal et al.) — retrieved content describes distributed file access via path names, i-nodes, file descriptors and system file tables, plus special files where "requests to read or write result in activation of the associated device," with one special file per communication line, disk, tape and main memory, and inter-process channels ("pipes") created by a file-descriptor-returning call. (US4825354A) This is a direct teaching that a file-system access can be the trigger that activates an associated processing operation — the core mechanism by which the '670 claims convert a stored request file into an executed program.
• US 4,500,960 (AT&T Bell Labs) — retrieved content describes nodes whose "each processor supports a plurality of processes," with "customized" processes developed for particular customers' needs alongside universal-capability processes, and a process that "can initiate a call to any other process and define the communication parameters … by selecting the appropriate internal link." (US4500960A) That supplies "processing means for executing a selected one of a plurality of predetermined service programs."

Motivation: Same field (AT&T network data-processing services), same era, and the references are complementary by design — 4,825,354 supplies the file abstraction and the file-access-triggers-action behavior; 4,500,960 supplies a multi-process node that selects and executes one of many customer-specific programs on request. A POSITA seeking to offer network-accessible processing services would predictably use the file system as the request/result transport and the process scheduler as the executor.

5. Ground 3 — Admitted prior art in the specification + US 4,780,821

The '670 specification itself describes, as prior art, (a) FIG. 6: a mail server 13 with a mail box 15 for a newly added server 14 that "executes a predetermined process for the mail message and stores the result"; and (b) FIG. 7: a file server 21 with "a storage area 22 to serve the newly added service 14." These admissions establish that, before the invention, the art already knew network-accessible service programs that execute on request and write results back. (Applicant admissions are § 103 prior art and need no separate reference.) The only structural delta in claim 1 — putting the "file server" and the service-program storage/processing means inside one "server" — is supplied by 4,780,821, whose server already stores both data files and application programs and runs them for remote requestors. This is the strongest ground because it requires no assumption about what an unverified reference teaches.

6. Claim 6 — same grounds

Claim 6 is claim 1 in means-plus-function dress (store request files ↔ store service programs ↔ process by reading out and executing the requested program ↔ store results). It is met by:

• 4,780,821 for the request-file storing means (storage 16, file/record-locking interception), the program-storing means (application programs available on the network; program matrix), the processing means (server executes the requested licensed program), and the result-storing means (file control block read/write and close routines writing back to server storage); and
• 4,887,204 for result storage under synchronization modes (client write → server cache → server node file).

No additional motivation is needed beyond Ground 1; claim 6 removes even the "file server" label, so it is broader and correspondingly easier to invalidate.

7. Dependent claims 2, 3, 4, 5, 7

Motivation for the scheduler claims: The specification itself states the problem — "When there are sets of data waiting to be serviced by the server 31, a user's request for first priority service execution cannot always be satisfied immediately" (col. describing temporary storage area 34). Once multiple queued requests and multiple selectable service programs coexist on one node, ordering/timing selection is not an optional nicety; it is a recognized necessity, and 4,387,427 addresses precisely that problem in a network/multiprocessing context. KSR: a known problem addressed by a known, predictable technique.

8. Motivating rationales (consolidated)

1. Same field / same problem: all primary references are LAN or distributed-data-processing server disclosures from the same decade, addressing remote users' access to centrally held programs and data.
2. Complementary, not overlapping, teachings: file-serving request/result transport (4,887,204 / 4,825,354) + multi-program execution and selection (4,780,821 / 4,500,960) + scheduling/buffering (4,387,427).
3. Predictable results, no new physics: the claims recite generic memory areas, a CPU, and a program selector; nothing in the claims requires an unexpected result. The specification's asserted benefits (fault isolation, "passivity," ease of add/remove) are attributed to colocation and registration, not to claim-recited structure.
4. Design incentive: reducing node count, avoiding per-client interface changes, and re-using the existing file-server interface were expressly recognized objectives in the art (the specification's own background states the goal of adding services "without any new interface at the client side").
5. KSR "obvious to try": with a finite, identified set of known server architectures (mail-box-based vs. file-based vs. colocated), placing the service program on the same node as the filing function is one of a small number of predictable options.

9. Where the case for obviousness is weakest (counterarguments)

• Colocation is the invention's real point. In the FIG. 7 admitted prior art, the file server 21 and the service server 14 are separate nodes. Claim 1, by contrast, recites a server comprising a file server — i.e., the filing function and the service function integrated in one unit. The strongest non-obviousness argument is that the art of record taught distribution (4,887,204's explicit stateless-vs-statefull discussion, the distributed services model generally), so combining file serving with application execution inside one node was a deliberate departure. Counter: 4,780,821's server already holds both the data files and the application programs and executes them for remote requestors — which is very close to that integration.
• The "identifying data are stored in said file server" limitation (so services are discoverable through the ordinary file-server interface, avoiding client changes) is the most distinctive element and is not squarely verified in any single reference. A court could find this the point of novelty.
• No secondary-considerations record. There is no evidence of record (and none found in the earlier searches) of commercial success, long-felt need, failure of others, or copying tied to the claims. The specification's "8000 INS" reference describes the assignee's own pre-existing system, and no nexus to the claimed structure is established.
• US 4,937,284 (Neste Oy, modified polyolefin) is non-analogous art; it must be excluded from every ground. Its presence in the citation list does not strengthen any obviousness theory.
• EP 0,278,317 is date-questionable (published after the 1988-03-16 priority date); any ground relying on it should be re-checked against its US counterpart before being asserted.

10. Bottom line and confidence

Assessment: Claims 1 and 6 are, in my view, more likely than not obvious over US 4,780,821 in view of US 4,887,204 (Ground 1), with US 4,825,354 + US 4,500,960 as a viable alternative (Ground 2) and the specification's own FIG. 6/FIG. 7 admissions as a supporting § 103 foundation (Ground 3). Dependent claims 2, 3, 4, 5 and 7 fall with the addition of US 4,387,427 for the temporary-storage and scheduling limitations, and 4,887,204/4,780,821 for result storage.

Confidence grading:

• High confidence in the factual content of 4,780,821, 4,887,204, 4,825,354, 4,500,960 and 4,387,427 as recited above (retrieved from Google Patents / Unified Patents / idiyas this session).
• Moderate confidence in the ultimate § 103 conclusion for claims 1, 2 and 6; moderate-to-lower for the scheduling claims 4, 5 and 7.
• Low confidence / unverified: the contents of EP 0,205,946, US 4,819,159, US 4,769,772, GB 2,187,009, US 4,754,395, US 4,835,685, EP 0,077,008, US 4,800,488, CA 1,244,142, JP S61-150436 and JP 2,574,784. I did not retrieve these; I have described them only by title, assignee and date, and any ground resting on them should be validated against the full text before being relied upon. Note also that the earlier-generated sections found no litigation in which these claims were ever tested, so there is no adjudicated claim construction to anchor this analysis.