US 7925763

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

Let me search for litigation/docket information and the USPTO record specifically.

Summary of US 7,925,763 B2

Important caveat up front: I searched for the exact number 7925763 / 7,925,763 across web-accessible patent databases (Google Patents, Justia) and general web results for PTAB, district-court, and CAFC activity. I found no litigation, PTAB proceeding, or CAFC 2026 docket that references US 7,925,763. I do not have direct USPTO PatentCenter/PACER/CM-ECF query access in this session, so I cannot certify that no such proceeding exists — I can only report that none surfaced in my searches. Treat the litigation/docket point as "not found," not as "confirmed nonexistent."

All bibliographic and claim detail below is taken from the patent text supplied in your message (Google Patents rendering, fetched 2026-09-18), which I treat as authoritative over any conflicting search snippet.

Bibliographic data

Abstract

Disclosed is a method and device for keeping persistency while switching between modalities, which include: (1) searching out the last operation state of the application service accessed by the user from the user history, in response to a request for accessing an application service from a user; (2) connecting to the application service and continuing to execute the application service from the last operation state; (3) updating the last operation state in the user history based on the execution result of the application service. A user could flexibly select different favorable terminals to continue with an unfinished application service.

Plain-language overview of the independent claims

Claim 1 — Method (device-to-device continuity). A user who previously used a first device logs in from a second device and asks for the same given application service. That service is made of many logic components, each producing its own execution result and having its own state. In response:

1. The system recalls from a user history the last operation state reached on the first device, and specifically retrieves an ID tied to that state, where the ID identifies one or more of the logic components.
2. It connects the user and executes those logic component(s) in accordance with the ID, so the user picks up where they left off rather than restarting.
3. It monitors the state of each logic component to determine the latest operation state.
4. It updates the user history with that latest state.
5. The final limitation requires that the "given operation state" be determined based at least in part on the respective states of each of the logic components — i.e., the state is an aggregate over all components, not just a single cursor value.

Claim 5 — Apparatus. Mirrors claim 1, phrased as "at least one computer based device operative to" perform the same obtain / recall (including the ID) / connect-and-execute / monitor-per-component / update steps, with the same aggregate-state limitation.

Claim 9 — System (four-module architecture). Claims a system comprising:

• A user history module that records historical call information per user ID for at least a given application service (again, the service is modularized into stateful logic components).
• A logon control module that identifies/verifies the user (on the second device), retrieves from user history the historical call information captured on the first device, and extracts the ID of the last operation.
• A session management module that connects the second-device user to the service, executes the component(s) via the ID, monitors each component's state to determine the latest operation state, and updates the stored historical call information.
• A connection management module that establishes the connection to the session management module based on the extracted ID.
• Same closing aggregate-state limitation as claims 1 and 5.

Notable dependent claims (context, not independent)

• 2–4 / 6–8 / 10–13: Returning/presenting the execution result via a "proper" contact method — either the method the user used to request access, or one selected from pre-registered contact methods tied to devices used during a defined time period (the spec gives an 8:00 AM–5:00 PM telephone / otherwise mobile-phone example).
• 14–16: The ID corresponds to a serial number of the logic component(s).
• 17–19: Logic components are executed recursively until the user issues an interrupt request or the service finishes.

Notes on the "CAFC 2026 docket" request

Nothing in my searches (CAFC scheduled-cases PDFs, CAFC opinions/orders listings, CourtListener dockets, and the PTAB documents that surfaced) ties any 2026 Federal Circuit appeal to patent 7,925,763. The 2026 CAFC items that appeared (e.g., Greenthread IPR2024-00553 re U.S. 11,316,014; Implicit/Sonos; ParkerVision v. Qualcomm 26-1033; Contour IP v. GoPro 26-2012) involve entirely different patents. Given the patent lapsed in 2015 for non-payment of maintenance fees, a 2026 appeal is unlikely, but I am flagging this as an absence of evidence rather than a verified negative.

I'll search for litigation involving US patent 7925763. Let me run several targeted searches.

Those searches returned unrelated cases. Let me run more targeted queries.

Litigation Search Results for U.S. Patent No. 7,925,763

No litigation involving U.S. Patent No. 7,925,763 was found.

I ran multiple targeted searches across patent-litigation resources (including general queries aimed at the Unified Patents litigation portal, PACER-style docket aggregators, CourtListener, drugpatentwatch, and general web/legal indexes) using the exact identifiers "7925763" and "7,925,763," as well as the patent's title ("Keeping persistence while switching between modalities"). I did not locate any district court complaint, CAFC appeal, ITC action, or PTAB proceeding naming this patent as a patent-in-suit. The hits returned were for unrelated patents/cases whose numbers merely overlapped on partial digits, and I excluded them per your instruction not to return similar numbers.

Why this result is consistent with the record

The patent's own front-page data (Google Patents, https://patents.google.com/patent/[US7925763](/patent/US7925763)/en) shows a profile typical of an unasserted asset:

• Owner: International Business Machines Corporation (original and current assignee); inventor Hong Cai.
• Priority: 2001-08-13 (CN 01124795.9); PCT filed 2002-07-08; U.S. granted 2011-04-12.
• Legal status: Expired – Fee Related. The maintenance-fee record shows a reminder mailed 2014-11-21, lapse for failure to pay on 2015-04-12, and the patent expired under 37 CFR 1.362 (lapsed 2015-04-12). The listed "adjusted expiration" of 2025-11-04 reflects the nominal term, not an active patent.
• Family: Parallel national filings (EP 1428114A2, JP 2005500615A, KR 100690290B1, CN 1407780A, CA 2451412A1, TW 577210B, AU 2002345210A1, IL 160073A0, WO 2003017089A2) — i.e., prosecution was pursued broadly, but no enforcement activity surfaced in those jurisdictions either.
• Citations/"cited by": The "Cited By" and "Families Citing this family" lists show later patents (e.g., ETRI US2011/0125888, Mitel proximity-session-mobility patents) that cite the '763 patent as prior art. Being cited by others is not litigation and does not indicate the '763 patent was ever asserted.

Confidence level and caveat

I have high confidence that the patent was never asserted in U.S. district court or at the PTAB, based on the absence of any docket reference in the sources searched. However, I want to be explicit about limits: I could not directly return a live query against the Unified Patents litigation caselist URL you cited, and free search indexes are not guaranteed to be exhaustive for very old, 2011–2015-era, low-profile suits. If a single short-lived case existed and left no searchable electronic footprint, it could conceivably be missed. A definitive negative check would require a subscription database (LexisNexis/Lex Machina, Darts-IP, or Bloomberg Law) run against the exact patent number.

Bottom line

• Plaintiff(s): None identified.
• Defendant(s): None identified.
• Jurisdiction / case number / filing date / status: Not applicable — no case found.

In short: on the record available, U.S. Patent No. 7,925,763 appears to have no known litigation history; it was an IBM-owned patent that lapsed in 2015 for nonpayment of maintenance fees.

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

Let me run additional targeted searches to rule out PTAB activity under alternative formats and litigation-driven filings.

Proceedings overview

Zero AIA trial proceedings on file. The USPTO Open Data Portal returns no IPR, PGR, or CBM for US 7,925,763, and my independent web searches (Google Patents PTAB panel, PTAB E2E document text, CAFC/District Court docket text, and the family records for the CN/EP/WO counterparts) surfaced no petition, no institution decision, no Final Written Decision, and no Federal Circuit appeal involving this patent. The structured count is 0 proceedings (0 active / 0 invalidated / 0 sustained / 0 settled / 0 institution-denied) — I am flagging that explicitly rather than inventing docket numbers.

This is not the typical "hardened patent survived IPRs" posture. The bottom line for a defendant is stronger than an invalidity-win posture, and it comes from a different source: the patent lapsed for failure to pay maintenance fees, effective 2015-04-12, and Google's legal-status field for US 7,925,763 reads verbatim "Expired - Fee Related." The absence of PTAB activity is consistent with the patent never having been a live assertion target — the usual reason patents attract IPRs (they get asserted; defendants respond) never materialized here.

Sources checked: US7925763B2 on Google Patents, PTAB E2E / PTAB Center public filings. No FWD or CAFC opinion exists to link to, because none exists.

No proceedings to itemize

Because the structured PTAB dataset is empty and my searches corroborate it, there are no {PROCEEDING_NUMBER} blocks to complete. I want to be precise about what I verified vs. what I could not:

• Verified as absent: any party-caption combination naming US 7,925,763 / "7925763" as the challenged patent in an IPR/PGR/CBM caption. Searches for the patent number against PTAB petition and FWD text returned only unrelated hits (e.g., the Netflix/Hulu IPRs on US 8,472,792 — a different patent whose number superficially resembles it, and which I am not conflating with 7,925,763).
• Caveat, stated honestly: PTAB E2E's free-text index is not perfectly searchable by patent number through general web search, and a terminally disclaimed or lapsed patent's sparse record can be hard to surface. I could not query the ODP API directly. My confidence that there is no instituted trial is high; my confidence that no petition was ever filed and denied/settled pre-institution is high but not absolute. If you are already in litigation, pull the Patent Center "PTAB trials" tab for the patent directly, or run a PTAB E2E patent-number search, before relying on this in a brief.

The thing that actually matters (and it is not PTAB)

Even with an empty PTAB list, a defendant facing a demand letter on this patent has a dispositive defense that is cheaper and cleaner than an IPR:

Two practical consequences:

1. No prospective relief. A patent that lapsed in 2015 cannot support an injunction or ongoing-royalty theory for conduct in 2026.
2. The § 286 damages window has run out. Damages reach back only six years from complaint filing. A suit filed now (2026) reaches back to roughly 2020 — five years after the 2015 lapse — so recoverable past damages for post-lapse conduct are effectively nil. Only a pre-2015 infringement suit that is somehow still alive (it isn't; no litigation surfaced in my searches) could reach the pre-lapse period.

Note the discrepancy that could be exploited or mistaken: Google's summary line shows an "Adjusted expiration 2025-11-04," which is a projected-term artifact, not the operative event. The Legal Events table is authoritative and shows termination in 2015. Do not let opposing counsel quote the 2025 line at you — it misstates the patent's actual status. Reviving a 2015 lapse via 37 CFR 1.378 (unintentional-delay petition) is theoretically available but, at eleven years' remove, is not a realistic risk; treat the patent as dead.

Strategic summary

Claim status. All 19 claims (independent claims 1, 5, 9; dependents 2–4, 6–8, 10–13, 14–16 serial-number variants, 17–19 recursive-execution variants) are UNTESTED at the PTAB — none canceled, none sustained, none construed by the Board. There is therefore no IPR-based estoppel record, no Board claim construction, and no PTAB win to cite. The only "narrowing" on this patent came from prosecution, not adjudication: the examiner cited US 6,349,337 (Microsoft — "Maintaining a first session on a first computing device and subsequently connecting to the first session via different computing devices"), US 6,076,108 (I2 Technologies — session state on a global session server), and WO 2000/079391 (Art Technology Group — session backup) against it. Those are now your best prior-art footholds if you ever need invalidity contentions — but the lapse defense obviates the need.

Estoppel landscape. There is nothing to be estopped by. § 315(e)(2) estoppel only attaches to petitioners after an FWD (§ 315(e)(1) for Office proceedings, § 315(e)(2) for civil actions/ITC). With no FWD, no petitioner is estopped, and equally, no prior-art ground is foreclosed to you. Every reference — including the examiner-cited Microsoft '337, I2T '108, and ATG '391, plus the two non-patent citations of record (Cannataro et al., An Object-Based Architecture for WAP-Compliant Applications, DEXA'00; Welling et al., Session Mobility Support for Multimedia Applications, 1999) — remains affirmatively available in district court. That said, § 325(d)(1) discretion cuts against instituting an IPR on art the examiner already considered; if you nonetheless wanted a PTAB challenge, you would need materially new art (e.g., later-dated session-mobility or WAP-state-persistence references) and, given the 2015 lapse, there is little economic reason to spend the filing fee.

Pattern signals. No repeat petitioner (there are no petitioners). No patent-owner appeal activity (no adverse PTAB decision to appeal). No defensive aggregator involvement (Unified Patents, RPX, etc.) surfaced anywhere in the chain. The counterpart family went passive in parallel: EP1428114A2 — "Ceased," WO2003017089A2 — "Ceased," CA2451412A1 — "Abandoned," AU2002345210A1 — "Abandoned," while JP2005500615A and KR100690290B1 remain separately tracked and the CN parent CN1407780A is listed "Pending." IBM let the entire Western portfolio go quiet — the classic signature of a patent that was never in the assertion revenue stream. The four "cited by" documents are sparse and mostly late (2007+), further evidence the disclosure never became a licensing bottleneck.

One durable-validity caveat worth keeping in your back pocket. The claims as granted are integrally tied to the "plurality of logic components... each having a respective state" and to "an ID... corresponding to an identifier of one or more of the plurality of logic components" (claim 1, and mirrored in claims 5 and 9). If this patent were ever revived, that component-level ID/state-monitoring limitation is a narrow hook — modern "resume across devices" implementations push state to a server or use cookie/token-based session resumption, not a call-ID indexing into discrete server-side logic components. But this is a fallback argument only; the lapse ends the inquiry first.

Recommended next steps

1. Lead with expiration, not invalidity. If you have received a demand letter citing US 7,925,763, the response is: the patent lapsed 2015-04-12 for nonpayment of maintenance fees under 37 CFR 1.362 and is "Expired - Fee Related" per USPTO records. Link the Google Patents legal-events entry and pull the Patent Center maintenance-fee history as the primary exhibit. There is no FWD to quote because no PTAB proceeding was ever filed — say that plainly; the absence is the signal, not a gap in your research.
2. Reject the "adjusted expiration 2025-11-04" line if opposing counsel cites it. It is a term-calculation artifact; the operative event is the 2015 lapse shown in the Legal Events table.
3. If (and only if) a live pre-2015 infringement suit somehow exists, then the invalidity path reopens and you should build contentions around the examiner-cited Microsoft '337 / I2T '108 / ATG '391 trio plus Cannataro and Welling — and note that no § 315(e) estoppel bars any of it.
4. Do not budget for an IPR. With no live damages exposure post-2015, the cost-benefit of a PTAB petition does not close. Reserve IPR analysis for a scenario in which the patent is revived (§ 1.378) or a pre-2015 claim surfaces — both remote.
5. Verification step before filing anything. Run a PTAB E2E patent-number search on 7,925,763 and the Patent Center "PTAB trials" tab yourself. My web-based negative finding is corroborated but is not a substitute for the authoritative Office record, and the ODP API's silence (per your prompt) is the strongest evidence I have.

I'll verify the assignment record against live sources before reconstructing the chain.

Assignment & Ownership Reconstruction — US 7,925,763 B2

Verification note: I could not directly load the USPTO Assignment Center UI/API for this proceeding in this session; the reel/frame and legal-event data below is taken from the authoritative patent text supplied (Google Patents legal events, which mirror USPTO assignment and maintenance-fee records) and cross-checked against Unified Patents' patent portal and Espacenet. Where a field is not represented in those records, I say so rather than guess.

Inventors

• Employer at filing: IBM is the original assignee and the assignment record names Cai as the assignor, so IBM employment at filing is well-supported.
• Location signal (suggestive only): the specification's worked example uses the literal sender identity From: caihong@cn.ibm.com and a To: office@myservice.com service. That is an illustrative example in the patent text, not a verified inventor address, so it indicates an IBM China / IBM China Research Lab footprint at most. Do not treat it as a record finding.
• Departure pattern: Not determinable. I found no evidence of when (or whether) Cai left IBM. The "all inventors departed within 12 months" tell therefore cannot be scored either way — and it is in any event inapplicable here, since there was no post-issuance portfolio sale to trigger a fire-sale inference.

Original assignee

International Business Machines Corporation (Armonk, NY) — named on the issued patent and still the assignee of record today.

• Product embodying the claims: Unclear / not established. The patent claims a session-persistence and modality-switching middleware layer (logon control module, user history module, session/connection management, user register module). IBM is an operating company with a very large middleware and WebSphere/Lotus product line, and the spec's examples use Lotus Notes and WAP/WML clients, but I found no evidence in the record tying these specific claims to a shipped IBM product, and I will not infer one.
• Primary line of business: enterprise computing, software, hardware and services.
• Current status: operating; not acquired, dissolved, or in bankruptcy. No IBM Chapter 7/11 event is relevant to this patent.
• Patent status: Lapsed 2015-04-12 for nonpayment of maintenance fees (37 CFR 1.362). The Google Patents header's "Adjusted expiration 2025-11-04" is the nominal 20-year adjusted term and is moot — the enforceable term ended in 2015.

Assignment timeline

One assignment of record, and it is the original inventor-to-employer assignment. There are no post-issuance assignments — no security agreements, no mergers, no change-of-name filings, no transfers to any LLC or aggregator.

• 2004-03-18 (executed / effective) / recorded 2004-07-19 — Reel 014867 / Frame 0853
  • Conveyance: Assignment
  • Assignor: Cai, Hong
  • Assignee: International Business Machines Corporation (New York)
  • Correspondent: Not determinable. The assignment abstract in the record shows the reel/frame, assignor, assignee and effective date but does not expose the filing correspondent in the sources I could reach. I cannot name an attorney or firm without fabricating it. (In practice a 2004 IBM in-house filing of this type would be handled by IBM's own IP law department, but that is an expectation, not a record — so it is not scored below.)
  • Context: Original employment/obligation-of-assignment transfer to the corporate employer. Not a sale, not a fire-sale, not a reorg.

Maintenance-fee legal events (not assignments, but they terminate the chain):

• 2014-11-21 — REMI: maintenance fee reminder mailed
• 2015-04-12 — LAPS: lapse for failure to pay maintenance fees (effective date)
• 2015-05-08 / 2015-05-11 — STCH: patent discontinuation (expired under 37 CFR 1.362)
• 2015-06-02 — FP: lapsed due to failure to pay maintenance fee, effective 2015-04-12

Foreign-family status (relevant to monetization appetite): CN1407780A (active/pending per listing), EP1428114A2 (ceased), WO2003017089A2 (ceased), KR100690290B1 (expired/fee-related), CA2451412A1 (abandoned), AU2002345210A1 (abandoned), IL160073A0 (unknown), TW577210B (IP right cessation), JP2005500615A (listed active/pending). Abandonment of CA and AU counterparts is consistent with an operating company pruning a non-core filing, not with an assertion program.

Timeline diagram

timeline
    title Ownership of US 7925763
    2001 : Priority filing CN01124795
    2002 : PCT and US national filing
    2004 : Inventor to IBM assignment recorded
    2011 : Patent issues as US 7925763 B2
    2015 : Lapsed for unpaid maintenance fee

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. No assignment off IBM to any "IP / Holdings / Licensing / Ventures" entity exists in the record. The only reel/frame entry (014867/0853) runs to IBM, not away from it.
2. Known asserter in the chain — Not present. The sole assignee is IBM. IBM does not appear on the Acacia / Marathon / IV / Wi-LAN / Conversant / Pendrell / Round Rock / MPHJ type lists, and Unified Patents' portal entry for US-7925763-B2 records the assignee as International Business Machines Corp with no intervening owner. No plaintiff matching those directories is in this chain.
3. Repeat correspondent across the chain — Unclear / not scoreable. A recurrence test requires two or more links; there is exactly one recorded link, and its correspondent is not exposed in the sources available to me. I decline to name one.
4. Cascading transfers — Not present. Zero transfers after issuance, let alone chained LLC hops in <24 months.
5. Pre-litigation transfer — Not present. No infringement suit naming this patent surfaced in any search. The only assignment (executed 2004-03-18 / recorded 2004-07-19) long predates issuance and is an employment transfer, not a standing-cleaning transfer.
6. Bankruptcy fire-sale — Not present. IBM has not filed Chapter 7/11; this patent was not sold in any Kodak/Nortel/Polaroid-style proceeding.
7. Privateering — Not present. No operating-company→NPE transfer; IBM did not place this patent with an assertion vehicle.
8. Defensive aggregator (anti-NPE) — Not present as an assignment (the chain does not end at RPX/AST/LOT/Unified/OIN). However, the patent is effectively neutralized by abandonment: it lapsed 2015-04-12 for unpaid maintenance fees, so no one can assert it. That is a different mechanism than defensive aggregation, but the practical outcome (unassertable) is the same.

Corroborating cross-references:

• Unified Patents portal entry listing IBM as assignee: https://portal.unifiedpatents.com/patents/patent/US-7925763-B2
• Google Patents record with legal events: https://patents.google.com/patent/US7925763/en
• Espacenet original document: https://worldwide.espacenet.com/patent/search?q=US7925763

Verdict

Insufficient data — in the rubric's specific sense of "only the original assignment." The record contains exactly one assignment (Reel 014867 / Frame 0853, executed 2004-03-18, recorded 2004-07-19, Cai → IBM) and nothing thereafter; IBM retained ownership until the patent lapsed on 2015-04-12 for nonpayment of maintenance fees. None of the eight NPE signals is affirmatively present — no shell LLC, no known asserter, no cascading transfers, no pre-litigation transfer — so this is not an NPE chain, but the absence of any post-issuance record also means there is no assignment chain to score beyond the original grant-to-employer transfer.

Caveat: this verdict reflects the records I could reach (Google Patents legal events, Unified Patents, Espacenet). To close out the correspondent field — the one genuinely missing datapoint — run the patent number directly at https://assignmentcenter.uspto.gov/ and open the Reel 014867 / Frame 0853 entry to capture the filing correspondent; that single field is the only thing that would let signal #3 be scored rather than marked unclear.

I'll research this systematically. Let me start with searches on the patent itself and its key cited references.

I have the patent text with its citation list. Let me gather details on the remaining significant cited references and the non-patent literature.

Scope and method

Patent under analysis (literal identifiers preserved): US 7,925,763 B2 — "Keeping persistence while switching between modalities when accessing an application service in a network"; Application No. US 10/486,716; PCT/GB2002/003121; earliest priority CN 01124795.9 (2001-08-13); PCT/US filing 2002-07-08; granted 2011-04-12; assignee International Business Machines Corp; inventor Hong Cai. Source: https://patents.google.com/patent/US7925763/en and Espacenet original document (NR=7925763B2, B2, 2011-04-12).

Search note. I could not open a native USPTO PatentCenter record in this session; the authoritative content above (fetched from Google Patents, which mirrors USPTO full text) plus Espacenet confirm the number, claims, and citation list. I have not substituted any similar number — all analysis is for 7925763 only.

Important limiting caveat on the § 102 question. Anticipation under 35 U.S.C. § 102 requires a single reference to disclose every limitation of a claim, arranged as claimed (Verizon v. Cox; MPEP 2131). Independent claims 1, 5 and 9 each contain a distinctive cluster: (i) the application service is composed of a plurality of logic components that independently generate respective execution results, each having a respective state; (ii) recall of a last-operation-state ID corresponding to an identifier of one or more of those logic components on reconnection from a second device after the user previously used a first device; and (iii) monitoring the respective states of each logic component to compute a latest aggregate operation state. None of the 24 cited references, read alone, appears to disclose that entire combination. The references below are therefore best characterized as grounds for § 103 (obviousness) or as anticipatory only of narrower dependent claims; I flag where a § 102 mapping is genuinely plausible versus where it is not. These are preliminary analyst judgments, not legal conclusions.

A. The cited references most relevant to the independent claims

1. US 6,349,337 B1 — [Microsoft Corp.](/litigations/by-plaintiff/Microsoft%20Corp.)
"Maintaining a first session on a first computing device and subsequently connecting to the first session via different computing devices and adapting the first session to conform to the different computing devices system configurations." Priority 1997-11-14; published/granted 2002-02-19. URL: https://patents.google.com/patent/US6349337 (see also family member US 6,085,247, Parsons et al., "Server operating system for supporting multiple client-server sessions and dynamic reconnection of users to previous sessions using different computers," filed 1998-06-08, granted 2000-07-04).
Description: A host maintains a session for a user; when the user reconnects from a different second computing device, the server recognizes the user's affiliation with the prior session, adapts/reconfigures the session to the new device's configuration, and re-associates the connection so the user continues the same session.
§ 102 assessment: This is the closest single reference to the device-to-device persistence core of claim 1 ("user previously accessed … using a first device" then obtains a request from "a second device"; continue the service). It is a credible § 102 candidate against that concept, but it does not disclose the claimed "plurality of logic components independently generating respective execution results, each having a respective state," nor "monitoring the respective states of each of the plurality of logic components." Those gaps mean it is unlikely to fully anticipate claim 1/5/9, but is strong § 103 art. Its session-reconfiguration disclosure is most relevant to claim 1's connecting/continuing steps.

2. US 6,076,108 A — I2 Technologies, Inc. (Courts et al.)
"System and method for maintaining a state for a user session using a web system having a global session server." Priority 1998-03-06; granted 2000-06-13 (continuation US 6,480,894 B1). URL: https://patents.google.com/patent/US6076108.
Description: A global session server stores a single set of session data representing the state of a user session; for each subsequent request the data is retrieved, the request processed, the data changed, and the single set updated. Explicitly addresses HTTP statelessness and the fact that "the user does not interact with the same web server from transaction to transaction."
§ 102 assessment: Relevant to the "recalling … last operation state … from a user history" and "updating the last operation state" limitations of claim 1 (and the "user history module" of claim 9). It does not address switching between two user devices or logic-component-level state monitoring, so it does not fully anticipate any independent claim; solid § 103 art in combination.

3. WO 2000/079391 A1 — Art Technology Group, Inc. (Abramson & Chung); US 6,539,494 B1; EP 1 192 545 B1
"Internet server session backup apparatus." Priority 1999-06-17; WO published 2000-12-28 (U.S. grant 2003-03-25; EP granted 2004-10-06). URL: https://patents.google.com/patent/WO2000079391A1.
Description: Three-tier web/application/backup-server architecture; each session is assigned a session ID encoding the backup-server and application-server IP addresses plus a unique session identifier so that a session "switched" to another application server is detected from the ID and the session data retrieved and reconstituted. Includes migration "in the absence of a failure," explicitly so the customer "does not need to 'start over.'"
§ 102 assessment: Strongly relevant to the "obtaining an ID associated with the last operation state, the ID corresponding to an identifier of one or more … logic components" limitation (claim 1) and the corresponding "extracting an ID of a last operation" language (claim 9), because it discloses an ID that both identifies the session and locates the stored state. However the "switch" here is server-to-server, not device-to-device, and there is no logic-component state monitoring — so it cannot alone anticipate claim 1/5/9, but maps to the session-ID/state-recovery features.

4. US 6,826,755 B1 — Microsoft Corp.
"Systems and methods for switching internet contexts without process shutdown." Priority 2000-06-30; granted 2004-11-30. URL: https://patents.google.com/patent/US6826755.
Description: Switching between internet/browsing contexts while preserving running process state, so a session/context can be re-entered rather than restarted.
§ 102 assessment: Marginal. It supports the general notion of preserving and resuming an in-progress context (relevant background to claim 1's "continuing … from the last operation state"), but discloses nothing about multiple user devices or logic components. No full anticipation; at most § 103 background.

5. US 6,049,831 A — GTE Laboratories Inc.
"System for transmitting network-related information where requested network information is separately transmitted as definitions and display information." Priority 1996-11-08; granted 2000-04-11. URL: https://patents.google.com/patent/US6049831.
Description: Separates transmitted information into definitions and display information across a network.
§ 102 assessment: Relevant only tangentially to rendering to heterogeneous client devices ("modalities"). No anticipation of any claim; background/§ 103 only.

B. Remaining U.S. patent citations (full list)

C. Non-patent literature citations

1. M. Cannataro et al., "An Object-Based Architecture for WAP-Compliant Applications," Proc. 11th Int'l Workshop on Database and Expert Systems Applications (DEXA '00), 8 pages, September 2000.
Description: Object/component-based architecture for building WAP-compliant applications (i.e., modularized application components served to wireless clients).
§ 102 assessment: Most relevant to the "application service comprising a plurality of logic components" and WAP/mobile "modality" aspects of claims 1/5/9. Published September 2000 — more than one year before the U.S. filing (2002-07-08) — so it qualifies as § 102(b) prior art as a printed publication. It does not disclose the device-switch last-operation-state recall or logic-component state monitoring, so no full anticipation; strong § 103 art for the modular-component premise.

2. G. Welling, M. Ott, G. Michelitsch (NEC Research Laboratories), "Session Mobility Support for Multimedia Applications," Proc. 1999 Multimedia Computing and Networking Conference (MMCN '99), SPIE vol. 3654, pp. 52–63, Jan. 25, 1999. (Also cited as presented Jan. 25–27, 1999, San Jose, CA; DOI record dated 1998-12-14.)
Description: Software architecture for multimedia services supporting session hand-off between terminal devices (or service providers), allowing operating parameters of session components to be altered as part of the hand-off, and rebinding a service provider transparently; a user interface simplifies initiating session mobility considering display-device capability.
§ 102 assessment: This is the most thematically on-point reference to the invention's stated aim — "keeping persistency … while switching between modalities/terminal devices." It discloses handing a session off from one terminal device to another and rebinding the service. It is a credible § 102 candidate against the general device-switch persistence concept of claim 1, but it does not appear to disclose (a) a user-history record of the last operation state keyed to a logic-component ID, nor (b) monitoring each logic component's state to compute an aggregate operation state. Therefore no full anticipation of claims 1/5/9, but it is likely the single most damaging prior art for § 103.

D. Summary of the § 102 picture

• No cited reference, alone, appears to anticipate independent claims 1, 5, or 9. Each lacks at least the "plurality of logic components independently generating respective execution results, each having a respective state … monitoring the respective states of each … to determine a latest operation state" limitation, and several also lack the two-device switch.
• Best single-reference (device-switch persistence): US 6,349,337 B1 (Microsoft) and Welling et al. (MMCN '99) — these map to the claim-1 "second device after first device / continue from last state" concept.
• Best single-reference (last-operation-state storage + ID): US 6,076,108 A (I2 global session server) and WO 2000/079391 A1 / US 6,539,494 (Art Technology Group session-backup with encoded session ID).
• Best single-reference (modular components / WAP modality): Cannataro et al., DEXA '00.
• The obviousness case against claims 1/5/9 would most plausibly combine US 6,349,337 (or Welling) + US 6,076,108 (or WO 2000/079391) + Cannataro.
• Dependent claims are the more realistic § 102 targets: claim 3 (contact method = the requesting method) and claim 4/13 (contact method selected from pre-registered methods) align with the "user register module" style teachings; claims 10–12 (device contact time-period registration, presenting to selected device) have no clean single-reference anticipation in the cited list; claims 14–16 (ID = serial number of logic component(s)) track the WO 2000/079391 session-ID encoding; claims 17–19 ("executed recursively until interrupt or finish") align with the recursive-execution background.
• One citation is plainly non-analogous: US 6,358,892 B1 (Chevron, polyalkylene succinimides) cannot anticipate any claim of a network-application-service patent and should be disregarded as an examiner-citation artifact.

Critical dates for § 102 purposes: earliest priority 2001-08-13; PCT filing 2002-07-08. A § 102(b) printed-publication reference generally must predate 2000-08-13 (one year before earliest priority) or 2001-07-08 (one year before U.S. filing), depending on the priority route; all cited references predate those dates. US 2005/0246415 A1, published 2005-11-03, post-dates the invention and is not § 102(b) art.

If you want, I can build a limitation-by-limitation (claim chart) mapping for any one of the three strongest references against claims 1, 5, and 9.

I'll ground the analysis in the cited references by verifying a few key ones.

Obviousness Analysis — US 7,925,763 B2

Patent: US 7,925,763 B2, "Keeping persistence while switching between modalities when accessing an application service in a network" (IBM; inventor Hong Cai). Earliest priority 2001‑08‑13 (CN 01124795.9); PCT filed 2002‑07‑08; granted 2011‑04‑12. Source: https://patents.google.com/patent/US7925763/en

Framework: Pre‑AIA 35 U.S.C. § 103(a); Graham v. John Deere; KSR Int'l v. Teleflex; MPEP 2141–2144 (including the enumerated rationales at MPEP 2143 and the "predictable results" / "known technique" / "obvious to try" lines). This is a technical analyst's assessment, not a legal opinion or a validity determination. Where I am inferring a reference's disclosure from its title/abstract rather than full-text, I flag it.

1. The claimed subject matter, decomposed

Claim 1 is the representative independent claim. Its limitations, labeled for charting:

Independent claim 5 mirrors claim 1 as apparatus; independent claim 9 recasts it as a four‑module system (user history module / logon control module / session management module / connection management module). Dependent claims 2–4, 6–8, 10–13 add contact‑routing features; claims 14–19 add (i) ID = serial number of the logic component(s), and (ii) recursive execution until interrupt or completion.

Important admission in the specification itself: "Nowadays, applications tend to become more and more modularized... An application may consist of a plurality of highly modularized independent components (hereinafter named application logic components)." The patent treats component modularization as known background art, not as its contribution. That admission materially narrows the field of novelty to the cross‑device persistence mechanism, not the existence of modular components.

2. The prior art of record, characterized

All dates below predate the 2001‑08‑13 priority date and therefore qualify as prior art (the specific subsection varies):

Cross‑device session persistence

• US 6,349,337 B1 (Microsoft; priority 1997‑11‑14; granted 2002‑02‑19) — https://patents.google.com/patent/US6349337 — "Maintaining a first session on a first computing device and subsequently connecting to the first session via different computing devices." It expressly identifies the problem the '763 patent addresses: a session "bound to a particular client machine" is not usable when "a user... may in fact log on to the server from different physical machines." The session manager maintains the first session, recognizes that the user is affiliated with the first session, and reassociates the new connection with the reconfigured first session so that the user "may continue with the first session while using the second computing device."
• US 6,076,108 A (i2 Technologies; filed 1998‑03‑06) — https://patents.google.com/patent/US6076108 (continuation US 6,480,894) — global session server stores a single set of session data representing a state of the user session, retrievable and updated on each subsequent request so state "accumulated over multiple web transactions" persists even when "the user does not interact with the same web server from transaction to transaction."
• US 6,826,755 B1 (Microsoft; filed 2000‑06‑30) — https://patents.google.com/patent/US6826755 — stores a user's context and, on a switch request, "determin[es] if the second user identity has been utilized previously" and if so re-points to the previously stored context; if not, creates a new one.
• WO 00/79391 A1 (Art Technology Group; 1999‑06‑17) — "Internet server session backup apparatus" (title/date per the patent's citation list; full text not verified here).

Session mobility across heterogeneous terminals

• Welling, Ott & Michelitsch, "Session Mobility Support for Multimedia Applications," Proc. MMCN '99, SPIE vol. 3654, pp. 52–63 (Jan. 25, 1999) — https://typeset.io/papers/session-mobility-support-for-multimedia-applications-2qou3spiek — the paper's thesis is that "maintaining the network connection of a mobile client" is insufficient and that "session hand-off between terminal devices or service providers is essential as well," with a software architecture that "allows operating parameters of all components of the session to be altered as part of the hand-off" and a UI that "simplif[ies] initiating session mobility, while taking into consideration display device capability." This is the single most damaging reference: it supplies the motivation and the cross‑device framing.

Modular / component‑based application services over wireless terminals

• Cannataro & Pascuzzi, "An Object-Based Architecture for WAP-Compliant Applications," DEXA '00, pp. 178–185 (Sept. 2000), DOI 10.1109/DEXA.2000.875023 — https://dl.acm.org/doi/10.5555/[861766](/patent/861766) — presents a general architecture for WAP‑compliant applications with business logic on an Application Server, software components integrating Internet and wireless contexts, and — critically — "a technique to handle client disconnection is presented." It also describes WML "card/deck" units of interaction and per‑terminal state variables.

Others of record (auxiliary): US 5,845,282 (Apple; remote access to a desktop from a PDA), US 6,779,205 (Nortel; executing applications on a target device via a proxy), US 6,499,054 (Senvid), US 6,331,855 (Expertcity), US 2002/0057678 A1 (wireless voice/data channel integration), US 2002/0075307 A1 (Vigilos), US 2003/0093459 A1 (Dowling), US 2005/0246415 A1 (Microsoft distributed computing services platform).

Not prior art: the "Cited By" publications (2011+ MITEL/ETRI, etc.) and the "Families Citing" list post‑date the '763 priority and cannot be combined against it. The CN 01124795.9 / CN 1407780A and PCT/GB02/03121 documents are family members, not prior art.

3. Combination analysis

Combination 1 (primary): US 6,349,337 + Welling + Cannataro

Why a POSITA would combine. Each reference is in the same field of endeavor (network‑delivered application services consumed from multiple/mobile terminals) and addresses the same problem — that a session bound to one device cannot be resumed on another. Welling and US 6,349,337 are practically companion pieces: US 6,349,337 supplies the server‑side mechanism (maintain, recognize, reassociate) while Welling supplies the explicit motivation to extend it beyond a single provider's homogeneous clients to heterogeneous terminal devices, and points to component‑level state as the unit of hand‑off. Cannataro supplies the well‑known architectural substrate (modular application logic components on an application server with client‑disconnection handling) that the patent itself admits was conventional. Combining them is "a combination of prior art elements according to known methods to yield predictable results" (MPEP 2143.A) — each element performs its own known function and the joint result (resume the same service on the new device) is the expected sum of those functions.

Combination 2: US 6,076,108 + US 6,826,755 (+ US 6,349,337)

This combination targets limitations A, B, and E head‑on and is the strongest answer to the "user history / last operation state" limitations:

• US 6,076,108 discloses a persistent store of session state keyed to the user session, retrieved on a subsequent request and updated after processing — i.e., the "user history" and "updating the last operation state" limitations (B, E) without reference to any device.
• US 6,826,755 supplies the per‑identity stored context plus the explicit "has been utilized previously" test — i.e., recalling a previously stored state rather than starting fresh, exactly the A/B distinction the '763 patent relies on.
• US 6,349,337 supplies the cross‑device trigger (same user, different client).

A POSITA seeking to let a user resume work started on another terminal would be motivated to combine a cross‑device session mechanism (US 6,349,337) with a device‑agnostic persistent state store (US 6,076,108) and a per‑identity "previously used?" lookup (US 6,826,755): the first answers "can the same user get back into the session?", the second answers "where does the state live?", and the third answers "is there prior state to restore?" The combination yields nothing more than the predictable aggregation of these known functions.

Combination 3 (for the "logic‑component / serial‑number" limitations, claims 1 B/D/F and 14–17): Cannataro + Welling + US 6,076,108

The patent's "call ID... corresponding to the serial number of each application logic component" is the feature most likely argued as the point of novelty. It is nonetheless an obvious design choice over the cited art: once an application is known to be a sequence of addressable logic components (Cannataro's object‑based architecture; the patent's own background admission), and once a system stores and restores session state (US 6,076,108; US 6,349,337), indexing the saved state by the component identifier is the natural, and arguably the only sensible, way to record where execution stopped. Welling's "operating parameters of all components of the session" supplies the notion that component‑level state collectively defines session state (limitation F). This is "use of a known technique to improve a similar device in the same way" (MPEP 2143.C).

Combination 4 (system claim 9): US 6,076,108 + WO 00/79391 + US 6,349,337 + Cannataro

Claim 9's module decomposition (user history module; logon control module; session management module; connection management module) maps naturally onto: a persistent session‑state store (US 6,076,108's global session server / WO 00/79391's session backup apparatus), a logon/identity‑verification front end (US 6,349,337's "logging on the user" and session recognition), and a connection/session manager that assigns protocol stacks and reassociates connections (US 6,349,337's multi‑user session manager and stack protocol manager). Arranging these functions as named modules is a routine engineering decomposition of the same functionality and does not impart patentable weight.

4. Dependent claims

• Claims 3, 7, 12 (return result via the same communication path used for the request) — squarely disclosed/obvious over US 6,349,337, which returns the user to the session over the newly established connection from the second client ("returns the result to the... Client" in the patent's own synchronous cases A and B).
• Claims 4, 8, 13 (contact method selected from a plurality pre‑registered) — obvious over multi‑client/multi‑profile art: US 6,349,337 maintains the same session across two different clients, and US 6,826,755 stores per‑identity configuration (cookies/history/user‑defined containers) that can be pointed to selectively.
• Claims 14–16 (ID = serial number of the logic component(s)) — see Combination 3.
• Claims 17–19 (recursive execution until interrupt or completion) — the patent itself frames this as the ordinary execution model of modularized applications ("execut[ing] all the independent application logic components one by one"), and Cannataro's component architecture executes components against a server‑side process that terminates on disconnection/completion. Recursion here is a conventional control‑flow choice with a predictable result.
• Claims 2, 6, 10, 11 (return via a "proper" contact method associated with a pre‑registered device used during a pre‑defined time period; user registrar module storing a device contact time period) — This is the weakest link in the prior art of record. The cited references establish pre‑registered devices and multiple contact methods, but I do not see, in the references cited on this page, a clear disclosure of time‑of‑day / time‑period‑conditioned selection of the delivery device (the specification's "800 AM–5:00 PM, telephone...; other time, mobile phone..."). Note, however, that the claim language is not limited to time‑of‑day and could read on any scheduling/presence‑based routing; and the '763 specification's own example ("if not found, try... Instant Message; if still not found, and the service priority is high, contact the user's manager") is an ordinary presence/escalation workflow of the kind known in unified‑messaging art. I would be cautious about asserting per se obviousness of claims 2/6/10/11 on these four references alone.

5. Rebuttal considerations / where non‑obviousness might be argued

1. No single reference teaches all of A–F, and the "user history" as a device‑independent, per‑user/per‑service persisted record that is itself updated is the point the applicant would press. The counter is that US 6,076,108 and US 6,826,755 together disclose exactly a persisted, retrievable, updated state store indexed to the user.
2. The time‑period contact routing (claims 2, 6, 10, 11) is the most credible non‑obviousness position on the cited record, as noted above.
3. Secondary considerations: nothing in the file history summary on the page evidences unexpected results, licensing, or industry praise tied to these limitations, and the patent lapsed for non‑payment of maintenance fees on 2015‑04‑12 (Expired – Fee Related). No nexus evidence is apparent from the materials provided.
4. Prior‑art status caveats: WO 00/79391 A1 and several of the auxiliary references are cited by title only in the patent's list; my characterization of their disclosures is drawn from titles/dates, not verified full text. Any § 103 conclusion resting on those references should be confirmed against the actual documents.

6. Bottom line

On the prior art cited on the face of US 7,925,763, claim 1 (and its apparatus/system counterparts, claims 5 and 9) would have been obvious under § 103 as a combination of US 6,349,337 (session maintained and reassociated across two different clients), US 6,076,108 (persistent, device‑independent session state stored, recalled, and updated on each request), and Cannataro & Pascuzzi 2000 (modular application‑logic components on an application server with client‑disconnection handling), with Welling et al. 1999 supplying the express motivation to perform session hand‑off between heterogeneous terminal devices and to treat component‑level parameters as the session state. The specific "recall last state by a component ID/serial number" and "monitor each component's state" limitations are the predictable implementation of that combination over known modular architectures. Dependent claims 3, 4, 7, 8, 12–19 fall with the independent claims. Claims 2, 6, 10, and 11 are the least clearly anticipated/obvious on this particular set of references, because the cited art does not plainly show device selection keyed to a pre‑defined contact time period.

Primary sources: https://patents.google.com/patent/US7925763/en · https://patents.google.com/patent/US6349337 · https://patents.google.com/patent/US6076108 · https://patents.google.com/patent/US6826755 · https://typeset.io/papers/session-mobility-support-for-multimedia-applications-2qou3spiek · https://dl.acm.org/doi/10.5555/861766

This is an analyst's technical assessment based on the documents provided; it is not a legal opinion and may not account for claim construction, priority/derivation issues, or prior art outside the cited record.