US 8832654

I'll search for authoritative information on this patent, including any litigation/docket activity.

The second result is unrelated (an Oasis Tooling § 101 petition). Let me run more targeted searches.

US Patent 8,832,654 — Analyst Summary

Source & verification note

I retrieved the patent from the authoritative full text at patents.google.com and cross-checked against the public record (Justia, patentimages PDF). Independent searches of Unisys patent litigation and Federal Circuit activity returned no CAFC 2026 docket and no district‑court or IPR proceeding that I could confirm involves US 8,832,654. I report that as a negative finding, not as proof of non-existence — PACER/CAFC docket coverage in my search results is incomplete.

(Minor housekeeping: the task prompt states "current date April 26, 2026," while my operating environment date is 2026‑09‑30. Searches were run on the latter date; nothing below turns on the discrepancy.)

---

Bibliographic data (literal, no auto-correction)

Security interests / assignments of record (do not change the assignee): General Electric Capital Corporation as agent (2011-06-27; again 2014-11-04), Wells Fargo Bank as collateral trustee (2017-04-27), JPMorgan Chase as administrative agent (2017-10-06), and Computershare Trust Company, N.A. as collateral trustee under an amended and restated patent security agreement (2025-06-27, Reel/Frame 071759/0527). Several of these were released (reels 030004/0619, 030082/0545, 044416/0358, 054231/0496).

---

Abstract (as granted)

The disclosure relates generally to autonomous generation of software documentation, and more specifically to autonomous generation of documentation that describes user-defined rules implemented for a computer management system. In one embodiment, a system for generating documentation comprises a database storing user-defined rules that specify management actions to be triggered, for managing a computer system, responsive to receipt of a system message from the managed computer system. The system further comprises a documentation generation device configured to autonomously generate documentation describing the rules, wherein the documentation contains identification of at least the system message and the corresponding management actions triggered by the rules upon receipt of the system message. In certain embodiments, the system further comprises a database utility device configured to generate a highly-structured representation of the user-defined rules from an autoaction database, and the documentation generation device processes the highly-structured representation to autonomously generate the documentation.

---

Plain-language overview of the independent claims

Claim 1 — Method

A method performed by a "documentation generator device":

1. Receive user-defined rules that specify management actions to fire in response to an event, for managing a computer system.
2. Process those rules to produce documentation describing them.
3. Wherein the documentation generator comprises an autoaction database containing patterns that specify messages to be identified plus the corresponding autoactions to be taken.
4. Wherein the generated documentation covers (a) the user-defined rules including the message-matching criteria used to compare patterns against messages, and (b) the corresponding autoactions, which are limited to three named types: raising alerts to a console of the computer management system, activating paging devices external to the management system, and generating cross-system commands that may be sent to another managed computer management system.

Practical read: this is the "auto-document your message-matching automation rules" claim. The three enumerated autoaction types in the final wherein clause are a narrowing limitation — a system that documents only, say, database-activation actions would arguably not read on claim 1 as granted.

Claim 9 — Computer program product

A non-transitory computer-readable medium carrying program logic that mirrors claim 1: (a) code for receiving user-defined rules specifying management actions triggered by an event, and (b) code for generating, from processing those rules, documentation describing them — bounded by the same autoaction-database / pattern / message-matching-criteria limitations and the same three autoaction types (console alerts, external paging devices, cross-system commands).

Practical read: substantively claim 1 recast in product form; it stands or falls with claim 1.

Claim 16 — Computing system

A computing system having a programmable processor, memory and mass storage, comprising:

1. A database storing user-defined rules that specify management actions to be triggered, for managing a computer system, responsive to receipt of a system message from the managed computer system.
2. A documentation generation device configured to autonomously generate documentation describing those rules, where the documentation identifies at least (i) the system message and (ii) the corresponding management actions triggered on receipt of that message.
3. Bounded again by the autoaction-database/pattern limitation, the message-matching-criteria element, and the three enumerated autoaction types.

Practical read: the apparatus counterpart, and the claim with the clearest "autonomous generation" (no manual authoring) and "system message → action" mapping language.

Dependent claims in brief: 2/10 add "identify the event and the actions triggered"; 3/11 add the highly-structured-representation step (the DADB dump); 4/12 specify the event as a system message; 5–6/13–14 add storing to a computer-readable medium/file; 7/15 add outputting the documentation; 8 recites the generator as software on at least one processor; 17 adds a database utility device generating the highly-structured representation from the stored rules.

---

Prosecution observation (relevant to scope)

Comparing the granted claims with the 2011 pre-grant publication (US 2011/0307863 A1), the granted claim 1 is narrower: the published claim 1 ended at "processing … for generating documentation describing the rules," with no autoaction-database or enumerated-autoaction limitations. In the granted version, the "wherein the documentation generator device comprises an autoaction database …" and the three-autoaction-type clause were added. The published application also contained means-plus-function system claims 18–20 ("means for receiving," "means for autonomously generating," "means for outputting") that do not appear in the granted patent — consistent with cancellation during prosecution. Anyone assessing infringement or validity should read the claims as granted, not the published version.

---

Grounding sources

• Full patent text: https://patents.google.com/patent/US8832654/en
• Grant PDF: https://patentimages.storage.googleapis.com/46/d1/1f/484d4f28be1ceb/US8832654.pdf
• Justia claims page: https://patents.justia.com/patent/8832654
• Pre-grant publication: https://patents.justia.com/patent/20110307863
• Patent term / maintenance data: as reflected in the Google Patents legal events for US 12/815,530

Explicit uncertainties

• No litigation found. Searches surfaced Unisys-adjacent patent cases (e.g., Patent Armory Inc. v. Unisys Corp., N.D. Tex. 3:25-cv-02115, voluntarily dismissed Aug. 2025; MCOM IP LLC v. Unisys Corp., N.D. Tex. 3:21-cv-02288), but these involve third-party patents asserted against Unisys, and I could not confirm that US 8,832,654 appears in any of them. I found no CAFC appeal, IPR, or PGR involving US 8,832,654.
• I cannot reliably confirm the current status of the 2025 Computershare security agreement or any 2026 fee events beyond what the legal-events table shows; treat fee/expiration status as a snapshot, not legal advice.
• CPC/US classification values come from third-party listings and may not reflect every class on the printed front page.

I'll search for litigation involving US patent 8832654 across patent litigation databases.

The initial searches returned irrelevant results (unrelated '265 patents). Let me search more targeted litigation sources.

Based on my searches of patent litigation sources, I was unable to locate any litigation involving U.S. Patent No. 8,832,654.

Findings

No known litigation involving US Patent 8,832,654.

I searched multiple litigation-oriented sources and general web results for the specific number "8832654" combined with litigation terms. The results returned:

• Google Patents / Justia / FreePatentsOnline entries — these describe the patent itself (title: "System and method for autonomous generation of software documentation"; inventor James Malnati; assignee Unisys Corporation; filed 2010-06-15; granted 2014-09-09). No litigation information is associated.
• Unrelated '265 patents — searches surfaced several other patents ending in "265" (e.g., a NEC '265 patent asserted against Dell; Alcon's '265 patent in an ANDA case; Maxell/Samsung '086 etc.). Per your strict instruction, I did not treat these as the same patent; none involves 8,832,654.
• Unisys litigation disclosures (10-K/10-Q filings) — Unisys's own litigation and contingency disclosures list securities, contract, environmental, and employment matters (e.g., Belgian Ministry of Justice, Lufthansa, Molina, a stockholder class action), but no patent assertion or defense involving 8,832,654.
• The patent's "Cited By" and "Citations" lists on Google Patents show only patent-prosecution citations (e.g., US7302674, US7404177, US7565643, US7657866, US7913228) and later citing documents — this is not litigation.
• Unified Patents PTAB caselist / portal — I did not find any PTAB proceeding (IPR/PGR) challenging 8,832,654.

Important caveats

• I could not access a direct, authoritative docket-level record (e.g., Unified Patents litigation caselist, PACER/Docket Navigator) for this specific patent. My conclusion is based on the absence of any hit in the searches performed, not on a definitive registry query.
• Public litigation databases can be incomplete, especially for older patents or sealed/non-electronic-filed cases.
• The patent's legal status shows it is active (adjusted expiration 2031-07-05), and it has been used as collateral in Unisys security/collateral agreements (GE Capital, Wells Fargo, JPMorgan, Computershare Trust Company) — these are financing events, not infringement litigation.

Recommended verification steps

If you need a definitive answer, the following sources should be queried directly:

• Unified Patents Litigation Portal: https://portal.unifiedpatents.com/litigation/caselist
• PACER (search party "Unisys Corporation" combined with the patent number)
• Docket Navigator (patent-level litigation lookup for 8,832,654)
• CAFC opinions search: https://www.cafc.uscourts.gov/opinions-orders (no result surfaced here)

Bottom line: On the record currently available, there is no known litigation (district court or Federal Circuit) involving U.S. Patent No. 8,832,654.

Proceedings overview

Total AIA trial proceedings on file for US 8,832,654: 0 (zero). The structured PTAB data block in this prompt — sourced from the USPTO Open Data Portal — lists no IPR, PGR, or CBM petitions against this patent, and my independent web checks (Google Patents «PTAB»/litigation panels, PTAB E2E/PTACTS petition text, Docket Alarm and CourtListener hits, and Law360/NatLawReview style PTO filing roundups) surfaced nothing naming either 8,832,654 or application 12/815,530 as the subject of a Board proceeding.[^1][^2] The breakdown is therefore: 0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution denials. Defensive posture bottom line: no help from the Board — this patent has never been tested in an AIA trial, all 17 claims are presumptively valid and in force, and a defendant is looking at a clean-slate fight. The absence of PTAB activity is not a sign of weakness in the patent; it is a sign that nobody has found it worth attacking (see "Pattern signals" below).

Because there are no proceedings to serialize, I have replaced the per-proceeding section with a short evidentiary record of the negative finding, followed by the strategic analysis that a defendant actually needs.

---

No proceedings — evidentiary basis

• Type: N/A (no IPR / PGR / CBM filed)
• Filed: N/A
• Status: N/A — no petition has ever been docketed against this patent in the sources searched as of 2026-09-30
• Judge panel: N/A
• Petition grounds: N/A
• Institution decision: N/A
• Final Written Decision: N/A
• Settlement / termination: N/A
• Appeal: N/A — no FWD exists, so nothing has been appealed to the Federal Circuit; the patent owner (Unisys Corporation) appears nowhere as an appellant on this patent
• Defensive value: Neutral-to-negative for a defendant. There is no IPR record to mine for the patent owner's admissions, no Board claim construction, and no § 315(e)(2) estoppel to exploit. Any invalidity attack would be first-in-time.

Caveats I want to be explicit about (per the operating rules — I do not fabricate proceeding numbers):

1. I did not find any PTAB proceeding, and I am not asserting one exists. I also cannot prove a universal negative; the ODP ingest is the canonical source here and it says zero.
2. "No proceeding found" is not the same as "no litigation." I did not locate a public infringement suit asserting 8,832,654 in the searches run, but I did not run a full district-court docket sweep, and a negative search on litigation is much weaker than the ODP-based negative on PTAB trials.
3. The patent's prosecution-citation list (US 7,304,674; US 7,404,177; US 7,565,643; US 7,657,866; US 7,913,228) and its "Cited By" list are not proceedings — those are patents/publications only, and should not be confused with an IPR record.

---

Strategic summary

Claim status: all 17 claims are UNTESTED and, subject to maintenance-fee payment, all remain in force.
Nothing has been narrowed by IPR. Independent claims are 1 (method), 9 (computer program product), and 16 (computing system), with dependents 2–8, 10–15, and 17. The Google Patents legal-status panel lists the patent as Active, with an adjusted expiration of 2031-07-05, which is later than the nominal 20-year term from the 2010-06-15 filing date and reflects PTA. Practical implication: this is a live, assertable patent for roughly another five years on the patentee's own term calculation, and there is no claim that a defendant can write off as already-canceled.

Two claim-level observations worth carrying into a defense, drawn from the claim text itself rather than from any proceeding:

• The independent claims recite the documentation-generation function broadly and functionally ("processing … the received user-defined rules for generating documentation describing the rules"), with the "autoaction database / patterns / message-matching criteria" elements pulled in via a "wherein" clause. That breadth is a double-edged sword: it makes infringement easier to prove over a range of log/rule-documentation products, but it also makes the claims vulnerable under § 103 to rule-engine + template/report-generator combinations. A prior-art search aimed at the 2005–2010 window for "autoaction database → symbolic dump → templated human-readable documentation" is the highest-value work product here.
• Claim 1's wherein clause bundles three disparate action categories — raising console alerts, activating external paging devices, and generating cross-system commands. A single prior-art reference or combination that teaches templated documentation of a rules database plus any one of those action types is a serious § 103 candidate. The patent's own incorporated-by-reference siblings (Ser. Nos. 12/644,517 and 12/637,928, on object alerts and audible alerts) may themselves be relevant prior-art-adjacent material depending on their publication dates.

Estoppel landscape: empty.
Because no petitioner has ever appeared before the Board against this patent, no § 315(e)(2) estoppel attaches to anyone — not to a hypothetical prior petitioner (there is none), and not to their privies. For a defendant being asserted against today, that is the good news: every prior-art ground a petitioner "raised or reasonably could have raised" is still on the table, unconstrained by any estoppel arising from a prior IPR. The only estoppel-relevant clock you face is your own: § 315(b) bars an IPR petition filed more than one year after you are served with a complaint alleging infringement of the patent. If a demand letter has arrived but no complaint has been served, your clock has not started; if a complaint has been served, count backward from the service date immediately and calendar the petition deadline. Note also that a prior ex parte reexamination request (a third-party prosecution-side vehicle, not an AIA trial) would not create § 315(e)(2) estoppel, and I found no indication one was filed either.

Pattern signals: no serial petitions, no aggregator, no appellee/appellant activity.

• Same-petitioner repeats: none — there is no petitioner at all.
• Defensive aggregator (Unified Patents, RPX, etc.): I found no evidence that a defensive aggregator has ever challenged this patent or taken it into a member-defense program. The assignee chain is the patent owner itself (Unisys Corporation, with a long line of security-interest/release assignments to General Electric Capital, Wells Fargo, JPMorgan Chase, and most recently Computershare Trust Company as collateral trustee as of 2025-06-27). Those are financing liens, not assertions or challenges — do not mistake the collateral-trustee assignments for a defensive program.
• Patent owner appeal aggressiveness: inapplicable — Unisys has never had an adverse Board decision on this patent to appeal.
• Maintenance-fee status as a signal: the record shows the 4th-year fee paid 2018-03-09 and the 8th-year fee paid 2022-03-09, and a maintenance-fee reminder mailed 2026-04-27 (large-entity reminder for the next, 12th-year, fee). Verify payment. If the 12th-year fee is not paid within the six-month grace window, the patent will lapse for non-payment well before the listed 2031-07-05 adjusted expiration — a materially better outcome for a defendant than winning an IPR, and far cheaper to monitor.

---

Recommended next steps

If you are a defendant and are hoping for a canceled-claim shortcut: there isn't one.
Do not write a demand-letter response that assumes any claim of 8,832,654 has been canceled. There is no Final Written Decision to cite and no PTAB E2E record to link, because none exists. Any argument built on "IPR-validated" or "IPR-invalidated" claims for this patent would be fabricated; don't make it.

If you are contemplating an IPR: you would be the first petitioner, which is an advantage.

• Grounds available: § 102 and § 103 on patents and printed publications only (§ 311(b) limits IPR to those). § 112 and § 101 are not available in an IPR.
• PGR is unavailable: the application was filed 2010-06-15, a pre-AIA effective filing date, so first-inventor-to-file PGR review under § 321 does not apply.
• CBM is unavailable: the claimed subject matter (generating human-readable documentation from a system-management rule database) is not a "covered business method" tied to financial products or services, and the CBM program sunset on 2020-09-16 in any event.
• Therefore IPR under §§ 311–319 is your only AIA trial route. Because no earlier petition exists, there is no General Plastic serial-petition exposure and no Fintiv-plus-estoppel stacking against you — but do not waste the one-shot filing; the § 315(e)(2) estoppel your own IPR creates is real and will bar you in the district court on any ground you raised or reasonably could have raised.
• Watch your § 315(b) clock from the date you are served with a complaint. If service has already occurred and more than a year has passed, the IPR door is closed for you and your invalidity case must live in the district court (where § 101 and § 112 arguments remain available).

Trial-stage milestones: not applicable — with zero petitions, there is no institution-decision deadline, no 12-month statutory FWD deadline, and no oral hearing to track. If you file, the standard timeline applies: a § 314(b) institution decision is due within six months of the petition's filing date, and a final written decision under § 318(a) is due within twelve months of institution (extendable to eighteen months for good cause).

Monitor the maintenance fee. Set a docket reminder around the 2026-04-27 reminder notice / the 12th-year fee window. A lapse for non-payment would end this patent's enforceability at the source; that is a cheaper and more certain outcome for an accused infringer than any AIA trial.

If you want to double-check the negative: the authoritative places to confirm are USPTO Patent Trial and Appeal Board E2E / PTACTS (https://ptacts.uspto.gov/ptacts/), the Google Patents "Litigation"/trial panels for the patent (https://patents.google.com/patent/US8832654/en), and CourtListener's docket search for any Federal Circuit appeal (https://www.courtlistener.com/?q=%228832654%22&type=r&order_by=score+desc). If a proceeding has been filed since the ODP ingest and has not yet propagated, that is the triangulation path; a PTAB E2E search by patent number is the definitive check and should be re-run before you commit to a defense theory that depends on the patent being untested.

[^1]: Google Patents, US8832654B2, https://patents.google.com/patent/US8832654/en — shows no "PTAB" trial entries; the "Cited By" list (e.g., US20140173562A1 → SAP's US9069646B2, and the Unisys "automation degree of implementation metrics" family US20160323207/08/09A1) are subsequent patents, not proceedings.
[^2]: PTAB E2E/PTACTS petition text search for the patent number and for the title "System and method for autonomous generation of software documentation" returned no petition; Docket Alarm and CourtListener queries for the same returned no matching trial or appeal.

I'll verify the assignment record and ownership context with targeted searches before writing up.

Analyst Note on Sources and a Data Gap

I reconstructed the chain from the Google Patents legal‑events table for US 12/815,530 (which mirrors the USPTO assignment records by reel/frame), cross‑checked against the Unisys 8‑K of 2025‑06‑27 (EDGAR accession 0001104659‑25‑063658) and Unisys IR filings.

Material limitation — correspondents: the task asks specifically for the correspondent of record on each recording. I could not retrieve correspondent names for any reel/frame in this chain. Google Patents legal‑events tables do not expose the correspondent field, and my searches on the reel/frame numbers (including 025174/0557) returned no indexed assignment‑summary pages. I therefore mark the correspondent field not retrievable from available sources rather than inventing names. Anyone verifying should pull each reel/frame at https://assignmentcenter.uspto.gov/ and read the "Correspondent" block directly. Given the chain is entirely financing‑collateral work by large institutional lenders, I would expect the correspondents to be lender counsel and Unisys in‑house/outside IP counsel — not NPE repeat players — but that is an expectation, not a finding.

Second caveat: the USPTO data I can see is a snapshot. Legal‑events tables lag recordings by weeks; treat dates below as recorded data, not as live docket truth.

---

Inventors

Unusual‑pattern check — no adverse pattern found. The classic fire‑sale precursor (all inventors leaving the assignee within 12 months of filing) is absent:

• Malnati is a serial Unisys inventor whose name recurs on Unisys patents filed well after 2010 — e.g., US 2011/0307904 A1 ("Method and apparatus for automation language extension," filed 2010‑06‑14, one day before this filing, also Malnati), and the 2015‑filed automation‑metrics family (US 2016/0323207–0323209, US 9,667,573, US 9,686,220, US 10,153,992) where his assignment to Unisys was executed 2015‑04‑30 with co‑inventors Robert Jamieson and John Troini (recorded Reel 036060/0910 and 043099/0843).
• He was still signing Unisys assignments five years after this filing. There is no departure signal and no evidence of a personal‑portfolio carve‑out.
• Sole inventorship also means no co‑inventor splits that could create later co‑ownership encumbrances. The chain is clean on the inventor side.

---

Original assignee

• Entity on the issued patent: Unisys Corporation, a Delaware corporation, Blue Bell, Pennsylvania.
• Recorded chain of title: the original assignee recorded 2010‑10‑21 (Reel 025174/0557) remains the current assignee today. Google Patents lists "Current Assignee: Unisys Corp." and no post‑issuance transfer of title appears anywhere in the record.
• Product embodying the claims — yes. Unisys shipped the product the specification is written about: Operations Sentinel, whose SP‑AMS (for MCP/UNIX/Linux partitions) and CP‑AMS (for OS 2200 partitions on ClearPath Plus) autoaction databases implement the pattern/message‑match/autoaction model the claims recite. The specification expressly identifies these as Unisys products and incorporates the Operations Sentinel Autoaction Message System Administration Guide (7862 6900‑006, Sept. 2008) by reference. The claimed documentation generator is the DADB‑dump → documentation post‑processor described in the spec.
• Primary line of business: IT services and enterprise computing solutions — digital workplace solutions, cloud & infrastructure, and enterprise computing (ClearPath forward). Publicly traded (NYSE: UIS).
• Current status: operating, not bankrupt, but financially stressed. Revenue ~$1.98B (2022) and ~$2.02B (2023); 2022 net loss attributable to Unisys of $106.0M after a $448.5M loss in 2021 driven largely by $499.4M of pension settlement losses; material global pension deficits (US qualified plans ~$366M deficit at YE2022); a $54.9M restructuring charge in 2022 including $35.8M of asset impairments. On 2025‑06‑27 Unisys issued $700M of 10.625% senior secured notes due 2031, secured by substantially all assets including IP. That is a genuine operating company under leverage pressure — relevant context, but it is secured borrowing, not a portfolio sale.

No change‑of‑name, merger, or license record appears against this patent.

---

Assignment timeline

Chronological, all entries from the recorded assignment data for US 12/815,530 / US 8,832,654. Note the distinction that matters most: exactly one of these ten recordings conveys title. The other nine are liens, lien releases, or lien confirmations.

1. 2010‑10‑08 (executed) / recorded 2010‑10‑21 — Reel 025174/0557

   • Conveyance: Assignment (ASSIGNMENT OF ASSIGNORS INTEREST)
   • Assignor: James Malnati (MALNATI, JAMES)
   • Assignee: Unisys Corporation (Pennsylvania)
   • Correspondent: not retrievable from available sources (see note above)
   • Context: standard inventor‑to‑employer assignment — the only title transfer in the entire chain.

2. 2011‑06‑23 (executed) / recorded 2011‑06‑27 — Reel 026509/0001

   • Conveyance: Security Agreement
   • Assignor: Unisys Corporation
   • Assignee: General Electric Capital Corporation, as Agent (Illinois)
   • Correspondent: not retrievable
   • Context: securitization — Unisys grant of a security interest in its patent portfolio to GE Capital as collateral agent under a credit facility. Not a transfer of ownership.

3. 2012‑11‑27 (executed) / recorded 2013‑03‑15 — Reel 030004/0619

   • Conveyance: Release By Secured Party
   • Assignor/releasing party: Deutsche Bank Trust Company
   • Assignee/released party: Unisys Corporation (Pennsylvania)
   • Correspondent: not retrievable
   • Context: lien release — clearing an earlier Deutsche Bank collateral position. Title unchanged.

4. 2012‑11‑27 (executed) / recorded 2013‑03‑26 — Reel 030082/0545

   • Conveyance: Release By Secured Party
   • Assignor/releasing party: Deutsche Bank Trust Company Americas, as Collateral Trustee
   • Assignee/released party: Unisys Corporation (Pennsylvania)
   • Correspondent: not retrievable
   • Context: lien release — second Deutsche Bank collateral release, same effective date. Title unchanged.

5. 2014‑10‑31 (executed) / recorded 2014‑11‑04 — Reel 034096/0984

   • Conveyance: Security Interest
   • Assignor: Unisys Corporation
   • Assignee: General Electric Capital Corporation, as Agent (Nebraska)
   • Correspondent: not retrievable
   • Context: securitization — GE Capital collateral position re‑taken/confirmed shortly after the 2014‑09‑09 grant. Title unchanged.

6. 2017‑04‑17 (executed) / recorded 2017‑04‑27 — Reel 042354/0001

   • Conveyance: Patent Security Agreement
   • Assignor: Unisys Corporation
   • Assignee: Wells Fargo Bank, National Association, as Collateral Trustee
   • Correspondent: not retrievable
   • Context: securitization — collateral trustee migration to Wells Fargo, tied to the October 2020 collateral trust architecture (Wells Fargo appears as original Collateral Trustee under the Collateral Trust Agreement ultimately dated 2020‑10‑29 in the Unisys IR filings).

7. 2017‑10‑05 (executed) / recorded 2017‑10‑06 — Reel 044144/0081

   • Conveyance: Security Interest
   • Assignor: Unisys Corporation
   • Assignee: [JPMorgan Chase Bank, N.A., as Administrative Agent](/asserters/jpmorgan-chase-bank-n-a-as-administrative-agent) (Illinois)
   • Correspondent: not retrievable
   • Context: securitization — ABL/administrative‑agent collateral layer, layered alongside the Wells Fargo collateral‑trustee layer.

8. 2017‑10‑05 (executed) / recorded 2017‑11‑09 — Reel 044416/0358

   • Conveyance: Release By Secured Party
   • Assignor/releasing party: Wells Fargo Bank, N.A. (successor to General Electric Capital Corporation)
   • Assignee/released party: Unisys Corporation (Pennsylvania)
   • Correspondent: not retrievable
   • Context: lien release — closing out the GE Capital‑era security agreement upon the 2017 refinancing. Title unchanged.

9. 2020‑03‑19 (executed) / recorded 2020‑10‑28 — Reel 054231/0496

   • Conveyance: Release By Secured Party
   • Assignor/releasing party: Wells Fargo Bank, National Association
   • Assignee/released party: Unisys Corporation (Pennsylvania)
   • Correspondent: not retrievable
   • Context: lien release — clearing the 2017 Wells Fargo position in connection with the October 2020 refinancing. Title unchanged.

10. 2025‑06‑27 (executed) / recorded 2025‑06‑27 — Reel 071759/0527

    • Conveyance: Amended And Restated Patent Security Agreement (collateral trustee appointment)
    • Assignors: Unisys Corporation; Unisys Holding Corporation; Unisys NPL, Inc.; Unisys AP Investment Company I; and others (grantors)
    • Assignee: Computershare Trust Company, N.A., as Collateral Trustee (Minnesota) — successor in interest to Wells Fargo Bank, N.A.
    • Correspondent: not retrievable
    • Context: securitization — the entire collateral package re‑pledged to Computershare as collateral trustee securing the $700M 10.625% senior secured notes due 2031. Note this recording sweeps in affiliate entities as co‑grantors; it is a joint collateral grant across the Unisys group, which is normal in a secured‑notes financing and is not evidence of a spinoff or asset sale. Title unchanged — Unisys still owns the patent.

Finding on the record as a whole: there have been zero post‑issuance conveyances of title to this patent. Every post‑2010 recording is a lien, a lien release, or a lien re‑grant between Unisys and an institutional lender. The patent has never left Unisys.

Maintenance‑status flag (adjacent but important): the 4th‑year fee was paid 2018‑03‑09 and the 8th‑year fee 2022‑03‑09. The legal‑events table shows a 12th‑year maintenance fee reminder mailed 2026‑04‑27 with no corresponding payment entry as of this snapshot. The 12th‑year window opened 2026‑03‑09 and the unextended due date has now passed (2026‑09‑09); payment is possible with surcharge in the grace period. If the fee is not paid, the patent will lapse — which is a real ownership‑value event and one worth re‑checking at USPTO Patent Center / fee‑payment records before relying on this patent as live collateral. I cannot confirm payment or non‑payment from the indexed sources; treat this as a flag, not a conclusion.

---

Timeline diagram

timeline
    title Ownership of US 8832654
    2010 : Filed by Unisys
         : Malnati assigns to Unisys reel 025174 0557
    2011 : GE Capital security agreement
    2013 : Deutsche Bank liens released
    2014 : Patent issued
         : GE Capital security interest re taken
    2017 : Wells Fargo collateral trustee lien
         : JPMorgan agent lien
         : GE Capital era lien released
    2020 : Wells Fargo lien released
    2025 : Computershare collateral trustee
    2026 : 12th year fee reminder mailed

---

NPE / troll‑pattern signals

1. Shell‑entity transfer — NOT PRESENT. No record shows a transfer to any entity with an "IP / Patents / Licensing / Holdings / Ventures" suffix. The only assignees appearing anywhere in the chain are Unisys Corporation and four institutional financial counterparties (GE Capital, Deutsche Bank, Wells Fargo, JPMorgan, Computershare) — all acting as secured parties/collateral agents, not as owners taking title. The final assignee of record for title purposes is still Unisys (Reel 025174/0557, plus the "Current Assignee: Unisys Corp." designation in the patent record). One nuance worth not over‑reading: the 2025 Reel 071759/0527 recording names Unisys AP Investment Company I, Unisys NPL, Inc., and Unisys Holding Corporation as co‑grantors. Those are Unisys group affiliates pledging collateral — a normal secured‑financing joinder, and they appear as grantors to the lender, not as assignees of this patent. Naming alone is not a finding; the instrument type here is dispositive.

2. Known asserter in the chain — NOT PRESENT. No assignee or grantor matches any public NPE list. I checked the chain against the named universe (Acacia, Marathon, Intellectual Ventures, IPNav, Wi‑LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities) — no match. Note the superficial near‑miss worth explicitly ruling out: "Document Generation Corp" appears on NPE lists and this patent is about document generation, but there is no corporate, assignee, or assignment‑record connection whatsoever. That is a coincidence of subject matter, not a link. Similarly, no RPX or Unified Patents asserter‑directory entry surfaced for Unisys or for any entity in this chain.

3. Repeat correspondent across the chain — UNKNOWN / NOT ESTABLISHED. This is the signal the task most wants, and the honest answer is that it cannot be scored with the data I retrieved. No correspondent field was available for any of the ten recordings. What I can say structurally: nine of ten recordings are lender‑side collateral instruments, which in Unisys's financing history are handled by the collateral trustee's counsel and the company's financing counsel, and the lone title assignment (Reel 025174/0557) is an inventor assignment typically prepared by Unisys in‑house or its then‑outside prosecution firm. I found no single appearing attorney to flag. A single appearance would not be a finding anyway — the signal is recurrence, and I have no recurrence data. Recommend pulling the correspondent block on all ten reel/frame entries at Assignment Center before drawing any conclusion here.

4. Cascading transfers — NOT PRESENT. There are no consecutive transfers of title at all — one assignment in sixteen years. The apparent clustering of 2017 recordings (Reels 044144/0081, 044416/0358, 042354/0001) is a single simultaneous refinancing — new ABL/administrative‑agent lien in, collateral‑trustee lien in, prior GE‑era lien out — not a chain through intermediate holders. The same is true of the paired 2012‑11‑27‑executed Deutsche Bank releases (Reels 030004/0619 and 030082/0545), which are two trustee layers released on one effective date. Same instrument family, same beneficiary class, no intermediate assignee ever takes title. This signal requires title moving through chained LLCs; here title never moves at all.

5. Pre‑litigation transfer — NOT PRESENT. I found no infringement suit naming US 8,832,654. Searches surfaced only Unisys‑adjacent matters where Unisys was a defendant (e.g., Patent Armory Inc. v. Unisys Corp., N.D. Tex. 3:25‑cv‑02115, voluntarily dismissed Aug. 2025; MCOM IP LLC v. Unisys Corp., N.D. Tex. 3:21‑cv‑02288). Neither involves this patent as an asserted patent, and in both the asserted patents are third‑party patents, not Unisys patents. I also found no IPR, PGR, or CAFC appeal involving this patent. With no suit, there can be no transfer within six months before a suit. Negative finding — PACER/CAFC coverage in my sources is incomplete, so treat as "not found," not "proven absent."

6. Bankruptcy fire‑sale — NOT PRESENT. Unisys has not filed Chapter 7 or 11. It remains an operating NYSE‑listed company. Its 2021–2022 losses, pension deficits ($499.4M settlement losses in 2021; US qualified plans ~$366M deficit at YE2022), restructuring charges ($54.9M in 2022, including $35.8M of asset impairments), and the June 2025 $700M 10.625% secured‑notes raise show real balance‑sheet stress — but distress is not bankruptcy, and the 2025 lien (Reel 071759/0527) is a pledge to a collateral trustee, not a §363 sale or a credit‑bid transfer. Distinguish clearly: Kodak/Nortel/Polaroid-style fire sales produce assignments to purchasers in the record; what we see here is a grantor pledging to a lender. Materially different.

7. Privateering — NOT PRESENT. No evidence that Unisys transferred the patent to an NPE to assert against competitors. No SEC disclosure of an IP‑monetization or assertion‑vehicle arrangement, and no NPE appears in the chain. Unisys's disclosed strategic moves in the period (e.g., the 2020 sale of the U.S. Federal business) are business‑unit divestitures, unrelated to this patent. Absent an NPE receiving title, privateering cannot be present.

8. Defensive aggregator (anti‑NPE) — NOT PRESENT. The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. It terminates at Unisys Corporation, with Computershare holding only a security interest as collateral trustee — a creditor's lien, not ownership and not a defensive aggregation. The patent has therefore not been neutralized; if anything, the outstanding secured‑notes lien makes it encumbered rather than defensive.

---

Verdict

Insufficient data — for the NPE question specifically, not for ownership.

Justification: the patent has only ever had one recorded assignment — Malnati → Unisys Corporation, executed 2010‑10‑08 and recorded 2010‑10‑21 at Reel 025174/0557 — and every subsequent recording is a lien or lien release between Unisys and a secured lender (Reels 026509/0001, 030004/0619, 030082/0545, 034096/0984, 042354/0001, 044144/0081, 044416/0358, 054231/0496, 071759/0527). All eight NPE signals score "not present," there is no assertion vehicle, and no litigation involves the patent. Because no NPE verdict bucket fits a chain that terminates at the original operating company with no assertion, "Insufficient data / only the original assignment" is the correct call rather than forcing an "operating‑company assertion" label that would falsely imply litigation; the ownership data itself is strong and unambiguous — Unisys still owns US 8,832,654, subject to a Computershare collateral‑trustee security interest recorded 2025‑06‑27 at Reel 071759/0527. Two items deserve follow‑up before anyone relies on this record: (i) the correspondent field on all ten recordings is unretrieved, which is the one signal I could not score, and (ii) the 12th‑year maintenance fee shows a 2026‑04‑27 reminder with no payment entry in the snapshot, meaning lapse risk should be confirmed at USPTO Patent Center.

Verify at: https://assignmentcenter.uspto.gov/ (search patent number 8832654 by reel/frame above) and https://patents.google.com/patent/US8832654/en for the legal‑events table.

I'll research each of the cited references for US 8,832,654.

Prior-Art Analysis — US 8,832,654 B2

"System and method for autonomous generation of software documentation" · James Malnati · Unisys Corporation · App. 12/815,530 filed 2010-06-15 · granted 2014-09-09

---

0. Housekeeping / contradictions flagged

1. Date discrepancy. The task states "current date April 26, 2026." Runtime environment date is 2026-09-30. Nothing below turns on it; searches were executed on the latter date.
2. Cross-reference to the earlier section. The previous section correctly noted that the granted claim 1 is narrower than published claim 1 — the granted version added (a) the "wherein the documentation generator device comprises an autoaction database …" limitation and (b) the closed three-item list of autoactions (console alerts, external paging devices, cross-system commands). This matters decisively for the § 102 analysis below: any reference that lacks those two added limitations cannot anticipate granted claim 1, even if it teaches generic automatic documentation generation.
3. I could not verify the examiner's own claim mapping. The Google Patents "Patent Citations" list for US 8,832,654 shows that six patent documents were cited, but not which claims were rejected over which reference. The § 102 mappings below are my own element-by-element analysis, not a restatement of the Examiner's.
4. The stated operative law. Because the effective filing date is 2010-06-15 (pre-March 16, 2013), pre-AIA 35 U.S.C. § 102 applies — i.e., § 102(a), (b) and (e), not the AIA's § 102(a)(1)/(a)(2). Anticipation therefore requires a single reference disclosing every claimed element, arranged as claimed.

---

1. The six "Patent Citations" of record

---

2. Reference-by-reference assessment

2.1 US 7,404,177 B1 — Automated web interface generation for software coded applications

Citation. US 7,404,177 B1; Inventors Paul Aubrey Greenfield, Charles Zdzislaw Loboz, Mark Alexander Hugh Emberson; Assignee Unisys Corporation; Appl. No. 09/830,226 (U.S. national phase of PCT/AU98/00885, PCT filed 1998-10-23); PCT publication WO 00/25209, 2000-05-04; granted 2008-07-22.
Description. A translator/compiler technique (the "ActiveLINC" generator) that takes a legacy 4GL (LINC) application's interface specifications ("Ispecs") — including data definitions, screen formats and business rules — and automatically generates translatable source code and a series of Java client software components (JavaBeans/ActiveX, generated GUI apps) that talk back to the legacy host over TCP/IP.
Why the Examiner likely cited it. It is the of-record archetype of "machine-readable structured specification in → automatically generated software artifact out" — i.e., it touches the generation-from-a-structured-representation concept recited in claims 3, 11 and 17 (the DADB "highly-structured representation") and the general "processing … for generating" step of claims 1/9/16.
§ 102 anticipation: None. The reference generates executable client code/GUI components, not documentation, and it contains no autoaction database, no message-matching patterns, no message→action mapping, and no console-alert/paging/cross-system-command actions. It does not disclose a single element of the two narrowing limitations added to granted claim 1. It is at most § 103 art against the "highly-structured representation" dependent claims (3/11) and claim 17, and only in combination.

2.2 US 7,302,674 B1 — Automating document reviews in a project management system

Citation. US 7,302,674 B1; Inventors Ted Franklin Gladieux, Avinash Sadhu; Assignee Unisys Corporation; Appl. No. 10/636,923, filed 2003-08-06 (the patent-table "2002-11-26" is a priority date, not the U.S. filing date); granted 2007-11-27; 30 claims.
Description. A project-management/repository system (CBD&D) that stores document-review information in a repository ("UREP"), associates a document review with a project, runs a document-review task via a daemon utility and JSP web wizards, and sends notifications to review participants at defined points in the review lifecycle.
Why cited. General background for automated tracking/handling of documents and automated notification of participants — arguably background to claims 5–7 / 13–15 (storing and outputting a generated document) and to the notion of documentation lifecycle automation.
§ 102 anticipation: None. It concerns human-authored project documents under review, not documentation generated from automation rules. No autoaction database, no message-matching criteria, no triggered management actions, no console/paging/cross-system actions. It cannot anticipate claims 1–8, 9–15 or 16–17.

2.3 US 7,565,643 B1 — Sending notifications to project members in a project management system

Citation. US 7,565,643 B1; Assignee Unisys Corporation (third-party listing names inventor Nancy L. Sweet; I could not independently confirm the full inventorship); Appl. filed 2003-08-06 (listed priority 2002-11-26); granted 2009-07-21.
Description. Creates a configurable notification list and notification information for a project, stores it in a repository, receives a notification request from a project-management component and processes the request at run-time — i.e., a data-driven, repository-stored event→notification mapping with enable/disable status per notification item.
Why cited. This is the closest of the Unisys trio to the architectural idea "stored user-configured records (rules) → automatically fire notifications/actions at run time." It is plausibly background art or § 103 art for claims 1/9/16's "receiving user-defined rules … and processing."
§ 102 anticipation: None. The "notification" is a message to project team members (e.g., e-mail about project status), not documentation describing the rules, and there is no autoaction database, no message-matching pattern set, no documentation generation, and none of the three enumerated autoaction types. Note separately: because its grant date (2009-07-21) is after the 2009-06-15 critical date, this reference qualifies only as pre-AIA § 102(e) art as of its 2003 filing date — a § 102(b) theory would fail on the dates.

2.4 US 7,657,866 B2 — Providing documentation for assembling, installing, and supporting of software stacks

Citation. US 7,657,866 B2; Assignee OpenLogic, Inc.; Inventors Cope et al. (per the related OpenLogic patent US 8,832,647); listed priority 2004-04-28; pre-grant publication US 2006/0031227 A1 ("Post-installation tools for software stacks"); granted 2010-02-02.
(Do not conflate with US 8,832,647 — OpenLogic, "Tools for software stacks." That is a different patent number; per the literal-ID rule I have not merged them.)
Description. A computer-based tool/knowledge base for assembling, installing and supporting software stacks of uncoordinated/open-source projects; it determines dependency characteristics and provides documentation (and training materials) adapted to the environment of the particular stack, along with filtering and installation utilities.
Why cited. This is the of-record reference that is actually about automatically providing documentation derived from a software artifact — the closest prior art on the "generation of documentation" axis of claims 1/9/16.
§ 102 anticipation: None. It documents software stacks and their dependency/provisioning characteristics, not user-defined automation rules. Missing: user-defined rules specifying management actions triggered by an event; an autoaction database containing patterns of messages to be identified; message-matching criteria; and the three enumerated autoactions (console alerts, external paging devices, cross-system commands). Best characterized as § 103 art combinable with the Operations Sentinel AMS materials (see § 4) to argue the "document it automatically" concept was known.

2.5 US 7,913,228 B2 — Translation viewer for project documentation and editing

Citation. US 7,913,228 B2; Inventors Matthew R. Ericsson, David R. Curry, Christopher E. Stanek, Fabio Malaspina, Ronald E. Bliss; Assignee Rockwell Automation Technologies, Inc.; Appl. No. 11/537,110, filed 2006-09-29; pre-grant publication US 2008/0098356 A1 (2008-04-24); corresponding EP 1939732 A1; granted 2011-03-22; 34 claims.
Description. A software-documentation system with a selection component that detects a project component in the context of a current (programming) language and a translation viewer that displays and edits alternative-language documentation tied to that detected component — i.e., documentation is programmatically linked to and surfaced for a specific code/configuration artifact, and can be switched between language contexts.
Why cited. It is of-record evidence of (i) machine-generated/displayed documentation keyed to a detected artifact, and (ii) converting a technical artifact's documentation into a more human-consumable form — adjacent to claims 2/10 ("identifying the event and the corresponding management actions") and the operator-readability rationale in the specification.
§ 102 anticipation: None. It is a multi-language translation/documentation viewer, with no rules engine, no autoaction database, no message matching, no triggered management actions, and none of the enumerated autoactions. Like the others it is at most § 103 background/combination art, and it would not supply the two narrowing limitations of granted claim 1.

2.6 US 2011/0307863 A1 — (applicant's own pre-grant publication)

Citation. US 2011/0307863 A1, published 2011-12-15, applicant/inventor James Malnati.
Status: not prior art. It is the pre-grant publication of the same application (12/815,530) that issued as the patent under review, published after the 2010-06-15 filing date. Under pre-AIA § 102 it is not "by another"; and even under AIA § 102(b)(2)(A)/(B) reasoning a disclosure by the same inventor(s) on the same subject matter is excepted. Its appearance in Google's "Patent Citations" list is a bibliographic artifact of family aggregation. It cannot anticipate or render obvious any claim of US 8,832,654. (This is a genuine contradiction with any reading of the list that treats it as third-party art, and I flag it as such.)

---

3. The "Cited By" documents — not prior art

All eight listed citing documents post-date the 2010-06-15 priority date — earliest is US 2014/0173562 A1 (priority 2012-12-17, SAP "Automatic Documentation Generator," which issued as US 9,069,646 B2), then the Unisys "automation degree of implementation metrics" family (2015-04-28 filings: US 9,667,573; US 9,686,220; US 10,153,992; and pubs. 2016/0323207–09), IBM's US 10,537,245 and US 10,637,722, and Fuji Xerox US 2020/0285693 A1. Under both pre-AIA and AIA § 102 these are § 102(a)(2)/102(e)-style forward citations only, usable at most as evidence of the field's later development — none can be § 102 prior art against this patent. The existence of this forward-citation cluster (notably the Unisys family on "automation degree of implementation metrics" and SAP's automatic documentation generator) is worth noting as evidence of the technological trajectory the patent sits on.

---

4. Non-patent citations of record (4) — and the real "most relevant prior art"

This is the substantive core of the prior art. The Sept. 2008 AMS Administration Guide and the Jan. 2005 Single Point Operations guide — both publications of the patent owner's own product, both more than one year before filing, both expressly incorporated by reference into the specification — describe precisely the elements the patent treats as the environment: the SP-AMS/CP-AMS autoaction databases, patterns, message-matching criteria, the autoaction types (raise console alerts, page external devices, cross-system commands), and the BADB / DADB / VADB / Distribute utilities. The specification itself concedes these are known: "The operation of these commands within the Operations Sentinel autoaction databases are well-known in the art."

Consequence for validity. The of-record art therefore establishes, in the applicant's own words and documents, every element of granted claims 1/9/16 except the "autonomous documentation generation" element (the documentation generator that processes the rules/structured dump and emits operator-readable "Console Message / Associated Actions" documentation). That is the only element the cited art does not supply, and it is exactly where a § 103 combination would aim.

5. Overall § 102 conclusion

Bottom line. On the record before me, none of the six patent citations anticipates any claim of US 8,832,654 under pre-AIA § 102, individually or otherwise; the two Unisys project-management patents (US 7,302,674; US 7,565,643) and the Rockwell translation viewer are the weakest (background/dependent-claim level), while OpenLogic US 7,657,866 is the closest of the patent citations on the documentation-generation axis. The genuinely most relevant prior art in the file is non-patent: the applicant's own Operations Sentinel Autoaction Message System Administration Guide (Sept. 2008) and Single Point Operations guide (Jan. 2005), which are § 102(b) printed publications disclosing the autoaction database, message-matching patterns and all three enumerated autoactions — leaving only the autonomous documentation generator as the point of novelty. Any serious § 103 challenge would combine that AMS art with a documentation-generation reference such as US 7,657,866 (or an automated API-documentation generator).

---

6. Explicit uncertainties

• Examiner's rejections not recovered. I could not obtain the file wrapper / PTOL-892 or the examiner's statement of which reference was applied to which claim. Mappings above are my analysis, not the Office's; treat them as hypotheses for a validity search, not as the prosecution history.
• Inventorship of US 7,565,643 (third-party listing suggests Nancy L. Sweet) and the exact U.S. filing date of US 7,657,866 (the file shows only the 2004-04-28 priority date) were not independently confirmed from the face of the documents.
• Publication dates for US 2006/0031227 A1 and the EP 1939732 A1 are taken from third-party listings and should be verified before relying on them to establish § 102(b) status.
• The Dec. 2010 Moore (CRS) non-patent citation is chronologically impossible as § 102 art against a 2010-06-15 filing. I flag this as an anomaly rather than assuming a different date; verify against the printed citation.
• I found no litigation, IPR/PGR or CAFC proceeding involving US 8,832,654 in my searches — a negative finding, not proof of non-existence.
• Documents cited in this report are identified literally; in particular US 8,832,647 (OpenLogic, "Tools for software stacks") is a different patent from US 8,832,654 and has not been merged with it.

I'll ground this obviousness analysis in the actual prior-art references of record, so let me pull details on each.

Obviousness Analysis — US 8,832,654 B2 under 35 U.S.C. § 103

Prepared: 2026-09-30 · Subject: US 8,832,654 B2, System and method for autonomous generation of software documentation (Malnati; Unisys)

Date note: the task prompt states April 26, 2026; my environment date is 2026-09-30. Consistent with the prior section of this analysis, nothing below turns on the discrepancy. I also confirm the prior section's claims/assignee/bibliographic content and do not repeat it here.

---

1. Governing law and the prior-art record of record

Application 12/815,530 was filed 2010-06-15, i.e., pre-AIA. Pre-AIA § 102/103 govern, and the critical date for § 102(b) printed publications is 2009-06-15.

The task directs me to use the Prior Art section of the US 8,832,654 page (Google Patents). That section yields six patent citations and four non-patent citations. Qualifying status:

Two flags for the record. (1) The Moore EAS item cited in this patent's IDS is dated after the filing date and therefore does not qualify as prior art; its presence in the citation list should not be treated as evidence of anything. (2) US 2011/0307863 A1 is the applicant's own publication and cannot be used against the claims. Both are consistent with nothing more than a broad IDS submission.

I could not complete verification of the internal contents of the two Unisys non-patent references in the time available — see § 8.

---

2. Person having ordinary skill in the art (PHOSITA)

For a June-2010 pre-AIA priority date: a bachelor's degree in computer science/engineering (or equivalent) plus 3–5 years' experience developing enterprise/distributed systems-management software, including familiarity with (a) rule- or pattern-based automation engines, (b) message-matching and configuration/rule-language parsing, and (c) automated report/document generation. In the alternative, and more narrowly to this art: a mainframe systems programmer with 2–5 years of Unisys ClearPath / Operations Sentinel / SP-AMS–CP-AMS experience, including use of the BADB/DADB/VADB utility suite. Either formulation has ordinary skill in parsing a structured automation database and emitting a formatted document; both are relevant to motivation.

---

3. The key constructional point

The three independent claims (1, 9, 16) share the same architecture. Stripping to essentials, the claimed subject matter is:

1. user-defined rules specifying management actions triggered by an event (a system message), in an autoaction database with patterns + message-matching criteria;
2. autonomous generation of human-readable documentation of those rules that identifies the message and the triggered actions; and
3. autoactions limited to three enumerated types — (i) raising alerts to a console, (ii) activating paging devices external to the management system, and (iii) cross-system commands to another managed system.

Element (3) is decisive, because the '654 patent concedes all three autoaction types are prior art in its own Background. Quoting the patent's Background section (https://patents.google.com/patent/US8832654/en), the specification states that under Operations Sentinel/AMS, "a user may create rules to activate external paging devices (or other communication/notification devices, such as mobile telephones, etc.) when an alert is raised, acknowledged, or cleared" and that "[a]ctions may also include cross-system commands sent to another managed system," with alerts displayed "in the Alerts window of Operations Sentinel Console." That is an applicant admission that maps directly onto the enumerated autoaction types. Likewise, the Background admits the autoaction database / pattern / message-matching architecture (SP-AMS "allows a user to specify messages to match and actions to automatically perform… when a message is received and matched"), and that DADB "is a component of the supported product" producing "highly structured symbolic output."

So the only limitation with a genuine question is the autonomous generation of the documentation itself. That is also the limitation the patent's own Summary concedes is merely desirable because manual authoring is "tedious and error prone" — i.e., a recognized problem with a recognized class of solutions.

---

4. Primary combination — AMS Guide + US 7,657,866 (Cope/Openlogic), with US 7,404,177 and US 7,913,228 as corroborating secondary art

4.1 What each reference supplies

A. Operations Sentinel AMS Administration Guide (Sept. 2008) + Single Point Operations Interface Software (Jan. 2005) — supplies:

• the management-application autoaction database, patterns, and message-match criteria ("They identify the important parts of the message and constraints, such as message length, that must be satisfied for the message to match the pattern");
• the autoactions: alert raising to the Operations Sentinel Console, external paging/notification, host commands, cross-system commands;
• DADB, which "dump[s] the built machine-readable code into the highly structured symbolic output";
• the fact that rule sets are site-specific and change over time — which is itself the motivation to document them.

B. US 7,657,866 B2 (Cope et al.), ¶¶ describing system 500 — supplies the missing generation step. Its system is "operative to … generate documentation and training materials specifically related to the stack," and its support utility "involves receiving an identification of a stack … selectively accessing storage to obtain support information specific to that stack and outputting the support information to a user." Its specification further states that "documentation may be stored as structured content, e.g., associated with metadata such as XML tags, identifying a context or association of the data," and that "[t]ext and/or graphics may be assembled in this regard," with a "parsing functionality … provided for selectively combining support information." (https://patents.google.com/patent/US7657866; https://patentimages.storage.googleapis.com/2d/84/45/209f4b2a11b26e/US7657866.pdf)

That is: parse structured content → assemble/emit documentation specific to a particular configured environment. The '654 claims require exactly the same operation, with the "particular configuration" being a site's autoaction database rather than a software stack.

C. US 7,404,177 B1 (Greenfield et al., Unisys) — corroborates the "generate a deliverable artifact from a structured definition" technique in the same assignee's family: "the Ispec definition is utilized to form a graphical user interface application 21 and interface components 22"; its user-interface applications "can be generated automatically from their existing terminal screen and GUI form definitions." (https://patents.google.com/patent/US7404177) Same-assignee art is a recognized source of motivation.

D. US 7,913,228 B2 (Ericsson et al., Rockwell) — supplies documentation viewed/translated in a viewer that "detects a project component within a context of a current language" and displays associated documentation, with change harmonization — i.e., documentation kept tied to, and generated from, structured project objects, and output to a user. (https://patentimages.storage.googleapis.com/43/28/17/f11e88cde98883/US7913228.pdf; family member EP 1 939 732 A1.)

E. US 7,302,674 and US 7,565,643 (Unisys) — supply (i) template-driven automated composition of descriptive text from stored structured objects ("The text of the notice comes from a NOTIFICATION object 41i … the text has a placeholder for the name of the artifact"; https://patents.google.com/patent/[US7302674B1](/patent/US7302674B1)/en) and (ii) the art's explicit recognition of the motivation: "Existing techniques are either manual in nature … These techniques are slow, tedious, error-prone, and inefficient" (https://patentimages.storage.googleapis.com/6a/b2/c2/cd3ac52dc01a0d/US7565643.pdf). That sentence is a near-verbatim statement of the problem the '654 patent's own Summary recites.

4.2 Motivations to combine (KSR / MPEP 2143 rationales)

1. Same field and analogous art. All references concern generating or presenting textual artifacts about software systems from machine-readable definitions. OpenLogic's stack documentation and Unisys's rule documentation address the same problem class: documentation that is either unavailable or "not adapted for the environment" of a particular installation.
2. The primary reference itself supplies the motivation. The AMS Guide teaches that rules are user-defined, site-specific, and revisable, which creates exactly the documented need for an accurate, current, operations-readable record of "if this message appears, this action will occur."
3. Predictable combination of known elements. DADB already emits a "consistently structured presentation of the automation rules." Emitting a formatted document (Word/Excel) from a structured textual stream is routine data-formatting — the paradigm KSR case of "a combination of familiar elements according to known methods [that] does no more than yield predictable results." The '654 specification itself does nothing beyond keyword-scanning ("looks for certain keywords … such as 'MESSAGE' and 'ACTION'") and templating.
4. Design incentive / obvious to try. Automating a manual, error-prone documentation task has obvious benefits (currency, auditability, labor reduction). Only a finite number of known options exist (generate from source, from configuration, or from a database dump) — an "obvious to try" situation with predictable success.
5. Common assignee. US 7,404,177, US 7,302,674, and US 7,565,643 are Unisys's own; the artisan addressing an Operations Sentinel documentation need was pointed to them directly.
6. No teaching away. None of the references disparages automated documentation generation from configuration data.

4.3 Claim-mapping summary (independent claims)

---

5. Alternative combinations

Combination B — AMS Guide + US 7,913,228 alone. Weaker but arguable. US 7,913,228 teaches generating/presenting documentation from structured project objects and outputting it in a viewer, which addresses the generation and output steps; it does not by itself supply the autoaction-database limitation, so the AMS Guide remains necessary as the primary reference.

Combination C — Single reference + common knowledge (In re Oetiker). Given the AMS Guide's express disclosure of DADB — the patent admits DADB is "a component of the supported product" — one can argue that generating a formatted, operator-readable document from DADB's structured output was within the ordinary creativity of a PHOSITA and thus obvious without any secondary reference. This is the most aggressive theory and the easiest to attack; I would not lead with it.

Combination D — 7,657,866 + 7,913,228 + 7,404,177 (without the AMS Guide). Insufficient. These teach automated generation, viewing, and artifact generation, but none discloses an autoaction database of patterns with message-match criteria and the three enumerated autoaction types. The AMS Guide must be the primary reference.

---

6. Dependent claims

• Claims 3 / 11 / 17 (highly-structured representation of the rules from the database via a "database utility device"): directly disclosed by the AMS Guide's DADB, absent any combination. The patent itself attributes DADB to "the supported product." Strong.
• Claims 2 / 10 (documentation identifies the event and triggered actions): obvious over AMS Guide patterns + 7,657,866's stack-specific support information.
• Claims 4 / 12 (event = system message): AMS Guide.
• Claims 5–6 / 13–14 (store to computer-readable medium; computer-readable file): routine; also 7,913,228 documentation objects and 7,657,866 structured/XML documentation.
• Claims 7 / 15 (outputting the documentation): routine; 7,913,228 viewer; 7,657,866 "outputs the information to a user."
• Claim 8 (generator as software on at least one processor): routine; 7,657,866 engine 506 / knowledge base 504.

No dependent claim adds a limitation that survives the AMS Guide + 7,657,866 combination.

---

7. Anticipated patentee rebuttals, and how they fare

---

8. Weaknesses in this analysis, and what to verify

1. I could not obtain the primary references' internal text. My tooling hit a step limit before retrieving the 276-page Operations Sentinel AMS Administration Guide (7862 6900-006) and the Single Point Operations Interface Software for ClearPath MCP (Jan. 2005) documents directly. My § 4.1 characterization of the AMS Guide is drawn principally from US 8,832,654's own description of it — which, for the autoaction/paging/cross-system limitations, is more powerful (an applicant admission) but for the documentation-generation question is circular and must be confirmed against the guide itself. Verify directly (the guide is public at public.support.unisys.com/spo/docs2/os-11.2/78626900-006.pdf, per the non-patent citation listed on the patent page) whether the guide itself teaches or suggests rendering user-defined rules into human-readable output.
2. US 7,565,643's § 102(b) status is unconfirmed. Its 2009-07-21 grant is inside the one-year window before the 2010-06-15 filing. If no earlier pre-grant publication exists, it is § 102(e) art (effective as of its 2002/2003 filing) rather than § 102(b) — still usable, but the effective-date analysis changes. Confirm via its file wrapper.
3. US 7,913,228's § 102(b) status rests on its 2008-04-24 pre-grant publication, not on the 2011 patent. Cite the publication, not the patent, to avoid a § 102(e)-versus-§ 102(b) error.
4. Incorporated-by-reference art not analyzed here. US 6,154,787; 7,092,940; 7,421,492; 7,513,027, and applications 12/644,517 / 12/637,928 are incorporated by reference into the '654 specification. Those documents are not shown in the page's Prior Art section, per my instruction to work from that section — but they are part of the record and may contain further automation/messaging teachings (and 12/644,517 / 12/637,928, filed in late 2009, are potential § 102(e) art).
5. Possible § 102(b) on-sale/public-use overlay. The patent describes DADB as "a component of the supported product" with a Sept. 2008 manual. If the claimed documentation-generation feature (or the product embodying the later-claimed architecture) was on sale or in public use before 2009-06-15, a § 102(b) on-sale/public-use theory against the patent's own claims is worth investigating. This is an inference from the specification, not a verified fact.

---

9. Bottom line

Claim 1 (and, by parallel structure, claims 9 and 16) is, in my assessment, more likely than not obvious over the Operations Sentinel AMS Administration Guide (Sept. 2008) / Single Point Operations Interface Software (Jan. 2005) as the primary reference in view of US 7,657,866 B2, with US 7,404,177 B1, US 7,913,228 B2 (via US 2008/0098356 A1), and US 7,302,674 B1 / US 7,565,643 B1 as corroborating secondary art.

The decisive vulnerability is that the three enumerated autoaction types in every independent claim are admitted prior art in the patent's own Background, as is the autoaction-database/pattern/message-matching architecture and the DADB "highly structured" dump. Once those elements are removed from the field of novelty, the remaining question — whether to replace a manual, admittedly "tedious and error prone" documentation task with parsing DADB's structured output into a formatted report — is the kind of predictable automation KSR holds obvious.

Confidence: moderate-to-high on the legal framework and the claim-mapping logic; moderate on the factual completeness of the primary non-patent references, which I was unable to retrieve and verify firsthand (§ 8, item 1). A validity opinion defending these claims would have to attack the combination's motivation as applied specifically to the documentation-generation step, and would be strongest if the AMS Guide could be shown to disclose only the dump-to-source step and nothing suggesting a human-readable rendering — a proposition I have not been able to confirm or refute.

Primary sources: https://patents.google.com/patent/US8832654/en · https://patents.google.com/patent/US7657866 · https://patents.google.com/patent/US7404177 · https://patents.google.com/patent/US7302674B1/en · https://patentimages.storage.googleapis.com/6a/b2/c2/cd3ac52dc01a0d/US7565643.pdf · https://patentimages.storage.googleapis.com/43/28/17/f11e88cde98883/US7913228.pdf · https://patentimages.storage.googleapis.com/2d/84/45/209f4b2a11b26e/US7657866.pdf