US 7840619

I'll run targeted searches against USPTO/patent databases and litigation dockets, holding the patent number literal (7840619).

US Patent 7,840,619 — Analyst Summary

Search note / scope: Searches were run against patent databases (Google Patents, FreePatentsOnline, Justia, uspto.report, SumoBrain), the Stanford NPE Litigation Database, and Federal Circuit 2026 docket/opinion listings. I found no CAFC 2026 appeal docketed under or referencing 7,840,619. I also could not retrieve the issued claim text from the sources returned, so the claim discussion below is flagged accordingly.

1. Bibliographic data (high confidence — corroborated by multiple sources)

Assignment chain (recorded, per Google Patents): Individual → Technology Innovations Associates, LLC (license, 2013-11-13) → Bruce Horn (2017-03-08) → Ehierarchy LLC (2017-03-08).
Source: https://patents.google.com/patent/US7840619/en

Family/continuations (from Google Patents timeline): US 8,055,692 B2; US 8,280,932 B2; US 9,229,945 B2; US 9,483,169 B2; and published application US 2016/0117071 A1. Note the continuation filings post-date the '619 grant, which is consistent with the priority claim above.

2. Abstract (as published)

"A computer data processing system including a central processing unit configured with a novel integrated computer control software system for the management of data objects including dynamic and automatic organization, linking, finding, cross-referencing, viewing and retrieval of multiple objects regardless of nature or source. The inventive system provides underlying component architecture having an object-oriented database structure and a metadata database structure which is unique in storing only one instance of each object while linking the object to multiple collections and domains by unique metadata links for the grouping into and retrieval from any of the collections. The system employs configurable, extensible attribute/properties of data objects in metadata format, and a truly user-friendly configurable interface that facilitates faster, more unified, comprehensive, useful and meaningful information management. Additional features include a sticky path object hierarchy viewing system, key phrase linking, viewing by reference, and drag-and-drop relationship link creation."

⚠️ Note: the abstract text as fetched from Google Patents is truncated at "Additional"; the fullest version I located (via FreePatentsOnline and Golden) supplies the sentence beginning "Additional features include a sticky path object hierarchy viewing system…," which I have quoted above. Do not treat the truncated form as authoritative.

3. Disclosure overview (high confidence — grounded in the specification)

The system is internally called MFS ("Metadata Filing System"). Key architecture elements (all defined in the patent's "Definition of Terms" section):

• Object Store — an object-oriented database storing object data by unique UID; stores only one instance of each object.
• Catalog — a metadata database (B-Tree based) storing reference objects addressed by UUID and their metadata properties; runs queries, issues change notifications, and maintains a dependency graph.
• Mirrored Object System — external files/emails are "mirrored" as Reference Objects with a UUID providing one-to-one mapping back to the external source.
• Collections / Containers — logical groupings by (1) user categorization, (2) Boolean metadata query, (3) key-phrase matching. Objects can belong to many collections simultaneously without duplication (only link metadata is added).
• Domains — plug-in/extensibility mechanism (File, EMail, Music, Personal Information, Image Management, Time domains).
• Consistency maintenance — threaded pipeline of Updater, Synchronizer, Notifier, Classifier processing an event queue; the key-phrase classifier uses a single-pass Aho-Corasick finite state machine to sort an object into an unlimited number of collections at once.
• Sticky Paths — list-view hierarchy display with a dynamically updating path pane at the top.
• Viewing by Reference — cross-reference display combining dynamic, time-based collections (e.g., "Today").

4. Independent claims — uncertainty flag

I do not have authoritative claim language for this patent. The full text provided to me contains the specification, definitions, figure descriptions, and abstract, but the fetch ends before any "What is claimed is" / claims section, and none of the external sources returned in my searches reproduced claim 1.

Accordingly, I will not paraphrase claims as if they were claim text. Based on the specification, the independent claims of a patent of this character would most plausibly be directed to one or more of the following, but treat this as inference, not quotation:

1. A computer data processing system (CPU + memory + display) configured with the object store + catalog architecture, storing a single instance of each object and linking it to multiple collections via metadata (annotation + link metadata) rather than duplication.
2. A method of automatic organization/indexing/viewing of objects from multiple sources: creating reference objects with UUIDs mapping to external data, extracting/attaching metadata, and classifying objects into collections via metadata query and/or key-phrase match.
3. Apparatus/software claims for the sticky path hierarchy display and/or viewing-by-reference cross-referencing UI.

If verbatim claim text is required (e.g., for infringement or validity analysis), it must be pulled from the USPTO PatentCenter/Patent Public Search record for 11/903,304 or the printed patent at columns following the specification — I could not verify it from the sources available here.

5. Litigation / post-grant activity (as reported by databases)

Per Google Patents and the Stanford NPE Litigation Database:

• eHierarchy LLC v. [Microsoft Corp.](/litigations/by-plaintiff/Microsoft%20Corp.), No. 2:16-cv-00564 (E.D. Tex.) — 7,840,619 listed as asserted patent.
  • https://npe.law.stanford.edu/patent/7840619
• Delaware District Court cases filed 2013: 1:13-cv-00348, -00349, -00350, -00351, -00352, -00353, -00355, -00356 (cluster of related filings).
• E.D. Tex. case 2:17-cv-00225 (2017).
• Family litigation also reported via Darts-IP (family 30116028).

On CAFC 2026: I found no Federal Circuit docket, oral-argument listing, or opinion in 2026 citing 7,840,619. The patent's legal status is expired (adjusted expiration 2024-05-19), which reduces the likelihood of new appellate activity, but I cannot positively confirm the absence of a 2026 appeal from the databases returned. Treat "no CAFC 2026 activity found" as a search limitation, not an affirmative clearance.

Also flagged as uncertain/unverified: a PTAB proceeding referenced as IPR2025-00393 ([LG Electronics Inc.](/litigations/by-plaintiff/LG%20Electronics%20Inc.)) appeared in search results with an exhibit list containing US 2004/0177319 A1 to Horn — a family member of 7,840,619 (the parent publication), not the '619 patent itself. I could not confirm that the '619 patent is challenged in that IPR; do not assume a nexus without checking the PTAB docket directly.

6. Bottom line

US 7,840,619 is Bruce L. Horn's "Metadata Filing System" patent — an early, unusually complete disclosure of a metadata-catalog-plus-object-store architecture supporting single-instance objects linked into many collections, dynamic auto-classification (Boolean metadata queries and Aho-Corasick key-phrase matching), mirrored reference objects keyed by UUID, sticky-path hierarchy views, and viewing-by-reference. Bibliographic data, abstract, and assignee chain are well corroborated. Claim language is unverified, and no CAFC 2026 activity was located.

I'll search for litigation involving US 7840619 across patent litigation databases.

Let me dig into the specific case numbers listed on the patent record and identify parties.

Litigation Involving US Patent 7,840,619

Research note / caveat up front: I was able to retrieve and cross-check some but not all of the case records before hitting tool-call limits. Below I clearly separate (a) what is affirmatively confirmed by a source I retrieved, from (b) what appears only as a case-number listing on the patent's own "family has litigation" record, and (c) what I could not verify. I have not auto-corrected any numbers; all case numbers and patent numbers are reproduced literally.

1. Confirmed case

Source: Stanford NPE Litigation Database, patent page for 7840619 — https://npe.law.stanford.edu/patent/7840619. This is the only case that database lists for this patent, and it classifies eHierarchy, LLC as an "Acquired patents" asserter.

2. Cases listed on the patent's own litigation record (parties not fully verified)

The Google Patents page for US7840619B2 (https://patents.google.com/patent/US7840619/en) carries a "Family has litigation" block drawn from Unified Patents' litigation data. It lists the following U.S. case dockets for the patent family:

Literal observations (per the strict no-correction rule):

• The Delaware 2013 cluster runs 00348, 00349, 00350, 00351, 00352, 00353, 00355, 00356 — note that 1:13-cv-00354 is absent from the listing, so this is an 8-case cluster, not 9.
• The record also contains a Darts-ip cross-reference: "First worldwide family litigation filed," family 30116028.

Links given on the Google Patents page (Unified Patents portal):

• https://portal.unifiedpatents.com/litigation/Delaware%20District%20Court/case/1%3A13-cv-00348
• .../case/1%3A13-cv-00349
• .../case/1%3A13-cv-00350
• .../case/1%3A13-cv-00351
• .../case/1%3A13-cv-00352
• .../case/1%3A13-cv-00353
• .../case/1%3A13-cv-00355
• .../case/1%3A13-cv-00356
• https://portal.unifiedpatents.com/litigation/Texas%20Eastern%20District%20Court/case/2%3A17-cv-00225
• https://portal.unifiedpatents.com/litigation/Texas%20Eastern%20District%20Court/case/2%3A16-cv-00564

I did not successfully retrieve the Unified Patents case pages themselves, so plaintiff, defendant, filing date, and outcome for items 1–9 above remain unverified.

3. A discrepancy you should be aware of

Case number 1:13-cv-00353 (D. Del.) — the number listed on the 7,840,619 family record — also appears in search results as:

Princeton Digital Image Corp. v. Ubisoft Entertainment, SA, 13-cv-00353 (D. Del. Apr. 8, 2019) (slip op.)

That is a different plaintiff (Princeton Digital Image Corp.), a different defendant (Ubisoft), and a different patent family (Princeton Digital's image/compression patents). I am flagging this because it means either (i) the same docket number is being reported for two unrelated matters in different databases, or (ii) the family-litigation mapping on the patent page is over-inclusive.

Consequently, I do not treat the 2013 D. Del. numbers as reliably established as 7,840,619 litigation without pulling the dockets. Given that (a) Stanford's NPE database lists only the Microsoft case, and (b) the Google Patents block is expressly keyed to the worldwide family (Darts-ip family 30116028) rather than to this single patent, it is materially possible that the 2013 Delaware cluster and E.D. Tex. 2:17-cv-00225 involve other family members — e.g., US 8,055,692; US 8,280,932; US 9,229,945; or US 9,483,169 — rather than 7,840,619 itself.

4. Ownership context (relevant to who could be the plaintiff)

Per the Google Patents assignment/legal-status record for US7840619B2:

• Current assignee: Ehierarchy LLC
• 2013-11-13 — assigned to Technology Innovations Associates, LLC (recorded as a LICENSE); assignor: Horn, Bruce
• 2017-03-08 — assigned to Horn, Bruce (from Technology Innovations Associates, LLC)
• 2017-03-08 — assigned to EHIERARCHY LLC (from Horn, Bruce)
• Legal status: Expired – Fee Related; adjusted expiration 2024-05-19

So a 2013 Delaware plaintiff could plausibly have been Technology Innovations Associates, LLC, and a 2016–2017 plaintiff eHierarchy, LLC — but I have not verified this, and I will not state it as fact.

5. Bottom line

• One case is confirmed on the patent in the Stanford NPE database: eHierarchy, LLC v. Microsoft Corporation, No. 2:16-cv-00564 (E.D. Tex.) — filing date and outcome not available in the sources I retrieved.
• Nine additional docket numbers are listed on the patent's family-litigation record (eight 2013 D. Del. cases and E.D. Tex. 2:17-cv-00225), but I could not verify the parties, filing dates, or outcomes for those, and at least one of them (1:13-cv-00353) collides with an unrelated case number in other databases.
• No appellate (CAFC) or PTAB proceedings on 7,840,619 surfaced in my searches. I did not find any IPR, PGR, CBM, or reexamination tied to this patent.

If you need court-grade accuracy, the next step is to pull the Unified Patents case pages and the PACER dockets for each number in Section 2 to confirm plaintiff, defendant, filing date, and disposition, and to confirm whether each docket asserts 7,840,619 specifically or another member of family 30116028.

Sources retrieved:

• https://npe.law.stanford.edu/patent/7840619
• https://patents.google.com/patent/US7840619/en
• https://portal.unifiedpatents.com/litigation/Delaware%20District%20Court/case/1%3A13-cv-00353 (link listed; page not retrieved)

I'll start by checking the canonical source and then searching for any PTAB activity on this patent.

Let me search more specifically for PTAB proceedings involving this patent and its owner.

Let me run more targeted searches on the specific parties and the PTAB docket.

Proceedings overview

Total AIA trial proceedings on US 7,840,619: 0. The structured "PTAB proceedings on file" block — sourced from the USPTO Open Data Portal — returns no IPR, PGR, or CBM proceedings for this patent, and my independent web searches surfaced none either. The bottom-line defensive posture is therefore not "the patent has survived two IPRs and is hardened," but rather the far more useful: no petitioner has ever put these claims through an AIA trial, there is zero § 315(e) estoppel on the books, and the patent itself is showing an adjusted expiration of 2024-05-19 with a current legal status of "Expired – Fee Related." If you are being asserted against today, your problem is almost certainly a backward-looking damages problem, not an injunction problem — and every ground of invalidity remains unburdened by estoppel.

What I searched, and what I found

Because the canonical list is empty, the deliverable is the negative finding plus the corroboration trail. Stating this up front so nothing below is over-read:

Search-noise warning (this matters for diligence). A large fraction of "’619 + IPR" results on the web are different patents that merely share the last three digits:

• US 7,302,619 ("Error Correction in a Cache Memory") — subject of IPR2021-00872 / IPR2021-01042 (petitioner HD Silicon Solutions; Patent Owner's Preliminary Response arguing Haswell lacks an instruction cache). Unrelated to Horn.
• US 9,085,619 (Fraunhofer, "Anti-TNF Antibody Formulations") — IPR2017-00823, IPR2017-01008 (Gokarn ’011 anticipation). Unrelated.
• US 8,421,619 (LBT IP I LLC) — IPR2020-01192 (Apple; claims 1–20 unpatentable under § 103(a)). Unrelated.

Do not let a vendor's IPR summary sheet import any of those outcomes onto US 7,840,619. They are separate patents, separate owners, separate art.

District-court litigation context for the family (not PTAB, but relevant to your posture)

The Google Patents record for US7840619B2 carries a "Family has litigation" flag with these entries — note these are family-level dockets and I cannot confirm from this record which patent in the family was actually asserted in each:

• eHierarchy, LLC v. Microsoft Corporation, 2:16-cv-00564, E.D. Tex. — confirmed in the Stanford NPE Litigation Database as a case involving this patent (patent asserter category: "Acquired patents"). NPE DB record
• Eight Delaware District Court cases, 1:13-cv-00348, -00349, -00350, -00351, -00352, -00353, -00355, -00356 (2013). These post-date the 2013-11-13 reassignment to Technology Innovations Associates, LLC (license) and the 2017-03-08 assignments to Horn, Bruce and then EHierarchy LLC.
• Two E.D. Tex. cases: 2:16-cv-00564 and 2:17-cv-00225.
• A Darts-ip "First worldwide family litigation filed" entry.

The striking fact is the pairing: a patent litigated at least ten times across 2013–2017 drew zero AIA petitions. That is not the normal profile of a patent that stayed in active assertion. The likely explanations — and I flag these as hypotheses, not findings — are (a) quick settlements at the 2013 wave, (b) reliance on district-court invalidity rather than PTAB (2013 filings predate the Fintiv era and largely predate the post-Cuozzo IPR volume surge), and (c) a family that went dormant as the term ran down. I did not find evidence of an ex parte reexamination either, but I also did not run the reexam control-number search to exhaustion — verify in Patent Center.

Strategic summary

Claim status: entirely UNTESTED at the PTAB. No independent claim and no dependent claim of US 7,840,619 has been canceled, confirmed, or even reviewed in an AIA trial. There is no FWD to cite, no claim-level disposition to quote, and no Board construction to argue about. Anyone who tells you "claims 1–5 are already dead" is conflating this patent with one of the unrelated '619s above. I also note that the claim set itself is not reproduced in the authoritative text supplied to me (the fetch is truncated before the claims), so I will not state a claim count or claim numbers I have not verified — pull the claims directly from the Google Patents page or Patent Center before relying on any.

Estoppel landscape: wide open. Because no IPR, PGR, or CBM was ever instituted, 35 U.S.C. § 315(e)(2) estoppel does not attach to anyone. There is no petitioner and no privy barred from raising any § 102 or § 103 ground in your district-court case. Every printed publication, every patent, every system-art theory that was available in 2013 and every reference that has become citable since is fair game. Practically, this is the single best piece of news in this memo: you inherit the entire prior-art universe, minus nothing. The only frictions are the ordinary ones — invalidity contentions disclosure requirements, the § 282 burden of clear and convincing evidence, and whatever IPR estoppel applies to co-defendants or unrelated parties (it doesn't help you, but it doesn't hurt you either).

Pattern signals: none to report. No repeat petitioner exists because there are no petitioners at all. The patent owner has never had occasion to defend an FWD on appeal to the Federal Circuit, so there is no CAFC docket to read for how the claims fare on appeal. And critically, no defensive aggregator appears in the PTAB chain — Unified Patents shows up all over the modern dockets (e.g., as the ex parte reexam requestor against VirtaMove's US 7,784,058), but I found no Unified Patents or RPX filing against 7,840,619. The absence of an aggregator challenge to a ten-times-litigated patent is itself a signal that the family was no longer economically worth attacking by the time the aggregators scaled up.

The expiration point deserves its own paragraph, because it may end the analysis. The Google Patents legal-status data records an adjusted expiration of 2024-05-19 and a current status of "Expired – Fee Related." Google expressly disclaims this as an assumption rather than a legal conclusion — but if it holds, the patent term has run, and the practical consequences for a defendant asserted against in 2026 are: no prospective injunction, no ongoing royalty, and a damages case cabined to pre-expiration infringement, further cut back by the six-year lookback of 35 U.S.C. § 286 measured from the complaint's filing date. Verify this first, in USPTO Patent Center, before you spend money on anything else. Note that a fee-related lapse and an end-of-term natural expiration are different events with different consequences for a pre-lapse damages window, so read the Patent Center maintenance-fee history and any sui generis revival petitions rather than trusting the summary label.

Recommended next steps

1. Confirm the expiration status before anything else. Pull the maintenance-fee and term history in USPTO Patent Center for US 7,840,619 and the family continuations (US 8,055,692; US 8,280,932; US 9,229,945; US 9,483,169 — the later ones are the live risk if you are being asserted today). If the '619 has lapsed and the continuations differ in claim scope, your real fight may be with a sibling patent, not this one.

2. Run the PTAB E2E / ODP docket search yourself as a belt-and-suspenders check. The canonical block says zero, and I corroborated zero, but confirm at PTAB E2E and via the ODP API for the patent number and for the patent-owner names (EHierarchy LLC, Technology Innovations Associates, LLC, Horn, Bruce). Check both the patent number and the owner names, since early petitions are sometimes indexed by owner.

3. Screenshot the empty PTAB result set now. If you are in litigation, the absence of any prior AIA challenge is a fact you will want to be able to prove cleanly — and it defeats any argument that the claims "have been tested and survived."

4. Treat the prior art as unconstrained. With no § 315(e)(2) estoppel anywhere in the chain, build your invalidity case without the usual carve-outs. If the patent is expired, evaluate ex parte reexamination as a lower-cost path to a validity adjudication that does not depend on the district court — but confirm first whether an expired patent's claims remain worth canceling given no prospective infringement.

5. If you do find an active assertion, the trial-stage milestones are inapplicable here — there is no institution decision deadline, no oral hearing, and no statutory one-year FWD clock running, because no trial was ever instituted. Anybody who hands you a PTAB scheduling order for this patent has pulled the wrong docket.

6. Do not cite an FWD for US 7,840,619. There isn't one. If a demand letter or an expert report asserts claim-level PTAB outcomes for this patent, that is a factual error worth putting in writing immediately — and it is exactly the kind of conflation (7,302,619 / 8,421,619 / 9,085,619) that practitioners are making in the wild.

Confidence statement. I have high confidence in the central finding — no AIA trial proceeding has been filed against US 7,840,619 — because the canonical USPTO ODP list is empty and two independent search passes found nothing. I have lower confidence in the litigation-family details (the eight Delaware 2013 dockets are family-level flags that I could not map to specific patents in this fetch), and I have not verified the expiration status beyond Google's summary legal-status field. Treat items 1–3 above as the verification work, not as optional diligence.

I'll research the assignment record for US 7840619 across the sources you specified.

Let me dig into the specific assignment records and the litigation history.

US 7840619 — Ownership Chain Reconstruction

Patent: US 7,840,619 B2 — "Computer system for automatic organization, indexing and viewing of information from multiple sources"
App: 11/903,304 · filed 2007-09-21 · priority 2002-07-16 · granted 2010-11-23
Family members: US 8,055,692 · US 8,280,932 · US 9,229,945 · US 9,483,169

⚠ Data-gap notice (read before using this report). Assignment records for this patent do exist — three post-issuance reassignment events are indexed on Google Patents. However, the USPTO Assignment Center abstract of title (which carries the reel/frame numbers and the correspondent of record) was not retrievable in this research session; the Assignment Center search endpoint returned no content I could verify. I therefore do not have reel/frame numbers or recorded correspondent names for this chain. I have not invented them. Every reel/frame field below is marked not retrieved. The dates, conveyance types, and party names below are taken from the Google Patents legal-events record, which reproduces the assignment data but strips reel/frame and correspondent. Verify at assignmentcenter.uspto.gov and the legacy assignment.uspto.gov/patent/index.html before relying on this for anything.

Inventors

Pattern notes:

• Single-inventor patent, originally owned by the inventor personally. Google Patents lists the Original Assignee as "Individual," i.e. no corporate assignee ever stood between Bruce Horn and the patent. The "inventors depart the assignee within 12 months of filing → fire-sale" pattern is inapplicable here because there was never a corporate assignee to depart from.
• The operative pattern is instead inventor-as-monetizer: personally held rights were (a) exclusively licensed to an assertion LLC in 2013, then (b) pulled back to the inventor and (c) re-assigned to an inventor-affiliated LLC in 2017. That is the signature of a self-directed assertion campaign, not of a corporate portfolio divestiture.

Original assignee

• Entity on the issued patent: the individual inventor, Bruce L. Horn (Google Patents: Original Assignee — Individual). No company is named on the face of US 7,840,619.
• Product embodying the claims: The specification describes a fully specified desktop application, the Metadata Filing System (MFS) — object store + metadata catalog, classifier, notifier, updater, synchronizer, "sticky paths," view-by-reference — stated to be "preferably implemented… executable on the Apple Macintosh OS9 and OSX operating systems." Whether a commercial MFS release actually shipped is not established by any record I retrieved. Mark this unclear, not "no product."
• Primary line of business: N/A (natural person).
• Current status: Patent Expired – Fee Related; Google Patents shows an "adjusted expiration" of 2024-05-19. The patent lapsed rather than being maintained to full term — consistent with a monetization program that had wound down, though this is an inference from status alone and I flag it as such.

Assignment timeline

All three entries below are the complete set of post-issuance reassignment events indexed for this patent. Reel/frame and correspondent fields: not retrieved (see data-gap notice).

1. Executed/recorded 2013-11-13 — Reel not retrieved

• Conveyance: LICENSE (Google Patents record text: "LICENSE (SEE DOCUMENT FOR DETAILS)")
• Assignor: HORN, BRUCE
• Assignee: TECHNOLOGY INNOVATIONS ASSOCIATES, LLC
• Correspondent: not retrieved — the recurring-correspondent test could not be run (see Signal 3)
• Context: Transfer-to-asserter. An exclusive license out of the inventor to a licensing vehicle, created roughly contemporaneously with an eight-plus-defendant assertion campaign. Note the sequencing problem flagged below.

2. Executed/recorded 2017-03-08 — Reel not retrieved

• Conveyance: ASSIGNMENT OF ASSIGNORS INTEREST
• Assignor: TECHNOLOGY INNOVATIONS ASSOCIATES, LLC
• Assignee: HORN, BRUCE
• Correspondent: not retrieved
• Context: Unwinding of the 2013 license / internal reorg. Rights are pulled back from the LLC to the individual inventor.

3. Executed/recorded 2017-03-08 — Reel not retrieved

• Conveyance: ASSIGNMENT OF ASSIGNORS INTEREST
• Assignor: HORN, BRUCE
• Assignee: EHIERARCHY LLC (current assignee of record)
• Correspondent: not retrieved
• Context: Transfer-to-asserter (same-day, second hop). Entries 2 and 3 are a two-hop chain executed on the same date, with Bruce Horn as the common principal (assignee in hop 2, assignor in hop 3) — i.e. the chain was engineered as a single transaction. Fifteen days later eHierarchy filed its next suit.

Documented discrepancy in the record

The recorded transfer of rights to eHierarchy LLC is dated 2017-03-08, yet **eHierarchy, LLC was already the plaintiff of record in eHierarchy, LLC v. [Microsoft Corp.](/litigations/by-plaintiff/Microsoft%20Corp.), 2:16-cv-00564 (E.D. Tex.), filed 2016-05-25** — ten months earlier. Either (a) the Google Patents date is the recordation date of an earlier-executed agreement, or (b) eHierarchy asserted in 2016 on a right not yet recorded. Both readings are material to the "clean standing record" analysis in Signal 5, and neither can be resolved without the actual assignment documents (reel/frame). Flag as unresolved.

Timeline diagram

timeline
    title Ownership of US 7840619
    2002 : Priority filing by Bruce L Horn
    2007 : Continuation application filed
    2010 : Patent issues to inventor as individual
    2013 : Exclusive license to Technology Innovations
         : Eight or more Delaware suits filed
    2016 : eHierarchy LLC sues Microsoft in EDTex
    2017 : Rights returned to Bruce L Horn
         : Assigned to eHierarchy LLC
         : eHierarchy sues BlackBerry
    2020 : eHierarchy sues Tumblr in SDNY
    2024 : Patent expired for unpaid fees

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT

Two distinct licensing-only transferees, each supported by concrete, non-name-based evidence:

• Technology Innovations Associates LLC (per the March 2013 complaint it filed, Technology Innovations Associates LLC v. Pantech Corp., D. Del.): a Delaware LLC whose stated principal place of business is 21301 S. Tamiami Trail, #320 MS 337, Estero, FL 33928 — a suite/mail-stop address of the type used as a correspondence drop. The complaint describes it as "the sole and exclusive licensee of the '619 Patent" holding "the exclusive right to take all actions, including the filing of this patent infringement lawsuit." Its only visible activity is filing at least eight lawsuits, numbered 1:13-cv-00348 through 1:13-cv-00356 (D. Del.), against handset makers within roughly two days. No product.
• eHierarchy LLC — filed a Rule 7.1 corporate disclosure statement stating "No Corporate Parent" in eHierarchy LLC v. Tumblr, Inc., 1:20-cv-00886 (S.D.N.Y.). A standalone, no-parent entity that is the plaintiff in three separate infringement actions (2016, 2017, 2020).

Rights moved inventor → LLC → inventor → LLC (2013-11-13 and 2017-03-08 entries). No transferee in this chain sells anything.

2. Known asserter in the chain — PRESENT (moderate)

eHierarchy, LLC is indexed in the Stanford NPE Litigation Database, patent page npe.law.stanford.edu/patent/7840619, with Asserter Category: "Acquired patents." Both eHierarchy matters are also carried in RPX Insight (e.g. eHierarchy v. Microsoft, 2:16-cv-00564).

Caveat: I found no evidence tying either Technology Innovations Associates LLC or eHierarchy LLC to the specifically enumerated rosters (Acacia, Marathon, IV, IPNav, Wi-LAN/Mosaid-Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Spangenberg entities). The signal rests on the Stanford NPE-database classification, not on a name match to a high-frequency-plaintiff list.

3. Repeat correspondent across the chain — UNCLEAR (cannot be evaluated)

This is the signal the report was supposed to carry, and it is the one I could not test. The assignment correspondents of record were not retrieved — no reel/frame, no attorney/firm name on any of the three recordings. I will not guess at them.

What I can report is a separate field — litigation counsel — which is not the same thing and should not be treated as a substitute:

• 2013 Delaware suits (Technology Innovations): Richard D. Kirk (D. Del. No. 0922) and Stephen Brauerman (No. 4952), Bayard, P.A., per the complaint signature block in 1:13-cv-00355.
• eHierarchy v. Microsoft (2:16-cv-00564): plaintiff represented by Richard C. Weinblatt.
• eHierarchy v. Tumblr (1:20-cv-00886): plaintiff represented by Richard Charles Weinblatt.
• eHierarchy v. BlackBerry (2:17-cv-00225): plaintiff's attorney listed as Stafford Grigsby Helm Davis.

Weinblatt recurs as eHierarchy's litigation counsel across at least two matters. That is a litigation-counsel recurrence, and it is not evidence of a repeat recording correspondent. Signal remains unclear pending the Assignment Center abstract of title.

4. Cascading transfers — PRESENT (moderate)

Two consecutive recorded links on a single day, 2017-03-08 (Technology Innovations Associates LLC → Horn; Horn → eHierarchy LLC), with a common principal, Bruce Horn, appearing on both sides of the pair. Functionally this is a same-day chained conveyance. Weaknesses: the intermediate link runs to a natural person, not to a second LLC, and the 2013-11-13 → 2017-03-08 interval is ~40 months, so the "multiple LLCs in <24 months" fact pattern is only partly met. Marked PRESENT on the strength of the same-day double-hop and the shared principal.

5. Pre-litigation transfer — PRESENT

• eHierarchy LLC assignment dated 2017-03-08 → eHierarchy, LLC v. BlackBerry Corporation, 2:17-cv-00225 (E.D. Tex.), filed 2017-03-23 = 15 days. On its face this is a pre-filing title-cleaning transfer.
• Counter-current fact that cuts the other way: eHierarchy filed suit against Microsoft on 2016-05-25, before the recorded 2017 assignment. If 2017-03-08 is a recordation date rather than an execution date, the "arranged to enable assertion" reading weakens considerably. Unresolvable without the assignment documents.
• Also note the 2013 license is dated/recorded 2013-11-13 — after the March 2013 Delaware complaints. If that is a recordation date, the license was recorded post-suit (a "sue first, paper later" ordering), which is itself a standing/venue artifact.

6. Bankruptcy fire-sale — NOT PRESENT

No bankrupt or insolvent assignor identified in this chain. Technology Innovations Associates LLC voluntarily licensed rights back; no Chapter 7/11 sale, no §363 auction, no court-supervised transfer appears in any record I reached.

7. Privateering — NOT PRESENT

Privateering requires an operating company transferring to an NPE that asserts against the operating company's competitors. Here the only non-inventor transferor is a licensing LLC, and the only individual transferor is the inventor himself. No SEC 10-K/8-K disclosure of a patent-transfer or assertion-support agreement was located. The subject matter (a Mac desktop information-management application) is not a competitor's product line.

8. Defensive aggregator (anti-NPE) — NOT PRESENT

The chain terminates at eHierarchy LLC, an asserter. There is no RPX, AST, LOT Network, Unified Patents, or OIN link anywhere in the record. The patent was not neutralized defensively; it lapsed for non-payment in 2024 instead.

Verdict

NPE — high confidence.

Three independent signals are supported by citable evidence rather than naming intuition: (i) the 2013-11-13 license to Technology Innovations Associates LLC, a Delaware LLC at a Florida mail-stop address that self-identified as "sole and exclusive licensee" and immediately filed at least eight suits (1:13-cv-00348 through 1:13-cv-00356, D. Del.) against handset makers; (ii) the same-day double transfer on 2017-03-08 (Technology Innovations → Horn → eHierarchy LLC) followed 15 days later by the BlackBerry complaint, 2:17-cv-00225; and (iii) eHierarchy LLC's classification as an "Acquired patents" asserter in the Stanford NPE Litigation Database at npe.law.stanford.edu/patent/7840619, with no-parent Rule 7.1 disclosure and serial filings in 2016/2017/2020. No entity in the chain sells a product, and the patent was abandoned for unpaid fees in 2024.

Two caveats I will not paper over: the correspondent-recurrence signal (Signal 3) is untested because the Assignment Center reel/frame and correspondent fields were not retrieved, and the 2017-03-08 execution-vs-recordation ambiguity could materially soften Signal 5. Confirm both against the Assignment Center abstract of title before treating this as a final finding.

Verify:

• USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search patent 7840619)
• Legacy Assignment Search — https://assignment.uspto.gov/patent/index.html
• Google Patents legal events — https://patents.google.com/patent/US7840619/en
• Stanford NPE Litigation Database — https://npe.law.stanford.edu/patent/7840619

I'll run targeted searches for the '619 patent's own front-page citations and the prior art cited against it, holding the number 7840619 literal.

The generic queries aren't surfacing the front-page citation list. Let me pull the patent's own citation data from Google Patents, FreePatentsOnline, and the USPTO full-text record.

Prior Art Analysis — US 7,840,619 B2 ("7840619")

0. Search log and ID discipline

• I held the identifier 7840619 literal throughout and excluded non-patent hits with the same number string — in particular PMID 7840619 (Gremse DA et al., "Effect of pyridoxal 5'-phosphate on the function of the purified mitochondrial tricarboxylate transport protein," Arch Biochem Biophys 316(1):215–9, 1995), which appears repeatedly in keyword search results and is not the patent.
• I also discarded results for similar patent numbers that surfaced (e.g. US 7,840,617, US 6,840,619-family noise, US 7,840,630-type neighbors, the '139/'649 shorthand in other patents' citation lists). None of those is 7,840,619.
• The one source that reproduced the front matter of 7840619 was uspto.report: https://uspto.report/patent/grant/7840619 (retrieved 2026-09-21). That is the basis for the citation list below.
• I could not retrieve the full issued claim set from the authoritative full text supplied to me (the fetch stops before the claims), but I did obtain the text of claim 1 from the uspto.report record.

1. ⚠ Correction to the previously generated summary: claim 1 is now verified, and it is NOT what we inferred

The earlier section flagged claim language as unverified and inferred that the independent claims would be directed to the object-store/catalog/collections architecture. That inference is contradicted by the record I retrieved. For 7840619, claim 1 as published in the grant is a "sticky path" user-interface claim:

"1. A computer data processing system having a central processing unit configured with an integrated computer control software system for the management of informational objects relating to multiple levels of objects organized in at least one of an expandable hierarchy structure and a group structure … a) a computer readable memory including a storage structure … b) a computer display connected to said memory for displaying said objects … in a view; c) a computer-user interface for selective display for viewing of at least a portion of said expandable hierarchy or group structure in said view; d) an applications program having component architecture code processed by said central processing unit so as, upon user scrolling of said structure, to continuously render visible, with dynamic updating, in at least one dynamically-updating sticky path display portion at a margin of said view; and e) said dynamically-updating … so as to provide the function of at least one of: i) when the scrolling enters the beginning of a new open branch … an identifier thereof is rendered in a sticky path display area as a new identifier … adjacent any visible identifier of a previously-entered and still open branch that directly contains said new branch …; and ii) when the scrolling continues past the end of the entries of said new branch … said new identifier of said new branch is removed … while any identifiers of previously-entered branches remain until they in turn are exited; thereby to provide continuous automatic dynamically-updated display … of at least one branch identifier of said structure."
— US 7,840,619 B2, claim 1 (as reproduced at https://uspto.report/patent/grant/7840619)

This is corroborated by the continuation family. In the '619 continuation US 9,483,169 B2 (app. 14/984,960), independent claim 1 recites "property identifiers … organized within a plurality of groups," a "first scrollable area," and a "second, independently scrollable area" whose organizational identifiers update dynamically on scrolling — the same sticky-path concept (claim text at https://patexia.com/us/application/14984960).

Consequence for this task: the § 102 inquiry for 7840619 must be run against a hierarchy-display/scrolling claim, not against the metadata-catalog claim I earlier predicted. Front-page art directed to file/metadata organization is therefore, on its face, weaker anticipation art for claim 1 than it would have been for the architecture claims. I flag this squarely so the prior sections are not relied upon.

2. The references cited on the face of US 7,840,619

From the grant front page (examiner: John E. Breene, Primary; Dennis Myint, Assistant; attorney Jacques M. Dulin, Innovation Law Group; Field of Search 707/102, 829, 830):

Important caveat on provenance. uspto.report presents this list under a heading rendered as References Cited [Referenced By]. Because every listed item pre-dates the 2003-07-16 filing of parent application 10/621,689 (while patents citing 7840619 all post-date the 2010-11-23 grant), these are unambiguously references cited in/against the '619 family, not later patents referencing it. That said, I could not independently confirm whether a given item was cited in 11/903,304 itself or carried over from the parent 10/621,689 file; the substantive effect on the § 102 analysis is the same either way.

3. Per-reference § 102 assessment

Analytical frame. 7840619 is pre-AIA (priority 2002-07-16 provisional 60/396,439; parent 10/621,689 filed 2003-07-16; '619 filed 2007-09-21 as a divisional). The critical date for § 102(a)/(b) is the invention/priority date. A single reference anticipates only if it discloses every limitation of the claim as arranged; front-page citations such as these are more commonly applied in § 103 combinations (indeed, the asserted litigations over this family turned on obviousness and § 112 issues as much as on anticipation). I therefore state, per reference, both (i) what could be read on claim 1 and (ii) what is plainly missing.

3.1 US 5,425,139 (Williams et al.), June 1995 — Earliest citation on the page. A 1995 GUI/file-display patent from the Apple/Mac-era art. It can plausibly be read on claim 1 elements (a) memory, (b) display, (c) user interface for selective display of a hierarchy. It cannot, on the face of the record, disclose claim 1(d)–(e) (a "sticky path display portion at a margin of said view" whose branch identifiers are added/removed as scrolling enters and exits branches). → No anticipation of claim 1; potential § 103 base reference for a display/hierarchy claim, or § 102 against a dependent claim reciting only a scrollable hierarchical view (if any).

3.2 US 5,544,360 (Lewak et al.), August 1996 — Mid-1990s hierarchical data-storage art. Bears on claim 1's "storage structure for storing information relating to said objects" and the container/path limitations. Says nothing about a dynamically-updating margin pane driven by scroll position. → No anticipation of claim 1.

3.3 US 5,899,997 (Ellacott), May 1999 — Late-1990s data/hierarchy management art. Same analysis: possibly element (c); not (d)–(e). → No anticipation of claim 1.

3.4 US 6,094,649 (Bowen et al.), July 2000 — Examined by the same art unit as work on document organization/classification, i.e., the "properties/metadata" side of the '619 disclosure. This is the most plausible § 103 partner for the metadata-query and key-phrase-classifier subject matter, and would be relevant if any claim recited Boolean metadata querying or key-phrase collection membership — but not to the sticky-path claim 1. → No anticipation of claim 1.

3.5 US 6,401,097 (McCotter et al.), June 2002 — Just predates the 2002-07-16 priority date, so potentially § 102(a)/(b) art. Otherwise same gap as 3.1–3.3. → No anticipation of claim 1.

3.6 US 6,457,017 (Watkins et al.), September 2002 — Issued after the 2002-07-16 priority date. If the '619 claims are entitled to the provisional's date (they should be, if the sticky-path disclosure appears in 60/396,439), this reference is not § 102(a)/(b) art against them; it would be at most § 102(e) art as of its own filing date. → Not prior art as of the 2002-07-16 date unless the priority claim fails; no anticipation of claim 1.

3.7 US 6,523,022 (Hobbs), February 2003 — Post-dates the 2002-07-16 priority date by ~7 months. Same § 102(e)-only status as 3.6. → No anticipation of claim 1.

3.8 US 6,691,282 (Rochford et al.), February 2004 — More than a year after the 2003-07-16 parent filing and ~19 months after the provisional. This reference can only be § 102(e) art (as of its US filing date, if that precedes the invention date), never § 102(a)/(b). → No anticipation of claim 1.

3.9 US 2003/0088593 A1 (Stickler), published May 2003 — A published application. Its publication date post-dates the 2002-07-16 priority date, so it is not a § 102(b) printed publication against the '619 claims. Its only § 102 route is § 102(e) (published application deemed prior art as of its US filing date). ⚠ I could not retrieve Stickler's filing date in this session, so I cannot determine whether the § 102(e) date precedes the critical date. Only if it does, and only if it discloses the sticky-path limitation, could it anticipate — and on its face it appears directed to metadata/organization subject matter, not the scrolling margin pane. → No anticipation of claim 1 demonstrated.

3.10 Dourish et al., "Using Properties for Uniform Interaction in the Presto Document System," UIST 1999 — The only non-patent literature on the face of the patent, and the most substantive citation on the page. Presto (Xerox PARC) is a document system in which properties — including user-defined ones — provide uniform interaction across a document store, and grouping/collections are property-driven. This is genuinely material § 102(b) art (printed publication, 1999, >1 year pre-priority) for any claim reciting user-definable properties attached to document objects and used to gather objects into groups. It is not sticky-path art. → Potential anticipation of property/metadata-grouping claims (if any exist in this family); no anticipation of claim 1.

4. Explicit uncertainty flags (per the operating rules)

1. Reference subject matter is not verified. My searches returned the citation list but not the text of most cited references. The "brief description" column is therefore a genre-level calibration, not a verified characterization, and any § 102 mapping drawn from it must be confirmed against each reference (e.g., via USPTO Patent Public Search / Google Patents) before reliance. I have not stated any reference title I could not support, and I have not fabricated titles.
2. Day-precision dates (issue dates) were not retrieved — uspto.report gives month/year only.
3. Claim scope is only partially verified. Only claim 1 was obtained. If 7840619 contains further independent claims (e.g., a system/method claim to the catalog-and-collections architecture), the relevance of items 1–10 rises sharply. The remaining claim set should be pulled from the USPTO PatentCenter record for 11/903,304 before this analysis is finalized.
4. Cross-family caution — do not confuse art. Two Horn-family items surface in searches that are not prior art against 7840619 and must not be treated as such: (a) US 2004/0177319 A1 (Horn) and the parent US 7,275,063 — same family, so they cannot anticipate; and (b) the LG Electronics IPR petitions at ptacts.uspto.gov (petitions 1557189 / 1557190) that use Horn's US 2004/0177319 A1 as prior art (exhibit "LG-1021") in combination with "Sporer." Those petitions challenge a different patent (work there references US 9,818,449), not 7840619, and the reference relied on is Horn's own family member. The earlier section's warning on this point stands: no verified IPR nexus to 7840619.
5. The most likely missing prior art for claim 1 is pre-2002 outline/tree/breadcrumb UI art (expandable list views with persistent ancestor-path display), which is not among the front-page citations I retrieved. Because claim 1 is a sticky-path claim, the cited art above does not appear to be the art that would decide claim 1 under § 102.

5. Bottom line

• The references cited on the face of US 7,840,619 are: US 5,425,139 (Williams, 6/1995); US 5,544,360 (Lewak, 8/1996); US 5,899,997 (Ellacott, 5/1999); US 6,094,649 (Bowen, 7/2000); US 6,401,097 (McCotter, 6/2002); US 6,457,017 (Watkins, 9/2002); US 6,523,022 (Hobbs, 2/2003); US 6,691,282 (Rochford, 2/2004); US 2003/0088593 A1 (Stickler, 5/2003); plus the non-patent publication Dourish et al., UIST 1999 (Presto).
• Verified correction to the earlier sections: claim 1 of 7840619 is the sticky-path scrolling-display claim, so the front-page art above is not anticipatory art for claim 1 on the record available. Items 6, 7, 8 and 9 post-date the 2002-07-16 priority date and are, at most, § 102(e) art.
• The only citation with obvious § 102(b) punch is Dourish et al. (1999), and it targets the property/metadata-grouping aspects, not the claimed scrolling feature.
• If precise § 102 anticipation contentions are needed, the next steps are mandatory: (i) retrieve the complete issued claims for 11/903,304 from PatentCenter; (ii) retrieve the front pages/text of the nine patent references to confirm titles, filing dates (for the § 102(e) items 6–9) and disclosures; and (iii) identify pre-2002 outline/tree-view art, which the examiner did not cite and which is the art actually aimed at claim 1.

I'll ground this analysis in the actual prior-art record for the '619 patent. Let me pull the front-page references and the PTAB/litigation art that has been asserted against the family.

Let me pull the actual reference lists and any post-grant challenges that identify prior art against this family.

Obviousness Analysis — US 7,840,619 B2 under 35 U.S.C. § 103

Analyst note on scope and contradictions (flagging per instructions):

1. The task instructs me to "use the results from the Prior Art section of this page," but the page as fetched contains no Prior Art / "References Cited" section. The Google Patents fetch of https://patents.google.com/patent/US7840619/en supplied bibliographic data, the "Definitions" extract, the specification body, figure list, classifications, and abstract — but it is truncated before the citation/reference lists and before the claims. I therefore could not use an examiner-supplied art table; I built the art set from (a) the patent's own admissions in the specification, (b) the family data I could retrieve, and (c) external searches. This is a substitution, and I flag it as such.
2. Date conflict: the system context says today is 2026-09-21 (the fetch timestamp), while the task header says "Current Date: April 26, 2026." I have treated 2026 as the operative year and note the inconsistency rather than resolving it silently.
3. Claim language remains unverified (consistent with the earlier-generated section). I do not treat any hypothetical claim text below as quotation. The analysis is organized as claim-element hypotheses keyed to the specification's own limiting language, so it can be re-run against the issued claims once pulled from USPTO PatentCenter for application 11/903,304.

1. Governing legal framework and the critical-date question

The '619 patent was filed 2007-09-21 as a continuation, claiming benefit back to application 10/621,689 (filed 2003-07-16) and provisional 60/396,439 (2002-07-16). Because the filing date precedes 2013-03-16, pre-AIA § 102/§ 103 applies. That matters in two ways:

• § 102(b) statutory bar runs from the earliest U.S. filing to which the claims are entitled. If the claims are supported by the 2002-07-16 provisional, the bar date is 2002-07-16; if not, the parent filing date of 2003-07-16 governs. Every reference I use as a primary ground below is comfortably before 2002 (Gifford 1991; Freeman/Gelernter 1999; Aho-Corasick 1975), so the analysis is robust either way.
• § 102(e) art (published applications/patents by filing date) becomes available only against the later date. This is where US 2004/0215600 A1 (the "XML-aware file system," retrieved at https://patentimages.storage.googleapis.com/78/4d/0d/12daae98bbd115/US20040215600A1.pdf) sits: its publication is 2004-10-28, after both candidate critical dates, so it is not § 102(b) art and is usable only if its own filing date precedes 2003-07-16 and the '619 claims are not entitled to the provisional — I could not confirm its filing date from the retrieved text. Treat XMLFS as confirmable-by-docket art, not as a ground I can assert today.

One more structural point that materially helps a § 103 challenge: the specification's "characterizing" paragraph reads like a paraphrase of the system claim — "at least one central processing unit configured with client operating system and applications software… said central processing unit includes at least one memory coupled to a bus; said memory including selected program structures stored therein, including an operating system program structure, at least one client and/or server system management program structure, at least one hierarchical data storage management system program structure, and selected application program code structures including the novel MFS code structure…" If the independent claim tracks this language, the hardware recitations are conventional and the patentability must reside in the programmed functionality — precisely the posture in which KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) most readily sustains obviousness.

2. The reference set

Could not verify this session (flagged, not relied upon): BeOS BFS live queries on typed file attributes (c. 1997–2000); Extensis Portfolio (1992), Canto Cumulus (1992), and iView MediaPro (1999) as catalog-and-multiple-category digital asset managers; Apple Data Detectors / Sherlock (1996–98) for in-text entity hyperlinking. Each is, in my prior knowledge, squarely on point for "multiple logical groupings of one stored asset" and "key-phrase detection in text." They should be verified against primary documentation before being asserted in a filing — I am not treating them as grounded here.

3. Element-by-element mapping and the grounds

Ground A — R1 + R3 + a conventional GUI file browser (R4)

Why a POSITA would combine R1 and R3: R1 supplies the complete organizing skeleton — auto-extracted attributes in a separate index, and groupings that are queries rather than physical directories — but implements queries over a bounded field:value syntax. R3 is a 1975 general-purpose tool for scanning text for many patterns in one pass, and R1's own transducer model already treats the unique words of a text file as attributes, i.e., SFS's "texts:" transducer is literally full-text indexing. Substituting a multi-pattern automaton for SFS's per-word text extraction is "the use of a known technique to improve similar devices in the same way" (KSR). Nothing in either reference teaches away; R1's conclusion invites richer query languages and additional transducer types.

Ground B — R1 (or Ground A) + R2 ('227) for the time-based / "viewing by reference" elements

The "Today" collection, the dynamically maintained time collections ("Last Week," "Last Month," "Last Year"), and the cross-reference shelf are, element-for-element, R2's disclosure: a substream created by a query that is thereafter automatically maintained, plus time-indexed streams. R2a even discloses substreams of substreams by booleans — the "Today ∩ Received email ∩ Financial" refinement described at FIGS. 21–22.

Motivation: R2's stated object is "to organize information as needed instead of at the time the document is created," and that documents "may belong to as many streams as seems reasonable." That is the same design goal the '619 specification states as its unmet-need motivation. Combining a query-defined dynamic grouping (R1/R2) with time as one more queryable property is a predictable variation, and the spec itself concedes the need was "long felt."

Ground C — R1 + R2 + R3 as a three-way ground against the flagship "single instance, many collections" limitation

The '619 specification's own words for the alleged improvement: "storing only one instance of each object while linking the object to multiple collections… by unique metadata links." R2 describes precisely this behavior in stream terms ("the new calling card comes equipped with a system-created icon which is marked on all documents that are part of the new stream"). R1's virtual directory likewise references matching entities rather than copying them. The claim of novelty then narrows to the bookkeeping data structures — the pContainers/pObjects, pInsiders/pOutsiders, pInclusions/pExclusions link-metadata sets, and the catalog that "notifies other processes of the metadata property changes." Those are conventional many-to-many relational modeling plus a change-notification/dependency mechanism. A POSITA combining R1's attribute index with an OODB (the spec's own "object store"/"catalog" partition is standard ODBMS-plus-index practice, and R1 already separates descriptors from data) would arrive at this without invention.

Ground D — Consistency maintenance (Updater/Synchronizer/Notifier/Classifier + event queue)

The invalidate → re-fetch → SetValue → diff-against-cached-value → notify-dependents pipeline (FIGS. 32–35) is the classic dependency-graph/recompute-on-invalidate pattern known from spreadsheets and model-view update architectures, and the spec expressly analogizes to it: "the values of checks written must be taken into account when balancing the checkbook; reconciling the checkbook involves propagating values from reconciled checks to the current balance." R1 supplies the convergence requirement in express terms — SFS is characterized as needing query consistency and "convergent consistency." A POSITA seeking query-consistent virtual collections would necessarily adopt exactly this invalidation/notification loop.

Ground E — Sticky paths (FIG. 12a–b)

This is the weakest linkage to the art I could ground today. Hierarchical list/outline views with expand/collapse triangles were ubiquitous (Finder list view, Windows Explorer, NeXT/OpenStep browsers), and persistent location/breadcrumb panes existed in file browsers. The asserted novelty is narrower: a two-pane scrollable outline where the path pane dynamically re-renders to the topmost visible item as the body scrolls. That is an incremental UI refinement combining a breadcrumb with an outline control. Absent a verified pre-2002 reference teaching the dynamic sticking behavior, I would rate this element as the most defensible part of the patent and would not assert a § 103 ground on it without better art.

4. Motivation-to-combine synthesis (the KSR / Graham factors)

1. Same field, same problem. All of R1, R2, R3 and the admitted art address "organize and retrieve heterogeneous information without copying it." The '619 specification frames the problem identically and admits a "long felt need" — an admission that cuts against non-obviousness.
2. Express exhortation in the art. R1's conclusion proposes (i) richer query languages including disjunction, (ii) user-assigned attributes alongside transducer attributes, (iii) audio/video transducers, (iv) a "confederation" of semantic file systems. R2's claims are directed at cross-referencing streams. These are the very extensions the '619 claims — textbook "obvious to try" with a finite number of identified, predictable solutions.
3. Predictable combination of known techniques. Aho-Corasick applied to object text to drive group membership is a known algorithm applied to a known classification task, with no unexpected result; the automaton's output state list (R3) maps naturally onto a collection list (the spec's own description of the FSM nodes holding "a list of collections that specify that phrase" is functionally R3's output-set mechanism plus a lookup table).
4. No teaching away, no criticality. Nothing in R1/R2/R3 disparages the combination, and the specification asserts no unexpected result from the specific multiplex-sort design.

Best § 103 formulations to assert (subject to claim text):

• Primary 3-way ground: Gifford SFS + Freeman/Gelernter '227 + Aho-Corasick, further in view of the applicant's admitted Outlook/Office/ID3/EXIF/XMP art.
• Alternative: Gifford SFS + BeOS BFS live queries (pending verification of BFS documentation) + Aho-Corasick.
• For the catalog/object-store split only: Gifford SFS in view of any conventional OODBMS with secondary indexing and dependency notification.

5. Where the patent most likely survives — and what must be verified

Defensible limitations (do not attack without better art):

• The sticky-path dynamic path pane (element(s) keyed to FIG. 12a–12b). I have no verified pre-2002 reference teaching dynamic re-rendering of the path pane on scroll.
• The multiplex key-phrase classification with a differential set update — scanning an object's text once and then computing the symmetric difference between the object's prior collection set and the final set, adding/removing only the deltas rather than re-running per-collection queries. R3 gives the scan; the delta-update-on-membership bookkeeping is a narrow implementation choice that may read as specific to the claims.
• The UUID-based one-to-one mirroring plus refresh-on-external-change, if claimed with the specific reconciliation semantics (compare folder contents to folder object; create/delete to match exactly). R1 has no external-change reconciliation thread — although the added subject matter is thin.
• Whether the claims are entitled to the 2002-07-16 provisional. If they are not, the effective date moves to 2003-07-16 and a materially larger art set (including US 2004/0215600 A1 subject to its filing date, and any 2001–2003 WinFS/"Storage+" disclosures) becomes available. This single question is the highest-leverage item in a validity workup.

Required verification before filing anything:

• Verbatim independent claims from USPTO PatentCenter for 11/903,304 (the page as fetched truncates before the claims).
• The actual examiner-cited references — the Google Patents "References Cited"/"Cited By" tables were not present in the fetched page; this analysis substitutes external art.
• Filing dates for US 2004/0215600 A1 and the BeOS/DAM/Sherlock materials to fix their § 102 category.
• Status of IPR2025-00393 (LG Electronics), previously flagged with an exhibit list containing US 2004/0177319 A1 — a family member, not the '619. Do not assume a nexus without the PTAB docket. Docket context: https://npe.law.stanford.edu/patent/7840619 (eHierarchy, LLC v. Microsoft Corp., 2:16-cv-00564, E.D. Tex.; Delaware 1:13-cv-00348–00356 cluster).

6. Bottom line

On the record I could ground, US 7,840,619's disclosed architecture — one stored instance per object, cross-linked into many query-defined and key-phrase-defined collections, with a mirrored object store and a metadata catalog — was rendered obvious by the 1991 Semantic File System paper in combination with Aho-Corasick (for the classification engine) and Freeman/Gelernter's document-stream patent (for dynamically maintained, time-based collections and cross-referenced substreams), with the specification's own admissions supplying the computing platform, the multi-grouping problem, and the long-felt need. The strongest residual claim scope is the sticky-path pane and the specific differential membership-update mechanics of the classifier; the weakest are the hardware recitations and the generic "collection" concept.

Two things must be fixed before this becomes an opinion rather than a hypothesis: the issued claim language (not present in the authoritative text I was given) and the examiner's actual cited art (the page contained no Prior Art section despite the task's premise). Absent those, the mapping above is directional and should be re-verified element-by-element against the printed claims.