US 6725427

I'll search for authoritative information on this patent and cross-check against the full text provided.

Let me check for any recent (2026) litigation or CAFC docket activity and confirm current ownership.

US Patent 6,725,427 B2 — Analyst Summary

Sourcing note: My primary source is the full patent text of US 6,725,427 as published on Google Patents (https://patents.google.com/patent/US6725427/en), cross-checked against USPTO reexamination records and Federal Circuit / district court documents surfaced by web search. The claim listing in the text I was given truncates mid-claim 27, so my coverage of claims 28–39 is necessarily incomplete — flagged below.

Bibliographic data

Assignee caution: The USPTO's 2009 reexamination order (control no. 95/001,172) states the '427 patent was then "assigned to MIRROR WORLDS, LLC. of TYLER, TEXAS." Apple's 2008 counterclaim filings state the patent "on its face identifies as assignee Mirror Worlds Technologies, Inc. of New Haven, Connecticut." There is a long chain of nunc pro tunc assignments in the Google Patents record (Recognition Interface, Inc./LLC, Plainfield Specialty Holdings I, Mirror Worlds LLC, Mirror Worlds Technologies LLC). I do not have high-confidence confirmation of the current 2026 owner of record; treat the assignee field as uncertain.

Abstract (as printed)

A document stream operating system and method is disclosed in which: (1) documents are stored in one or more chronologically ordered streams; (2) the location and nature of file storage is transparent to the user; (3) information is organized as needed instead of at the time the document is created; (4) sophisticated logic is provided for summarizing a large group of related documents at the time a user wants a concise overview; and (5) archiving is automatic. The documents can include text, pictures, animations, software programs or any other type of data.

Plain-language overview of the independent claims

The provided text shows independent claims 1, 8, 16, and 25, plus a fifth apparent independent claim beginning at 27 (truncated in my source). Notably, every independent claim ends with the same structural hook: the stream-based/controlling operating system utilizes subsystems from another operating system for writing documents to storage media, interrupt handling, and input/output. This is the "layer-on-top-of-Windows/Apple" architecture (dependent claims 3, 4, 11, 12, 20, 21 name Windows and Apple operating systems specifically).

Claim 1 — Stream-based OS with diverse-format ingestion, auto-archiving, and hover "glance view" over a stack
A stream-based operating system built on another OS's subsystems, comprising:

• a document organizing facility that (a) receives documents created by diverse applications in diverse formats, (b) automatically associates selected indicators with them (time-based per claim 2), (c) automatically archives documents and indicators in a consistent format for selective retrieval, and (d) automatically creates data specifying each document's glance view and document representation;
• a display facility that shows document representations as a receding, foreshortened stack of partly overlapping representations, so only part of each representation after the first is visible;
• the display facility shows a cursor/pointer and responds to the user sliding the cursor without clicking over the stack by displaying the glance view of whichever document is currently touched — i.e., hover-to-preview;
• the OS reuses the host OS's subsystems for storage, interrupts, and I/O.

Claim 8 — "Controlling operating system" variant (broader on display)
Same core architecture (diverse-application ingestion, automatic indicator association, automatic consistent-format archiving, auto-generated glance views and representations, hover-without-clicking glance view, host-OS subsystem reuse), but the base claim only requires the display facility to display at least selected document representations. The stack feature is moved into dependent claim 10. So claim 8 is the broader "glance view via hovering" claim.

Claim 16 — "Controlling operating system" variant without the explicit archiving phrase
A controlling operating system built on another OS, comprising: a document organizing facility that associates selected indicators (time-based per claim 17) with received or created documents and creates glance-view information and document-representation information; a display facility showing at least selected representations; and sliding the cursor/pointer without clicking over a portion of a displayed representation to show that document's glance view. Claim 19 adds that documents arrive in heterogeneous-application formats and the glance-view information is created so the views display in a consistent format. Again, the stack is relegated to dependent claim 18.

Claim 25 — Document stream OS with chronological indicators and the stack display
A document stream operating system built on another OS, comprising: a document organizing facility that associates chronological indicators with documents received from diverse applications in diverse formats and creates glance-view and representation information; a display facility showing the representations as a receding, foreshortened, partly overlapping stack (only part of each representation except the first visible); and sliding the cursor without clicking over the stack to display the glance view of the represented document currently touched. Claim 26 adds that chronological indicators are associated at receipt or creation without requiring a user to name the documents — the "no filenames needed" concept from the specification.

Claim 27 (truncated in my source) begins "A stream-based opera[ting system]..." and appears to be a further independent claim. I cannot characterize claim 27 or any independent claims in the 28–39 range with confidence, and I was unable to locate a clean full claim set in the search results. Treat the independent-claim count as "at least five, possibly more."

Common thread across all independents: the novelty is presented as (i) a document store that transparently ingests heterogeneous formats and auto-organizes/archives them, (ii) auto-generated "glance view" previews triggered by cursor hover (sliding without clicking), and (iii) implementation as a layer that borrows the underlying OS's storage/interrupt/I/O subsystems.

Litigation and post-grant history (as found)

• Inter partes reexamination 95/001,172, requested April 23, 2009 and granted July 1, 2009, raising a substantial new question of patentability against claims 1, 2, 5, 7–10, 13, 15–19, 22, 24–26, 29, 31–34, 37, and 39 over Mander et al. (US 6,243,724). Source: USPTO order, reproduced at https://cases.justia.com/federal/district-courts/texas/txedce/6:2008cv00088/[108627/221](https://assignmentcenter.uspto.gov/search/patent/reelFrameDetail?reelFrame=108627-0221)/18.pdf. I did not find the resulting reexamination certificate in my search results, so I cannot state the final confirmed/cancelled status of those claims.
• Mirror Worlds, LLC v. Apple, Inc., E.D. Tex. No. 6:08-cv-00088 (patents included the '427 patent). The 2012 Federal Circuit decision (Mirror Worlds v. Apple, 692 F.3d 1351) reversed a ~$625M jury award on damages grounds and held no infringement under the doctrine of equivalents because Apple's system lacks a "cursor or pointer," and equating absence-of-cursor with presence-of-cursor would vitiate that claim limitation.
• Mirror Worlds Technologies, LLC v. Facebook, Inc./Meta, Fed. Cir. (appeal no. 22-1709; the precedential December 4, 2024 decision). The asserted patents were US 6,006,227, 7,865,538, and 8,255,439 — the '427 patent was not among them. That panel affirmed summary judgment of non-infringement (including on the "glance view" limitation) and dismissed Facebook's cross-appeal.
• 2026 CAFC docket search for '427: I found no Federal Circuit docket activity in 2026 specifically involving US 6,725,427. Given the patent expired September 12, 2016, any remaining enforcement would be limited to past damages, so fresh appellate activity on this specific patent is unlikely. I flag this as a search limitation rather than a definitive negative.

Uncertainties / things I could not verify authoritatively

1. Full text of claims 27–39 is truncated in my source; I cannot summarize independent claims beyond 1, 8, 16, and 25 confidently.
2. Current 2026 owner of record for the '427 patent (Recognition Interface vs. Mirror Worlds Technologies, LLC vs. others).
3. Outcome of reexamination 95/001,172 (final certificate status per claim).
4. Any 2026 CAFC docket entry for this patent — none found.

If you need definitive answers on (1)–(4), I'd recommend pulling the USPTO PatentCenter "Reexamination Certificate" for 95/001,172 and the current assignment abstract directly from the USPTO Assignment Database, plus a PACER docket query under the patent number rather than by party name.

I'll research litigation involving US patent 6725427. Let me start with the leads already embedded in the patent's Google Patents page and then broaden out.

Good leads. Let me dig into the related cases (Mirror Worlds II and III), the Federal Circuit appeals, the Supreme Court petition, and any Network-1 follow-on suits.

Let me pin down the specific patents asserted in the follow-on cases and check for any other suits touching the '427 patent specifically.

Litigation involving US Patent 6,725,427 ("'427 patent")

Bottom line: I found one confirmed litigation campaign in which the '427 patent was actually asserted — Mirror Worlds, LLC v. Apple Inc., filed in the Eastern District of Texas in March 2008 and litigated through the Federal Circuit and the Supreme Court. The later Mirror Worlds/Network‑1 cases (2013–2024) that I could verify asserted the '227 patent (and its descendants '538/'439), not the '427 patent. I found no evidence of any other case in which the '427 patent (as opposed to its siblings) was asserted. Details and caveats below.

1. Mirror Worlds, LLC v. Apple Inc. — E.D. Tex. (the core '427 case)

Outcome / current status:

• Oct. 1, 2010 — Jury returned a verdict for Mirror Worlds, finding the '227, '313 and '427 patents valid and willfully infringed, awarding $208.5M per patent ($625.5M total).
• April 4, 2011 — Judge Davis granted Apple's JMOL, vacating the $625.5M verdict; the court found insufficient evidence of direct/induced infringement (and, as to the '427 specifically, no doctrine-of-equivalents infringement of the "cursor or pointer" limitation).
• Sept. 4, 2012 — Federal Circuit affirmed in Mirror Worlds, LLC v. Apple, Inc., 692 F.3d 1351 (Fed. Cir. 2012), Nos. 2011‑1392 and 2011‑1393 (http://patentlyo.com/media/docs/2012/09/11-1392.pdf). As to the '427 and '313 patents, the panel held Mirror Worlds failed to show the accused Cover Flow interface had an equivalent of the claimed "cursor or pointer"; the majority also found insufficient evidence of direct/induced infringement of the '227 method claims (a dissent disagreed).
• June 24, 2013 — Certiorari denied: Mirror Worlds LLC v. Apple Inc., U.S. Supreme Court No. 12‑1158 (https://portal.unifiedpatents.com/litigation/U.S.%20Supreme%20Court/case/12-1158). This ended the case. Reported by The Register (https://www.theregister.com/2013/06/24/supreme_court_denies_review_of_apple_mirror_worlds_patent_case/).

Related agency proceeding (not court litigation): Apple‑requested inter partes reexamination 95/001,172, filed April 23, 2009, covering the '427 patent. The USPTO granted the request, finding a substantial new question of patentability for claims 1, 2, 5, 7‑10, 13, 15‑19, 22, 24‑26, 29, 31‑34, 37 and 39 (see the PTO order in the 6:08-cv-00088 record, Dkt. 221 exh.). Note also the '427 patent is listed as Expired – Lifetime, adjusted expiration 2016‑09‑12, so it is no longer enforceable regardless of case outcome.

2. Follow‑on cases in the same family — '427 involvement NOT confirmed

After the Apple case, the portfolio passed to Network‑1 Technologies, Inc. (via its subsidiary Mirror Worlds Technologies, LLC, buying from Mirror Worlds, LLC/Looking Glass LLC in May 2013), and further suits were filed. From the record I reviewed, the asserted claims in these cases were the '227 patent and its later descendants ('538 and '439), not the '427 patent — but I want to be explicit that I did not conclusively verify the full list of patents‑in‑suit in each:

Network‑1's own SEC filings (e.g., Q3 2014 10‑Q; 2023 10‑K) recount that the Apple case was the prior litigation and that a $25M settlement gave Apple a fully paid‑up non‑exclusive license to the '227 patent. They do not describe the later suits as '427 cases.

3. Caveats / what I could not confirm

• I performed targeted searches and reviewed the patent's own litigation links, the EDTX docket excerpts, the Federal Circuit opinion, and Network‑1 filings. I did not find any other case in which the '427 patent was asserted — e.g., no separate NPE assertion of '427 against Google, Microsoft, HP, Samsung or others that I could verify.
• Some of these sources (Justia/CourtListener/RPX/UniCourt) are secondary aggregators, and docket coverage of the 2013–2016 cases is incomplete. I am therefore not able to rule out an additional suit that asserted '427 but left no searchable trace; I simply found none.
• If you need certainty, the authoritative check is the PACER docket for 6:08‑cv‑00088 plus USPTO Patent Center's "Litigation" / assignment history for 6,725,427, and Network‑1's 10‑K/10‑Q litigation footnotes for the post‑2013 period.
• I have taken patent and case numbers literally as they appear in the sources; where a source's identifier looked inconsistent (e.g., "6,725,427" vs. 6,725,427) I have not silently corrected it in substance, but note that the '427 patent appears in the record as U.S. Patent No. 6,725,427.

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

Key finding to run down: an apparent PTAB exhibit referencing US 6,725,427 (Freeman) in IPR2023-00242, plus the sibling-patent CBM activity. Let me verify carefully.

Two clarifications needed: (a) which patent the 2015–16 CBM petitions actually targeted, and (b) the outcome of the pre-AIA inter partes reexam 95/001,172.

I hit my tool-call budget, so here is the analysis based on the structured PTAB data block plus what my searches confirmed (and explicitly flagged where they didn't).

Proceedings overview

Total AIA trial proceedings on US 6,725,427: 0. The USPTO Open Data Portal structured block returned no IPR, PGR, or CBM proceeding for this patent, and every candidate my web searches surfaced turned out on inspection to be either (a) a proceeding against a different patent in the same family (US 6,006,227), or (b) a proceeding involving a different patent that merely shares the "'427" shorthand. The only government-side validity challenge ever mounted against US 6,725,427 was a pre-AIA inter partes reexamination (95/001,172, requester Apple) — which is not an AIA trial and produced no institution decision at the Board. Bottom-line defensive posture: this is not a "hardened" patent (nobody ever tried to kill it at the PTAB), and it is not a "claims already canceled" patent either. It is an untested, expired patent — expiration 2016-09-12 means any 2026 assertion is past-damages-only, and the Federal Circuit's 2012 "cursor or pointer" holding already gives a strong non-infringement defense.

AIA proceedings: none

(No IPR / PGR / CBM proceedings — none found in ODP or on web search)

• Type: n/a
• Filed: n/a
• Status: No AIA trial on file. This is the canonical ODP result and it is corroborated by targeted searching.
• Sources / link: USPTO PTAB E2E and Patent Center, https://ptacts.uspto.gov/ ; https://patentcenter.uspto.gov/

Near-misses and misattributions (important — do not let these be miscited against you)

These are the three things that look like PTAB activity on 6,725,427 but are not. If opposing counsel or a search vendor's docket report hands you one of these, it is wrong.

IPR2023-00242 — Apple Inc. and Google LLC v. SpaceTime3D, Inc. (U.S. 8,881,048, not the '427)

• Type: Inter Partes Review
• Filed: 2022-11-22 (patent 8,881,048; challenged claims 1–18)
• Relevance to '427: US 6,725,427 appears only as an exhibit. Docketalarm lists "Exhibit 2005 — US Patent No 6,725,427 Freeman et al," filed 2023-03-23 in IPR2023-00242. Exhibit 2001/2005-numbering is the Patent Owner's exhibit range, i.e., SpaceTime3D filed the Freeman patent as evidence — not as the challenged patent. US 6,725,427 was not under review. That is the whole of the connection.
• Outcome (for context, as to the '048 patent): The Final Written Decision found claims 14–18 of the '048 patent unpatentable; Petitioners' Request for Director Review (2024-07-15) argued the Board left the "rendered first webpage" limitation unresolved as to claim 1. Panel (after a panel-change order): APJs David C. McKone, Sheila F. McShane, and Frederick C. Laney. Appeal at Fed. Cir. No. 25-1022.
• Link: https://www.docketalarm.com/cases/PTAB/IPR2023-00242/Apple_Inc._v._SpaceTime3D_Inc/
• Defensive value: zero for or against US 6,725,427. Cite it only to rebut a claim that the '427 patent has been through the PTAB.

IPR2023-01240 — Peloton Interactive, Inc. v. NEC Corp. (a different "'427 patent")

• Indexing engines will return this as an "IPR on the '427 patent." It is not US 6,725,427. The claims recited in that proceeding ("play-list information," "user equipment," SIP/SDP authentication, "main content" / "sub content") bear no resemblance to the document-stream claims of US 6,725,427. The petition challenged "claims 1-9, 14-18, 20, 22, 32, and 34-36 of the '427 patent" — a different patent number abbreviated identically.
• Defensive value: a warning about number collisions, not about this patent.

CBM review of US 6,006,227 — the sibling, not the '427

• Type: Covered Business Method review
• Petitioner: a Mirror Worlds defendant (the IPWatchdog report does not name the petitioner in the passage I retrieved — flag: petitioner and CBM number not verified)
• Outcome: Institution DENIED, ~2016-05-26. Panel: APJs Thomas Giannetti, David McKone, and Barbara Parvis. The Board relied on the then-fresh Enfish v. Microsoft decision and held the challenged claims were not directed to an abstract idea, i.e., an improvement in computer functionality; it also agreed the '227 patent was not CBM-eligible (with Judge McKone concurring on the latter point).
• Link: https://ipwatchdog.com/2016/05/26/ptab-cafc-enfish-covered-business-method-mirror-world-patent/
• Why it matters for the '427: It is the closest thing to a PTAB merits read on this family's subject matter, and the Board came out for the patent owner on § 101. Note this is a different patent (6,006,227) and a different statutory vehicle (AIA § 18 CBM, now sunset). Do not represent it as a proceeding on 6,725,427.

The one real agency challenge: pre-AIA inter partes reexamination 95/001,172 (not an AIA trial)

95/001,172 — Apple Inc. (third-party requester) v. Mirror Worlds, LLC (patent owner)

• Type: Inter partes reexamination under pre-AIA 35 U.S.C. §§ 311–318 (this is not an IPR).
• Filed: 2009-04-23 (requester: Apple Computer, Inc., naming itself as "Third Party Requester and Real Party in Interest").
• Granted: 2009-07-01 — Decision Granting Inter Partes Reexamination; examiner Christopher E. Lee; Art Unit listed as 3992/7590 area (the transmittal is garbled — treat examiner/art unit as low-confidence).
• Status: I could not verify the final reexamination certificate in my searches. Flag: the terminal disposition — claims confirmed, amended, or canceled — is unverified. Given the patent's adjusted expiration of 2016-09-12 and the 2012–2013 appellate activity, a certificate very likely issued at some point, but I will not state its contents.
• Judge panel: n/a (Central Reexamination Unit, not the Board).
• Grounds (verbatim from the SNQ order and claim charts):
  • SNQ found as to claims 1, 2, 5, 7-10, 13, 15-19, 22, 24-26, 29, 31-34, 37, and 39.
  • Anticipation by U.S. Patent No. 6,243,724 (Mander): claims 16, 17, 18, 19, 22, 24, 25, 26, 29, 31, 32, 33, 34, 37, and 39.
  • Obviousness over Mander in view of the Retrospect User's Guide ("Retrospect"): claims 1, 2, 5, 7, 8, 9, 10, 13, and 15.
  • Secondary references in the exhibit list: Lucas (U.S. 5,499,330), Magellan Explorer's Guide, Using Lotus Magellan.
  • Note the claim structure this implies: the SNQ reached every one of the four independent claims I can read with confidence — claim 1 and 8 on obviousness, claims 16 and 25 on anticipation — plus the truncated claim 27 family (29, 32, 34, 37, 39).
• Patent assignment noted in the order (2009): "'427 Patent is currently assigned to: MIRROR WORLDS, LLC. of TYLER, TEXAS."
• Link: https://cases.justia.com/federal/district-courts/texas/txedce/6:2008cv00088/[108627/221](https://assignmentcenter.uspto.gov/search/patent/reelFrameDetail?reelFrame=108627-0221)/18.pdf
• Defensive value: The Mander-based anticipation and Mander+Retrospect obviousness theories are public, fully charted, and free to reuse — no AIA estoppel attaches to an inter partes reexamination. But apply one caveat below on pre-AIA § 315(c) estoppel for Apple and its privies. Also note the examiner granted the request, meaning the art in those charts cleared the "substantial new question of patentability" bar — a meaningful quality signal if you need to plead invalidity under Twombly/Iqbal.

Strategic summary

Claim-level status: everything is UNTESTED; nothing is CANCELED. No claim of US 6,725,427 has ever been canceled by the Board. The only claim-level invalidity record is administrative: the 2009 SNQ finding (a threshold, non-binding finding, not a rejection that stood) covering claims 1, 2, 5, 7-10, 13, 15-19, 22, 24-26, 29, 31-34, 37, and 39 over Mander, with Mander alone mapped against the claim-16/25 independent claims. Everything the Board has ever said about this family's subject matter has been favorable to the patent owner (the Enfish-based CBM denial on the '227 patent). And on the civil side, the only final appellate word on the '427 is non-infringement: in Mirror Worlds, LLC v. Apple, Inc., 692 F.3d 1351 (Fed. Cir. 2012) (Nos. 2011-1392, 2011-1393; http://patentlyo.com/media/docs/2012/09/11-1392.pdf), the court held Apple's Cover Flow lacked the claimed "cursor or pointer," and that equating its absence with its presence would vitiate that limitation; cert denied 2013-06-24 (S. Ct. No. 12-1158).

Estoppel landscape. Because there has never been an IPR or PGR, no § 315(e)(2) / § 325(e)(2) estoppel exists against anyone on this patent. The one estoppel-adjacent risk to check is pre-AIA 35 U.S.C. § 315(c) (inter partes reexamination), which estops the third-party requester — Apple — and its privies from later asserting, in a civil action, invalidity of any claim finally determined to be valid and patentable on any ground Apple raised or could have raised in 95/001,172. That estoppel is conditional on the certificate outcome I could not verify; if the certificate confirmed claims 1/8/16/25 unchanged, Apple and its privies are boxed in on the Mander grounds. If you are not in privity with Apple, the entire Mander / Mander+Retrospect / Lucas / Retrospect / Magellan record is available to you with no estoppel at all.

Pattern signals. No defensive aggregator (no Unified Patents filing appears in the record for this patent). The patent owner side was aggressive but in court, not at the Board: the 2008 EDTX case, the 2011–2012 Federal Circuit appeals, and the 2013 cert petition. Notably, the mirror-worlds defendants chose the CBM route on the sibling '227 rather than an IPR on the '427 — and the CBM route closed when PTAB denied institution. For the '427, the IPR window opened 2012-09-16; by then the Apple case was on appeal and the patent's asserted claims were down to 1/8/16/18/25. Whatever the reason, the window closed unused on 2016-09-12.

Recommended next steps

1. Pull the reexamination certificate for 95/001,172 (USPTO Patent Center → "Reexamination Certificate," or the Image File Wrapper for control no. 95/001,172). This is the single highest-value open item: it determines whether claims 1/8/16/25 issued confirmed, amended, or canceled, and it drives (a) any pre-AIA § 315(c) argument against Apple/privies and (b) possible intervening rights under 35 U.S.C. § 307(b) if claims were substantively amended. I could not establish the answer from public search and will not guess.
2. Exploit the expiration. Status is Expired – Lifetime, adjusted expiration 2016-09-12. If a 2026 demand letter cites US 6,725,427, respond that the patent is expired and that recovery is limited to past damages for pre-2016 conduct — and then note that the six-year lookback of 35 U.S.C. § 286 from a 2026 complaint date reaches back only to 2020, after expiration, leaving no compensable window at all unless the claimant already had a timely-filed suit.
3. Lead with the Mirror Worlds v. Apple "cursor or pointer" holding. Every independent claim (1, 8, 16, and 25) requires a cursor or pointer over which the user slides without clicking to summon a glance view. If the accused product has no cursor/pointer interaction with a document-representation stack, you have a non-infringement defense that the Federal Circuit has already endorsed, and an equivalents theory is barred as vitiating the limitation. FWD/appellate texts and links: Fed. Cir. opinion, http://patentlyo.com/media/docs/2012/09/11-1392.pdf; full claim listings at https://patents.google.com/patent/US6725427/en.
4. No PTAB milestones to track. There is no pending trial, so no institution deadline, no oral hearing, no 1-year § 316(a)(11) FWD clock. If a defendant wants a Board ruling on this patent today, the door is effectively closed: it is expired, so an IPR petition would face a threshold "petitioner will not prevail on at least one claim" / redressability problem (and the Office's current discretionary-denial posture, per the June 2025 Director memo, disfavors disturbing settled expectations on long-in-force patents — see the Director's discretionary-denial decision reproduced in Fed. Cir. No. 26-115, https://fedcircuitblog.com/wp-content/uploads/2026/03/26-115_Kangxi-Communication_Petition-Only.pdf).
5. Verify the zero-IPR negative yourself before relying on it in a filing: PTAB E2E search by patent number, https://ptacts.uspto.gov/. The absence of any IPR is itself the signal here — this patent was asserted against a deep-pocketed defendant for five years and was never once taken to the Board.

Explicit uncertainty flags (do not treat as verified): (i) the final certificate status of reexamination 95/001,172; (ii) the CBM petition number and petitioner identity for the '227 CBM denial; (iii) whether any court finally adjudicated validity of the '427 after the Federal Circuit's non-infringement affirmance (the jury found it valid in 2010, but I cannot confirm a final appellate or judgment-level validity determination); (iv) current 2026 owner of record — the earlier patent-summary section flagged this as unresolved, and I did not resolve it here.

I'll research the assignment chain for US 6,725,427 through the USPTO Assignment database and cross-references.

Let me dig deeper into the specific assignment records, reels, and correspondents.

Good progress. Let me try to pull the actual reel/frame numbers and correspondent details.

Assignment & Ownership Chain — US 6,725,427 B2

Sourcing limitation up front (read this first): I was unable to query the USPTO Assignment Center database directly in this session — it is an interactive search application and my tooling returned no record-level extract for patent 6,725,427. Accordingly I have no verified reel/frame numbers and no verified assignment-correspondent fields for any link in this chain. I will not invent them. Every entry below is grounded in one of the following retrieved sources, and I mark the reel/frame and correspondent fields as not retrieved rather than filling them with plausible-looking data:

Verification path for the missing fields: Assignment Center → search by patent number 6725427 → run the Patent Assignment Abstract of Title view, which prints each chain link in the exact "Reel/Frame — Conveyance — Assignors — Assignee — Correspondent" format. That single view will fill every bracketed gap below.

Inventors

Pattern note (unusual, and the inverse of the usual tell): There is no record in anything I retrieved of the inventors departing the assignee within 12 months — because at the filing date there was no operating assignee to depart from. The company (Mirror Worlds Technologies) was formed after the priority filing, which is the classic university-spin-out sequence: inventors → newly formed company. The distinctive feature here is that both inventors remained economically aligned with the assignment chain all the way to 2013, taking consulting fees from the ultimate acquirer (Network-1) and equity in the licensee (Lifestreams). When the original inventors are inside the monetization structure, the "abandoned inventors" fire-sale signal is absent — but the asset is still being asserted by a non-practicing entity.

Not determinable from my sources: whether Yale University ever took or assigned an interest. The issued patent names a company assignee and no government-interest statement; I found no Yale assignment record. Treat "no university interest" as unverified rather than established.

Original assignee

• Entity on the issued patent (Google Patents): Mirror Worlds Technologies LLC (New Haven, Connecticut per its own literature).
• Contradiction to flag: the recorded assignment event of 2005-02-22 lists the **assignor as "MIRROR WORLDS TECHNOLOGIES, INC."**. Google Patents' front-page "original assignee" string says "LLC". The 2008 Apple counterclaim separately describes "Mirror Worlds Technologies, Inc. ('Mirror Worlds (Del.)')" as a Delaware corporation. I cannot reconcile the Inc./LLC forms from the sources I have — do not treat the corporate form as settled; this may be a normalisation artefact at Google Patents, or two distinct entities (a Delaware Inc. and a Connecticut LLC) used at different times.
• Did it ship a product embodying the claims? Yes. Its product was Scopeware, released March 2001, described by Network-1 as the commercialization/proof-of-concept for the portfolio. Network-1's development timeline lists "03/01 Scopeware released."
• Primary line of business: document-stream / "lifestreams" personal information management software.
• Current status: Unclear / wound down. The portfolio was sold out of the family in February 2005, and Network-1's timeline places "06/97 Mirror Worlds Technologies Founded" with no later operating milestone. However, a Mirror Worlds Technologies, Inc. (Delaware) was still a live litigant in 2008–2012 (Apple counterclaimed against it on a separate patent, the '101). I found no Chapter 7/11 filing, no SEC filing, and no dissolution record for it. I therefore mark status as unclear rather than dissolved.

Assignment timeline

Legend for missing data: [reel/frame not retrieved] and [correspondent not retrieved] mean exactly that. I did not obtain any assignment-correspondent fields for this patent. The single attorney name I did retrieve (Ivan S. Kavrukov / Cooper & Dunham LLP) is a reexamination-proceeding correspondence address, not an assignment correspondent — see the caveat under the timeline.

• YYYY-MM-DD [not retrieved] / recorded [not retrieved] — Reel [reel/frame not retrieved]

  • Conveyance: Assignment of inventors' interest
  • Assignor: Eric Freeman; David H. Gelernter (inventors)
  • Assignee: Mirror Worlds Technologies, Inc./LLC (form disputed — see above)
  • Correspondent: [correspondent not retrieved]
  • Context: Company formation — university-spin-out assignment from the Yale inventors to the company formed to commercialise the work. Date is inferred from the 1997 entity formation; I did not retrieve this record and it may have been recorded in more than one link.

• 2005-02-22 (recorded; execution date [not retrieved]) — Reel [reel/frame not retrieved]

  • Conveyance: Assignment
  • Assignor: Mirror Worlds Technologies, Inc.
  • Assignee: Recognition Interface, Inc.
  • Correspondent: [correspondent not retrieved]
  • Context: De-operationalising transfer — the operating company hands the portfolio to the entity that would thereafter finance and monetise it (Recognition Interface is described by Network-1 as "a New York based investment partnership that financed the commercialization of the Patent Portfolio").

• 2007-11-28 (recorded; executed 2007-12-24 per Apple's counterclaim) — Reel [reel/frame not retrieved]

  • Conveyance: Assignment
  • Assignor: Recognition Interface, Inc.
  • Assignee: Recognition Interface, LLC
  • Correspondent: [correspondent not retrieved]
  • Context: Internal reorganisation / conversion of the financing vehicle from Inc. to LLC, executed two days before the next hop. Note the Apple counterclaim's date range: "Plainfield Specialty Holdings I, Inc. was the assignee of the patents-in-suit from December 24, 2007 to March 5, 2008" — a 10-week holding period.

• 2008-03-12 (recorded) — Reel [reel/frame not retrieved]

  • Conveyance: Nunc pro tunc assignment (per Google Patents conveyance text)
  • Assignor: Mirror Worlds Technologies, Inc.
  • Assignee: Recognition Interface, Inc.
  • Correspondent: [correspondent not retrieved]
  • Context: Chain repair — a retroactive re-recording of the 2005 transfer, filed back-to-back with the pre-suit assignment stack. A nunc pro tunc re-record is what an owner does when it needs to show a clean title record.

• 2008-03-13 (recorded) — Reel [reel/frame not retrieved]

  • Conveyance: Nunc pro tunc assignment
  • Assignor: Recognition Interface, Inc.
  • Assignee: Recognition Interface, LLC
  • Correspondent: [correspondent not retrieved]
  • Context: Chain repair — retroactive confirmation of the 2007 Inc.→LLC hop.

• 2008-03-13 (recorded) — Reel [reel/frame not retrieved]

  • Conveyance: Nunc pro tunc assignment
  • Assignor: Recognition Interface, LLC
  • Assignee: Plainfield Specialty Holdings I, Inc.
  • Correspondent: [correspondent not retrieved]
  • Context: Transient conduit — Plainfield takes title only to pass it on; it held the patents-in-suit for roughly ten weeks (2007-12-24 → 2008-03-05).

• 2008-03-13 (recorded; executed 2008-03-05 per Apple's counterclaim) — Reel [reel/frame not retrieved]

  • Conveyance: Nunc pro tunc assignment
  • Assignor: Plainfield Specialty Holdings I, Inc.
  • Assignee: Mirror Worlds, LLC (Tyler, Texas)
  • Correspondent: [correspondent not retrieved]; PlainSite lists the entity at 4540 Kinsey Dr, Tyler, TX 75703
  • Context: Pre-litigation transfer — the patent lands in the plaintiff entity. Nine days later (2008-03-14) Mirror Worlds, LLC filed Mirror Worlds, LLC v. Apple, Inc., E.D. Tex. 6:08-cv-00088, asserting the '427 among others.

• (2013-05-21 executed / 2013-05-22 recorded per PlainSite — partially retrieved row) — Reel [reel/frame not retrieved]

  • Conveyance: Assignment (confirmatory/clean-up, inferred — flagged)
  • Assignor: Plainfield Specialty Holdings I, Inc.
  • Assignee: [not retrieved — possibly Mirror Worlds, LLC/Looking Glass LLC]
  • Correspondent: [correspondent not retrieved]
  • Context: Chain repair ahead of sale. PlainSite shows Plainfield appearing again as assignor on 5/22/2013, immediately before the Network-1 purchase closed. I only have a partial row, so I flag this as probable pre-sale clean-up, not a confirmed finding.

• 2013-06-21 (recorded; deal closed 2013-05-21 per Network-1) — Reel [reel/frame not retrieved]

  • Conveyance: Assignment
  • Assignor: Mirror Worlds, LLC — by then renamed Looking Glass LLC (formerly Mirror Worlds, LLC) per Network-1's S-1
  • Assignee: Mirror Worlds Technologies, LLC — wholly-owned subsidiary of Network-1 Security Solutions, Inc. / Network-1 Technologies, Inc. (OTC: NSSI, later NYSE American: NTIP)
  • Correspondent: [correspondent not retrieved]
  • Context: Transfer-to-asserter. Consideration: $3,000,000 cash + warrants for 875,000 shares @ $1.40 + warrants for 875,000 shares @ $2.10. Network-1 also cut a deal with Recognition Interface, LLC, which "retained an interest in the licensing proceeds."

Assignment correspondent — the honest answer: Google Patents' legal-event rows for this patent do not surface a correspondent, and I could not reach the Assignment Center to extract one. I therefore cannot run the "repeat correspondent" test that this exercise is designed around, and I will not manufacture a name. The one attorney of record I recovered is from the reexamination file : the 2009-07-01 Order Granting Inter Partes Reexamination for control no. 95/001,172 carries a correspondence block for IVAN S. KAVRUKOV, COOPER & DUNHAM LLP (attorney docket 8157.012.427). Caveat: I could not determine whether that block is the patent owner's address or the requester's (Apple's) — on a PTOL-2063 order either is possible — and in any event a reexamination correspondence address is not an assignment correspondent. I am recording it as a lead to chase, not a finding.

Timeline diagram

timeline
    title Ownership of US 6725427
    1996 : Priority application filed by Yale inventors
    1997 : Company founded after the filing
    2004 : Patent issued to Mirror Worlds Technologies
    2005 : Assigned to Recognition Interface Inc
    2007 : Assigned to Recognition Interface LLC
    2008 : Assigned to Plainfield Specialty Holdings I
         : Assigned to Mirror Worlds LLC of Tyler Texas
         : Apple suit filed nine days later
    2013 : Mirror Worlds LLC renamed Looking Glass LLC
         : Assigned to Network 1 subsidiary
    2016 : Patent expired

NPE / troll-pattern signals

Threshold note: every signal below is assessed on dates, party names and transaction context — not on reel/frame and not on correspondent, because I could not retrieve those. Signals that specifically depend on reel/frame or correspondent are marked unclear even where the surrounding evidence is suggestive.

1. Shell-entity transfer — PRESENT (strong on dates and function, unconfirmed on the entity-details prongs).
The chain moves an operating company's patents into a licensing-only financing vehicle (Recognition Interface, Inc. → LLC, 2005-02-22 and 2007-11-28), then through a serial holding company whose very name — Plainfield Specialty Holdings I, Inc. — implies a numbered series and which held title for only ~10 weeks (2007-12-24 → 2008-03-05), then into Mirror Worlds, LLC of Tyler, Texas (PlainSite: 4540 Kinsey Dr, Tyler, TX 75703), which sued nine days later. What I cannot confirm: whether Recognition Interface or Plainfield used a registered-agent service address, whether they were single-member Delaware/Texas LLCs, or whether they had any employees. No products came out of Recognition Interface, Inc./LLC, Plainfield Specialty Holdings I, Inc., or Mirror Worlds, LLC (Tyler) in any source I retrieved — that part of the prong is met.

2. Known asserter in the chain — PRESENT (final link only).
The current assignee of record is Mirror Worlds Technologies, LLC, a wholly-owned subsidiary of Network-1 Technologies, Inc., which self-describes as "a company engaged in the acquisition, development, licensing, and monetization of intellectual property" (8-K Ex-99.1, 2013-05-22/28). Network-1 derives revenue from licensing/settlement (Microsoft $4.65M Nov 2015; Apple $25M Jun 2016). Caveat: Network-1 is not on the enumerated list in the brief (Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities), and I did not find it on an RPX or Unified Patents high-frequency-plaintiff list in this session. I mark the signal present on the licensing-only, no-product, self-described monetization basis, not on directory membership. Note also that the post-2013 Network-1 suits I could verify asserted the '227, not the '427 — the '427's assertion was by Mirror Worlds, LLC in 2008.

3. Repeat correspondent across the chain — UNCLEAR / not assessable.
This is the signal I was specifically asked to look for and the one I could not test. No assignment-correspondent field was retrievable for any of the eight-plus links. The only attorney name recovered (Ivan S. Kavrukov, Cooper & Dunham LLP, 95/001,172 reexam correspondence block, 2009) is from a different record type. Do not treat the absence of a hit as evidence of absence — pull the Assignment Abstract of Title and rerun this test; the eight-plus events between 2005 and 2013 were almost certainly papered by a small number of firms.

4. Cascading transfers — PRESENT.
Six recorded events inside a ~3.5-month window: 2007-11-28 → 2008-03-12 → 2008-03-13 (×3) → 2008-03-13 executed 2008-03-05. Four of the links were re-filed nunc pro tunc, i.e. recorded retroactively in a two-day burst (2008-03-12/13) — a hallmark of a chain assembled or repaired against a filing deadline. A further Plainfield recording appears on 2013-05-22, immediately before the Network-1 close, consistent with another clean-up pass.

5. Pre-litigation transfer — PRESENT (very strong).
Mirror Worlds, LLC v. Apple, Inc., 6:08-cv-00088 was filed 2008-03-14. Mirror Worlds, LLC's title was executed 2008-03-05 and recorded 2008-03-12/13 — 9 days before the complaint. This is squarely inside the 6-month window and is the textbook pattern of arranging the chain to establish standing and to fix venue in the Eastern District of Texas.

6. Bankruptcy fire-sale — NOT PRESENT (as far as I can find).
No Chapter 7/11 proceeding involving Mirror Worlds Technologies appears in any source I retrieved, and the 2005 transfer to Recognition Interface reads as a negotiated monetisation deal, not a court-supervised sale. The later Mirror Worlds, LLC → Looking Glass LLC → Network-1 transfer (2013) was a negotiated $3M-cash-plus-warrants purchase (8-K, 2013-05-22), not a bankruptcy sale. Marked not present, with the caveat that I did not run a PACER bankruptcy search.

7. Privateering — UNCLEAR.
The commercialisation was financed by Recognition Interface, LLC, "a New York based investment partnership that financed the commercialization of the Patent Portfolio prior to its sale to Mirror Worlds, LLC and also retained an interest in the licensing proceeds" (Network-1 8-K Ex-99.1). That is litigation/portfolio financing, and it is close to privateering — but classic privateering requires an operating company transferring to an NPE to assert against its own competitors on its behalf. Here the operating company had already exited in 2005. Unclear.

8. Defensive aggregator — NOT PRESENT.
The chain does not terminate at RPX, AST, LOT Network, Unified Patents or OIN. It terminates at a for-profit public patent-monetisation company (Network-1) that then asserted the portfolio and settled with Microsoft and Apple. Inverse of the neutralised signal.

Verdict

NPE — high confidence.

Two independently strong signals drive this. First, pre-litigation transfer + cascading transfers: the '427 was executed into Mirror Worlds, LLC of Tyler, Texas on 2008-03-05, recorded in a nunc pro tunc burst on 2008-03-12/13, and asserted nine days later in 6:08-cv-00088 (Apple counterclaim Dkt. 48; USPTO reexam order 95/001,172). Second, shell-entity transfer into a licensing-only chain: the portfolio left the Scopeware operating company in 2005-02-22 for Recognition Interface, Inc., passed through Recognition Interface, LLC and a ~10-week stop at Plainfield Specialty Holdings I, Inc., and today sits with Mirror Worlds Technologies, LLC, a subsidiary of the self-described IP-monetisation company Network-1 Technologies (8-K Ex-99.1, 2013-05-22; purchase price $3M cash + 1,750,000 warrants). The one signal that could have upgraded this to a correspondent-confirmed finding — recurrence of a recording attorney/firm across the chain — is unresolved, because I could not retrieve any assignment-correspondent field.

Verify at: https://assignmentcenter.uspto.gov/ (search patent number 6725427, open Patent Assignment Abstract of Title) and the mirror at https://assignment.uspto.gov/patent/index.html. Pull the reel/frame + correspondent for each of the events dated 2005-02-22, 2007-11-28, 2008-03-12, 2008-03-13 (×3), 2013-05-22 and 2013-06-21, then rerun signal #3 — that is the missing link in this analysis.

I'll research the specific prior-art citations for US 6,725,427. Let me start with targeted searches.

Excellent — I found the reexamination SNQ references. Now let me retrieve the face-of-patent citations and the full reexamination grounds.

Let me pull the face-of-patent citation list and the full reexamination grounds document.

Prior Art for US Patent 6,725,427 — Analyst Findings

Sourcing note: I searched for the specific number 6,725,427 and did not substitute sibling numbers. The most authoritative prior-art record I could retrieve is the USPTO Decision Granting Inter Partes Reexamination in Mirror Worlds, LLC v. Apple (E.D. Tex. 6:08‑cv‑00088, Dkt. 221 exh.) and the Mirror Worlds/Apple invalidity contentions (Dkt. 221). Before the per-reference table, two important framing notes:

• The Google Patents full text I was given for the '427 does NOT include the "References Cited" (front‑page citation) block. Therefore I cannot present a complete, verified enumeration of every patent cited on the face of the '427 from that source. What I can confirm is limited to references the reexamination order and the litigation record explicitly tie to the '427. I flag each place where my list may be incomplete rather than inventing references.
• Beware of "6725427" search hits that are actually forward citations. Numbers like 6,725,427 appear in the "References Cited" lists of later Apple patents (e.g., US 8,564,543 "Media player with image based browsing"; US 7,747,968; US 11,281,843). Those are patents that cite the '427 — they are not prior art to the '427, and I have excluded them.

A. The four references the USPTO relied on to grant reexamination

These are the references that raised a substantial new question of patentability (SNQ) against claims 1, 2, 5, 7‑10, 13, 15‑19, 22, 24‑26, 29, 31‑34, 37, and 39 of the '427 (inter partes reexamination 95/001,172, requested 23 Apr 2009, granted 1 Jul 2009).

1. Mander et al. — U.S. Patent No. 6,243,724 B1

Brief description: Mander discloses a computer system that organizes information in a time‑based/stream‑like arrangement and provides user-interface organization of stored items. It is the single most important piece of prior art for the '427 — it is the only reference both (a) on the patent's face and (b) relied upon by the PTO in the reexamination.

Caveat on the § 102 mapping: My retrieved fragment of the reexamination order shows the claim list associated with the Mander ground but not the full element‑by‑element rationale. Treat the "claims 16‑19, 22, 24‑26, 29, 31‑34, 37" attribution as reliable (it is quoted from the order) but note the order also refers to multiple "Grounds" (#1–#4) that I could not fully extract.

2. Lucas et al. — U.S. Patent No. 5,499,330 A

Brief description: Lucas '330 (the MAYA "Workscape" work) discloses displaying documents as three‑dimensional "strands" — a time‑ordered, receding sequence of document objects — expressly as an improvement over "folders or directories" used to organize files. The Apple contentions describe it as showing "documents along a 'strand' through a three‑dimensional display space," with repositories/workspaces accessible locally or over a network.

Direct hit on the '427: This reference is aimed squarely at the '427's "receding, foreshortened stack of partly overlapping document representations" display limitation (claims 1, 25) because the '330 "strand" display is a time‑ordered, perspective rendering of document objects. Apple's invalidity expert relied on it for exactly this purpose.

Note: Independently, Apple contentions also asserted Lucas '330 as prior art to the sibling '227 patent and the other patents‑in‑suit. The reexamination treated Lucas as not "of record" in the '427 file — consistent with it being a new reference.

3. "Retrospect User's Guide," v. 3, 1st ed. — Dantz Development Corp. (1989–1995)

Brief description: Documentation for Retrospect, a Macintosh automatic‑backup application. Its relevance is to the '427's automatic archiving / automatic backup limitations (the "automatically archiving the documents … in consistent format" element of claims 1 and 8).

4. "Magellan Explorer's Guide" (Lotus Development Corp., 1989) and 5. David P. Gobel, "Using Lotus Magellan" (QUE Corp., 1989)

Brief description: Documentation for Lotus Magellan, a DOS file viewer/organizer known for automatic indexing and "glance"-style preview of file contents across heterogeneous formats. Its relevance is to the '427's glance‑view / document‑representation generation limitations (claims 1, 5, 6, 7, 8, 13, 14, 15, 16, 22, 23, 24, 25) — i.e., auto‑generated previews of documents from diverse applications/formats.

B. Additional prior art asserted in the Apple litigation (not part of the PTO SNQ, but relevant)

Apple's invalidity contentions (Dkt. 221) added the "Workscape References" as prior art to the '227 and the other patents‑in‑suit (including the '427):

These were expressly asserted against the '227 and "the other patents in suit — U.S. Patent Nos. 6,638,313, 6,725,427, and 6,768,999 … under 35 U.S.C. § 102." They attack the same "stream/strand, time‑ordered display" concept that underlies the '427's independent claims.

C. Non‑patent literature cited in the specification of the '427 itself

The '427 specification (references [1]–[4]) cites four NPL items. These are "cited" in the patent's own text rather than on the front page, and they are background rather than anticipatory:

Ref [4] deserves a caution: because it is by the same inventors (Gelernter) and predates the 28 Jun 1996 filing, it could in principle be raised as § 102(a)/(b) art against the broader stream concept — but the patent affirmatively distinguishes it as incomplete, and it was the inventors' own disclosure.

D. What I could NOT verify (explicit limitations)

1. The complete face‑of‑patent "References Cited" list for the '427 is not available in my source. The Google Patents full text provided contains no U.S. patent citation block. I therefore cannot guarantee that Mander is the only patent on the face of the '427. (The reexamination order implies Mander is the notable face citation, but does not catalogue the rest.)
2. Per‑reference § 102 claim mapping is only partial. I have a reliable mapping for Mander (claims 16‑19, 22, 24‑26, 29, 31‑34, 37) and the SNQ claim set (1, 2, 5, 7‑10, 13, 15‑19, 22, 24‑26, 29, 31‑34, 37, 39). I do not have the element‑by‑element breakdown for Lucas, Retrospect, or Magellan, nor the full text of Grounds #1, #2, #4 in the order.
3. Final outcome of reexamination 95/001,172 is unconfirmed in my results — I did not retrieve the reexamination certificate, so I cannot state which claims were finally confirmed, amended, or cancelled. (This is consistent with the uncertainty already flagged in the previously generated Patent/Litigation summaries.)
4. I did not find the '427's own examiner citations beyond what the reexamination order mentions — a PatentCenter lookup of the '427's "References Cited" and the reexam certificate would close this gap.

E. Bottom line

The most relevant prior art for US 6,725,427, ranked:

1. Mander, US 6,243,724 — filed 8 Aug 1994, issued 5 Jun 2001. The only reference on the '427's face and relied on by the PTO; the order finds claims 16‑19, 22, 24‑26, 29, 31‑34, and 37 unpatentable over it.
2. Lucas, US 5,499,330 — filed 17 Sep 1993, issued 12 Mar 1996. Directly targets the "strand"/receding‑stack time‑ordered display limitation (claims 1, 25).
3. Magellan Explorer's Guide + Using Lotus Magellan (1989) — target the auto‑indexing / glance‑view generation limitations.
4. Retrospect User's Guide (1989–1995) — targets the automatic‑archiving limitations.
5. Workscape/CHI '94 references (1994) — litigation‑only art targeting the time‑ordered strand display.

All four SNQ references were found to raise a § 102 substantial new question against the broad claim set 1, 2, 5, 7‑10, 13, 15‑19, 22, 24‑26, 29, 31‑34, 37, and 39 — i.e., including the independent claims 1, 8, 16, and 25 that carry the "glance view via cursor‑hover over a foreshortened stack" limitation. If you need the definitive per‑claim, per‑reference rejection map and the final certificate status, the authoritative documents are USPTO PatentCenter's reexamination file for 95/001,172 and the Dkt. 221 exhibits in 6:08‑cv‑00088.

I'll research the prior art landscape for this patent, starting with the references already embedded in the record and the reexamination art.

Excellent — the reexamination record gives me the exact prior-art grounds. Let me pull the specific references and the reexam outcome.

The litigation record contains the actual obviousness expert reports. Let me pin down the priority dates and the exact disclosures of the key references.

I have enough to complete the §103 analysis. Note: my searches hit the step limit before I could confirm two things (Mander '724's exact filing/priority date, and the final reexamination certificate outcome). I flag both as open items rather than assume them.

Obviousness analysis — US 6,725,427 B2 under 35 U.S.C. § 103

A. Framework and governing law

• Effective priority date / prior-art cutoff: June 28, 1996. The '427 patent is a divisional of 09/398,611 (now US 6,638,318), itself a continuation of 08/673,255 (now US 6,006,227), all claiming the June 28, 1996 filing. Because the effective filing date precedes the AIA's March 16, 2013 change, pre-AIA § 103 governs (AI 2007 KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398, still applies to pre-AIA claims).
• One of ordinary skill in the art (POSITA): a software engineer/UI designer with 2–4 years' experience building graphical desktop applications — file managers, document viewers, or information-retrieval interfaces — on a general-purpose OS. This is a combination-of-known-elements case, not an unpredictable-arts case.
• The controlling inquiry: whether the differences between the claimed subject matter and the prior art are such that the subject matter as a whole would have been obvious. Because the claims are directed to a predictable combination of known GUI and file-management elements (piles/stacks + hover preview + automatic indexing/archiving + reusing host-OS services), KSR's "combination of familiar elements according to known methods" rubric applies with force.

B. The prior-art corpus (from the reexamination request and the record)

The examiner's decision granting inter partes reexamination 95/001,172 (filed 4/23/2009, granted 7/1/2009) identifies this art:

Sources: PTO reexam order reproduced in the E.D. Tex. record, https://cases.justia.com/federal/district-courts/texas/txedce/6:2008cv00088/108627/221/18.pdf; the Piles disclosures were also confirmed in the Apple expert report at https://cases.justia.com/federal/district-courts/texas/txedce/6:2008cv00088/108627/221/25.pdf; Lucas '330 text at https://uspto.report/patent/grant/5499330.

Statutory status of each: Lucas (filed 1993, issued 3/12/1996) is §102(a)/(e) art; the Magellan materials (1989) and the CHI '92 Piles paper are §102(b) art; Retrospect (1989–1995) is §102(b) art. Mander '724 issued after the priority date, so it can only be §102(e) art if its application was filed before the applicant's invention date — I could not confirm its exact filing/priority date in the sources available, so treat that as an open item. This does not weaken the analysis, because the same Piles disclosures are independently §102(b) art through the 1992 CHI paper and are cited on the '427's face.

C. What each independent claim actually requires

The four fully-visible independents share a common skeleton:

1. An operating system (or program) that utilizes subsystems of another OS for storage writes, interrupt handling, and I/O (claims 1, 8, 16, 25; dependent claims 3/4, 11/12, 20/21 name Windows and Apple).
2. A document organizing facility that:
   • ingests documents from diverse applications in diverse formats (claims 1, 8, 19, 25);
   • automatically associates selected/chronological indicators (time-based per claims 2, 9, 17; without requiring a user to name documents per claim 26);
   • automatically archives documents + indicators in consistent format for selective retrieval (claims 1, 8);
   • automatically creates glance-view data and document-representation data.
3. A display facility that renders representations as a receding, foreshortened stack of partly overlapping representations (claims 1, 10, 18, 25).
4. A cursor/pointer that, when slid without clicking over a representation, displays the glance view of the document currently touched (all independents).

Claim 8 (broader — stack moved to dep. claim 10) and claim 16 (drops the "archiving" phrase; adds "received or created") are progressively easier to invalidate than claim 1. Claim 27 and claims 28–39 are truncated in the source text and I cannot characterize them with confidence — see §H.

D. Combination 1 — Mander '724 / Piles alone (or + Retrospect): claims 8, 16 and the time-based dependents

The reexamination requester asserted, and the PTO found an SNQ for, Mander alone against claims 16–19, 22, 24–26, 29, 31–34, and 37, and Mander + Retrospect against claims 1, 2, 5, 7–10, and 13.

Element mapping (Mander '724 as described in its own abstract and the US7849035 continuation):

• Diverse-format ingestion + automatic organization. Mander discloses "a computer filing system" that displays "a plurality of documents ... of an electronic mail network," "automatically divide[s] a pile ... into subpiles on the basis of the content of each document," and "automatically file[s] away documents into existing piles ... on the basis of a similarity match between the content ... of the document and the content ... of existing piles." This is automatic organization of heterogeneous documents (email, and document types generally), addressing the identical problem the '427 background laments ("pointless" filenames; "hierarch[ies] that quickly become obsolete").
• Automatic indicators. The Piles design supports pile visualization "according to various criteria (e.g. date)" — i.e., automatically associating and using a temporal indicator without a user naming files. This meets claims 2/9/17 (time-based indicators) and claim 26 (no user-imposed names).
• Automatic creation of glance views — the key limitation. Mander expressly discloses "browsing the collection of documents by pointing a cursor at a particular item in the collection of documents to reveal an indicia [i.e., a proxy] for the particular item" and "reveal a proxy within a viewing cone for the item positioned under the cursor." A "proxy" is, by definition, an abbreviated stand-in for the document — a glance view. Notably the '427's own specification describes the identical browse-card concept ("an abbreviated version of a document which ha[s] been compressed into a micro-document like an index card"). Mander's pointer-reveal therefore reads on the "sliding without clicking the cursor ... to display the glance view of the document ... touched" limitation.
• Document representations. Mander displays each document as a "graphical representation of the document," and after piling, "a side-ways view of the graphical representation of the document replaces the original graphical representation" — the representations are drawn as an overlapping collection (a pile). Combined with the "viewing cone" (= a perspectival/foreshortened view), this reads on the "receding, foreshortened stack of partly overlapping document representations."
• Host-OS subsystem reuse. The Piles work was implemented on the Apple Macintosh desktop environment; a macOS/Macintosh filing-system implementation inherently uses that OS's storage/IO/interrupt services. The '427 specification itself concedes that its system "can be implemented ... utiliz[ing] subsystems from other operating systems such as DOS, WINDOWS, and OPERATING SYSTEM 7."

Adding Retrospect supplies the "automatically archiving the documents and indicators in consistent format for selective retrieval" limitation (claim 1): Retrospect is an automatic backup/archive product that copies files to long-term media and maintains a catalog/index in a consistent format for later retrieval — precisely the automatic-archiving function the '427 claims, and the exact pairing the reexaminer found raised an SNQ.

Result: claims 8, 16, 18, 19, 22, 23, 24 and their time-based dependents are strongly vulnerable to Mander alone; claim 1 is vulnerable to Mander + Retrospect.

E. Combination 2 — Mander/Piles + Lucas '330: the "receding, foreshortened stack" (claims 1, 10, 18, 25)

If one accepts that Mander's "viewing cone" is not a sufficiently explicit receding stack, Lucas '330 fills the gap precisely.

Lucas discloses a three-dimensional workspace in which documents are positioned as screen objects along strand paths (a "two dimensional line through a three dimensional display space"), rendered through a perspective process that "converts the three dimensional coordinates into two dimensional screen space coordinates." Critically, Lucas defines:

"A pile of documents is a strand having a strand path defined by a function causing the strand to be oriented substantially parallel to the Z axis of the display, that is, going straight back from the surface of the display device that is closest to the user." (https://uspto.report/patent/grant/5499330)

That is a receding stack, rendered via perspective (i.e., foreshortened), of partly overlapping rectangular screen objects. Lucas's "screen objects are generally rectangular," and the perspective process necessarily occludes/overlaps the farther objects behind the nearer ones. Lucas also discloses document attributes, unique identifiers, a FIND tool that generates output strands, and knots that divide a strand into substrands (a substring/filter analog) — and, per the Apple expert report, annotation of newly arrived email/fax/scan documents.

Why a POSITA would combine Mander + Lucas: Both are in the identical field (graphical desktop document organization), both were authored in the same Apple research environment and both were motivated by the same failing — Lucas states the problem is that "documents in containers or directories are hidden from the user" and users "typically cannot see the documents inside a container without opening up the container." Mander's pile suffers the same occlusion problem; Lucas supplies the explicit 3-D receding, perspective-rendered pile that lets users see the document edges in depth. Combining a known pile metaphor (Mander) with a known depth/foreshortened rendering engine (Lucas) to produce a more legible pile is a predictable, design-driven combination and is exactly the sort of "arrangement of old elements" KSR holds obvious. A POSITA would have a reasonable expectation of success, since both already render overlapping document representations and both run on the same class of desktop OS.

F. Combination 3 — + Magellan (heterogeneous formats, auto-indexing, Speedview, filtered subsets). The broadest combination

Claim 1 (and independent claim 25) requires diverse formats and automatic association of indicators + glance-view creation. The Lotus Magellan materials (1989) close these:

• Diverse applications/format ingestion. Magellan "index[es] and search[es] all the documents ... on the hard drive ... spreadsheets, text, graphics, databases," including files from "1-2-3, WP, e-mail." Trial testimony: "the types of documents that were handled by the system and indexed and searched using Lotus Magellan, there are a diverse set ... like spreadsheets, text, graphics" (https://cases.justia.com/federal/district-courts/texas/txedce/6:2008cv00088/108627/423/0.pdf).
• Glance view / preview. Magellan's advertising and the record describe Speedview: "shows you the complete content of your files, even though the applications that created them are not loaded," i.e., a preview of a document's content in a consistent format without opening the native application — this is the "glance view" limitation for heterogeneous documents. The trial transcript expressly calls it "glance view": "as I moved down the list of documents, you saw you got a quick glance of the data that was in each document."
• Automatic indicators. Magellan's index "identifies the time and date that each document was created or last modified," enabling chronological ordering with no user-supplied names.
• Filtered subsets. Magellan's Explore/Concept/Path searches produce a subset list from the whole-disk index — the "substream"-style filtering the '427's specification describes, and directly relevant to claim 19 (consistent-format glance views across heterogeneous sources).
• Host OS. Magellan ran on MS/DOS — a program that necessarily uses the host OS's file/storage/IO subsystems.

Motivation to combine: Mander's piles and Magellan's indexed file list solve the same "I can't find what I've got" problem and target the same user. A POSITA seeking a single personal-information interface would naturally pair Magellan's cross-format index + Speedview (to supply content for the proxy/glance view and to draw documents in from many applications) with Mander's pile display (to supply the stacked, cursor-browsable presentation). The '427 specification itself cites Malone (1983), Nelson (1990), and Lansdale (1988) — the very desktop-organization literature that frames Piles and Magellan — confirming these are the references a POSITA would consult. The combination yields the predictable result of a browsable, auto-organized, multi-format document space.

G. Combination 4 — '852 (Thompson-Rohrlich) + Inside Macintosh + Piles + Magellan + Retrospect

Apple's expert (Dr. Feiner) also contended that '852 + Inside Macintosh, in view of Mander '724/Piles, Lucas/Workscape, Magellan, Retrospect, SDM/SDMS, On Location, and/or Memoirs, renders '427 claims 1, 2, 5, 7–10, 13, 15–19, 22, 24–26, 29, 31–34, 37, 39 obvious — and the reexaminer found the SNQs are not cumulative to the art of record.

• '852 discloses a "Viewer [that] acts as an intelligent folder that continually searches for files meeting a specification supplied by the user" — i.e., persistent/dynamic filtering (a live, automatically-updated result set) plus auto-created "aliases" (representations) organized in a display window. This is material to the persistent-substream / "documents organized as needed" concept and to any dependent claims reciting dynamic updating.
• Inside Macintosh supplies HFS's automatic date/time metadata (creation/modification timestamps) and the Mac OS file/IO services — supporting the "automatic indicators" and "utilizing subsystems from another operating system" limitations.
• The trial record further shows On Location (an Apple/MS-DOS-era indexer) providing date-sorted lists plus double-click "glance view[s]" of files including "graphical file[s]," with the index automatically updated when files change — corroborating the Magellan disclosures.

Motivation: each of these is a known, off-the-shelf mechanism directed to the same goal (indexing/organizing/previewing a heterogeneous document collection), and KSR makes it obvious to combine known techniques to meet a known need.

H. Why the combination is "obvious" — the KSR rationales

1. Predictable combination. Piles (Mander), 3-D receding pile rendering (Lucas), cross-format indexing/Speedview (Magellan), and auto-archive (Retrospect) are all known elements; the invention merely arranges them in a single interface with expected results.
2. Common problem, common field. Every reference attacks the same deficiencies recited in the '427's own "Background" — unnamed files, obsolete hierarchies, lost history, no auto-archiving. Where the problem is known and the elements are known, the combination is obvious.
3. Finite number of predictable solutions. The design space for "show a pile of documents and let the user peek at one" was small and well-trodden by 1996 (piles, stacks, tool-tip proxies, viewing cones, Speedview).
4. Market/design pressure (screen real estate & browse-before-open). Foreshortened stacks conserve screen space and cursor-reveal previews reduce the cost of opening files incorrectly — ordinary design incentives expressly recognized in KSR.
5. Explicit "design with the host OS" concession. The '427 specification itself states the system may "utilize subsystems from other operating systems such as DOS, WINDOWS, and OPERATING SYSTEM 7" and may "operate as a document stream utility" — so the host-OS-subsystem limitation is admitted to be a conventional implementation choice, and claims 3/4, 11/12, 20/21 (Windows/Apple) are squarely disclosed by the Mac/Macintosh and MS/DOS implementations of the art.

I. Dependent-claim vulnerabilities

• 2, 9, 17 (time-based indicators) — Magellan (create/modify timestamps), HFS metadata, and piles-by-date all disclose time-based automatic indicators.
• 5, 6, 13, 14, 22, 23 (abbreviated glance views / header-stripping) — Magellan's Speedview (content preview without the native application) and Mander's proxy. "Header-stripping" is a routine content-summarization step; note that Mander et al.'s sibling work (US 5,867,164, Interactive document summarization) and [4] of the '427's own reference list address automatic summarization.
• 7, 15, 24 (complex analysis of important words/pictures/sounds) — Magellan's Concept/Explore ("related ideas") and the Mander vector-based content representation (term-weighting, "most significant words"), which is literally "complex analysis ... so that 'most important' words ... are presented."
• 19, 26 (heterogeneous formats; no user naming) — Magellan indexes across formats without requiring user-assigned names.
• 3, 4, 11, 12, 20, 21 (Windows/Apple host OS) — the Piles project (Apple) and Magellan (MS/DOS) implementations; plus the specification's own admission.

J. The counterarguments a patent owner would (and did) raise — and their strength

The Mirror Worlds rebuttal expert report (.../221/25.pdf) argued that Lucas lacks any "stream"/"main stream," that strands are displayed in their entirety (so no selective segment display), and that Lucas teaches away by criticizing containers. That rebuttal is directed primarily at the '227/'313 "stream"/"substream"/"persistent stream" limitations — which are not limitations of the '427's independent claims 1, 8, 16, and 25 (those claim a "stream-based / controlling operating system," not the main-stream-plus-substream data structure emphasized in the '227 claims). So the strongest non-obviousness argument against the '427 specifically is weaker than against its sibling patents.

That said, the most credible non-obviousness argument is:

• No single reference discloses the cursor-slide-without-clicking over a receding stack combination. Mander uses cursor pointing/dwell on a pile; Lucas renders a receding pile but its interaction model is sliding a document along a strand (drag), not hover-preview. A patent owner would argue no reference teaches combining Lucas's receding depth-stack rendering with Mander/Magellan's hover-without-click glance view. The counter is KSR predictability: hover-to-preview (Mander proxy, Magellan Speedview, which updates as the user moves down the list) applied to a pile display is the natural, expected combination, and the Federal Circuit's Mirror Worlds decision shows this limitation is the claim's crux — but a §103 invalidity finding would not be defeated by the doctrine-of-equivalents reasoning that defeated infringement.
• Secondary considerations. Any argument of long-felt need or commercial success (Apple's Spotlight/Cover Flow/Time Machine) is weak on nexus, because those products embody many features unrelated to the specific "glance view on hover over a foreshortened stack" limitation — exactly the nexus gap that supported the JMOL/Fed. Cir. ruling in Mirror Worlds v. Apple, 692 F.3d 1351 (Fed. Cir. 2012).

K. Bottom line

If the prior-art dates hold, the '427 patent's independent claims are highly vulnerable under pre-AIA § 103. The two strongest attacks are:

1. Mander '724 (or the CHI '92 Piles publication) alone — against claim 8, 16 and the time-based/no-naming dependents; and
2. Mander/Piles + Lucas '330 + Magellan (Gobel / Explorer's Guide), optionally + Retrospect — against claim 1 and claim 25, mapping:
   • diverse formats, auto-indexing, glance view, filtered subsets → Magellan;
   • piles, cursor-reveal proxy/glance view, automatic organization, date criteria → Mander/Piles;
   • receding, foreshortened, partly overlapping stack → Lucas '330;
   • automatic archiving in consistent format → Retrospect;
   • using host-OS subsystems → the Mac/Macintosh and MS-DOS implementations plus the specification's own admission.

The combination is supported by a clear, articulable motivation (same field, same stated problem, predictable substitution of known GUI/file-management elements) and by the reexaminer's own finding of a substantial new question of patentability. The principal genuine weakness in the obviousness case is the "slide cursor without clicking over the stack" interaction limitation, which no single reference discloses in combination — but §103 permits its assembly from Mander's cursor-reveal-proxy teaching and Magellan's list-traversal glance view.

L. Explicit caveats and open items

1. Claims 27–39 are truncated in the supplied text. Claim 27 begins "A stream-based opera…" and appears to be a further independent claim; I cannot map its limitations or those of claims 28–39 (e.g., the claim numbers the reexaminer listed as 29, 31–34, 37, 39) to specific prior art with confidence. Several of those may recite stream/substream or summarize/"squish" features not present in claims 1/8/16/25, and would need reference-by-reference treatment once the full text is available.
2. Mander '724's filing/priority date is unconfirmed in my sources. Its printed-paper counterpart (CHI '92, June 1992) is unambiguously §102(b) art, so the combination survives regardless; but if only the '724 patent is relied upon, its §102(e) date must be verified against the '427's June 28, 1996 filing.
3. Final outcome of reexamination 95/001,172 (which claims were confirmed/cancelled) was not located and should be pulled from the reexamination certificate in USPTO Patent Center.
4. I have taken all patent and control numbers literally as they appear in the sources cited; where a source rendered the same number differently I did not silently alter it.