US 5892825

I'll verify against live sources, then summarize.

US 5,892,825 A — Summary

Verification note: I searched for the identifier 5892825 literally and did not auto-correct it. All sources I retrieved — Google Patents, the USPTO-hosted PDF, FreePatentsOnline, Unified Patents' portal, PlainSite assignment records, and third-party patent documents citing it — consistently return the same document. I found no Federal Circuit docket entry dated 2026 (or any year) naming this patent. What I did find in PTAB records (IPR2015‑01364, IPR2016‑00472) is directed at US 6,101,534, a different patent in the same Mages/Planet Graphics family — I am flagging that deliberately rather than treating it as this patent's record. I also could not verify the specific litigation referenced by Google Patents' "Family has litigation → First worldwide family litigation filed" link (Darts‑IP family ID 24587338).

---

Bibliographic data

Assignment chain recorded on the face of the file: Planet Graphics → Hyperlock Technologies (1998‑01‑12) → Broadbridge Media LLC (1999‑11‑17) → John Galt Media (2002‑07‑22) → Sharp Holding Corp. (2003) → Octopus Media, LLC (2003‑10‑06) → Samantha Inventions, LLC (2008‑01‑05) → Apple Inc. (2011‑03‑28).

Family members (per the Google Patents family listing): US 5,937,164; US 6,185,306; US 6,035,329; US 6,463,467; plus WO 98/24037, EP 0974217 A2, AU 52598/98, CA 2271012.

---

Abstract (as printed)

A method of triggering video imaging and/or audio data on a "HyperCD" (CD‑ROM) via a trigger through a network for instant local access of encrypted data on local media. The CD‑ROM contains video/audio files that have been crippled by removing the critical information thereof. The CD‑ROM also contains programs for directly and automatically connecting the end‑user's computer to a targeted server (URL) a network (such as the Internet). The method of the invention includes the separation of critical information/key from media data, the storage of encrypted keys on a remote server, the storage of crippled chunky media data on CD‑ROM's, the socket‑to‑socket connection between the end‑user's computer and the remote server and the request from user to server, the authorization and transmission of the uncrippling key through a network to the end‑user's computers, the combining in RAM of the trigger and the crippled data from CD‑ROM for instant rendering/playback, the storing of the uncrippling key on the end‑user's computers non‑volatile media for owning the data on the CD‑ROM for later access authorized by the remote server. Since the uncrippling trigger is much smaller than the entire media file, it saves considerable amount of transmission time, thereby instantly combining the trigger with the crippled data on the CD‑ROM, after which instantly playing the designated track from end‑user's computer.

---

Plain-language overview of the independent claims

Claim 1 — Network method (manufacture-to-playback pipeline).
Covers a method of invoking a "crippled" file on a local storage medium. The steps are: (a) digitize analog video/audio; (b) cripple the files by removing selected information (e.g., the header) so they are unusable absent an uncrippling trigger; (c) store the digitized media on a storage medium for an end user's computer; (d) keep the removed "selected information" as uncrippling trigger data at a host computer; (e) send that trigger over a network to the end user's computer; (f) receive it into volatile RAM; (g) uncripple "substantially instantly" by combining, in RAM, the host-sent trigger with the crippled data on the medium; (h) before step (e), encrypt the trigger; (i) after step (f), decrypt it. Step (g) must follow step (f) immediately, and playback must follow step (g) immediately. Step (f) itself is further defined by three sub-steps: (1) directing the incoming trigger to RAM for temporary storage, (2) combining it in RAM with the crippled files, and (3) preventing copying to non-volatile memory — with step (g) occurring while the uncrippling data remains in RAM.

Claim 8 — Network method (host/end-user side, narrower).
Drops the analog-conversion and media-manufacturing steps and recites the transaction over the Internet: (a) host stores the uncrippling trigger data (the removed selected information); (b) host encrypts and transmits it over the Internet; (c) end user receives and decrypts it — where step (c) includes directing the incoming trigger to volatile RAM, combining in RAM with the crippled file, and preventing copying to non-volatile memory; (d) immediately after (c), uncripple; (e) immediately after (d), play. Again, uncrippling must occur while the data is in volatile RAM.

Claim 17 — Apparatus/storage-device claim ("In a storage device … the improvement comprising").
A system-style claim to a storage device plus computer. Key elements: crippled data files representing video/audio, crippled by removal of selected data, and the crippled data files themselves being encrypted; an end-user computer that plays them back; a host computer storing encrypted uncrippling data (the removed selected data); and a network linking the two. On the host sending the uncrippling data, the storage device's crippled files are uncrippled in the end user's RAM and rendered playable after decryption. The end-user computer must include: means for decrypting both the trigger and the encrypted crippled files; volatile memory means to receive the encrypted trigger; means for immediately joining the decrypted trigger and decrypted data files in RAM for immediate playback; player means; and means for preventing copying of the trigger to non-volatile memory.

Note a drafting asymmetry worth recording: claim 1 encrypts the trigger, while claim 17 recites encryption of the crippled data files as well as the trigger.

Dependent-claim coverage (context, not independent): claims 2/9 add the "catcher" directing data to a specific cache directory; claims 3/10 recite header removal/restoration; claims 4/11 add binary→seven-bit ASCII encoding; claim 5 adds digital→analog conversion at playback; claims 6/13 add the storage medium carrying auto-dial/auto-connect routing information to the host server; claim 7 adds server-permission-only storage of the trigger on non-volatile media "for permanent ownership"; claim 12 requires steps (c)/(d) be substantially simultaneous; claim 14 lists alternative crippling schemes (header removal, hidden-status flag removal, ZIP/UNZIP, extension change); claim 15 lists CD‑ROM, floppy disk, and hard drive as the storage medium; claim 16 adds permanent ROM storage of the trigger for repeated uncrippling; claims 18/19 limit the removed data to A/V header information and the device to a CD‑ROM.

---

Points of uncertainty / drafting defects I observed

• No 2026 appellate activity confirmed. I could not locate any CAFC 2026 docket for US 5,892,825. Given the patent's 2016‑05‑15 anticipated expiration and "Expired – Lifetime" status, active 2026 enforcement would be unusual, but I state this as an absence of evidence, not proof of absence. The Google Patents "family has litigation" indicator is unelaborated on the page I retrieved.
• Internal reference-numeral inconsistencies in the specification (e.g., RAM appears as both 12 and 13, with numeral 13 also used for the audio system; "end-user's computer 7" versus "server 7"). I am reporting these as they appear, not correcting them.
• Priority/relationship ambiguity: the specification calls the application a CIP of Ser. No. 08/645,022 (filed 1996‑05‑15), while the Google Patents family table also lists Ser. No. 08/568,631 (filed 1995‑12‑07) as a CIP parent and shows family priority dates of 1995‑12‑07 for several siblings. Priority dates on that page are expressly labeled as assumptions.
• Claim transcription in the source document contains apparent typographical errors (e.g., "the end-users's computer," "claims … according to claim 1, wherein said step (c) comprises removing the header data" where claims 1/8 label the header steps differently). I have summarized claims as written rather than repairing them.
• The patent contains a copyright notice ("COPYRIGHT - 1996 PLANET GRAPHICS, INC.") over portions of the disclosure, and the disclosed source listings.

I'll research litigation involving US patent 5892825 across patent litigation databases.

Let me search more targeted litigation databases and the patent's known assignees.

I found a lead. Let me confirm the litigation details.

I'll start by verifying the structured block's "no PTAB activity" signal against what's publicly searchable.

Let me search more specifically for PTAB activity and the litigation family history.

Let me check the litigation history and any post-grant challenges more directly.

I ran multiple targeted searches against the structured block's signal. Here is what I can and cannot substantiate.

Tooling note / limitation: I was cut off partway through my search plan (step limit). I completed the searches below and report them; two planned follow-ups (a direct PTAB E2E docket pull and a CourtListener docket sweep for CAFC appeals of any FWD) did not run. Where that leaves a gap, I say so explicitly rather than filling it in.

---

Proceedings overview

Total AIA trial proceedings on file for US 5,892,825: 0. Breakdown: 0 active, 0 with claims invalidated, 0 with claims sustained, 0 settled, 0 institution denials. The structured "PTAB proceedings on file" block is canonical and states that the USPTO ODP API returns no AIA trial proceedings for this patent as of the most recent ingest; my independent web searches surfaced no IPR, PGR, or CBM proceeding naming US 5,892,825 as the challenged patent, and no Federal Circuit appeal of any PTAB FWD on it.

Bottom line for a defendant: there is no IPR-based offensive or defensive playbook on this patent, because none has ever been tried. That is not the same as "hardened by surviving IPRs." The correct reading is different and, on the current record, more favorable to a defendant than an IPR win would be: the patent reached its anticipated expiration on 2016-05-15 and Google Patents lists its legal status as "Expired - Lifetime." There are no canceled claims to point at — but there is also no live patent worth attacking. Any assertion letter landing today is, on its face, asserting an expired patent.

I want to be precise about what the absence of PTAB activity does and does not tell you, because the prompt's framing ("well-asserted patents eventually attract IPRs") cuts the other way here: US 5,892,825 was never a high-volume NPE assertion vehicle during the AIA era, which is consistent with its expiration timeline and with the fact that current assignee of record is Apple Inc. (assignment recorded 2011-03-28 from Samantha Inventions LLC) — an operating company, not a troll.

---

Proceedings overview — why there is nothing to itemize

Because there are zero proceedings, the per-proceeding template has no entries. Instead, here is the near-miss ledger — challenges I affirmatively checked and ruled out, plus one that superficially looks relevant but is not.

Near-miss 1 — IPR2016-00472 (challenger v. patent owner of US 6,101,534)

• Type: Inter Partes Review — but of a different patent (US 6,101,534), not US 5,892,825.
• Why it surfaced: US 5,892,825 ("Mages") is cited as the primary § 102(e) prior art against claims 1, 3, 7, 8 and 9 of the '534 patent. The petition/exhibit excerpt I retrieved contains a side-by-side mapping table: "Claim 1 of the '534 Patent" vs. "U.S. Patent No. 5,892,825 ('Mages')," citing '825 Fig. 1 (end-user computer 1 / server 7), col. 6 ll. 41–48 (CD-ROM auto-dialing the host server) and col. 6 ll. 53–56 (host sending back the decrippling trigger).
• Defensive value to you: Double-edged and worth knowing. Your patent is not the target — it is the weapon. If you are ever facing assertion of US 6,101,534 or its family, the '825 disclosure is a ready-made § 102(e) anticipation reference. Conversely, if Apple ever asserted '825 against a CD-ROM/network-hybrid product, the accused party would find that the patent's own specification language ("Such an automatic connection is well-known," Windows 95 / ActiveX, INTEL "Smart Video Reorder Pro") reads like a roadmap for obviousness arguments — though this is my inference from the specification text, not a holding of any tribunal.
• Source: PTAB petition exhibit retrieved via Docket Alarm (Exhibit 1003 in IPR2016-00472). I did not verify the institution outcome, FWD, or appeal of IPR2016-00472 — it is not a proceeding on your patent, so I did not chase it further.

Near-miss 2 — BroadBridge Media, L.L.C. v. Hypercd.com, 106 F. Supp. 2d 505 (S.D.N.Y. 2000)

• Type: Not a patent case. This is an Anticybersquatting Consumer Protection Act (ACPA) domain-name case, decided in the patent owner's favor over the hypercd.com domain (the domain through which HyperCD-protected content was reached). Cited as BroadBridge Media, L.L.C. v. Hypercd.com, 468 / 106 F.Supp.2d 505, 55 U.S.P.Q.2d (BNA) 1426 (S.D.N.Y. 2000).
• Why it matters: It is almost certainly the source of the Google Patents "Family has litigation" flag, which points at a Darts-IP family record (family=24587338). That flag is about commercial litigation touching the HyperCD business, not PTAB validity challenges. Do not read it as evidence of an IPR.
• Caveat: I could not retrieve the full Darts-IP family litigation list, so I cannot rule out that some 1998–2002-era district court patent suit on '825 exists. I found no evidence of one, and by the time the AIA created IPRs (2012) the patent had only ~4 years of life left.

Near-miss 3 — entity/ownership lineage (not a proceeding, but strategically load-bearing)

Recorded assignment chain on the face of the patent: Individual → Planet Graphics (1996-11-25) → Hyperlock Technologies, Inc. (1998-01-12) → Broadbridge Media LLC (1999-11-17) → John Galt Media, Inc. (2002-07-22) → Sharp Holding Corporation (2008-12-04) → Octopus Media, LLC (2003-10-06 / 2008-11-18/2009-01-05) → Samantha Inventions, LLC (2009-01-05) → Apple Inc. (2011-03-28). Note also the Frank A. Leo security interest (2003-04-15), released 2009-01-02 with Sharp Holding as assignor.

This chain is the single best explanation for the zero-IPR result. The patent passed through a bankruptcy/liquidation-era patent-monetization chain (HyperLOCK → BroadBridge → Octopus → Samantha) and was swept into Apple in 2011. It was never held by a serial litigant in the post-AIA window, and it expired five years after Apple acquired it.

---

Strategic summary

Claim status. All 19 claims of US 5,892,825 — independent claims 1, 8, and 17, and dependents 2–7, 9–16, and 18–19 — are UNTESTED at the PTAB. None has been canceled. None has been affirmed. There is therefore no narrowing to report and no "surviving claim set" to hand a litigator. What exists instead is a term expiry: the structured record shows "2016-05-15 Anticipated expiration" and legal status "Expired - Lifetime." I'd flag one record inconsistency rather than silently reconcile it: the structured block lists "1995-12-07" as the priority date in one family row (that is the § 120 benefit date of parent Ser. No. 08/568,631) while the patent's own front page gives [63] Continuation-in-part of Ser. No. 645,022, May 15, 1996. Under the URAA, the 20-year term runs from the earliest claimed U.S. filing — which is why the expiry landed on 2016-05-15, matching the May 15, 1996 date, not the December 1995 date.

Estoppel landscape. § 315(e)(2) estoppel is not a factor here at all — no petitioner ever instituted, so no petitioner or privy is estopped, and equally, no petitioner received the benefit of a PTAB validity ruling. Practically this means: if you are somehow facing assertion, you are not limited to grounds "reasonably could have been raised" by anyone. You may raise any § 102/§ 103/§ 112 ground in district court. But the better first move is not a validity defense at all — it is expiration and the § 286 damages bar. Under 35 U.S.C. § 286, no recovery lies for infringement more than six years before the complaint is filed. Measured from today (2026-09-30), that six-year window opens on 2020-09-30 — a date that falls after the 2016-05-15 expiry. On those facts there is no recoverable damages window at all for a suit filed now. That is my own § 286 application to the structured expiry date, not a holding from any tribunal — you should confirm it against the specific theory being asserted (e.g., any pre-expiry acts already in suit).

Pattern signals. (1) No serial petitioner — the same-petitioner-files-multiple-IPRs pattern is absent because there are no petitions. (2) No PTAB appeal history — nothing to appeal; no CAFC docket found. (3) No defensive aggregator. I found no evidence that Unified Patents or any similar entity ever filed an ex parte reexamination or IPR against '825. (A Unified Patents-captioned reexamination transmittal form surfaced in my search, but it was for US 8,898,260, an unrelated patent — I checked specifically so as not to misattribute it.) (4) Family breadth is the real risk vector, not '825 itself. The continuation family — US 5,937,164, US 6,185,306, US 6,035,329, US 6,463,467, plus WO 98/24037, EP 0 974 217, CA 2271012, AU 52598/98 — shares the 1995-12-07 / 1996 filing ancestry, so its members are on the same expiry clock or earlier. If a demand letter cites the family rather than '825, check each member's expiry independently.

---

Recommended next steps

Because no PTAB activity exists, the honest answer is: there is nothing to litigate at the Board, and the analysis shifts entirely to the courts. Concretely:

1. Lead with expiration, not invalidity. The record shows "Expired - Lifetime" with anticipated expiration 2016-05-15 — verify on the Google Patents page for US5892825A and against USPTO PatentCenter maintenance/expiry records for Application No. 08/756,162. An expired patent cannot support an injunction and supports no forward-looking royalty.
2. Run the § 286 arithmetic on whatever is actually asserted. If the letter or complaint is dated today, the six-year lookback (2020-09-30 → 2026-09-30) sits entirely outside the patent term, leaving no damages period. If a suit was filed earlier, recompute against that filing date.
3. If a family member is asserted, pull its own term expiry — don't assume it is coextensive with '825. The family rows in the structured data all carry 1995-12-07 / 1996-11-25 lineage, which points to expiry in the same 2015–2016 range, but I have not verified each member individually and you should not rely on my inference.
4. If you do need a prior-art defense on any CD-ROM/network-hybrid claims, the '825 specification is itself a strong reference against neighbours like US 6,101,534 — see the § 102(e) mapping in the IPR2016-00472 petition exhibit (Docket Alarm). For your own patent, note that the specification affirmatively concedes the connecting technology was "well-known" and commercially available (Windows 95, ActiveX), which is useful § 103 framing, not a tribunal holding.
5. Close the two gaps I could not close, since the absence of PTAB activity is the whole conclusion and it should be airtight:
   • PTAB E2E / PTAB Center search on Patent No. 5,892,825 (and on the family numbers) — confirm no post-grant paper exists under any trial number.
   • CourtListener / CAFC docket sweep for any appeal naming Mages, Feng, Hyperlock, Broadbridge, John Galt, Octopus, Samantha Inventions, or 5,892,825.
     I found no such records, but I did not complete those two queries before hitting the tool limit, so treat them as unverified rather than confirmed-negative.

One caveat I will not paper over: the "zero PTAB proceedings" conclusion rests on (a) the structured ODP block, which is authoritative for its ingest date, and (b) my searches, which found no contra evidence. I did not find an affirmative enumeration of zero from PTAB E2E itself. Given that the whole defensive thesis here is "no proceedings," I'd verify that directly before relying on it in a filing.

I'll research the ownership chain and NPE patterns for US 5892825.

Found the Apple assignment reel. Let me dig into the intermediate entities.

Let me pull the assignment detail page and research the intermediate entities.

Ownership & Assignment Analysis — US 5,892,825

Method / source limitation (read first): I was able to reach the USPTO Patent Assignment Search interface only indirectly. The reel/frame numbers below are given only where an independent mirror of the USPTO assignment database (plainSite, which reproduces USPTO assignment abstracts of title) surfaced them. For the remaining links I have the conveyance type, assignor, assignee and recording date from the record set reproduced in the patent's own legal-events table (Google Patents, which mirrors USPTO assignment records) but not the reel/frame or the correspondent field. I am flagging that as a data gap rather than guessing. Verify at USPTO Assignment Center — search 5892825.

---

Inventors

Pattern note: The inventors assigned to Planet Graphics on filing day, which is a normal employee-assignment pattern, not a departure/abandonment signal. I found no evidence of either inventor departing within 12 months of filing, and I did not find a later inventor-held record (i.e., no reversion to the inventors), which is what a portfolio fire-sale usually looks like at the inventor level. Both names recur across the patent family (US 5,937,164; US 6,185,306; US 6,035,329), suggesting a small, stable inventor team rather than a one-off.

Discrepancy to flag: Google Patents' header for this patent lists the original assignee as "Individual" and states "Application filed by Individual," while the recorded assignment of the same date names Planet Graphics as assignee. These are not reconcilable from the sources I could reach; the reel record (Planet Graphics) is the more reliable of the two.

---

Original assignee

Planet Graphics (also rendered "Planet Graphics, Inc." in the 1997 and 1998 instruments) — a small graphics/media company; the specification's commercial example is illinois-lottery video advertising delivered on a CD-ROM ("HyperCD") that auto-dials the advertiser's web server (col. 6, http://tekweb.com/hypercd/adver/lotto.html).

• Product embodying the claims: likely yes, in prototype/commercial form — the CD-ROM "HyperCD" with removed-header video files and an auto-dial URL stub is exactly what the claims recite, and the spec includes a full source-code appendix (SENDFILE.C and the "Wincode" encoder/decoder dialogs). I could not verify shipping volumes or retail distribution.
• Primary line of business: multimedia/Internet advertising production, not patent licensing.
• Current status: unknown / no independent evidence retrieved. The name disappears from the chain in January 1998, ~20 months after the patent's filing, when it assigned everything to Hyperlock Technologies. I found no bankruptcy filing, dissolution notice, or acquisition record for Planet Graphics.

---

Assignment timeline

13 recorded instruments. Everything below is from the record set reproduced in the patent's legal-events table unless a reel is cited.

1. 1996-11-25 (exec.) / recorded 1996-11-25 — Reel not retrieved

   • Conveyance: Assignment of Assignors' Interest
   • Assignor: Jie Feng; Kenneth G. Mages
   • Assignee: Planet Graphics
   • Correspondent: not retrieved. Flag: if the same correspondent recurs here and on the 1998 Hyperlock recording, that would show the original prosecution/transaction counsel carried the assets forward — worth completing.
   • Context: ordinary inventor-to-employer assignment, executed on the filing date.

2. 1997-09-11 / recorded 1997-09-11 — Reel not retrieved

   • Conveyance: Security Agreement
   • Assignor: Planet Graphics, Inc.
   • Assignee: The Saranow Group, L.L.C.
   • Correspondent: not retrieved.
   • Context: securitization — Planet Graphics pledged the patent as collateral roughly four months before the issued patent's filing-date sibling patents and seven months before it divested the portfolio. A lender taking the patent as collateral this early is a financial-distress tell, not a normal commercial step.

3. 1998-01-12 / recorded 1998-01-12 — Reel not retrieved

   • Conveyance: Assignment of Assignors' Interest
   • Assignor: Planet Graphics, Inc.
   • Assignee: Hyperlock Technologies, Inc.
   • Correspondent: not retrieved.
   • Context: transition to a dedicated operating/venturing vehicle. Hyperlock is the entity that appears as the parent company in third-party patent databases for this family (Unified Patents lists "Hyperlock Tech Inc" as parent company for family member US 5,937,164).

4. 1999-11-17 / recorded 1999-11-17 — Reel not retrieved

   • Conveyance: Assignment
   • Assignor: Hyperlock Technologies, Inc.
   • Assignee: Broadbridge Media, L.L.C.
   • Correspondent: not retrieved.
   • Context: outbound transfer of the whole family to a differently named media LLC, ~7 months after grant. This is the first link where the buyer's business is not documented as product-making.

5. 1999-12-06 / recorded 1999-12-06 — Reel not retrieved

   • Conveyance: Release of Security Agreement
   • Assignor: The Saranow Group, LLC
   • Assignee: Hyperlock Technologies, Inc.
   • Correspondent: not retrieved.
   • Context: release of the 1997 collateral lien, recorded three weeks after Hyperlock had already conveyed the patent to Broadbridge. The lien release follows the asset out the door — consistent with a lender being paid off out of sale proceeds.

6. Executed 2001-07-23 / recorded 2002-07-22 — Reel 13117/0441

   • Conveyance: Assignment
   • Assignor: Broadbridge Media, L.L.C.
   • Assignee: John Galt Media, Inc., 412 Eighth Avenue, 5th Floor, New York, NY 10001
   • Correspondent: not shown on the retrieved abstract. Flag: this is the highest-value field to complete — a single New York correspondent filing both this and the 2008 Sharp Holding / Octopus Media recordings would tie the whole 2001–2009 chain to one agent.
   • Context: sale to a New York media holding company, with a ~12-month execution-to-recording lag — the sort of delay seen when a transaction closes in distress and recordation is deferred.

7. 2003-04-15 / recorded 2003-04-15 — Reel not retrieved

   • Conveyance: Security Interest
   • Assignor: Sharp Holding Corporation
   • Assignee: Frank A. Leo
   • Correspondent: not retrieved.
   • Context: second securitization of the same asset, this time on the Sharp Holding side. The portfolio has now been pledged as collateral twice in six years.

8. 2003-10-06 / recorded 2003-10-06 — Reel not retrieved

   • Conveyance: Assignment
   • Assignor: Sharp Holding Corporation
   • Assignee: Octopus Media, LLC
   • Correspondent: not retrieved.
   • Context: transfer into a licensing-style LLC that holds title for the next ~5 years.

9. 2008-11-18 / recorded 2008-11-18 — Reel not retrieved

   • Conveyance: Option
   • Assignor: Octopus Media, LLC
   • Assignee: Samantha Inventions, LLC
   • Correspondent: not retrieved.
   • Context: option grant — the structural precursor to a sale. An option recorded before the assignment is the signature of a structured, negotiator-driven transaction rather than a spot sale.

10. 2008-12-04 / recorded 2008-12-04 — Reel not retrieved

    • Conveyance: Assignment
    • Assignor: John Galt Media, Inc.
    • Assignee: Sharp Holding Corporation
    • Correspondent: not retrieved.
    • Context: chain-of-title cleanup. John Galt's residual interest (from reel 13117/0441) is swept up by Sharp Holding in December 2008, one month after the Samantha option and one month before the Samantha assignment. This is title-curative work performed specifically to deliver clean ownership.

11. 2009-01-02 / recorded 2009-01-02 — Reel not retrieved

    • Conveyance: Release by Secured Party
    • Assignor: Frank A. Leo
    • Assignee: Sharp Holding Corporation
    • Correspondent: not retrieved.
    • Context: lien release, clearing the 2003 Frank A. Leo security interest two days before the assignment to Samantha Inventions.

12. 2009-01-05 / recorded 2009-01-05 — Reel not retrieved

    • Conveyance: Assignment
    • Assignor: Octopus Media, LLC
    • Assignee: Samantha Inventions, LLC
    • Correspondent: not retrieved.
    • Context: exercise of the 2008 option; title moves to the LLC that will hold it for the Apple transaction.

13. Executed 2010-03-15 / recorded 2011-03-28 — Reel 26034/0439

    • Conveyance: Assignment of Assignors' Interest
    • Assignor: Samantha Inventions, LLC
    • Assignee: Apple, Inc., 1 Infinite Loop, Cupertino, CA 95014
    • Correspondent: the retrieved abstract has a Correspondent field that was not populated in the text I could capture — do not treat as "none." Complete this field; the identity of the firm recording a six-patent package to Apple is the single best indicator of who ran the 2008–2009 cascade.
    • Context: package acquisition by an operating company. Reel 26034/0439 covers the whole family in one instrument: US 5,892,825; 5,937,164; 6,035,329; 6,185,306; 6,463,467 and 6,438,232. Note the ~12½-month gap between execution (2010-03-15) and recording (2011-03-28).

Also of record (family-level, not this patent's chain): the "option/assignment" and "security interest/release" pairs above are cross-recorded against the sibling patents, so the ownership history of '825 cannot be read in isolation from US 5,937,164 / 6,185,306 / 6,035,329 / 6,463,467. Apple took all of them at once.

---

Timeline diagram

timeline
    title Ownership of US 5892825
    1996 : Filed by inventors Mages and Feng
         : Assigned to Planet Graphics
    1997 : Security agreement to Saranow Group
    1998 : Assigned to Hyperlock Technologies
    1999 : Assigned to Broadbridge Media
         : Saranow security interest released
    2001 : Broadbridge agrees to sell to John Galt
    2002 : John Galt Media assignment recorded
    2003 : Frank Leo takes security interest
         : Assigned to Octopus Media LLC
    2008 : Option granted to Samantha Inventions
         : John Galt assigns to Sharp Holding
    2009 : Leo security interest released
         : Assigned to Samantha Inventions
    2011 : Assigned to Apple Inc
    2016 : Patent expired

---

NPE / troll-pattern signals

1. Shell-entity transfer — unclear. Facts that cut toward present: the chain passes through at least four successive non-operating holding vehicles — Broadbridge Media, L.L.C. (1999), John Galt Media, Inc. (2002, reel 13117/0441), Sharp Holding Corporation, Octopus Media, LLC (2003), Samantha Inventions, LLC (2009) — none of which I can associate with a shipped product. Facts that block a firm call: I did not retrieve registered-agent addresses, state of organization, or member identity for John Galt Media, Octopus Media, or Samantha Inventions, and I have no "no products in commerce" evidence beyond their absence from product records. John Galt's recorded address (412 Eighth Avenue, 5th Floor, New York, NY 10001) is an office building, not a known registered-agent service. Under the rule against inferring from naming alone, this stays unclear.

2. Known asserter in the chain — not present. No assignee in the record matches the supplied NPE list (Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or any Spangenberg entity). To be fair to the data: I did not complete a search of the RPX / Unified Patents asserter directories for Octopus Media, LLC or Samantha Inventions, LLC as plaintiff names, so "not present" rests on absence from the named lists plus absence of any suit I could surface. The Unified Patents portal entry for the family (US-5892825-A) lists only Hyperlock Tech Inc and Apple Inc as parent companies.

3. Repeat correspondent across the chain — unclear (data not retrieved). The correspondent field is the highest-value tell here and it is precisely the field I could not capture, for the 1996, 1997, 1998, 1999, 2003, 2008 and 2009 recordings. On reel 13117/0441 and reel 26034/0439 a Correspondent column exists in the abstract but was not populated in the text available to me. I am not going to assert recurrence I did not see. If a single firm appears as correspondent on both reel 13117/0441 (2002) and the 2008–2009 Octopus/Sharp/Samantha recordings, that converts signals 1 and 4 from "unclear" to "present." Recommend this as the first follow-up pull.

4. Cascading transfers — present. Four recorded instruments in roughly seven weeks — 2008-11-18 (Octopus→Samantha Option), 2008-12-04 (John Galt→Sharp Holding), 2009-01-02 (Leo→Sharp Holding Release), 2009-01-05 (Octopus→Samantha Assignment) — involving five distinct parties (Octopus Media, Samantha Inventions, John Galt Media, Sharp Holding, Frank A. Leo). The sequence is textbook title-curative: sweep residual interests back to the parent, release the collateral lien, then exercise the option and convey. The participants plainly shared a common objective and, on the face of the record, a common transaction. This is the one signal I can call with confidence.

5. Pre-litigation transfer — unclear, and I flag a contradiction. The Google Patents record for this patent states "Family has litigation — First worldwide family litigation filed" (Darts-IP family 24587338), yet the litigation section previously generated for this analysis is a stub — it contains research intent ("I'll research litigation…", "I found a lead") and no case caption, court, date or party. I therefore have no first-suit date to measure the six-month window against. This is a direct contradiction between the primary source and the earlier-generated section and should be resolved before any verdict on litigation timing is relied upon. On the timeline itself, the cascade of 2008-11-18 → 2009-01-05 sits roughly 13–14 months before the Apple assignment was executed and ~26 months before it was recorded — that reads as sale preparation, not assertion preparation.

6. Bankruptcy fire-sale — unclear. No Chapter 7/11 filing was found for Planet Graphics, Hyperlock Technologies, Broadbridge Media, or John Galt Media. Against that: the record is full of distress markers — a security agreement to Saranow Group in 1997 followed by divestiture of the entire portfolio within five months; a lien release recorded three weeks after the encumbered asset had already been sold; a ~12-month gap between signing and recording the 2001 Broadbridge→John Galt sale; and a second security interest taken by Frank A. Leo over the same portfolio in 2003. That pattern is consistent with repeated refinancing or a distressed disposition, but "consistent with" is not a recorded bankruptcy, and I will not upgrade it.

7. Privateering — not present. No evidence was found that any operating company transferred this patent to an NPE to assert against its own competitors. The 2011 acquisition by Apple runs the opposite direction: an operating company bought the family into its own portfolio, capping off a decade of third-party holding.

8. Defensive aggregator (anti-NPE) — not present. Apple Inc. is not RPX, AST, LOT Network, Unified Patents or OIN. The chain terminates at an operating company, not at a neutralizer — which is protective in practice but does not fit this signal's definition.

---

Verdict

NPE — moderate confidence.

Justification: exactly one strong signal — the cascading transfer cluster of 2008-11-18 through 2009-01-05 (four instruments, five parties, seven weeks) — supported by a chain in which at least five successive assignees (Broadbridge Media, John Galt Media at reel 13117/0441, Sharp Holding, Octopus Media, Samantha Inventions) show no evidence of ever shipping a product, plus two separate pledge-and-release securitizations (1997 Saranow Group; 2003 Frank A. Leo). I deliberately do not reach high confidence, because the two strongest corroborating signals are unverified: the correspondent-of-record field was not retrievable for any link, and the "shell entity" characterization lacks registered-agent, state-of-organization and ownership evidence for Octopus Media and Samantha Inventions. Critically, this is a historical chain-quality finding, not an active-troll finding — the chain terminates at Apple Inc. via reel 26034/0439 (executed 2010-03-15, recorded 2011-03-28), an operating company, and the patent's status is Expired – Lifetime with an anticipated expiration listed as 2016-05-15. No post-Apple assertion by any of these entities was surfaced.

Verification links:

• USPTO Assignment Center / Patent Assignment Search: https://assignment.uspto.gov/patent/index.html (query patent 5,892,825)
• Samantha Inventions, LLC → Apple Inc., reel 26034/0439: https://www.plainsite.org/patents/assignment.html?id=[5127219](/patent/5127219)
• Broadbridge Media, L.L.C. → John Galt Media, Inc., reel 13117/0441: https://www.plainsite.org/patents/assignment.html?id=[2386805](/patent/2386805)
• Google Patents legal events: https://patents.google.com/patent/US5892825/en
• Unified Patents portal: https://portal.unifiedpatents.com/patents/patent/US-5892825-A

Open items I could not close (do not treat as negative findings): (i) reel/frame and correspondent for the 1996, 1997, 1998, 1999, 2003, 2008 and 2009 recordings; (ii) the correspondent on reel 26034/0439 and reel 13117/0441; (iii) the litigation record behind the "Family has litigation" flag; (iv) the corporate status of Planet Graphics, Hyperlock Technologies, Inc., Broadbridge Media, L.L.C., John Galt Media, Inc., Sharp Holding Corporation, Octopus Media, LLC and Samantha Inventions, LLC; and (v) the identity of Frank A. Leo relative to Sharp Holding.

I'll research the USPTO record for US 5,892,825 and reconstruct its cited prior-art list, since the authoritative text in my working set is truncated mid-way through the "Citations (44)" table (it breaks off at US 4,305,098 A).

The authoritative text I was given truncates the "Citations (44)" table at US 4,305,098 A. Let me reconstruct the remainder from independent sources.

Prior-Art Analysis — US 5,892,825 A

0. Verification of the identifier and scope

Identifier check (interpreted literally, no auto-correction). I searched the string 5892825 and accepted only records matching it. Every source returned the same document: US 5,892,825 A, application 08/756,162, filed 1996‑11‑25, issued 1999‑04‑06, inventors Kenneth G. Mages and Jie Feng, title "Method of secure server control of local media via a trigger through a network for instant local access of encrypted data on local media." I did not substitute near-neighbours such as US 5,882,825, US 5,892,285, or US 6,101,534 (the latter being a different Mages/Planet Graphics-family patent that appears in PTAB records).

Interpretation of "each patent citation for 5892825." The patent's front page carries two distinct lists, and the distinction matters:

• "Citations (44)" — the 44 references cited in (against) US 5,892,825. This is what "prior art for 5892825" means, and it is what I analyse below.
• "Cited By (159)" — later patents that cite 5892825 as prior art. These are not prior art against it; they are its progeny. I exclude them.

Sources actually used, and an important limitation. The authoritative full text supplied to me truncates the "Citations (44)" table at entry 14 (US 4,305,098 A). I therefore reconstructed the balance from (i) Google Patents' rendering of the USPTO front-page reference data, (ii) the Unified Patents portal record for US-5892825-A (a third-party mirror of the reference data), (iii) PlainSite USPTO assignment records (which confirm application 08/756,162 → Patent 5,892,825), and (iv) search-report citations of 5892825 in later cases (e.g. WO 2001038954 A1, which cites it as an "X" reference against claims 1–3; EP 1593268 A1, citing it in examination). I did not retrieve the scanned USPTO front page itself within my search budget, so the list below is marked by verification tier.

Two caveats I will not paper over:

1. Tier 2 entries are provisional. The mirror gives filing/priority dates and titles for many, but not grant dates; a small number of fields were truncated in retrieval.
2. I have not read the full text of most of these references. My § 102 assessments for Tier 2 rest on the reference's subject matter and title, not a full-text comparison against claims 1, 8 and 17. They are screening judgements, not complete § 102 rejections.
3. § 102 subdivision depends on the effective priority date, which is contested in this family. The face of the file says priority 1996‑05‑15 (CIP of Ser. No. 08/645,022); the Google Patents family table also shows a 1995‑12‑07 lineage (Ser. No. 08/568,631). So the § 102(b)-critical date is either ~1995‑05‑15 or ~1994‑12‑07.

---

1. Tier 1 — References verified verbatim from the file text (items 1–14)

These are quoted directly from the authoritative document, in the ascending patent-number order in which the table presents them.

Note that in Tier 1 there is no identified reference-filing gap; the table runs in strict ascending numeric order and includes several references (object monitoring, point-of-sale, machine control) that sit in wholly unrelated arts.

---

2. Tier 2 — Remaining references (provisional reconstruction)

Reconstructed from the third-party mirror of the front-page data. Grant dates were not always retrievable and are marked "n/r". I account for 29 additional references here, bringing the total to 43 of the stated 44; I could not reconcile the final entry, and I flag that rather than invent one.

---

3. § 102 analysis — which claims each reference could reach

3.1 Methodological point first

The 44 references are the front-page "References Cited" list. Critically, the front page does not tell you whether a reference was used under § 102 (anticipation) or § 103 (obviousness), nor whether it was cited by the examiner or merely submitted in an IDS. In a Mages/Planet Graphics filing of this era, the list has the hallmarks of an applicant-submitted IDS: it sweeps in 1970s object-monitoring and point-of-sale art (items 1–5, 7–13) that has no § 102 bearing on claims 1, 8 or 17 whatsoever. I therefore separate the art into clusters.

3.2 Cluster A — 1970s object-ID / point-of-sale / machine control (items 1–14)

References: US 3,829,833; 3,911,397; 3,924,065; 3,946,220; 3,970,992; 4,071,911; 4,112,421; 4,209,787; 4,217,588; 4,220,991; 4,232,317; 4,270,182; 4,305,098.

§ 102 status: all are § 102(b) art (published well over one year before either candidate priority date).

Anticipation analysis: None of these anticipates any of claims 1, 8 or 17. Every independent claim requires (i) a crippled video/audio file on a storage medium, (ii) uncrippling trigger data held at a remote host computer and (iii) transmitted over a network, (iv) receipt into volatile RAM, and (v) combining in RAM with a prohibition on copying to non-volatile memory. Cluster A contains no networked server, no encryption, no A/V media, and no RAM-resident key handling. At most these teach the generic notion of "interrogate an identifier, then grant access" — which is § 103 fodder at best, and only against the broadest functional language.

3.3 Cluster B — Digital A/V capture, storage, compression and CD reading (items 6, 16, 21–23, 25, 27, 30, 33, 34)

References: US 4,070,692; US 4,555,803; US 5,321,750; US 5,375,068; US 5,388,097; US 5,410,354; US 5,450,489; US 5,499,252; US 5,535,199; US 5,535,327.

Anticipation analysis: these map onto claim 1, step (a) (converting analog video/audio to digital) and to claim 1's "combining in RAM" only in isolation. US 4,070,692 and US 4,555,803 teach digitising and filing images; US 5,410,354 and US 5,375,068 teach compressed video handling and networked video; US 5,499,252 (CD-ROM decoder reading selected data from CD into memory) is the most proximate, because claim 1(g) and claim 17 require joining the trigger with CD data in RAM — but 5,499,252 discloses no remote trigger, no encryption, and no anti-persistence measure, so it cannot anticipate claims 1, 8 or 17 in their entirety. It is relevant at most to step (a)/(g) individually.

3.4 Cluster C — Access control, licensing and pay‑per‑use of content (items 15, 17, 19, 20, 24, 28, 29, 31, 32, 36, 37, 40–43)

This is the only cluster with real § 102 exposure, and it is where I would focus if I were challenging the patent.

No single Cluster C reference, on what I can see, discloses the full conjunction of (1) crippled local media with (2) the removed critical information held as a small trigger at a remote host, (3) delivered over a network into (4) volatile RAM, (5) combined there with the local crippled file, and (6) with copying to non-volatile memory affirmatively prevented. Elements (4) and (6) — the catcher's RAM-only residency and the anti-persistence limitation in claim 1(f)(1)–(3) and claim 8(c) — are the narrowest and most defensible limitations, and they are the ones for which I found no prospective § 102 reference among the 44. That is the same reading later examiners reached: WO 2001038954 A1 and WO 2004051443 A1 each cite US 5,892,825 (and its sibling US 6,463,467 B1) as an "X" reference — i.e. anticipation-quality art against later applications — which confirms the disclosure's breadth without bearing on its own validity.

3.5 A drafting asymmetry worth carrying forward

Claim 1 encrypts only the trigger; claim 17 recites encryption of the crippled data files themselves as well as the trigger. Cluster C's encryption art (US 5,400,403; US 5,629,980; US 5,638,443) therefore bites harder on claim 17 than on claim 1. This asymmetry — already flagged in the Summary section — is the single most useful handle for a § 102 attack mounted on the cited references.

---

4. Bottom line

• Most relevant prior art overall: US 5,319,705 A (IBM, Halter) and US 5,457,746 A (Spex) — both § 102(b), both directed to remotely-enabled access control of content on local/portable media, and together the closest to independent claims 1, 8 and 17. Add US 5,222,134 A for claim 7 and US 5,400,403 A for the encryption limitations.
• No cited reference, on the evidence available, anticipates any independent claim in full. The RAM-only conjunction and the anti-copying-to-non-volatile-memory limitation (claim 1(f)(3); claim 8(c); claim 17 final clause) are not disclosed by any of the 44.
• The bulk of the list (items 1–14 and much of Cluster B) has no § 102 purchase and reads as IDS background.
• I did not verify full text, grant dates, or examiner-vs-applicant provenance for the Tier 2 entries, and I could account for 43 of the 44 listed references. US 4,633,207 A in particular looks facially irrelevant and may be a mirror-indexing artefact. These are open items, not findings.

I'll research the prior-art record for US 5,892,825 before analyzing § 103.

Now let me pull the actual examiner-cited references and the key analogous-art references used against the sibling patents.

Obviousness Analysis — US 5,892,825 under 35 U.S.C. § 103

Scope note. As requested, I anchor this analysis on the Prior Art section of the US5892825A page (https://patents.google.com/patent/US5892825/en) — i.e., the examiner-cited references, the CPC classifications, the "Prior art keywords," and the specification's own admissions. Where I rely on material I retrieved from outside that page (the IPR petition exhibits, the Unified Patents portal), I say so explicitly.

Two record contradictions to flag up front, per instruction:

1. Date discrepancy. My operating rules give today's date as 2026-09-30; the task header gives April 26, 2026. This does not affect the § 103 analysis (which is keyed to the 1995/1996 critical date) but it does affect any § 286 six-year damages arithmetic and I am not silently reconciling it.
2. My own evidence gap. The "Citations (44)" block on the page as supplied to me is truncated after US 4,305,098 A (issued 1981-12-08). I have only the 1972–1981 slice of the examiner citations visible. My follow-up searches for the full list and for specific references (Halpern US 5,319,705; Hellman US 4,658,093; Shear US 4,827,508; the Batchelor patent number) did not return results before I hit the tool limit. I therefore do not assert the numbers or contents of those references below, and I mark every such gap as unverified rather than filling it in.

---

I. Critical date — this is dispositive, so it goes first

Consequence: pre-AIA § 102(a)/(b) art must predate the invention date, presumptively the earliest supporting filing. That means:

• Refs dated on or before 1995-12-07 / 1996-05-15 — clean prior art.
• Refs dated after 1996-05-15 but before the 1996-11-25 filing — not § 102(a)/(b) art to the '825 claims (they would only matter if the priority claim to 1996-05-15 fails for a given claim, or against a later-filed family member).

This matters concretely because of the Non-Patent Literature surfaced on the Unified Patents portal for US-5892825-A (https://portal.unifiedpatents.com/patents/patent/US-5892825-A), which lists eight NPL items, including:

• Reisman, "CD-ROM/Online Hybrids—The Missing Link," CD-ROM Professional, vol. 8, no. 4, Apr. 1995 — clean art.
• Reisman, "All Power to the Web—CD-ROM is dead, or is it," N.Y. Media Assoc. News, Jan. 1996 — clean art (subject to the applicant swearing behind to a Dec. 1995 conception).
• Reisman, "Spin Webs around a CD-ROM: The Next Generation of CD/Web Hybrids," Mass High Tech, Sept. 1996 — date-marginal; not § 102(a) art to '825 on its face.
• "Overview of CD-ROM Encryption, Copy Protection, and Metering," Disc Mfg., Inc., Oct. 19, 1996 — date-marginal, same problem.

I flag this because a § 103 chart built on the Sept./Oct. 1996 items would be date-defective against '825 itself (though usable against family members filed in 1997–1998). The Apr. 1995 and Jan. 1996 Reisman items are the load-bearing ones.

Attribution caveat: the NPL list came from a search result (Unified Patents portal), not from the page's own Prior Art section. Treat its contents as second-hand.

---

II. What the page's Prior Art section actually tells us — including the classification signal

Three things in the page's Prior Art material are analytically important and are not about individual references at all:

(1) Prior art keywords: data, crippled, audio, uncrippling, video.
Remember that this list was generated from the patent's face/citations. The word "crippled"/"uncrippling" being a top extracted keyword means the vocabulary of deliberate file-degradation was already the indexing handle for this art. That is not by itself prior art, but it establishes that the field was organized around exactly the concept the claims are said to add.

(2) The CPC assignments are themselves a § 103 argument. '825 was classified into:

A patent classified squarely into G11B 20/00224 ("key obtained from a remote server") and G11B 20/00137 ("restriction to authorised users") is, by the Office's own classification scheme, sitting in an art area whose recognized techniques are (i) restricting access to authorized users and (ii) obtaining the key from a remote server. Under KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), that is the "identified, predictable" solution space for the problem, not an inventive departure from it. The classification makes the combination that the claims recite the default in this art unit, not an exception.

(3) The visible 1972–1981 examiner citations cluster into one function. The citations I can read — US 3,829,833 ("Code element identification method and apparatus"), US 3,911,397 ("Access control assembly"), US 3,920,065, US 3,946,220 (Transactron, point-of-sale), US 3,970,992 (IBM, transaction terminal), US 4,220,991 (electronic cash register with removable memory packs for cashier identification), US 4,270,182 (automated information input, storage and retrieval), US 4,234,317, US 4,202,987 / 4,212,421 / 4,112,421 (object monitoring), US 4,070,692 (video digitizing system), US 4,305,098 (UK Secretary of State for Defence) — are, collectively, an identification → authorization → access-grant cluster plus a video-digitization reference. That is precisely the skeleton of claim 1's steps (d)–(g) (store trigger at host → transmit → receive → unlock). The Office was reading this invention against forty years of challenge-response access control, which is a strong § 103 posture for the "authorize then release the key" elements.

I could not map the 1981–1996 citations. Those would almost certainly be the references that map to CD-ROM, network, and media encryption elements. That is the single largest hole in this analysis and I am not going to paper over it.

---

III. The specification's own admissions — the backbone of the § 103 case

Statements in the '825 specification about what was "well-known" or "conventional" are treated as admitted prior art (MPEP § 2144.03; In re Fout). The specification supplies most of the claim elements itself:

* Line cites here and below are as quoted in the IPR2016-00472 petitioner exhibit retrieved via Docket Alarm and via PTACTS (see § IV.C); I did not independently verify every column/line pin against the printed patent.

Why this matters: the admitted prior art supplies the transport, the codec/format, the encryption tool, and the motivation. What is left for the claims is the arrangement: bulk media sits crippled on local media; a small key sits on a server; the key is fetched on demand and assembled in RAM. Under KSR, that is a combination of known elements with predictable results.

---

IV. Specific § 103 combinations

A. Combination 1 — Reisman (CD-ROM/online hybrid) + conventional remote-key/authorization art

References: Reisman, "CD-ROM/Online Hybrids—The Missing Link," CD-ROM Professional 8(4), Apr. 1995; Reisman, "All Power to the Web—CD-ROM is dead, or is it," N.Y. Media Assoc. News, Jan. 1996.
Elements taught: the CD-ROM/web hybrid premise itself — a disc that carries content plus an embedded address, with the computer auto-connecting web-side. This reads on claims 6 and 13 (storage medium carrying "the necessary information for automatically and directly connecting … to a host computer") and on the premises of claims 1(c)–(e) and 8(a)–(a).
Motivation: identity of problem and field. Reisman's stated subject is the bandwidth bottleneck between local optical media and online content — the same problem the '825 Background spends four paragraphs on.

Strength: high for claims 6/13 and for the general architecture. Weakness: I have only the citation strings, not the article texts. I have not read them and cannot quote them.

---

B. Combination 2 — Header-removal crippling + network-delivered restoration key

References: the identification/access-control cluster from the examiner citations (US 3,829,833; US 3,911,397; US 4,220,991; US 4,270,182) for the authorize-then-release mechanism, plus the admitted art for the encryption/transport.
Elements taught: the three-step skeleton of claims 1(d)–(g) and 8(a)–(d): credential held remotely → request → authorization → release.
Motivation: the "crippling" here is a file-format defect, and file-format defects were the standard, cheap way to render a file unopenable. Removing a header so that a parser cannot identify the stream is the minimal-bandwidth lock: it costs almost nothing to store (because the bytes simply aren't there) and almost nothing to transmit (because only the header crosses the network). That economy — a few bytes instead of 2.5 MB, per the spec's own arithmetic — is an express, articulated reason to select header removal over, e.g., bulk re-encryption of the payload.

This combination is what makes claims 3, 10, and 18 (header removal / header restoration / header as the removed data) the weakest claims in the patent. Their entire content is "remove the parse-critical bytes; send them back."

---

C. Combination 3 — Mages ('825) + Batchelor + VEMMI, as actually pleaded in a PTAB proceeding

This is the one combination in the record where a real party wrote out the motivation language. In IPR2016-00472 / IPR2015-01364 (challenging US 6,101,534), the petitioner used US 5,892,825 ("Mages") as the primary § 102(e) reference, and relied on Batchelor and VEMMI for the server-control element. The petitioner's articulation (retrieved via PTACTS and Docket Alarm:

"Mages and Batchelor both recognize the desirability of locally storing data on a CD-ROM and a remote server directing access to that data… one of ordinary skill in the art would have recognized and appreciated that the remote server of Mages could access and control the specific tracks of the local CD-ROM as specifically provided by Batchelor… This would have been desirable because… it would enable the remote facility to directly control the utilization of specific tracks on the local CD-ROM such that two complementary pieces of information are displayed simultaneously."

And on VEMMI: "video data stored on a local disk or CD-ROM and whose display is controlled by the remote server."

Why this is directly relevant to a § 103 attack on '825: the petitioner's stated reason to combine is also the reason to combine against '825's own "server control of local media" objective — and the '825 specification itself recites that objective verbatim ("It is another objective of the present invention to enable server control of the local media data … Specific tracks on the CD-ROM can thereby be controlled by the remote server," col. 4 ll. 7–17). Petitioner's second combination, Mages + VEMMI, relied on the express finding that "Mages and VEMMI were seeking to address the same type of problem – avoid the drawbacks associated with transmitting large data files over a network – and did so with similar solutions." That is a textbook KSR "same problem, known solution" rationale.

Two important limits on this: (i) these are a party's contentions, not a tribunal holding, and they were directed at US 6,101,534, not at '825; (ii) I did not verify the institution outcome or any final written decision. I am using them as evidence of what a skilled practitioner would have found obvious and why, which is legitimate, not as res judicata.

Also noteworthy: during the '534 prosecution, the examiner twice rejected claims as obvious over Mages alone — "On pages 40 and 41 of the Second Office Action, the Examiner asserted that the claimed invention would have been obvious in view of Mages." A single reference being treated as teaching an entire interactive remote computer interface system tells you how much scope the Office attributed to the '825 disclosure.

---

D. Combination 4 — VDOLive (admitted) + local bulk storage

Reference: VDOLive, VDOnet Corp., admitted in the '825 Background.
Teaching: real-time-ish compressed video over POTS, buffered at the user's computer, but "choppy and drops as much as 96% of the video data over a conventional phone line."
Motivation, as a KSR "finite number of identified solutions" argument: given an admitted art that (a) already buffers media in the user's computer memory and (b) already fails because of bandwidth, the identified solutions were: compress harder; cache more; or stop shipping the bulk data over the wire at all. The '825 claims take the third path. That is not merely "obvious to try" — it is the relief the background itself sets up.

---

E. Combination 5 — MIME/base64 (admitted) for claims 4 and 11

Claim 4 recites "converting binary data into seven-digit ASCII code"; claim 11 recites "encrypting … from binary format into encoded text format data." The '825 Background names MIME, RFC 1421's 3-in-4 base-64 scheme, uuencode, pure hexadecimal, and ATK! as existing encodings for moving binary through text channels. On its face, claims 4 and 11 are anticipated-or-obvious over the applicant's own admitted prior art, with no combination required. Claim 4, in particular, is drafted so broadly ("seven-digit ASCII code") that it may not even exclude ordinary base64.

---

F. Combination 6 — Standard DOS/Windows file operations for claim 14

Claim 14 lists, as alternative crippling schemes: header removal; removing the hidden-status flag; unzipping a zipped file; changing the file extension. Every one of these is a standard operating-system file operation, and the '825 specification lists them itself:

"the file could be made a hidden file, with the trigger data from the host computer being a command to remove the hidden status. Alternatively, the video/audio file could have a changed extension… the crippling of the video/audio file may be achieved by the use of ZIP file, with the trigger data from the host computer being a command to UNZIP the data."

The motivation to use a password-protected ZIP, or a hidden/system attribute, or an extension change, as an access gate is not merely present — it is the ordinariness of the technique. Claim 14 survives only if the combination of "obscure the file" + "server holds the un-obscuring command" is non-obvious, which is difficult when the server simply sends "unzip."

---

V. Claim-by-claim vulnerability

---

VI. The two places the patent might actually resist — and why they may not hold

A. The negative limitation in claim 1(f)(3): "preventing the copying thereof to nonvolatile memory"

This is the only element I would not want to concede lightly, and it is worth being precise about why.

In 1996, the conventional mechanism for handling a downloaded object — the browser cache — wrote to non-volatile disk. So a limitation requiring the trigger to stay in RAM and not be copied to disk is, on its face, a departure from the ordinary HTTP/caching behavior of the day, and a defendant cannot simply say "the browser does it." That is the strongest non-obviousness argument available on this record.

Three reasons it is weaker than it looks:

1. The specification's stated reason for the "catcher" is data placement, not security. The '825 spec says the catcher "will direct the incoming key to a random location in RAM … and that the key will only be visible to the program," and explains the catcher is necessary because "it is the 'nature' of personal computers to randomly dump data which has not had a specific destination assigned to it. Thus, without the catcher, the incoming data may be strewn into a different directory and/or sub-directories, to, thus, be irretrievably lost." That is a RAM-disk/data-integrity rationale, not an anti-copying rationale. The security purpose the claim attaches to (f)(3) is not the purpose the specification assigns to the structure. That is a § 112(a) enablement/written-description exposure as much as a § 103 one.
2. The patent elsewhere expressly permits permanent non-volatile storage of the very same trigger. Claim 7 ("allowing, by server-permission only, the end-user the ability to store said trigger on non-volatile media for permanent ownership"), claim 16 ("permanently storing the incoming uncrippling trigger data in ROM"), and the FIG. 8 discussion ("the storing of the key 4 to non-volatile media 13, such as a hard drive, for permanent ownership") all contradict a literal "must never be copied to non-volatile memory" reading. Under In re Peters / Bell Communications Research v. Vitalink, this makes the "preventing" language vulnerable to a construction requiring only that the program-controlled assembly step happen in RAM — i.e., that the software not do it automatically, which is a much thinner limitation.
3. The sibling patents use "cache directory" language, where the catcher "will direct the incoming key, such as the header, to a random location in RAM 8 such as cache directory." A cache is, definitionally, a structure whose contents may be written to disk. That drafting undermines a hard anti-persistence reading of the '825 claim.

Net: (f)(3) is a real but soft limitation. If I were attacking claim 1, I would couple the primary-art combination with a secondary reference teaching volatile-memory-only key handling (a well-trodden security practice for key material — key material is conventionally zeroed from memory and never paged to disk), and pair it with a § 112 rejection on the "preventing" language.

B. Claim 17's encryption limitation — a § 112(a) problem masquerading as a § 103 problem

Claim 17 requires that the "crippled data files of said memory means being encrypted," and requires the end-user computer to include "means for decrypting … said encrypted crippled data files of said storage device." Note the asymmetry already flagged in the prior section of this analysis: claim 1 encrypts the trigger; claim 17 encrypts the files.

The '825 specification describes encryption only of the key/header. The FIG. 6 flow ends at encoding the "Header" trigger with RSA; the FIG. 5/6 discussion says the CD-ROM binary data is "crippled," not "encrypted." The Abstract says "the storage of encrypted keys on a remote server, the storage of crippled chunky media data on CD-ROM's" — encryption is attached to the keys, crippling to the media. The title's phrase "encrypted data on local media" is not corroborated by the body in which encryption is applied to the media files themselves.

So: claim 17's media-encryption limitation appears to lack written description / enablement in the '825 specification. That is a § 112(a) defect, not a § 103 defect, and it is potentially cleaner than an obviousness attack on that claim. (The later sibling WO 98/24037 does claim encrypted video and an encrypted player downloaded from a website — but a sibling's disclosure does not cure a defect in the earlier patent's own specification.) I flag this as my reading of the two texts; no tribunal has so held.

---

VII. Motivation-to-combine summary (consolidated)

No teaching away is present in the record I reviewed. The specification's arguments against E-mail/MIME transmission of video are arguments against a different solution, not against the art it claims to improve on.

---

VIII. Bottom line

1. On the record available to me, the '825 claims are highly vulnerable under § 103, and several dependent claims (4, 5, 11, 14, 15, 19) are vulnerable on the applicant's own admissions with no combination needed at all.
2. The only element that meaningfully resists is claim 1(f)(3) / claim 17's "preventing the copying … to nonvolatile memory," and even that is undercut by the '825 specification itself, which (a) gives the catcher a data-integrity rationale rather than a security rationale and (b) expressly permits permanent non-volatile storage of the same trigger in claims 7 and 16.
3. Claim 17 has what I regard as a stronger, non-§ 103 defect: the claim recites encryption of the crippled media files, which the '825 specification does not describe (it describes encryption of the key/header and crippling of the media). Treat this as § 112(a), and verify against the printed specification columns before relying on it.
4. Practical context, stated plainly: all of this is retrospective. The page records "1996-05-15 Anticipated expiration" and legal status "Expired - Lifetime," confirmed by the RPX/Insight record ("Status: Expired due to Term"). An expired claim supports no injunction; this § 103 analysis has value for (i) defensive use of the '825 disclosure as prior art (it is already functioning that way — the Office and a PTAB petitioner used it as § 102(e) art against US 6,101,534), (ii) evaluating the family — US 5,937,164; US 6,185,306; US 6,035,329; US 6,463,467; WO 98/24037; EP 0 974 217; CA 2271012; AU 52598/98 — and (iii) any historical validity dispute.

---

IX. What I could not verify (do not treat these as negative findings)

1. The full 44-item examiner citation list. The page text as supplied truncates after US 4,305,098 A. The post-1981 references — the ones that would map to CD-ROM, network key delivery, and encrypted media — are not in front of me. Any competent § 103 chart for this patent must start from that list.
2. Halpern US 5,319,705; Hellman US 4,658,093; Shear US 4,827,508; the Batchelor patent number. My searches for these did not return results before I hit the tool limit. I have therefore not asserted their numbers, dates, or contents anywhere in this analysis. If any of them is in the '825 citation list, it should be added to Combinations 1–2.
3. Reisman and the Disc Mfg. NPL texts. I have citation strings only. I have not read the articles and cannot quote them. Their source (Unified Patents portal) is second-hand relative to the page's own Prior Art section.
4. Outcome of IPR2016-00472 / IPR2015-01364. I retrieved petition exhibits only. I did not verify institution, any final written decision, or any appeal. The petitioner's motivations quoted above are party contentions about US 6,101,534, not holdings, and not directed at '825.
5. Any date-sensitive analysis. The operating date (2026-09-30) and the task date (April 26, 2026) conflict; I have not reconciled them, and any § 286 arithmetic should be recomputed against whichever is correct.