US 10771849

I'll search for the patent and any related 2026 docket activity.

Let me pull more detail on the claims and the 2026 litigation posture.

Let me confirm the full claim set (independent claims) for the '849 patent.

Summary of US 10,771,849

Note on scope: I searched for the exact identifier 10771849. Several results in the hit lists are unrelated noise (e.g., US 2,664,593 "Molding apparatus," whose 1950s-era application numbers happen to include "107718" / "10771849"). Those do not correspond to the patent below and I have excluded them.

Bibliographic data (per the patent document and Google Patents)

Because the 2018 filing claims 2005 priority, the term ran from the 2005 date and the patent expired 2025‑04‑26 — i.e., roughly eight months after the infringement suit below was filed.

Abstract (as printed)

"A method for multimedia playback and transmission to wireless clients is described. A host webserver transcodes a live digital or analog audio-visual or audio broadcast signal and splits the input stream into small multimedia objects of an efficient compression such as MPEG4/AAC, and then immediately deploys the objects to distributed content servers for a geographically dispersed population of wireless clients. A java applet object player, downloaded to wireless clients at the beginning of the multimedia on-demand session, interprets and decodes the multimedia objects as they are received, using multiple levels of optimization. The applet uses novel video and audio decoding optimizations which can be generically applied to many digital video and audio codecs, and specifically decodes Simple Profile MPEG4 video and Low Complexity AAC audio."

Plain-language overview of the independent claims

The specification's own focus (transcoding, object windowing, Huffman-codebook decoding, IMDCT optimization, quality-scaled "shortcuts") is largely not captured in the granted independent claims. The claims are materially broader than the disclosure:

Claim 1 — method (the claim the court treated as representative):

1. Receive audio and video segments encoded in a digital encoding format with an encoding rate, where those segments are associated with (a) object parameters and (b) a "supplied host path identification" so as to form "multimedia objects."
2. A multimedia player requests transmission of those objects; the objects are located using HTTP and received from servers over a wireless connection.
3. The player plays the objects back in a sequence that maintains "fluidity, video quality and audio quality" by (i) selecting a plurality of objects that reflect available network bandwidth, (ii) autonomously adjusting that selection and playback according to the object parameters and host path identification, and (iii) using "optimized decoding processes" to maintain quality playback.

In plain terms: client-driven adaptive streaming — the player, not the server, decides how to pick and pace object delivery based on bandwidth, using HTTP-hosted segments and metadata.

Claim 6 — apparatus ("a multimedia player"): the same subject matter as claim 1, recast as computer-executable instructions on a non-transitory computer-readable medium executed by a processor (request transmission → receive encoded A/V segments associated with object parameters and supplied host/path identification → objects located using HTTP over wireless → playback maintaining fluidity/video/audio quality via bandwidth-reflective selection, autonomous adjustment, and optimized decoding).

Possible third independent claim — uncertainty flagged: In the N.D. Cal. joint claim construction statement, the disputed-term groups for "fluidity, video quality and audio quality are maintained" and "maintain quality playback" / "optimized decoding processes" are attributed to claims 1, 6 and 7, and "optimal / optimal digital format / optimal encoding rate" to claims 3 and 8. That pattern is consistent with either (i) claim 7 being a further independent claim in the same family (with claim 8 depending from it), or (ii) claim 7 merely depending from claim 6 and inheriting the language. I could not verify the full claim set of the '849 patent (total claim count, and whether claim 7 is independent) from an authoritative source in this session — treat the claim-7 characterization as unconfirmed. The verified independent claims are 1 and 6; claims 2–5 are directly verified as dependent on claim 1 (claim 2 = directory-serving protocols; claim 3 = live-stream transcoding to optimal digital format/rate; claim 4 = I-frame splitting with separate audio segment; claim 5 = splitting multiple MPEG4 composite layers at I-frames).

Litigation and PTAB posture as of 2026‑09‑26

• District court: Withrow Networks, Inc. v. Google LLC and YouTube, LLC, N.D. Cal., Case No. 5:24‑cv‑03203‑PCP (VKD) (also docketed as 3:24‑cv‑03203), complaint filed 2024‑05‑28, before Judge P. Casey Pitts / Mag. Judge Virginia K. DeMarchi. The sole asserted patent is the '849 patent. Withrow characterizes the invention as the origin of adaptive multi-bitrate (ABR) streaming via stateless HTTP "multimedia objects" rather than stateful media servers. Google/YouTube's motion to dismiss (Dkt. 35) was denied on 2025‑01‑21; the § 101 analysis distinguished the claims from Hawk Technology Systems v. Castle Retail and Helios Streaming v. Warner Bros. Discovery on the ground that claim 1 recites a specific method (autonomous adjustment of object selection/playback per object parameters and host path identification), not merely a result-oriented generality. Claim construction briefing and a Markman hearing were set for 2025‑10‑28, with the parties identifying as potentially dispositive: "fluidity," "maintain quality playback," "optimized decoding processes," "optimal"/"optimal digital format"/"optimal encoding rate" (indefiniteness defenses). Earlier in 2025 the court addressed discovery/contentions disputes over Transcoder API, Live Stream API, and (unaccused) Google TV.
• PTAB: IPR2025‑00775, Google LLC v. Withrow Networks Inc. (P.T.A.B.) — challenged patent US 10,771,849. Docket entries show a Director discretionary decision ("refer") on 2025‑08‑14, a Board decision granting institution on 2025‑09‑29, Patent Owner's request for Director Review on 2025‑10‑10, and a Board order denying Director Review of the institution decision on 2025‑12‑15. Google Patents lists the case as "Pending – Instituted." (Note: the Google Patents page's litigation sidebar labels IPR2025‑00775's petitioner generically as "Unified Patents," while the PTAB docket identifies the petitioner as Google LLC — I follow the docket.) A final written decision would ordinarily be due about 12 months after institution, i.e., on or about late September 2026; I did not find a reported 2026 final written decision.
• CAFC: I found no Federal Circuit 2026 docket entry for US 10,771,849. To date the dispute has been litigated in N.D. Cal. and the PTAB; any CAFC appeal (from a final written decision or a district-court judgment) would not yet exist on the record I retrieved. I cannot confirm a 2026 CAFC docket for this patent — stating that explicitly rather than inferring one.

Confidence and caveats

• High confidence: title, inventor, assignee chain, application/filing/issue dates, 2005 priority date, expired status/expiration date, abstract text, and the text of independent claims 1 and 6 (verified against the patent document, Justia, and the district court's quotation of claim 1).
• Moderate confidence: litigation and PTAB chronology (from the Unified Patents PTAB portal summary and N.D. Cal. filings); exact current docket items may have advanced since the sources were captured.
• Low confidence / unresolved: total number of claims and whether claim 7 (and possibly further claims) are independent; the ultimate outcome of IPR2025‑00775; any appellate activity.

I'll search for current litigation records for this specific patent number.

Let me search for the latest 2026 developments and any additional cases.

Let me search for additional cases and the current 2026 status of both the district court and IPR proceedings.

Litigation Involving US 10,771,849

Bottom line: Searching the litigation portals and dockets for the exact identifier 10,771,849 (and application 16/181,285), I find one district-court action and one PTAB proceeding — both involving the same parties. I found no other infringement suits (no other defendants), and no Federal Circuit appeal docketed for this patent. Details below, with sources.

Date note (flagging a contradiction): this task states the current date as April 26, 2026, while my session context and the previously generated sections are dated 2026-09-26. The previously generated sections cross-reference court/PTAB events through December 2025 and note the FWD would ordinarily be due ~late September 2026. My most recent verified docket item is the IPR's 2026-01-13 Petitioner Updated Mandatory Notices. Treat the last ~8 months of activity as potentially unrefreshed.

---

1. District court — Withrow Networks, Inc. v. Google LLC and YouTube, LLC

Procedural history / current status (most recent verified events):

• 2024-05-28 — Complaint filed (Dkt. 1). Withrow alleges Google/YouTube infringe via HLS and MPEG-DASH streaming, framing the '849 patent as the origin of client-centric adaptive multi-bitrate (ABR) streaming over stateless HTTP "multimedia objects." Source: https://insight.rpxcorp.com/litigation_documents/15704692
• 2025-01-21 — Motion to dismiss (Dkt. 35) DENIED by Judge Pitts. The court distinguished Hawk Technology Systems v. Castle Retail and Helios Streaming v. Warner Bros. Discovery, holding claim 1 recites a sufficiently specific method (autonomous adjustment of object selection/playback per object parameters and host path identification) to survive Alice step one; denial was without prejudice to renewal after claim construction. Source: https://storage.courtlistener.com/recap/gov.uscourts.cand.[430146](/patent/430146)/gov.uscourts.cand.430146.44.0.pdf
• 2025-03-08 — Withrow serves Patent L.R. 3-1 infringement contentions against five accused instrumentalities (YouTube w/ HLS; YouTube w/ MPEG-DASH; Transcoder API w/ HLS; Transcoder API w/ MPEG-DASH; Live Stream API w/ HLS), asserting claims 1, 2, 3, 4, 6, 7 and 8 (claims 1–4, 8 method; claim 6 apparatus; claim 7 system).
• 2025-06-20 — Joint Claim Construction and Prehearing Statement filed; Markman hearing set for 2025-10-28. No terms agreed; Defendants flagged as potentially dispositive: "fluidity…"/"fluidity, video quality and audio quality are maintained," "maintain quality playback," "optimized decoding processes" (claims 1, 6, 7) and "optimal"/"optimal digital format"/"optimal encoding rate" (claims 3, 8) — i.e., indefiniteness defenses. Source: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1557614](/patent/1557614)/download-documents?artifactId=woVqmhp_jRcYPOySSmegOfPgFdpPSSxCuxhxAcFHTSUptuGkqSlo7ZI
• 2025-08-21 — Motion to strike infringement contentions (Dkt. 80) GRANTED IN PART / DENIED IN PART (Dkt. 100). Court declined to strike but ordered Withrow to amend its contentions for Transcoder API and Live Stream API (and later Google TV) to state its direct-infringement theories and properly map each limitation; amended contentions due 2025-09-05. Source: https://law.justia.com/cases/federal/district-courts/california/candce/5:2024cv03203/[430146/100](https://assignmentcenter.uspto.gov/search/patent/reelFrameDetail?reelFrame=430146-0100)/
• 2025-08-21/22 — Two further discovery orders: (i) permitting Withrow to serve Google TV contentions (Google TV not previously accused) with a joint status report due 2025-09-19 (Dkt. 101); and (ii) denying Withrow's bid to compel 2007–present design-consideration discovery, limiting the damages-relevant window to around September 2020 (date of first alleged infringement) (Dkt. 102). Sources: https://cases.justia.com/federal/district-courts/california/candce/5:2024cv03203/[430146/101](https://assignmentcenter.uspto.gov/search/patent/reelFrameDetail?reelFrame=430146-0101)/0.pdf and https://storage.courtlistener.com/recap/gov.uscourts.cand.430146/gov.uscourts.cand.430146.102.0.pdf

Outcome: Not yet resolved. No judgment, settlement, or dismissal found. As of my last verified filings the case was in fact/claim-construction phase (Markman 2025-10-28), with the outcome of claim construction — particularly the indefiniteness challenges to "fluidity"/"maintain quality playback"/"optimized decoding processes" — not confirmed in my sources.

2. PTAB — Google LLC v. Withrow Networks Inc., IPR2025-00775

Key docket entries (most recent first):

• 2026-01-13 — Petitioner's Updated Mandatory Notices (backup counsel)
• 2025-12-24 — PO Notice of Deposition of Dr. Nathaniel Polish
• 2025-12-15 — Board Order Denying Director Review of Institution Decision
• 2025-11-24 — PO Notice of Deposition
• 2025-11-04 — Patent Owner Power of Attorney
• 2025-10-30 — Petitioner's Notice of Intent to Designate Backup Counsel
• 2025-10-10 — PO Request for Director Review — Institution Decision
• 2025-09-29 — Board Institution Decision: Grant (+ Scheduling Order)
• 2025-08-14 — Director Discretionary Decision: Refer (Paper 10)

Source: https://gaeflexstaging-dot-docketupdate.appspot.com/cases/PTAB/IPR2025-00775/Google_LLC_v._Withrow_Networks_Inc/

Outcome: No final written decision found. Institution was granted (12-month statutory FWD deadline ≈ 2026-09-29), and Director Review of institution was denied 2025-12-15. Google Patents lists the case as "Pending – Instituted."

3. Federal Circuit — none found

I found no CAFC docket for US 10,771,849. No appeal from the § 101 denial (interlocutory orders are not appealable as of right) or from any final judgment exists on my record. Any appeal from an IPR FWD would arise only after the FWD issues.

---

Caveats and notes

1. Only one defendant group. Every hit for this patent traces to Withrow v. Google/YouTube (N.D. Cal.) and IPR2025-00775. I found no suits against other streaming defendants (e.g., Amazon, Netflix, Apple), and the Darts-IP "first worldwide family litigation" flag on the Google Patents page is consistent with this being the earliest-known litigation on the family.

2. Petitioner discrepancy (repeating an earlier flag). The Google Patents litigation sidebar generically labels IPR2025-00775's petitioner as "Unified Patents," while the PTAB filings identify Google LLC as petitioner; I follow the docket. Source: https://patents.google.com/patent/US10771849/en

3. Case-number duplication is not two cases. Google Patents lists both 3:24-cv-03203 and 5:24-cv-03203; these are the same N.D. Cal. action (re-docketed/assigned under the San Jose division number).

4. Expired patent. The '849 patent's term ran from the 2005-04-18 priority date and it shows "Expired – Lifetime; adjusted expiration 2025-04-26." Any surviving litigation (and the IPR) therefore concerns an already-expired patent — relevant to available damages, but it does not moot the IPR.

5. What I could not verify: the outcome of the 2025-10-28 Markman hearing; whether any post-September-2025 trial schedule was set; whether IPR2025-00775 reached a final written decision; the total claim count (note the record shows claims at least through claim 8, with claim 7 characterized as a "system claim," which resolves the earlier open question in favor of claim 7 being independent of claim 6); and any settlement. I state these as unresolved rather than inferring them.

I'll research the PTAB record for US 10,771,849, reconciling the structured ODP block with what web sources show.

I've confirmed IPR2025‑00775 exists. Let me dig into the panel, grounds, claim scope challenged, and any 2026 developments.

⚠️ Source conflict flagged up front

The structured block in my prompt states the USPTO ODP API returns no AIA trial proceedings for US 10,771,849. That is contradicted by the public record I retrieved: IPR2025‑00775 was filed on 2025‑03‑24, instituted on 2025‑09‑29, and is still active as of today (2026‑09‑26). I treat the ODP "no proceedings" result as a stale/not-yet-ingested feed, not as evidence of absence. Everything below is grounded in the PTAB petition papers on PTACTS, the public docket for IPR2025‑00775, and the N.D. Cal. filings. I could not rule out that a second petition exists that neither source surfaced; no second proceeding appeared in any search.

Note also that the Google Patents litigation sidebar lists the IPR2025‑00775 petitioner as "Unified Patents." The PTAB papers themselves caption the petitioner as Google LLC — I follow the papers. There is no defensive aggregator in this chain on the record I retrieved.

---

Proceedings overview

One AIA trial proceeding is on file — IPR2025‑00775 (Google LLC v. Withrow Networks Inc.), status: instituted and active, no Final Written Decision yet; zero claims canceled, zero claims adjudicated patentable, no settlement, no appeal. For a defendant facing assertion of the '849 patent today, the defensive posture is "pending, not yet hardened and not yet broken": all nine claims are in the crucible, the Board has found a reasonable likelihood of unpatentability on a Carmel-based obviousness theory, but nothing is final and the FWD is due within days (on or about 2026‑09‑29) — which makes this a watch item, not a settled question.

---

IPR2025‑00775 — Google LLC v. Withrow Networks Inc.

• Type: Inter Partes Review (35 U.S.C. §§ 311–318)
• Filed: 2025‑03‑24 (Petition). Google represented that the filing was "more than two months before the statutory deadline, less than three weeks after PO served its infringement contentions." Google LLC and YouTube, LLC are identified real parties‑in‑interest.
• Status: Instituted – Grant (verbatim: "Board Institution Decision: Grant — Decision Granting Institution of Inter Partes Review," Paper 12, 2025‑09‑29). Gloss: trial is ongoing; no FWD yet.
• Judge panel: Not confirmed. The panel composition was not disclosed in any source I retrieved. I am not going to guess APJ names.
• Petition grounds: Obviousness over Carmel in view of Mattavelli or Wang. Carmel appears as Exhibits 1004/Carmel family members — Ex. 1009 (U.S. Pat. No. 5,841,432 to Sharon Carmel et al.) and Ex. 1010 (U.S. Pat. No. 6,397,230 to Sharon Carmel et al.); the petition opposition brief also references U.S. Pat. No. 6,389,473 to Sharon Carmel et al. Mattavelli (Ex. 1004) and Wang (Ex. 1007) supply the "optimized decoding processes" / computational graceful degradation teaching. Google's papers state the Petition "presents prior art disclosure corresponding to each and every element of the nine challenged claims" — i.e., claims 1–9 of the '849 patent, the full claim set (the patent has nine claims per the patent document's claim listing). Statutory basis is § 103 (obviousness). I have not read the institution decision itself, so I cannot state claim‑by‑claim or ground‑by‑ground what was instituted versus denied; I can only confirm institution was granted on 2025‑09‑29.
• Institution decision: Instituted 2025‑09‑29 (Paper 12), with a Scheduling Order the same day (Paper 13). Pathway to institution:
  • 2025‑04‑25 — Patent Owner's Request for Discretionary Denial (PO Exs. 2001–2032, including the denial of Google's motion to dismiss and the district court stay papers).
  • 2025‑07‑17 — Patent Owner Preliminary Response.
  • 2025‑08‑14 — Board/Director Discretionary Decision: Refer.
  • 2025‑09‑16 — Board Exhibit 3001.
  • 2025‑09‑29 — Institution granted.
  • 2025‑10‑10 — Patent Owner Request for Director Review of the institution decision.
  • 2025‑12‑15 — Board Order Denying Director Review of the Institution Decision (Paper 23). This ended the discretionary‑denial front: the institution stands.
  • Panel reasoning, as reflected in Google's opposition papers (which argue under Fintiv/Twitch/Imperative and Advanced Bionics/Becton, Dickinson): the district court had denied a stay without prejudice and "welcome[d]" a renewed motion post‑institution; PO is not a competitor of Google/YouTube; Google filed a broad Sotera stipulation; and the Examiner materially erred by allowing the claims without ever rejecting over Carmel's multi‑level streaming teachings, which the petition maps onto the "reason for allowance" limitation (client player selecting/adjusting quality level based on available bandwidth per object parameters and host path identification).
• Final Written Decision: Not issued as of 2026‑09‑26. Statutory § 316(a)(11) deadline runs one year from institution = on or about 2026‑09‑29. Any claim‑level verdict (canceled / sustained) is therefore not yet available, and I will not speculate about it. No claim of the '849 patent has been canceled or held patentable by the Board.
• Settlement / termination: None. The proceeding is live; the record shows active merits work into 2026 — Patent Owner notice of deposition of Petitioner's expert Dr. Nathaniel Polish (2025‑12‑24, Paper 25) and Petitioner's Updated Mandatory Notices re backup counsel (2026‑01‑13, Paper 26), with counsel Reuben Chen designated as backup on 2025‑10‑30.
• Appeal: None. No Federal Circuit docket for IPR2025‑00775 or for the '849 patent was found. An appeal is not ripe because there is no FWD.
• Defensive value: A defendant gets a very recent, live Board finding of reasonable likelihood that all nine claims are obvious over Carmel + Mattavelli/Wang — the same Carmel family the examiner saw only in an IDS and never used in a rejection. That is a strong § 103 roadmap and effectively a ready‑made second‑petition template. But because no FWD has issued, the patent is not narrowed, and no estoppel attaches yet.

---

Strategic summary

Claim status. No claim of US 10,771,849 is canceled, and no claim has been adjudicated patentable. The Board instituted on a petition challenging all nine claims (1–9) — independent method claim 1, independent apparatus claim 6, independent system claim 7, and dependents 2–5, 8–9 — so effectively the entire patent is under challenge and nothing is yet "tested" to a final judgment. (This resolves an open item from the earlier sections: Espacenet's claims listing confirms the patent has nine claims with three independent claims — 1 (method), 6 (apparatus/player), and 7 (system) — and claim 8 depends from claim 1 while claim 9 also depends from claim 1; the earlier "claim 7 independent?" uncertainty is now confirmed as yes, independent.) Litigation‑wise the asserted claims are 1, 2, 3, 4, 6, 7 and 8 (claims 5 and 9 unasserted), so a cancellation of claims 1–9 would zero out the entire case.

Estoppel landscape. No § 315(e)(2) estoppel exists yet for anyone, because estoppel attaches only after a final written decision. Google, however, is contractually self‑bound: its Sotera stipulation commits that if IPR2025‑00775 is instituted (it was, and Director Review of institution was denied), Google/YouTube "will not pursue in this litigation as to the claims challenged in IPR2025‑00775: (i) the specific grounds raised in IPR2025‑00775 or (ii) any other grounds that reasonably could have been raised before the PTAB in that instituted proceeding." That is broader than statutory estoppel and will constrain Google's district‑court invalidity case regardless of the FWD outcome. For any other defendant being asserted (or threatened) on the '849 patent: you are not a privy of Google, so you retain the full § 102/§ 103 prior‑art field, including Carmel‑based combinations — but watch the § 315(b) one‑year bar from service of an infringement complaint. Note the patent is expired (2025‑04‑26), which does not bar IPR but limits remedies to pre‑expiration damages.

Pattern signals. Sole petitioner: Google (with YouTube as RPI); single petition, no joinder, no follow‑on IPRs (as of this writing). Patent Owner fought institution aggressively (discretionary‑denial request, POPR, then a Request for Director Review of the institution decision) and lost that fight on 2025‑12‑15, opting against settlement and pressing into the merits through at least early 2026. There is no defensive aggregator in the chain despite the Google Patents sidebar's "Unified Patents" label. The parallel N.D. Cal. case (Withrow Networks, Inc. v. Google LLC et al., No. 5:24‑cv‑03203‑PCP (VKD)) survived a § 101 motion to dismiss on 2025‑01‑21 and was headed to a Markman hearing on the indefiniteness terms "fluidity," "maintain quality playback," and "optimized decoding processes."

---

Recommended next steps

1. Mark your calendar to the day. The FWD in IPR2025‑00775 is due on or about 2026‑09‑29 (one year from the 2025‑09‑29 institution). If you are advising a defendant now, the single highest‑value action is to pull the FWD the moment it issues from PTAB E2E / PTAB Decisions and re‑run the claim status table:
   • All claims canceled → the '849 patent is dead and any demand or suit citing it collapses; cite the FWD's disposition verbatim.
   • Some claims canceled, some sustained → map the surviving claims against your accused product; the sustained set (often the independent system/apparatus claims) becomes the real battleground.
   • Claims sustained → an IPR‑based defense gets materially harder; pivot to the district‑court § 101 and indefiniteness theories the N.D. Cal. case already flagged.
2. Do not rely on estoppel yet. No § 315(e)(2) estoppel has arisen for anyone, and Google's Sotera stipulation binds only Google/YouTube. If you are a different defendant, the Carmel + Mattavelli/Wang combination (and the examiner‑error narrative that beat § 325(d)) remains open to you — but confirm your § 315(b) status first.
3. Track the district court, since a post‑institution stay is very much in play: the N.D. Cal. court denied Google's stay motion only without prejudice and expressly invited a renewed motion if IPR was instituted (it was). The claim‑construction track (Markman set for 2025‑10‑28) and the § 101 ruling (Dkt. 44, 2025‑01‑21) are the other live threads.
4. Verify the panel and the instituted scope directly. Two facts I could not confirm from retrieved sources and that matter to you: (a) the APJ panel hearing IPR2025‑00775, and (b) whether institution covered all nine claims and both Carmel‑based grounds or a subset. Pull Paper 12 (Decision Granting Institution, 2025‑09‑29) from PTAB E2E and the case's PTAB portal page at portal.unifiedpatents.com/ptab/case/IPR2025-00775; the underlying petition papers I relied on are on PTACTS. For the parallel litigation, the § 101 ruling is at CourtListener.
5. Watch for a Federal Circuit appeal. None exists today; if the FWD goes against Patent Owner, expect a Withrow notice of appeal within 63 days, and re‑check the CAFC docket at that point.

Confidence statement. High confidence: existence, number, partnership, filing and institution dates, the Director Review denial, the petitioner/RPI identity, and the Carmel‑based § 103 grounds. Moderate confidence: the "nine challenged claims" scope and the Sotera stipulation scope (both sourced from Google's briefs, not the institution decision). Low confidence / unverified: the APJ panel and the precise claim‑by‑claim reach of institution. Unknown: the FWD outcome — no FWD has issued as of 2026‑09‑26.

I'll research the assignment chain for this patent, starting with the USPTO assignment records and cross-references.

Key finding already: the complaint states Withrow Networks was founded in 2020 and traces to dissolved predecessors LogoVision Wireless Inc. and Movidity Inc. Let me dig into the actual assignment records and correspondent.

Let me pin down the recorded assignment details and Withrow's litigation/entity profile.

Note on method: I hit a hard tool-call limit before I could pull the Assignment Center detail page. Everything below is grounded in what I retrieved; I explicitly flag the two fields (reel/frame and correspondent) I could not obtain.

---

Inventors

Unusual pattern — and it is not the one you asked about. The suspect signal here is not "inventors departing within 12 months." It is that the patent never entered the operating company at all. Per the complaint (Withrow Networks, Inc. v. Google, LLC, N.D. Cal. 5:24-cv-03203, Dkt. 1 ¶ 7 n.1): "Withrow was founded in 2020, but its origin traces to LogoVision Wireless Inc. and its technology commercialization subsidiary, Movidity Inc. Those two latter businesses are now dissolved." The 2005 priority application was filed by Krebs as an individual and the two continuation filings were also made in Krebs's own name (Espacenet applicant: Krebs Mark Sinclair). So when LogoVision/Movidity dissolved, the patent was not in their estates — it sat with the founder and was later transferred into a fresh 2020-vintage assertion vehicle.

Chain of filings (Espacenet priority data): US priority number 2005-0107952, filed 2005-04-18 → continuation 15/016,821, filed 2016-02-05 → continuation 16/181,285, filed 2018-11-05 (the application that issued).

---

Original assignee

Mark Sinclair Krebs, an individual (the issued patent's stated assignee; Google Patents renders this as "Individual"). Espacenet confirms applicant(s): Krebs Mark Sinclair [CA].

• Products embodying the claims: Not by the assignee personally. Products existed only at the predecessor operating companies — the Movidity mobile ABR player (demonstrated with ARM/Qualcomm/Sprint in 2006, per complaint ¶¶ 14–17) and the Movy.tv mobile video sharing service (2007). The complaint concedes those businesses are dissolved.
• Primary line of business of the original assignee: individual inventor/technology founder.
• Current status: LogoVision Wireless Inc. and Movidity Inc. — dissolved (per Dkt. 1 ¶ 7 n.1). Krebs himself is the assignor of record; no operating business of his is identified in any filing I retrieved.

Contradiction flagged vs. the earlier section: the prior summary described "Original assignee: Individual" and separately noted the recorded 2021 reassignment. That is consistent with what I found, but the significance is worth restating plainly: because the operating companies never held title, there was no corporate assignment, no security interest, and no bankruptcy transfer in the operating era. The only recorded conveyance is inventor → assertion entity.

Also resolved from the earlier sections: the previously-unconfirmed claim 7 is now verified independent. Espacenet's claim listing shows "7. A system comprising at least one centralized content server transmitting multimedia objects…" So the patent has three independent claims (1 method, 6 apparatus/CRM, 7 system), and the N.D. Cal. joint claim construction groups attributed to "claims 1, 6 and 7" are explained. (New drafting irregularity spotted: claims 8 and 9 read "The method of claim 1" despite being numbered after the claim-7 system claim — an antecedent-basis irregularity worth noting alongside the pending indefiniteness terms.) Total claim count remains unconfirmed — the Espacenet listing truncates after claim 9. Minor conflict to log: Unified Patents' patent page gives the '849 priority date as 2005-04-17, while Google Patents and the patent document say 2005-04-18; the patent document controls.

---

Assignment timeline

I was able to establish the existence, parties, and recording date of one post-issuance assignment, but not the reel/frame numbers or the correspondent of record. I am not going to invent them.

• Execution date not captured / recorded 2021-09-20 — Reel not retrieved (USPTO Assignment Center detail page not reached before tool limit; Google Patents legal-events entry does not publish reel/frame)
  • Conveyance: Assignment ("ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)")
  • Assignor: Krebs, Mark Sinclair
  • Assignee: Withrow Networks Inc
  • Correspondent: not determined. No correspondent data appeared in any indexed source I retrieved. Do not treat "Withrow & Terranova" (a North Carolina IP firm that showed up on an unrelated 2006 reel 017601/0629 assignment for a fuel-systems patent) as related — that is coincidental name noise, not this chain.
  • Context: transfer-to-asserter. Krebs personally assigned the '849 patent to Withrow Networks Inc, an entity the complaint says was formed in 2020, roughly one year before the recording.
• No other recorded assignments found. In particular, no pre-issuance assignment from Krebs to LogoVision/Movidity, no security agreement, no merger/change-of-name, and no release or correction. Consistent with the patent having been prosecuted personally by the inventor and never held by the operating companies.

Verification link: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search by patent number 10771849 or by assignor "Krebs, Mark Sinclair"; the legacy index at https://assignment.uspto.gov/patent/index.html surfaces the same records). The one entry that is corroborated is the 2021-09-20 recordation shown in Google Patents' legal-events listing: https://patents.google.com/patent/[US10771849](/patent/US10771849)/en

---

Timeline diagram

timeline
    title Ownership of US 10771849
    2005 : Priority application filed by Krebs
    2016 : Continuation filed by Krebs
    2018 : Continuation filed by Krebs
    2020 : Patent issues to Krebs as individual
    2020 : Withrow Networks founded
    2021 : Assignment recorded to Withrow Networks
    2024 : Infringement suit filed against Google
    2025 : Patent reaches adjusted expiration
    2025 : IPR2025-00775 instituted

---

NPE / troll-pattern signals

1. Shell-entity transfer — present (qualified). The patent moved from the individual inventor into Withrow Networks Inc, an entity founded in 2020 whose only documented activity is acquiring the '849 patent (recorded 2021-09-20) and suing on it (2024-05-28). Concrete evidence of single purpose: the complaint identifies no products, no revenue, and describes the predecessors as dissolved; Withrow's own narrative places all commercialization (Movidity player, Movy.tv) at companies that no longer exist. Caveat: I did not verify Withrow's state of incorporation, registered-agent address, or member structure, so the "mail-drop address / single-member LLC" component is unconfirmed. Note also that it is an "Inc," not an "LLC," and it is inventor-controlled rather than a third-party-filed shell.

2. Known asserter in the chain — not present (with a note). Withrow Networks does not match any of the listed NPE families (Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation, Spangenberg entities). It is, however, indexed as an asserter in Stanford's NPE Litigation Database, classified "Individual-inventor-started" — https://npe.law.stanford.edu/party/withrow-networks-inc. A one-suit, one-patent plaintiff is not a "high-frequency plaintiff."

3. Repeat correspondent across the chain — unclear. With only one recorded conveyance in the chain there is no recurrence to measure, and I could not retrieve the correspondent name/firm on that single entry. For completeness: Withrow's litigation counsel of record is John V. Picone III, Spencer Fane LLP, San Jose (appearing on Dkt. 78/80 in 5:24-cv-03203). That is litigation counsel, not the assignment correspondent, and a single appearance is not a finding under this signal.

4. Cascading transfers — not present. Exactly one recorded assignment; no chained LLC hops and no shared-address pattern.

5. Pre-litigation transfer — not present. Recorded 2021-09-20; first suit filed 2024-05-28. That is roughly 32 months, well outside the 6-month window. (If the execution date turns out to be materially later than the recording date, this would need re-checking — but the recorded date is 2021-09-20.)

6. Bankruptcy fire-sale — not present. No Chapter 7/11 proceeding surfaced for Krebs, LogoVision, or Movidity. The predecessors are described as "dissolved," but I found no judicial-sale record — and in any event the patent was never an asset of those entities, so no estate could have sold it.

7. Privateering — not present. The transferring party is the inventor himself, not an operating company arming an NPE against competitors. This is closer to direct inventor-assertion than to privateering.

8. Defensive aggregator — not present. The chain terminates at an asserter that has sued Google LLC and YouTube, LLC; it has not been neutralized by RPX, AST, LOT, Unified Patents, or OIN.

---

Verdict

NPE — moderate confidence.

The chain is inventor → the inventor's own 2020-formed assertion vehicle (assignment recorded 2021-09-20, per the Google Patents legal-events entry for US 16/181,285), followed by assertion of the '849 patent against Google LLC and YouTube, LLC on 2024-05-28 (N.D. Cal. 5:24-cv-03203). That is one strong signal (single-purpose assertion-entity transfer, with no products and dissolved predecessor operating companies) plus several weak corroborators: the predecessor commercialization companies LogoVision/Movidity are dissolved, the current assignee holds no product line, and the suit was pressed on a patent whose adjusted expiration was 2025-04-26 — i.e., a late-life assertion. I stop short of "high confidence" because the two facts that most reliably confirm the shell pattern — the reel/frame record and the recording correspondent — could not be retrieved, and I did not confirm Withrow's incorporation state or registered-agent address.

Verify the two open fields here: https://assignmentcenter.uspto.gov/ (patent 10771849; assignor "Krebs, Mark Sinclair")

I'll search for the patent's citation records. Note: the previously-generated summary is timestamped 2026‑09‑26 while this task states 2026‑04‑26 — I'll flag that discrepancy but treat the authoritative patent text as controlling.

I have the citation list. Let me verify each reference's dates and confirm the citation set belongs to US 10,771,849.

Scope, verification, and one correction to the earlier section

Identifier check. I searched USPTO/patent-database sources for the specific number 10771849 and restricted results to US 10,771,849 B2 ("Multimedia system for mobile client platforms," Mark Sinclair Krebs; app. US 16/181,285; filed 2018‑11‑05; granted 2020‑09‑08). Numerically-similar hits that surfaced were excluded: e.g., US 2,664,593 ("Molding apparatus"), whose 1950s application numbers coincidentally contain "107718"/"10771849," and US 11,471,863 B2, US 10,348,040 B2, US 9,390,550 B2, US 5,414,455 — none of which are the '849 patent.

Date discrepancy (flagged as instructed). The previously-generated summary is timestamped 2026‑09‑26; this task states 2026‑04‑26. I cannot reconcile the two from the sources retrieved, so the prior-art analysis below is date-independent (it rests on 1999–2005 priority dates, not on today's date).

Correction to the earlier "unresolved" item — claim 7 is independent. The earlier summary flagged that it "could not verify... whether claim 7 is independent." The Espacenet claim set for US10771849 B2 resolves this. Claim 7 opens "A system comprising at least one centralized content server transmitting multimedia objects; said multimedia objects including segmented audio and video, and parameters and supplied host path identification; at least one multimedia player configured to: request transmission of multimedia objects..." It is therefore a third independent claim. Claims 8 and 9 both read "The method of claim 1..." (dependent on claim 1). So the verified claim set is: independent 1 (method), 6 (player/apparatus), 7 (system); dependent 2–5 and 8–9. This upgrades the previous "low confidence / unresolved" note to confirmed, and it matters for prior-art mapping because claim 7 is the broadest claim in some respects (no "optimized decoding processes" element, and the server is recited only generically).

---

Source and confidence of the citation list

The citation set below is drawn from the Google Patents family record for the '849 patent chain (continuity shown as US 11/107,952 → US 15/016,821 (US 10,171,873 B2) → US 16/181,285 (US 10,771,849 B2)), which lists a "Patent Citations (4)" table. Each row's first date column is the priority date, the second is the publication/grant date:

Duplicate filing dates confirmed independently: US 6,490,320 filed 2000‑02‑02 (app. 09/496,706); US 7,295,608 filed 2002‑09‑26 (app. 10/256,866; provisional 60/325,483 filed 2001‑09‑26); Neomtel's PCT/KR00/01511 published as WO 01/50766 A1.

Caveat (stated explicitly): I could not load the USPTO PatentCenter/Image File Wrapper "References Cited" page for 16/181,285 directly in this session, and the citation table surfaced via a Google Patents page whose URL was associated with a sibling publication. Treat the four-reference list as moderate confidence for the '849 patent specifically; it is at minimum the citation set for the immediately-continuing family member. The list appears to be a US‑only set — no foreign patent documents and no non‑patent literature were in the retrieved citation table.

---

Reference-by-reference analysis

1. US 7,120,194 B2 — Neomtel Co. Ltd.

• Full citation: U.S. Patent No. 7,120,194 B2, "System for moving image data using wireless communication and the method of the same," Neomtel Co. Ltd. (KR). Priority 1999‑12‑22; US national-phase/grant 2006‑10‑10. PCT counterpart WO 01/50766 A1 (PCT/KR00/01511).
• Disclosure: Transmits encoded moving-picture data to mobile stations; mobile station includes receiving means, decoding means, and display means. Data is organized into frames with a "basic picture type" and a "transition picture type" (i.e., I-frame/P-frame analogues), plus a frame header and a moving-picture identifier; the frame format carries per-frame timing (fixed vs. variable frame duration) and file-size/frame-count fields for a mobile client.
• §102 exposure: Qualifies under pre‑AIA §102(b) via the WO 01/50766 A1 printed publication (mid‑2001, more than one year before the 2005‑04‑18 priority) and under §102(a); a §102(e) theory would require the US filing date (PCT national-stage) to predate 2005‑04‑18, which is plausible but not verified here.
• Claims potentially affected: Best candidate for claim 4 ("a multimedia file is scanned for an I-frame, and a multimedia segment is split at the I-frame to create the video segment, together with a separate audio segment") on its basic/transition picture-type teaching — but Neomtel splits by picture type encoding, not by scanning for the next I-frame and separating an audio segment, so even this is weak. It touches claim 1(e)'s "optimized decoding processes" only generically (a mobile decoding means). It does not disclose HTTP object retrieval, object parameters/host path identification, or bandwidth-reflective autonomous selection → no anticipation of claims 1, 6, or 7.

2. US 6,490,320 B1 — Mitsubishi Electric Research Laboratories (Vetro et al.)

• Full citation: U.S. Patent No. 6,490,320 B1, "Adaptable bitstream video delivery system," Vetro, Divakaran, Sun; Mitsubishi Electric Research Laboratories, Inc. Filed 2000‑02‑02 (app. 09/496,706); granted 2002‑12‑03.
• Disclosure: A multimedia delivery system with a transcoder and a content-network-device (CND) manager that selects among multiple conversion modes based on content semantics, network characteristics (network congestion, available bit rate (ABR), packet loss), and user-device characteristics (processing power, memory, display constraints); generates optimal rate-quality functions for a given available bit rate of the network. Related sibling: US 6,542,546 ("Adaptable compressed bitstream transcoder," same assignee).
• §102 exposure: Qualifies under pre‑AIA §102(b) (granted 2002‑12‑03, more than one year before 2005‑04‑18) and §102(a).
• Claims potentially affected: The single most relevant reference on the bandwidth-reflection element. It maps onto claim 1(e1) ("selecting a plurality of said multimedia objects that reflect available network bandwidth") and onto claim 3 ("transcoded into an optimal digital format at an optimal encoding rate reflecting available network bandwidth"). But the selection is made by a server-side CND manager, not by a client "autonomously adjusting said selection and playback," and it is not HTTP-object-based. So it is strong §103 art (especially against claims 3 and 8) but does not anticipate claim 1, 6, or 7. Its disclosure of "user device characteristics" also helps a §103 attack on the "optimized decoding processes" element if combined with an MPEG‑4 decoder reference.

3. US 7,295,608 B2 — Reynolds et al.

• Full citation: U.S. Patent No. 7,295,608 B2, "System and method for communicating media signals," Reynolds, Jodie Lynn (and Ingraham); app. 10/256,866 filed 2002‑09‑26, provisional 60/325,483 filed 2001‑09‑26; granted 2007‑11‑13. PCT counterpart WO 03/027876 A1 (published 2003‑04‑03); pre-grant pub. US 2004/0045030 A1.
• Disclosure: Streaming media system that takes a library of separate, distinct CODECs (a searchable CODEC library) and uses a computer-implemented intelligence mechanism to learn/capture characteristics of the signal as it is streamed; compresses and decompresses media between source and destination devices. Describes streaming as "a series of discrete packets, pieces, or blocks that are interrelated and may be thereafter used... in a continuous, interrelated fashion."
• §102 exposure: Qualifies under pre‑AIA §102(b) via the WO 03/027876 A1 publication (2003‑04‑03, more than one year before 2005‑04‑18) and §102(e) via the US patent (app. filed 2002‑09‑26, before 2005‑04‑18).
• Claims potentially affected: Provides background for the segmentation/"multimedia object sequence" concept and for CODEC selection (relevant to claim 1(e3) / "optimized decoding processes"), but the "intelligence" is described at the streaming server/CODEC-management level, not as a client that locates HTTP objects via host-path identification. No anticipation of any independent claim; marginal §103 value, mostly cumulative to Mitsubishi.

4. US 2004/0194144 A1 — Chi‑Tai Lin

• Full citation: U.S. Patent Application Publication No. 2004/0194144 A1, "Method for asynchronously watching programs from the internet and the system thereof," Chi‑Tai Lin. Priority/filing 2003‑03‑25; published 2004‑09‑30.
• Disclosure: Client-side method/system for asynchronously viewing Internet programs (i.e., time-shifted/on-demand retrieval of program material by a client rather than a live linear feed).
• §102 exposure: Does not qualify under §102(b) — published 2004‑09‑30, less than one year before the 2005‑04‑18 priority. It is §102(a)/(e) art only if the reference's 2003‑03‑25 US filing date is used (§102(e) for a published application filed before the applicant's invention/filing).
• Claims potentially affected: Weak. Its subject matter (asynchronous/on-demand program retrieval) is a broad backdrop for claim 1(c)–(d), but it does not disclose HTTP object location via supplied host/path identification, object parameters, or bandwidth-responsive client selection. No anticipation; low §103 value.

---

Bottom line on anticipation

None of the four citations of record anticipates claims 1, 6, or 7 under §102. Each independent claim requires the conjunction of (i) A/V segments carrying object parameters and a supplied host path identification forming "multimedia objects," (ii) the player requesting transmission with objects located using HTTP and delivered over a wireless connection, and (iii) client-autonomous adjustment of selection and playback — a combination of which none of the four references discloses. The realistic posture is:

• §103, strongest pairing: US 6,490,320 (adaptive, bandwidth-reflective transcoding/delivery to constrained devices) in view of US 7,120,194 (mobile-station decode/display of segmented picture-type video) — aimed at claims 3, 4, 5, 8.
• §102 proximity, closest single reference: US 6,490,320 against claim 3 alone (but it lacks "optimal digital format" splitting of a live stream into intervals for wireless-client access as claimed).

---

The prior-art issue that likely dominates (priority-date risk)

This is more consequential than the four citations. The '849 patent claims a 2005‑04‑18 priority, but its granted claims are broader than the 2005 disclosure — the earlier summary correctly notes the specification's focus is transcoding, object windowing, Huffman-codebook decoding, IMDCT optimization, and quality-scaled decoding shortcuts, whereas claim 1 recites client-driven, bandwidth-reflective selection among HTTP-hosted objects with autonomous adjustment. If the claims are not entitled to 2005‑04‑18 (no §112 written-description support in the 11/107,952 specification), the effective date shifts to 2016‑02‑05 and/or 2018‑11‑05, and the following become available as prior art under AIA §102(a)(1)/(a)(2):

• Apple HTTP Live Streaming (HLS) — 2009 Internet‑Draft;
• Microsoft Smooth Streaming / "HTTP Smooth Streaming" — 2009–2010 (cited as prior art in the DISH patent family, e.g., in US 8,868,772/US 9,459,601 prosecution);
• 3GPP Adaptive HTTP Streaming / MPEG DASH (ISO/IEC 23009‑1) — 2009–2012 (Qualcomm "Use Cases and Examples for Adaptive httpstreaming," 2010; Stockhammer "WD 0.1 of 23001‑6 Dynamic Adaptive Streaming over HTTP," 2010);
• DISH multi-bitrate streaming family — US 7,818,444 → US 8,402,156 → US 8,612,624 → US 9,071,668 → US 9,571,551 → US 10,659,513 → US 11,470,138 → US 11,677,798 (all claiming a 2004‑04‑30 provisional), a line of art expressly directed to "apparatus, system, and method for multi‑bitrate content streaming."

Any §102 anticipation of claim 1 in this patent is far more likely to come from that post‑2005 HTTP adaptive-streaming corpus (asserted on the theory that the claims get only the 2016/2018 dates) than from the four references cited during prosecution. That is also the natural framing for IPR2025‑00775, whose institution the earlier section reports.

---

Explicit limitations of this analysis

• I did not retrieve the USPTO PatentCenter "References Cited" list for application 16/181,285 directly. The four-reference table above comes from the Google Patents family record and is moderate-confidence; a manual PatentCenter/IFW check (and a review of the parent application 15/016,821 and grandparent 11/107,952 file wrappers, where the same references likely originated) is needed to certify it.
• No non-patent-literature citations and no foreign patent-document citations appeared in the retrieved table; if the examiner cited NPL, it is not reflected here. I will not invent it.
• Two references the specification itself acknowledges as prior art are not in the citation table but are relevant to the background of the asserted claims and may appear in the IDS: Waese et al., U.S. Pat. No. 6,286,031 (transmission of video as objects through pre-fetched lists) and Stumm, U.S. Pat. No. 5,768,528 (instant/scheduled notification file downloading), plus U.S. Pat. No. 5,699,121 (pattern-marching residual correction, discussed as a non‑MPEG‑4 alternative). I could not verify their presence on the '849 IDS.
• Filing dates for the PCT national-stage entry of Neomtel's US 7,120,194 were not verified, so its pre‑AIA §102(e) status is asserted conditionally.

I'll gather the prior-art record for this patent — both the specification's cited references and the art asserted in the instituted IPR.

Good — I found the key IPR identifier (petition 1557614) and a mention of the asserted grounds. Let me pin down the actual references.

Now let me identify the two secondary references (Mattavelli and Wang) precisely.

Obviousness Analysis — U.S. Pat. No. 10,771,849 B2

"Multimedia System for Mobile Client Platforms" (Krebs; Withrow Networks Inc.)

Analysis date: April 26, 2026
Statute: 35 U.S.C. § 103(a) (pre-AIA), for the reasons in §1.3 below

---

1. Threshold matters, including two record conflicts I am flagging

1.1 Resolution of an open question from the prior section

The previously generated summary flagged as unconfirmed whether claim 7 is independent and what the total claim count is. The Espacenet claim set for this patent resolves it:

Nine claims total; three independent (1, 6, 7). This is consistent with the parties' Joint Claim Construction and Prehearing Statement, which groups the "fluidity / maintain quality playback / optimized decoding processes" terms under claims 1, 6, 7 and "optimal / optimal digital format / optimal encoding rate" under claims 3 and 8. Source: https://worldwide.espacenet.com/publicationDetails/claims?CC=US&NR=[10771849B2](/patent/10771849B2)&KC=B2&FT=D ; https://ptacts.uspto.gov/ptacts/public-informations/petitions/1557614/download-documents?artifactId=woVqmhp_jRcYPOySSmegOfPgFdpPSSxCuxhxAcFHTSUptuGkqSlo7ZI

1.2 Two date discrepancies in the record (flagged, not corrected)

1. Priority date. Google Patents states prior-art date 2005-04-18 (matching Espacenet's US20050107952, 2005-04-18). A Unified Patents portal page lists US-10771849-B2 with "2005-04-17." The one-day difference is immaterial to everything below, because the primary reference issued in 2002 and the co-references are 1990s/early-2000s art. I flag it only because the instructions require literal treatment of conflicting identifiers.
2. "Current date." The task header says April 26, 2026; the system framing says 2026-09-26. The previously generated section states the IPR final written decision was expected "on or about late September 2026." This memo is written as of April 26, 2026, so the FWD is treated as not yet issued.

1.3 Which § 103 applies — and a priority-chain vulnerability that changes the prior-art universe

The application at issue (US 16/181,285) was filed 2018-11-05, as a continuation in a chain running through US 15/016,821 (filed 2016-02-05) back to US 11/107,952 (filed 2005-04-18). Espacenet's priority list: US20050107952 20050418; US201615016821 20160205; US201816181285 20181105 (https://worldwide.espacenet.com/publicationDetails/biblio?CC=US&NR=10771849B2&KC=B2&FT=D).

Two consequences:

• If all nine claims are entitled to the 2005-04-18 priority, pre-AIA § 103(a) governs, and prior art must predate the date of invention/2005. That is the posture I assume below.
• If the 2005 priority fails (the granted claims are notably broader than the 2005 disclosure — see §5), then the effective filing date is 2016-02-05 or 2018-11-05 and the entire 2005–2016 adaptive-streaming literature and patent corpus becomes available art under AIA § 102(a)(1)/(a)(2). That corpus includes the post-2005 ABR references that surfaced repeatedly in the record I retrieved, e.g. US 9,264,475 (Carmel-et-al. family continuation, referenced as prior art in later patents: https://patentimages.storage.googleapis.com/cc/31/77/7363b617016e1d/US9264475.pdf) and the WAG Acquisition "streamlet" family. This is the single highest-leverage issue in the invalidity case, and it is a § 112 written-description question as much as a priority question.

---

2. What the "Prior Art" record actually consists of

The fetched Google Patents page did not include the machine-readable References Cited / Cited By / Similar Documents tables — only the "Prior art keywords" field (multimedia, audio, video, multimedia objects, objects) and the references the applicant itself discussed in the specification. I therefore built the record from three layers:

Layer A — references the applicant cited/characterized in the '849 specification (admissions):

Every one of those is an admission that the element was known in the art as of the 2005 priority date. Admissions in a patent's own specification are usable as prior art against it. This matters enormously, because it means the applicant's own document supplies the "wireless," "HTTP," "directory protocol," "codec," and "optimized decoding" elements.

*Layer B — the references actually asserted in the instituted IPR (IPR2025-00775, Google LLC v. Withrow Networks Inc.):*

The Board instituted on 2025-09-29 (per the previously generated section). Petitioner's Opposition to Patent Owner's Request for Discretionary Denial states the grounds plainly:

"The Petition presents a straightforward combination of the teachings of Carmel [EX1003] and Mattavelli [EX1004] or Wang [1007] that render obvious the challenged claims. (See Pet. at 3 (listing asserted grounds).)"

and

"The Examiner materially erred by overlooking (or misapprehending) the teachings of Carmel and by not rejecting the claims as obvious over Carmel in view of Mattavelli or Wang."

Source: https://ptacts.uspto.gov/ptacts/public-informations/petitions/1557614/download-documents?artifactId=0IecaQfTeUZL8u93uaqmhcSgEbOqulR1uAFiXj3ndEkH0nbGEWnokdk

I could not verify the bibliographic identity (patent number, venue, dates) of EX1004 "Mattavelli" or EX1007 "Wang" in this session. From the Petition's own characterization they teach "optimized decoding processes (e.g., computational graceful degradation for video decoding techniques) to maintain fluidity and quality playback" (Pet. at 45–55). I therefore do not attribute patent numbers to them, and nothing below depends on my guessing their identity. They should be confirmed from the exhibit list of the Petition.

Layer C — corroborating ABR/segmentation art that appears throughout the parallel PTAB record (secondary/context art):

Sources: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1556058](/patent/1556058)/download-documents?artifactId=rPBUUlp1bOsVUHpAFYw5smWddVzEFZdLh73aLnbgorGf_tBSKFl3mlo ; https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1556059](/patent/1556059)/download-documents?artifactId=E1h0SIQATIOrEL5g6zfFArOHh9SH1Ucf-etoShi0TlDu_Joab8UHrk0

---

3. The primary reference: Carmel, U.S. Pat. No. 6,389,473 B1

Bibliographic confirmation (from the reference itself, retrieved in the record):

• Title: "NETWORK MEDIA STREAMING"
• Inventors: Sharon Carmel, Tzur Dabooosh, Eli Reifman, Naftali Shani, Ziv Elirz, Dror Ginsberg, Edan Ayal
• Assignee: Geo Interactive Media Group Ltd., Givatayim (IL)
• Appl. No. 09/275,703, filed March 24, 1999; IL priority 123819, March 24, 1998
• Issued May 14, 2002; 41 claims, 11 drawing sheets
• Source: https://www.docketalarm.com/cases/PTAB/IPR2025-00470/Webgroup_Czech_Republic_a.s/docs/01-17-2025-Petitioner/Exhibit-1016-Ex_1016.pdf?download=true

Carmel is therefore § 102(b) art by more than six years relative to the 2005-04-18 priority date.

What Carmel discloses (all from the reference, as quoted in Board and court filings):

1. Segmented multimedia objects, each a separate file with an index. "each segment or slice is contained in a separate, respective file" (2:22–23); the transmitting computer "stores each slice as a corresponding file, having a running slice index 1, 2, 3 … N" and "each file also includes one or more time stamps" (7:27–31).
2. HTTP delivery. "Clients 30 preferably download the sequence [from server 36] using the Hypertext Transfer Protocol (HTTP)" (7:4–17); Fig. 1 shows clients connecting to server 36 "via a HTTP link"; Figs. 6A/6B show "Connect to Server" / "HTTP From Server."
3. Client-pull by request. Clients "read the index file" and select a starting slice (Fig. 6A, "SELECT SLICE"); PO's opponent concedes the client "would need to send an HTTP GET request to tell the server what slice it wanted."
4. Multiple quality levels (an encoding-rate ladder) within slices. In the single-indexed-file embodiment, "each slice in [data] stream 41 includes multimedia data at multiple quality levels" (8:56–57) — the reference expressly lists "audio level #1, audio level #2, video level #1, video level #2" in the slice header.
5. Autonomous bandwidth-responsive selection by the client, and upward and downward adjustment. "the client selects one of the available quality levels in the stream" and the server sends "data slices at the chosen quality level" (11:2–7); the client "determines a compression ratio by which to compress the data, based on the collective bandwidth of its open links with server 36" (11:40–42); and Carmel "discloses instances where the selected quality level is improved or increased, if/when allowed by the available network bandwidth" (10:64–11:22, Fig. 6B).
6. A sliding window of stored objects. "the memory available on server 36 is limited, and files 42, 44, 46, etc., will be stored on the server and erased therefrom in a 'first-in-first-out' sequence" (7:55–58); J = earliest stored slice, N = most recent (8:23–29).
7. Fluidity as the design objective, with real-time broadcast and download rate ≳ generation rate (2:51–59), and random access/jumping along the stream (8:32–41).
8. Alternative directory-serving protocols. Carmel discusses "utilizing multiple FTP links to maintain a reliable download path" among its adaptation options (Carmel 12:54–58, as quoted at https://ia800507.us.archive.org/14/items/gov.uscourts.utd.[141844](/patent/141844)/gov.uscourts.utd.141844.19.3.pdf).

And petitioner's summary of Carmel for this case:

"Carmel discloses techniques for dividing a stream of multimedia data into a sequence of slices (including instances where slices have multiple, different quality levels) for real-time broadcasting over a computer network, as well as a client multimedia player that makes dynamic adjustments to the selection of what slices are downloaded in response to changes in the available network bandwidth so as to maintain fluidity and the highest quality playback."

---

4. Person of ordinary skill in the art (POSITA)

For an April 2005 priority date, a POSITA would have: a B.S. in electrical engineering, computer science, or equivalent, plus 2–4 years of experience in networked multimedia delivery and/or embedded/mobile codec implementation — or a master's degree with less experience. That skill set is the reason the record art is combinable: a POSITA in this field is by definition someone who works at the seam between transport of compressed media (Carmel, Feig, Ogdon, Waese, Stumm) and decoding of compressed media on constrained hardware (Mattavelli/Wang, MPEG-4/AAC standards, Chen's IDCT) — which is precisely what the '849 claims recite.

Notably, the challenged claims themselves set the POSITA bar low: the applicant's own specification states that the invention's contribution is enabling existing MPEG-4 Simple Profile video and AAC LC audio to run in a downloadable J2ME applet on a handset, with trade-offs in quality — not new codecs, not new transport.

---

5. Claim-by-claim element mapping

Independent claim 1 — method

Bottom line on claim 1: every element is disclosed by Carmel, except that (a) the "wireless" carrier is supplied by the applicant's own admissions plus routine artisan knowledge, and (b) "optimized decoding processes" is supplied by Mattavelli or Wang. The Examiner's own Notice of Allowance identified the sole reason for allowance as the "wherein the multimedia player is configured to play multimedia objects in a sequence such that fluidity, video quality and audio quality are maintained by selecting a plurality of said multimedia objects that reflect available network bandwidth, autonomously adjusting…" limitation (EX1008 at 589, quoted in Petitioner's Opposition). That is the limitation the Petition is aimed squarely at — and it is the limitation Carmel's Fig. 6A/6B and 10:64–11:22 address.

Independent claims 6 and 7

Claim 6 is the same subject matter recast as computer-executable instructions on a non-transitory medium — same art, same mapping. Claim 7 adds "at least one centralized content server transmitting multimedia objects," which is Carmel's server 36 (a single server to which the transmitting computer uploads and from which all clients download); Waese (US 6,286,031) and Stumm (US 5,768,528) independently disclose the pre-fetched-list/notification distribution model, reinforcing claim 7.

Dependent claims 2–5, 8–9

---

6. Grounds of rejection and the motivation to combine

Ground 1 (primary): Carmel in view of Mattavelli — claims 1–9

Ground 2 (alternative): Carmel in view of Wang — claims 1–9

(These are the grounds actually presented in IPR2025-00775, Pet. at 3.)

Why a POSITA would have combined them — the four KSR rationales that apply:

1. Known technique improving a similar device in the same way (KSR, Rationale C). Carmel adapts delivery to the constraint the client reports (bandwidth). Mattavelli/Wang adapt decoding to the constraint the client experiences (compute budget). Both are in the same field (delivery/decoding of compressed multimedia to heterogeneous client devices), both operate on the same kind of bitstream, and applying the latter to the former yields a system responsive to both constraint classes. The Petition's expert expressly framed the motivation this way: "A POSA would have been motivated to combine the complementary methods and techniques taught by Carmel and Mattavelli to form a streaming system that is more robust and that addresses both bandwidth-based and client processing-based constraints together to provide an improved or best possible/practicable playback experience" (EX1002 ¶¶40–59).

2. Market forces and design incentives (Rationale D/F). The challenged patent's own Background supplies the motivation, verbatim: cell phones have "slow CPUs, high memory latencies, slow drawing capabilities and the absence of YUV to RGB conversion in the hardware"; most handsets "cannot support multimedia playback because they are only capable of supporting one or a few concurrent processing threads"; and North American wireless bandwidth ranged from 2–3 kB/s to 64 kbit. A 2005 POSITA confronting those two constraints — bandwidth and decode capability — would be driven to combine a bandwidth-adaptive delivery scheme with a complexity-adaptive decoder. This is not hindsight reasoning; it is the applicant's own statement of the problem.

3. Finite number of identified, predictable solutions / obvious to try (Rationale E). Given a need to keep a media sequence playing on a low-power handset over a variable wireless link, the two levers available are (i) request cheaper representations and (ii) spend less compute on each representation. Both were known, both had been applied in the same art, and combining them has a predictable effect (playback continues; quality degrades gracefully). The Federal Circuit has repeatedly upheld exactly this kind of two-lever combination in the streaming space — the Board's own FWDs in the WAG Acquisition IPRs found the "client pull via successive HTTP GET requests of independently stored slices" obvious over Carmel, and the appellee's brief in the consolidated Federal Circuit appeal (WAG Acquisition, LLC v. Walt Disney Co., No. 24-1627, briefing filed 2024-11-04) argues affirmance on both the Carmel-alone and Carmel+Feig theories. I did not retrieve the Federal Circuit's disposition of that consolidated appeal; I state the outcome as unknown.

4. Substitution of a known element for a known element (Rationale B) — for claims 2, 8. HTTP ↔ FTP/NNTP/IMAP4 for "serving files in a directory" is a mechanical substitution with a predictable result, and the specification itself concedes the equivalence.

Reasonable expectation of success. Both references are software-level techniques applied to standardized codecs (MPEG-4, AAC LC) whose decode flows were fixed by published specifications (the patent's own Fig. 6 is captioned as the MPEG-4 Committee's flow, and Fig. 14 as "the general steps of audio decoding"). A POSITA could implement the combination with no undue experimentation, and the only risk — some loss of output quality — is expressly embraced by the challenged claims themselves (see §6.1).

No teaching away. Carmel's disclosure that a slice may be dropped rather than sent when bandwidth is short (12:56–57) is not a teaching away from maintaining fluidity; it is an alternative mechanism for preserving continuity, which is the same objective as claim 1's "fluidity … maintained." The Board rejected the contrary argument in the WAG proceedings.

Ground 3 (my proposed additional/backup ground): Carmel in view of Waese (US 6,286,031) and Stumm (US 5,768,528) — claims 1, 2, 6, 7

Both are cited by the applicant in the '849 specification. Waese discloses transmission of video "as objects through pre-fetched lists," and Stumm discloses scheduled/instant notification file downloading to clients. These supply, with the applicant's own imprimatur: (a) the "multimedia object" abstraction delivered over file-based protocols, (b) the client-maintained list of objects to fetch, and (c) the "supplied host path identification" indexing. Motivation: both are directed to the same problem — delivering discrete, individually addressable media files to clients without a stateful media server.

Ground 4 (backup): Ogdon in view of Allen — claims 1, 3, 8 (and 6, 7)

Ogdon teaches "client-request technology, where each presentation member's browser requests information from a communication network server," a client that "select[s] an appropriate content webserver and an appropriate version of the next collection of segments to be retrieved," and a client that "monitors the time elapsed before transmission of the selected collection of segments is completed, and determines whether these segments are provided within an appropriate window of time." Allen teaches low/medium/high-quality streams at differing bitrates. Together they disclose the bandwidth-reflective selection of a quality-rung and the client-side timing/fluidity control. Motivation: to reduce startup latency and adapt to variable throughput.

---

6.1 The claim-construction pincer on the three disputed terms

The § 103 case and the § 112 case are mutually reinforcing here, and this should be exploited:

The practical lesson: the broader the construction Patent Owner advances to avoid the art, the more indefinite the claims become; the narrower the construction it advances to survive § 112, the less of the specification it can capture and the more squarely Carmel/Mattavelli (or the post-2005 art, if priority fails) reads on it.

---

7. Where the § 103 case is weakest (candid assessment)

1. "Received by said multimedia player from servers using a wireless connection." Carmel's clients are general-purpose computers on a network; the reference does not emphasize cellular/wireless or mobile handsets. For the 2005 priority date this element must be supplied either by (a) the applicant's own admitted prior art in the Background, or (b) a mobile-specific secondary reference. I did not retrieve a mobile-specific reference in this session whose citation details I can verify — this is a gap that should be closed with, e.g., contemporaneous WAP/J2ME mobile-media-delivery art.
2. "Autonomously adjusting." Patent Owner will argue the adjusting in Carmel is (i) user-triggered (start-point selection via indicator 58) and (ii) partly server-side in the single-file/push embodiment. The ITC record is genuinely mixed here: the ALJ in the DISH/Icon investigation found Carmel did not expressly disclose HTTP GET requests for individual slices and did not disclose "independently requestable streamlets," characterizing Carmel's client action as "selecting a quality level + a starting time" followed by a server push (gov.uscourts.utd.141844.19.3.pdf; quoted at https://ia800507.us.archive.org/14/items/gov.uscourts.utd.141844/gov.uscourts.utd.141844.19.3.pdf). That is an anticipation holding, and obviousness can bridge an express-disclosure gap — the Board and the Amazon appellee brief take the position that a POSITA would plainly implement Carmel's separate-file embodiment via successive HTTP GETs. But it gives Patent Owner usable ammo.
3. Claim 5 (multiple MPEG-4 composite-layer streams split at I-frames) — the mapping I can support from the retrieved record is the thinnest of the nine. If claims 1–4 fall, claim 5 should be separately worked up from MPEG-4 multiple-VOL/multiple-VOP-layer art, or conceded as a lesser point.
4. Priority/§112 fragility cuts both ways. If Patent Owner successfully narrows the claims to the 2005 disclosure's specific optimizations to defeat Carmel, it simultaneously invites an attack that the granted claims (which recite none of those optimizations) are not supported by the 2005 specification — which, if successful, opens the 2005–2016 ABR art and makes the case far easier. This should be pleaded as an alternative theory, not just a fallback.

---

8. Secondary considerations

Patent Owner's complaint (N.D. Cal. 5:24-cv-03203, Dkt. 1) telegraphs the objective-indicia narrative: an ARM Ltd. license to demonstrate high-performance MPEG-4 Part 2 streaming (Dec. 2005); joint ARM/Qualcomm/Sprint demonstrations (mid-2006); Movy.tv as an ABR platform (2007); industry-recognized "long-felt need" (Pillsbury's Gone in a Flash); and copying (Google/Adobe "Sliced Bread," then HLS 2009 and DASH 2012).

Assessment (for the § 103 rebuttal):

• Nexus is the weak link. The praise and the ARM/Qualcomm/Sprint activity attach to the Movidity player and its decoder optimizations and adaptive object model — features the granted claims do not recite. The claims recite generic client-driven adaptive selection and "optimized decoding processes." Any presumption of nexus requires the objective evidence to be tied to the claimed features, not to the specification's unclaimed decoder work.
• Post-dating. The launching and licensing activity occurs 2005–2007, i.e., at or after the priority date; it is evidence of commercial value, not of non-obviousness as of the invention date, unless tied back.
• "Copying" is attenuated. HLS (2009) and DASH (2012) are standards that post-date the patent by 4–7 years; Google's own account (the cited 2013 YouTube talk) attributes its ABR implementation to work begun around 2009. Copying evidence requires a nexus to the claimed invention, and standards adoption is a weak nexus showing.
• Primary reference is a 1998/1999 filing. The core idea — slice a stream into indexed files, store multiple quality levels, let the client pick and switch based on its own bandwidth — was published in Carmel by 2002, six years before the '849's 2005 priority and nearly a decade before the '849 issued. That fact substantially undercuts any "long-felt, unsolved need" narrative directed at the claim's central feature.
• Expired status. The patent expired 2025-04-26 (over 20 years from the 2005 priority). Petitioner made, and the record shows the Director credited, the point that "settled expectations in the validity and continued enforceability of the … now-expired, '849 patent" are absent — which matters to discretionary denial and to the equities, though not to the § 103 merits.

---

9. Conclusion

On the record retrieved, claims 1–4 and 6–9 of US 10,771,849 are, in my assessment, more likely than not invalid under pre-AIA § 103(a), on Carmel (US 6,389,473 B1) in view of Mattavelli (or, alternatively, Wang) — the combination actually asserted in IPR2025-00775, in which the Board instituted on 2025-09-29 and denied Patent Owner's Director Review request on 2025-12-15. Claim 5 is the weakest of the nine and requires separate, better-developed art mapping.

The single most important structural observation: the granted claims are far broader than the 2005 disclosure. Nothing in claims 1–9 recites the IMDCT/IFFT decimation, the Huffman codebook look-up scheme, the macroblock texture buffer, the YUV-to-RGB scaling loop, or the object-window indexing that occupy FIGS. 7–20 and most of the specification. What the claims do recite — segmented A/V objects with parameters and host/path identifiers, delivered over HTTP to a wireless client that autonomously picks among them according to available bandwidth and decodes with "optimized decoding processes" — is the state of the art as of 2005, with Carmel at its center and the applicant's own Background section supplying the motivation.

Confidence and explicit caveats

• High confidence: the bibliographic data for Carmel (US 6,389,473 B1) and its quoted disclosures; the text of claims 1–9 and the identity of the three independent claims (1, 6, 7); the identity of the asserted grounds (Carmel + Mattavelli / Carmel + Wang) as quoted from Petitioner's Opposition; the fact and date of institution.
• Moderate confidence: the specific pin cites for Carmel's passages as reproduced in third-party briefs and petitions (I did not read the Carmel PDF end-to-end in this session); the claim-construction posture, which may have moved since the October 28, 2025 Markman hearing.
• Explicitly unresolved — do not treat as findings:
  • The bibliographic identity (number, venue, dates) of IPR exhibits EX1004 "Mattavelli" and EX1007 "Wang." I could not verify these and have deliberately not assigned numbers to them.
  • Whether the Petition asserts additional grounds beyond the two quoted.
  • The outcome of the consolidated Federal Circuit appeal in the WAG Acquisition line, and of the ITC's Carmel findings on appeal — only briefing and the ALJ's findings appear in the record I retrieved.
  • The status of IPR2025-00775's final written decision. The prior section projected it for late September 2026; if the FWD has since issued (i.e., if the operative "today" is in fact 2026-09-26 rather than 2026-04-26, per the date conflict in §1.2), this analysis should be updated against it immediately.
• Flagged as a conflict with the prior section: the prior section listed priority/prior-art date as 2005-04-18 from Google Patents while a Unified Patents page shows 2005-04-17. I note it, and I do not resolve it.