US 6803909

I'll run targeted searches for this specific patent number, including any litigation/docket activity.

Let me try a few more targeted queries (Google Patents, USPTO Patent Center, and CAFC docket searches).

Search scope and caveats (read first)

• I queried general web search, not a live USPTO Patent Center / PACER / CAFC docket API. No CAFC 2026 docket entry, appeal, or district-court litigation naming US 6,803,909 surfaced in my searches. Absence of a docket hit here is not proof that none exists.
• Two near-miss identifiers appeared and must not be conflated with the patent you asked about:
  • BR DI 6803909-3 — a Brazilian industrial design registration (Yamaha Hatsudoki, motorcycle), an unrelated ID, not a US utility patent.
  • US 6,219,730 — asserted in Genuine Enabling Technology LLC v. Sony Group Corp., CAFC, decided Feb 19, 2026 (affirming summary judgment of non-infringement under §112(f)). That is a different patent; it is not US 6,803,909. I mention it only because it is the sole Sony-related 2026 CAFC item my search returned, and it is not relevant to this patent.
• Everything below is drawn from the authoritative full text of the patent.

Bibliographic summary — US 6,803,909 B2

Abstract (verbatim): "A data processing system including a data processing apparatus having an apparatus having relatively low picture image processing capacity. The system enables such a low-performance data processing apparatus to constantly display homogenous three-dimensional virtual reality spatial picture image. Specifically, a shared server reads VRML contents including a VRML file and a script program via a storage media before generating three-dimensional virtual reality spatial picture image pixel data. The shared server enables the three-dimensional virtual reality spatial picture image pixel data to be transmitted to a client PC from a satellite communication network via a WWW server. The client PC individually displays the received three-dimensional virtual reality spatial picture image pixel data on a monitor."

Plain-language overview of the independent claims

The 28 claims fall into six "aspects," each drafted as a trio: apparatus / method / storage-medium. This structure is important — the same inventive concept repeats across server-side, client-side, and system-level claims.

Aspect 1 — Server renders, server sends (claims 1, 5, 6)

• Claim 1 (apparatus): A server that supplies a 3D virtual-reality space to another device over a network. It has (a) a generating part that creates pixel data of the VR space from 3D VR spatial data (i.e., the server does the rendering, not just file serving), and (b) a transmitting part that sends that pixel data to the other device.
• Claim 5 (method): The same operations as steps — generate the pixel data, then control its transmission.
• Claim 6 (medium): A computer-readable program product performing generate + transmission-control.

Aspect 2 — Client receives pixels and displays (claims 7, 9, 10)

• Claim 7 (apparatus): The receiving-side device: a receiving part for pixel data of the VR image (where the image is represented by 3D VR spatial data upstream), and a display part for showing it.
• Claim 9 (method): Reception-control + display-control steps.
• Claim 10 (medium): Program performing reception-control + display-control.

Aspect 3 — Two-device system, first renders for second (claims 11, 13, 14)

• Claim 11 (system): Combines the above: the first apparatus generates pixel data from 3D VR spatial data and transmits it; the second receives and displays it.
• Claim 13 (method): Method split across both apparatuses (generate/transmit at the first; receive/display at the second).
• Claim 14 (medium): Program controlling both apparatuses of the system.

Aspect 4 — Server decides who renders (claims 15, 19, 20)

• Claim 15 (apparatus): A server with (a) a generating part that can produce pixel data, (b) an identifying part that decides whether the server itself or the other apparatus should generate the pixel data, and (c) a transmitting part that, depending on that decision, sends either the data needed to generate the image (e.g., the VRML file) or the pixel data.
• Claim 19 (method): Same decision logic as method steps.
• Claim 20 (medium): Program performing generate + identify + transmission-control.

Aspect 5 — Client asks for a format, then renders or just displays (claims 21, 23, 24)

• Claim 21 (apparatus): A client that (a) requests a particular deliverable format of the VR image from the other device, (b) receives either the generating data (3D VR spatial data) or pixel data in response, (c) generates pixel data itself if it received the generating data, and (d) displays whichever it ends up with.
• Claim 23 (method): Request-control + reception-control + generate + display-control steps.
• Claim 24 (medium): Program performing the same.

Aspect 6 — Full system with negotiation (claims 25, 27, 28)

• Claim 25 (system): The first apparatus generates, identifies whether it or the second apparatus should render, and transmits accordingly; the second apparatus requests a format, receives generating data or pixel data, generates if needed, and displays.
• Claim 27 (method): The combined two-party method of claim 25.
• Claim 28 (medium): Program controlling both apparatuses to do the above.

Dependent claims (context only): 2 (avatar image controlled by the other device), 3 (image captured from a fixed viewpoint), 4 / 8 / 12 / 18 / 22 / 26 (delivery via satellite communication network), 16 (storage part for data about the other apparatus), 17 (identification based on stored device data or a request from the other apparatus).

Notes / uncertainty flags

1. Drafting inconsistency in claim 12: it recites "A data processing apparatus as claimed in claim 11," but claim 11 is a data processing system. I am reporting the text literally rather than correcting it.
2. Mixed apparatus/system/medium language: several claims use "part" for method steps and "step" for apparatus elements (e.g., claims 15, 19). Noted as written.
3. Claim 11 recites "generating means" in one clause while the surrounding claim uses "generating part" — a potential §112(f) means-plus-function issue, but I have no prosecution-history or court construction available to confirm how it was treated.
4. No litigation found, with a strong caveat: I could not locate any CAFC 2026 docket (or any federal litigation) involving US 6,803,909. Given the patent's adjusted expiration of 2021-12-20 and "Expired – Lifetime" status, 2026 appellate activity on this specific patent would be unusual, but I cannot rule it out from the search tools available to me. If you need a definitive answer, this should be confirmed against PACER/CM-ECF and the CAFC docket system directly.
5. The Google Patents record lists cited prior art (US 5,877,759; US 5,982,372; US 5,987,480; US 6,405,249; US 6,414,677; US 6,487,557; US 6,493,001) and later citing documents (e.g., Ubitus GPU-resource patent family, IBM GUI-screen-sharing publication), but I did not independently verify those citations.

Litigation Search Results — U.S. Patent No. 6,803,909

Bottom line: I found no evidence of any litigation (district court, ITC, or Federal Circuit appeal) involving U.S. Patent No. 6,803,909. I also found no PTAB post-grant proceeding (IPR/PGR/CBM) or ex parte/ inter partes reexamination against it.

Because no case exists, I cannot provide plaintiff(s), defendant(s), jurisdiction, case number, filing date, or outcome — those fields are empty rather than merely unknown to me.

What the authoritative record shows

The patent itself is:

Why the absence of litigation is meaningful here: Google Patents' US 6,803,909 page includes the usual "Cited By," "Patent Citations," and "Family" tables, and it does surface litigation for patents where such records are known to Google. For this patent, no litigation table and no litigation entry appear anywhere on the page — only pre-grant publications, family members (JP 2001236525A; US 2001/0024230 A1), § 4/8/12-year maintenance-fee payments, and the 2021 adjusted-expiration event. The prosecution history likewise shows only a routine assignment from Tsukahara to Sony (reel 011836, frame 0285) — no adversarial activity.

A separate search of the litigation-aggregation sources you named (Unified Patents portal caselist, PTAB filings, Docket Alarm, Justia/PTAB dockets, and general web dockets) likewise returned no case caption in which US 6,803,909 is the asserted patent.

Two matching-number decoys I explicitly ruled out (per your instruction not to auto-correct)

• DI 6803909-3 (INPI Brazil, filed 2008-09-05, granted 2009-10-13) — a design registration for a motorcycle configuration, owned by Yamaha Hatsudoki Kabushiki Kaisha, inventor Seiko Ando. Unrelated to the U.S. utility patent by subject matter, office, and type.
• Notice No. 6803909 in a Lithuanian companies register publication (UAB "Autogazas") — an administrative filing number, not a patent.

Neither is the patent in question.

Context: why this patent likely was never asserted

US 6,803,909 is a server-side-rendering patent. Its independent claims (e.g., claim 1, claim 15, claim 25) contemplate "a data processing apparatus providing three-dimensional virtual reality space for another data processing apparatus linked to a network," in which the server "generat[es] pixel data of the three-dimensional virtual reality spatial picture image from three-dimensional virtual reality spatial data" and transmits that pixel data to the client (with satellite transmission recited in claims 4, 8, 12, 18, 22, 26). Sony is a practicing, operating company rather than an NPE; it pursued server-rendered VRML through its own "Community Place" products rather than through an assertion campaign, and the patent reached expiry (2021) with its maintenance fees paid but with no public record of enforcement.

Caveats on coverage

1. PACER is not a complete public index by patent number. PACER has no free full-text patent-number search across all districts; confirming a negative nationwide requires paid, district-by-district docket queries. My conclusion rests on aggregator databases plus the patent record, which is strong evidence but not mathematically exhaustive.
2. I could not retrieve a live, per-patent hit page from the Unified Patents portal caselist for this patent number. Searches returned Unified's general litigation/PTAB pages and unrelated patents instead. If you need a court-admissible negative, I'd recommend a paid Westlaw/Lexis KeyCite or Docket Navigator patent-family report.
3. Underlying complaint allegations could exist without a docketed, publicly indexed case — e.g., demand letters, licensing correspondence, or cases dismissed before service and never indexed. Those would not appear in any public litigation database.
4. Consistent with your ground-truth rule: nothing in the search results contradicts a "no litigation" finding — every "6803909" hit I retrieved was either the U.S. patent's own record or one of the unrelated decoys above. I am reporting the absence of litigation rather than inventing cases to fill the requested fields.

I'll verify the structured data's "no proceedings" default against web sources before finalizing.

No PTAB hits surfaced. Let me run a few more targeted checks before concluding.

PTAB / AIA Trial Analysis — US Patent 6,803,909

Patent: US 6,803,909 B2 — "Data processing apparatus, method, system, and storage medium"
Patent Owner (current assignee of record): Sony Corporation (original assignee; inventor Hiroki Tsukahara)
Priority date: 2000-02-23 · Filing date: 2001-02-24 · Grant date: 2004-10-12
Claims: 28 (independents: 1, 5, 6, 7, 9, 10, 11, 13, 14, 15, 19, 20, 21, 23, 24, 25, 27, 28)
Source of record: Google Patents US6803909B2

Proceedings overview

Total AIA trial proceedings on file: zero (0). The structured "PTAB proceedings on file" block drawn from the USPTO Open Data Portal states that the ODP API returns no AIA trial proceedings for US 6,803,909 as of the most recent ingest — no IPRs, no PGRs, no CBMs, in any status (active, instituted, FWD-invalidated, settled, or denied institution). The bottom line for a defendant is therefore not "the patent has been hardened by surviving challenges" and not "the claims are canceled" — it is that there is no PTAB record at all, and the more consequential defensive fact is that the patent's term has run out (adjusted expiration 2021-12-20, status listed as Expired – Lifetime). This is a 2001-filing-era Sony VRML/networked-3D patent that was never contested at the Board and, so far as public sources show, was never a high-volume assertion vehicle.

Caveat on the "zero" finding. I ran targeted web searches (patent number variants, "IPR"/"PTAB"/"CBM", Unified Patents, litigation pairings) and surfaced no petition, institution decision, FWD, or appeal referencing 6803909 or 6,803,909. That is absence of evidence, not proof of absence — an unindexed or very-recently-filed petition would not appear. If you need certainty, run the number directly through PTAB E2E and PTAB Decisions, and confirm in Patent Center. Do not rely on the empty ODP field alone for a litigation-critical conclusion.

Individual proceedings

None. The required per-proceeding template (proceeding number, petitioner, panel, grounds, institution decision, FWD, settlement, appeal) cannot be populated because no proceeding exists. I will not invent proceeding numbers, panels, or dispositions to fill the template.

Strategic summary

Claim status: all 28 claims are UNTESTED at the PTAB. No claim of US 6,803,909 has ever been canceled, confirmed, or amended in an AIA trial, because no AIA trial has been instituted. There is accordingly no narrowing to track: if you receive a demand letter or complaint citing claims 1–5 (the server-side generate-and-transmit claims), claims 7–8 (the receive-and-display claims), or the third-aspect claims 15–20 and 21–24 (the "identify who renders / request a format" claims), none of those claims has been knocked out for you — but none has been validated against a skilled challenger either. Every claim is a blank slate.

Estoppel landscape: § 315(e)(2) estoppel is a non-issue. Because no petitioner has ever been through an IPR/PGR on this patent, no party is estopped, and there is no petitioner/privy chain to worry about. Conversely, you get no free ride from someone else's prior work: there is no instituted ground, no FWD claim construction, and no Board-construed record to lean on. Any invalidity position you take is a first impression. Note the statutory constraints if you were to file: § 315(b)'s one-year bar runs from service of a complaint, and under § 315(a)(1) a filed DJ action of invalidity bars a later IPR by that party.

Pattern signals: none. There is no repeat petitioner (no Unified Patents or other defensive aggregator in the chain), no serial-filing patent owner, and no Federal Circuit appeal of any 6803909 decision — because there is no decision to appeal. The three forward citations in the record (US20090172557A1, IBM, GUI screen sharing; and the Ubitus GPU-cloud-gaming family US20130147818A1 / US8736622B2 / US9483996B2 / US10223997B2) show the patent sits in the ancestry of remote/cloud rendering art — it describes what practitioners now call cloud gaming — but citations are not assertions, and none of those citing parties challenged the patent.

The real defensive story is expiration, not invalidation. The record lists the adjusted expiration as 2021-12-20 (20 years from the 2001-02-24 filing plus PTA) and the status as Expired – Lifetime. Two practical consequences: (1) no prospective injunctive relief and no ongoing-infringement theory — an expired patent can only support past damages; and (2) § 286 caps recovery at six years before the complaint/counterclaim is filed, so the recoverable window shrinks every day. If suit were filed today (2026-09-30), the reachable infringement period is roughly 2020-09-30 through 2021-12-20 — a ~15-month tail. That math, not any PTAB estoppel, is your strongest leverage. One open item worth verifying: the indexed Legal Events show maintenance-fee payments at the 4-, 8-, and 12-year marks (2008-04-14, 2012-04-04, 2016-04-05) but no indexed payment for the final fee window, which for a PTA-extended term would have come due around 2020-04. The fetched ODP/Google record does not resolve whether the term ended naturally at 2021-12-20 or lapsed earlier for non-payment — check Patent Center before you commit to a date in a brief.

A note on whether an IPR is even worth filing. You can IPR an expired patent (the claims are still invalidatable; the patent owner simply cannot amend), and doing so could moot a parallel past-damages case. But the calculus is weak here: the damages tail is short, the claims are software-era VRML (US 6,405,249 and US 6,493,001, both Sony, are already in the record as cited art and would be natural § 103 starting points), and there is no litigation-driven reason to spend the money unless the patent owner actually sues. If no suit has been served, filing an IPR now mainly buys you a § 315(b)-free preemptive strike — usually not worth it against a dead patent.

Recommended next steps

• Verify the zero, don't assume it. Confirm on PTAB E2E and PTAB Decisions by application number 09/792,330 and patent number 6,803,909. If a petition has been filed since the ODP ingest, it will appear there first. If you find one, re-run this analysis with the proceeding number — I could not source one, and I will not fabricate one.
• If you are a defendant and no suit has been served: do nothing at the PTAB. Preserve your § 315(b) clock and your arguments. The patent's expiration date (2021-12-20) and the § 286 six-year lookback are your first-line defenses; build the non-infringement and damages-window rebuttals on those.
• If you are a defendant and suit has been served: (a) pull the filing date to fix your § 286 damages window, and (b) confirm the maintenance-fee status in Patent Center — if the term lapsed early for non-payment, your damages window may be shorter or non-existent. Assess infringement only against the ~15-month post-2020 window, and consider whether the asserted claims (especially the server-side claims 1–20, which were drafted for a 2001 client-server/satellite architecture) read on a modern distributed or edge-rendered system at all.
• If you are facing the patent as a licensor's portfolio item: the absence of any PTAB history is a mild signal that the patent was never worth attacking — consistent with a patent that expired before the cloud-gaming assertion wave matured. That is a point in your negotiating posture, but it is an inference, not a holding.

Nothing in this section should be filed without re-checking the ODP/E2E record on the day of filing — the single most important fact in this report is a negative, and negatives decay.

I'll verify the assignment record against live sources before writing up.

Let me check for any litigation or NPE activity, and the assignment record more directly.

Assignment & Ownership Analysis — US 6803909 B2

Patent: US 6803909 B2 — "Data processing apparatus, method, system, and storage medium"
Application: US 09/792,330 · filed 2001-02-24 · granted 2004-10-12 · priority JP 2000-045326 (2000-02-23)
Status: Expired – Lifetime (adjusted expiration 2021-12-20); maintenance fees paid at 4, 8, and 12 years (2008-04-14, 2012-04-04, 2016-04-05)

Source-access note: I was unable to open assignmentcenter.uspto.gov directly in this session (tool budget exhausted after the search phase). The record below is reconstructed from the Google Patents legal-events/assignment mirror (https://patents.google.com/patent/US6803909/en), which is populated from the same USPTO assignment data. Reel/frame, execution date, recording date, and assignor/assignee are as indexed there. The correspondent of record was not exposed in the retrieved legal-event entry, so I have not named one — see the honesty flag under "Repeat correspondent" below.

Inventors

• Basis for employer: the assignment (reel 011836/0285) was executed 2001-04-26 to Sony Corporation, and the specification repeatedly describes the "Community Place Browser" as "developed by Sony Corporation (which is the applicant of the present invention)." Tsukahara was a Sony employee-inventor on the Community Place / VRML shared-space line of work.
• Unusual patterns: None. Single inventor, no co-inventor departures to track, no inventor-held continuation chain, and no evidence of inventors departing the original assignee near filing. The patent also lists a Japanese priority (JP2000045326) and a JP sibling publication JP2001236525A (published 2001-08-31, since withdrawn / not active).
• Related-family caveat: several other Sony patents from the same era (e.g., US 6405249, US 6493001, and the US 2002/0118186 / US 6820112 shared-server family) cover adjacent Community Place subject matter and were also filed by Sony inventors. They are separate families — do not merge them into this chain.

Original assignee

Sony Corporation (1-7-1 Konan, Minato-ku, Tokyo, Japan) — named on the issued patent and on the only recorded assignment.

• Primary line of business: consumer electronics, computing, and interactive/networked entertainment software. The patent's own subject matter sits in Sony's late-1990s/2000 internet-3D push.
• Did they ship a product embodying the claims? Yes. The specification's "Community Place" is a real, shipped Sony product line: the Community Place Browser (VRML 2.0 client for Windows 95/NT) plus Community Place Bureau and Bureau World Location Server (multi-user server + WLS). Sony publicly announced OEM sales of Community Place in the U.S. on 1997-01-30 (https://www.sony.com/en/SonyInfo/News/Press/[199701/9701](https://assignmentcenter.uspto.gov/search/patent/reelFrameDetail?reelFrame=199701-9701)-30/), with the product in commercial release since September 1996 and the browser downloadable from vs.sony.co.jp. The patent's shared-server / pixel-data-distribution architecture maps directly onto the Bureau + browser client-server design.
• Current status: Operating. Sony Corporation remains an active global operating company (its legal-name history includes the 2013 change-of-name to "Sony Kabushiki Kaisha" in related trademark records, but the patent assignee of record is and remains Sony Corporation). No bankruptcy, no dissolution, no spin-off of this patent identified.

Assignment timeline

One recorded assignment exists. There are no post-issuance transfers in the chain.

• 2001-04-26 (executed) / recorded 2001-05-24 — Reel 011836/0285
  • Conveyance: ASSIGNMENT OF ASSIGNORS INTEREST
  • Assignor: Hiroki Tsukahara (individual inventor)
  • Assignee: Sony Corporation, Japan
  • Correspondent: not captured in the retrieved record — the Google Patents legal-event entry (code AS) lists only owner name "SONY CORPORATION, JAPAN" and the free-format text "ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:TSUKAHARA, HIROKI;REEL/FRAME:011836/0285; Effective date: 20010426." I am not naming a correspondent rather than guessing one. A direct Assignment Center pull on US 6803909 would surface the recording attorney.
  • Context: Ordinary employee invention assignment — an inventor-to-employer conveyance, executed ~3 months after the 2001-02-24 U.S. filing and ~14 months after the JP priority, then recorded ~1 month later. Not a fire-sale, not a securitization, not an NPE transfer.

Subsequent chain: none. The only other legal events after grant are the three maintenance-fee payments and the 2021-12-20 expiration. No assignment, security agreement, merger, change of name, license, release, or correction conveyance is recorded against this patent after 2001-05-24.

Timeline diagram

timeline
    title Ownership of US 6803909
    2000 : JP priority application filed
         : Sony files US 09 792 330 in Feb 2001
    2001 : Inventor assigns to Sony Corporation
         : Recorded reel 011836 frame 0285
    2004 : Patent US 6803909 B2 granted
    2008 : Maintenance fee paid at 4 years
    2012 : Maintenance fee paid at 8 years
    2016 : Maintenance fee paid at 12 years
    2021 : Patent expires 2021-12-20
         : Still held by Sony Corporation

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. The only recorded transaction is an inventor→operating-company assignment recorded 2001-05-24 at reel 011836/0285. There is no transfer to any "IP / Patents / Licensing / Holdings / Ventures" entity, no single-purpose LLC, and no registered-agent address anywhere in the chain.

2. Known asserter in the chain — Not present. Neither Sony Corporation nor the inventor appears on the comparison lists (Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities). No Unified Patents or RPX NPE listing for this patent surfaced in searching. Note for completeness: the patent is cited by Ubitus Inc. cloud/GPU-gaming patents from 2011–2017 (US 8,736,622; US 9,483,996; US 10,223,997 and their publications) — those are forward citations only and create no ownership link to this patent.

3. Repeat correspondent across the chain — Cannot be assessed / not present. With only one recorded assignment, there is no second link against which recurrence could be measured, and the correspondent field was not exposed in the retrieved entry (see note above). This is a data gap, not a negative finding — I am flagging it explicitly rather than inferring.

4. Cascading transfers — Not present. Zero consecutive assignments, let alone chained LLCs, in the 20-year life of the patent.

5. Pre-litigation transfer — Not present. No infringement litigation naming US 6803909 was found in any search (RPX litigation insight, general web, and Google Patents "Cited By" all came back clean). With no suit, there is no transfer timed against one. The lone assignment predates the 2004 grant by three years.

6. Bankruptcy fire-sale — Not present. Sony never filed for bankruptcy protection and did not divest this patent in any proceeding. Contrast the enumerated Kodak/Nortel/Polaroid/Sun examples — none applies here.

7. Privateering — Not present. No transfer from Sony to any assertion vehicle, and no Patent Progress / EFF / SEC-filing coverage indicating Sony placed this patent with an NPE to sue competitors.

8. Defensive aggregator (anti-NPE) — Not present. The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. The patent simply stayed with Sony to expiration.

Verdict

Insufficient data (only the original assignment).

The entire recorded chain is a single inventor-to-employer conveyance — Hiroki Tsukahara → Sony Corporation, executed 2001-04-26 and recorded 2001-05-24 at reel 011836/0285 — with no subsequent assignment, security interest, or transfer of any kind before the patent expired on 2021-12-20. Sony is an operating company that shipped a product embodying the claims (Community Place Browser / Bureau, commercially released 1996, U.S. OEM sales announced 1997-01-30) and is not asserting this patent; no litigation naming US 6803909 was found. Every NPE signal is not present; the only caveats are (a) the unnamed recording correspondent, which is a retrieval gap and not evidence, and (b) the fact that a clean single-link chain terminating at an operating company is itself the finding.

Verify at: https://assignmentcenter.uspto.gov/ (search patent number 6803909; the sole entry should be reel/frame 011836/0285) — and cross-check the mirror at https://patents.google.com/patent/US6803909/en under "Legal Events."

I'll research the patent and each of its cited references. Let me start with several parallel searches.

Let me research the remaining four cited references.

Prior-Art Analysis — U.S. Patent No. 6,803,909 B2

1. Patent-number verification (no auto-correction)

Per your strict rule, I searched for 6803909 literally and return only that number. The record I rely on is the Google Patents copy of the USPTO record for US 6,803,909 B2 (source URL: https://patents.google.com/patent/US6803909/en, fetched 2026-09-24), cross-checked against bibliographic sources for the cited references.

Consistent with the earlier litigation section, I ran into the same two matching-number decoys and again do not treat them as this patent: (i) DI 6803909-3 (INPI Brazil design registration, Yamaha motorcycle configuration) and (ii) Notice No. 6803909 (Lithuanian companies-register publication, UAB "Autogazas"). No look-alike number was substituted for the U.S. utility patent.

Important scope note: On the face of US 6,803,909, the only prior art cited is seven U.S. patent documents, each marked as cited-by-examiner. No foreign patent documents and no non-patent literature are cited. So the universe of "patent citations for 6803909" is exactly those seven.

2. The claim framework any §102 reference must meet

For anticipation under §102, a single reference must disclose every element arranged as in the claim. The independent claims cluster into three aspect groups. The critical common element across the server-side claims is:

"generating pixel data of the three-dimensional virtual reality spatial picture image from three-dimensional virtual reality spatial data" plus transmitting that pixel data to the other (client) apparatus.

Note a prosecution-relevant detail: the granted claims recite generating pixel data "from three-dimensional virtual reality spatial data," whereas the pre-grant disclosure's summary language (and pre-grant publication US 2001/0024230 A1) described only "generating pixel data of the three-dimensional virtual reality spatial picture image." This added limitation is a narrowing one, and it makes anticipation by the non-VR references below even less available.

Satellite transmission is claimed only in dependents 4, 8, 12, 18, 22, 26.

3. Reference-by-reference analysis

3.1 US 5,987,480 A — Donohue

• Full citation: U.S. Patent No. 5,987,480 A, Michael Donohue, "Method and system for delivering documents customized for a particular user over the internet using imbedded dynamic content," granted Nov. 16, 1999 (priority Jul. 25, 1996).
• Statutory basis: §102(a)/(b) (patented more than one year before the 2000 priority), and §102(e) (U.S. application filed pre-2000). Verified live (Google Patents, uspto.report, DE10051021A1 citation list).
• Description: Server-side generation of customized documents. A web server stores templates and a data source of name/value pairs; embedded dynamic tags and IF/LOOP flow directives populate templates at request time. Notably, it also discloses automatically identifying the client's browser type (by reading a browser signature) and selecting, from a plurality of templates, the one compatible with that browser, then delivering the resulting document; plus template inheritance through a directory hierarchy.
• Most plausible §102 target: Claims 15–17 (the "identifying part" that decides what to send, and the "storage part … storing data related to the another data processing apparatus"; dependent claim 17's identification "based on … data related to another data processing apparatus").
• Anticipation verdict — NO. The reference never generates three-dimensional virtual reality pixel data and never transmits pixel data of a 3D VR image; its "display data" is a populated HTML-style document. Under the added "from three-dimensional virtual reality spatial data" limitation, no independent claim (and therefore no dependent claim) is anticipated. Best treated as §103 material for the client-capability-adapted delivery concept.

3.2 US 5,982,372 A — IBM

• Full citation: U.S. Patent No. 5,982,372 A, Brush II, Cuomo, Redpath, Lection & Wood (assignee International Business Machines Corp.), "Visual metaphor for shortcut navigation in a virtual world," granted Nov. 9, 1999 (priority Nov. 14, 1996).
• Statutory basis: §102(a)/(b); §102(e). Verified live.
• Description: A VRML virtual world with selectable navigation "hot spots"; selection causes the user's avatar to move (walk or be transported) to predetermined "move-to" coordinates with a designated viewpoint; multi-user hot spots, distance enhancement, developer-configurable routes/speeds. Explicitly requires the user to have a VRML viewer/browser on the client to render the world.
• Most plausible §102 target: Claim 2 (picture image "including an avatar image controlled by the another data processing apparatus") and Claim 3 (image "captured from a fixed point in the three-dimensional virtual reality space," i.e., preset viewpoint).
• Anticipation verdict — NO. Two independent failures: (i) rendering is client-side (the whole point of the VRML viewer requirement), so there is no "generating part" on a server that generates pixel data and no transmission of pixels to another apparatus; and (ii) the "avatar controlled by the another data processing apparatus" element is absent — the avatar is controlled by its own local user. Strong as §103/background for VRML shared worlds, avatars and fixed viewpoints (claims 2–3).

3.3 US 5,877,759 A — Netscape Communications

• Full citation: U.S. Patent No. 5,877,759 A, "Interface for user/agent interaction," Netscape Communications Corporation, granted Mar. 2, 1999 (priority Mar. 26, 1997).
• Statutory basis: §102(a)/(b); §102(e). Verified live.
• Description: A GUI for Internet/intranet software agents — a "funnel" plus pop-up menu, a meta-agent that consolidates plug-in agent capabilities, parsers, and delegation of whole tasks to agents that autonomously collect server/network resources and return results (Java applets).
• Most plausible §102 target: None.
• Anticipation verdict — NO. No 3D/VR rendering, no pixel data, no networked 3D-space provision. Its only tangential bearing on the '909 disclosure is the concept of autonomous agents (the spec's AO servers 13/14 providing robots/electronic pets) — but agents are not claimed. Background only.

3.4 US 6,487,557 B1 — Casio Computer Co., Ltd.

• Full citation: U.S. Patent No. 6,487,557 B1, Shoichi Nagatomo (assignee Casio Computer Co., Ltd.), "Network-access management system and method applied to network and computer program product including computer program recorded on storage medium for creating display data," granted Nov. 26, 2002; U.S. App. 09/218,279 filed Dec. 22, 1998; JP priorities Dec. 26, 1997 and Mar. 18, 1998.
• Statutory basis: §102(e) as of its 1998-12-22 U.S. filing (and §102(a)/(b) via the JP priority/grant). Verified live (Espacenet bibliographic record; uspto.report/PDF).
• Description: An intermediary server between the client and the WWW. It keeps a per-user history table of accessed addresses and, on a connection request, creates display data "in accordance with the access frequency and type of the addresses … and the performance of the client terminal," pastes anchors (or anchor-function images) into that display data, and sends the created display data to the client terminal, which displays it.
• Most plausible §102 target: Claim 15 (identifying part deciding what to send; "storage part" of claim 16; claim 17's identification based on stored client data/request).
• Anticipation verdict — NO, but this is the strongest structural near-miss on the capability-adaptation concept. The server does tailor and generate delivered display data based on client performance — the closest thing in the cited set to claim 15's premise. It fails the claim because (i) the generated content is page/display data with anchors, not pixel data of a 3D VR spatial image generated from 3D VR spatial data, and (ii) it sends no "file required for generating the three-dimensional virtual reality spatial picture image." Best characterized as the lead §103 reference for claims 15–17 and 25's "certain format" negotiation aspect.

3.5 US 6,405,249 B1 — Sony Corporation

• Full citation: U.S. Patent No. 6,405,249 B1, Sony Corporation, "Information processing apparatus and method, information processing system and program providing medium," granted Jun. 11, 2002 (priority Jan. 23, 1998).
• Statutory basis: §102(e) (U.S. application by another filed before the applicant's 2000 invention date); also §102(a)/(b) via grant.
• Description / verification status: ⚠️ I could not retrieve the full text or claim set of this reference — my retrieval budget was exhausted before reaching it. Based only on the bibliographic record, it is a Sony information-processing-system/program-providing-medium patent directed to apparatuses exchanging shared information through a network. I am flagging this rather than describing content I did not verify.
• Most plausible §102 target: Claim 1's preamble ("providing three-dimensional virtual reality space for another data processing apparatus linked to a network") — the element most likely to overlap with a Sony shared-space patent.
• Anticipation verdict — CANNOT CONCLUDE / likely NO. Because it is a same-assignee (Sony) reference, the more probable §102 posture would be §102(e)/§103 within a shared "Community Place" family context rather than a clean anticipation of the pixel-data-generation-and-transmission element. Recommendation: pull the full text (US 6,405,249 B1 claims + spec) before relying on this reference for any anticipation theory.

3.6 US 6,493,001 B1 — Sony Corporation

• Full citation: U.S. Patent No. 6,493,001 B1, Sony Corporation, "Method, apparatus and medium for describing a virtual shared space using virtual reality modeling language," granted Dec. 10, 2002 (priority Sep. 3, 1998).
• Statutory basis: §102(e) (U.S. application filed 1998-09-03, before the applicant's invention date); also §102(a)/(b) via grant.
• Description / verification status: ⚠️ Full text not retrieved (same retrieval exhaustion). Per its title, it is directed to describing a virtual shared space using VRML — i.e., the VRML representation/schema for a shared multi-user 3D space. This is textually the closest subject-matter match in the citation set to the '909 shared server's VRML/Community Place environment.
• Most plausible §102 target: The "three-dimensional virtual reality spatial data" / shared 3D-space-server aspects of claims 1 and 11 (and the VR-file side of claim 21).
• Anticipation verdict — likely NO. On the face of the title, the reference addresses describing the shared space in VRML (the file/representation that a client renders), not generating pixel data on a server and transmitting the pixels to another apparatus. It therefore appears to be the "file required for generating" side of the art, not the pixel-transmission side. Treat as principal §102/§103 background for the VRML shared-space framework. Recommendation: verify against the full claim set.

3.7 US 6,414,677 B1 — Microsoft Corporation

• Full citation: U.S. Patent No. 6,414,677 B1, Microsoft Corporation, "Methods, apparatus and data structures for providing a user interface, which exploits spatial memory in three-dimensions, to objects and which visually groups proximally located objects," granted Jul. 2, 2002 (priority Sep. 14, 1998).
• Statutory basis: §102(e) (U.S. application filed 1998-09-14); also §102(a)/(b) via grant.
• Description: A client-side three-dimensional desktop user interface that exploits human spatial memory, arranging/clustering objects in 3D space and visually grouping proximally located objects.
• Most plausible §102 target: None.
• Anticipation verdict — NO. It is a local 3D user-interface/display arrangement patent with no networked server, no 3D VR space provision to another apparatus, no generation of pixel data to transmit, and no client/server file-vs-pixel negotiation. Purely background on 3D display technique.

4. Consolidated §102 mapping

Bottom line on §102: None of the seven cited references anticipates any claim of US 6,803,909. The reason is structural and consistent across the set: the cited art either (a) renders the 3D/VR world on the client (US 5,982,372; US 6,493,001), or (b) is entirely outside 3D VR (US 5,987,480; US 5,877,759; US 6,487,557; US 6,414,677), or (c) is same-assignee Sony shared-space art whose precise disclosure I could not verify (US 6,405,249). The claimed novelty point — a server that generates pixel data of the 3D VR spatial image from 3D VR spatial data and transmits those pixels to a thin client (and, in claims 15–28, negotiates who renders) — is not met by any one of them. Their collective presence in the file is best explained as §103 combination material (notably US 5,987,480 + US 6,487,557 on client-capability-adapted delivery, and US 5,982,372 + US 6,493,001 on VRML shared spaces/avatars).

5. Adjacent art surfaced in search but not cited on the '909 face — flagged for transparency

• US 6,161,140 (Moriya, Casio Computer) — "System for transferring information between a server and a data terminal through a network": a central facility discriminates a client's model code and sends information in a manner appropriate to the client's capabilities (e.g., display capability/size table). This is arguably an even closer match than cited US 6,487,557 to the "identify the client's capability and adapt the delivered data" concept behind claims 15–17. Caution: this reference is not in US 6,803,909's citation list (it appeared in my search only as an exhibit in an unrelated IPR, IPR2015-02008, of U.S. Pat. 6,924,727). I present it as potentially relevant art to consider, not as a citation on the patent.

6. References that are not prior art to US 6,803,909

The "Cited By" and "Families Citing" entries all post-date the 2000-02-23 priority and therefore cannot be §102/§103 art against this patent. They are useful only as evidence of the patent's technological footprint:

• US 2009/0172557 A1 (IBM, 2008-01-02) — GUI screen sharing between real and virtual PCs.
• Ubitus Inc. series (US 2013/0147818; US 8,736,622; US 2014/0225906; US 9,483,996; US 2017/0039985; US 10,223,997; priority 2011-12-07) — "leveraging GPU resources to enhance performance of an interact-able content browsing service," i.e., the modern cloud/GPU gaming descendants of the server-rendering idea claimed here.

These confirm that the '909 disclosure is an early instance of server-side (now "cloud") rendering, but they are chronologically disqualified as prior art.

7. Caveats

1. Two cited references (US 6,405,249 and US 6,493,001) were not full-text verified — my retrieval budget ended before I could pull their claims/specifications. Their assessments above rest on bibliographic data and titles and are marked accordingly; verify against the full documents before relying on them.
2. Anticipation vs. obviousness. My §102 conclusions are strict (single-reference, all-elements). The cited set is far more useful as a §103 combination, and my per-reference "best target claim" column is the best-case §102 hook, not a finding of anticipation.
3. Pre-AIA §102(e) dates. I keyed reference dates to each reference's U.S. filing/priority date; a rigorous §102(e) analysis would also confirm the applicant's actual invention date and any corresponding JP-family filing dates.
4. Consistent with your instruction, I did not auto-correct any identifier, and I did not substitute the Brazilian design DI 6803909-3 or the Lithuanian notice 6803909 for this U.S. patent.

I'll ground this analysis in the actual prior-art references listed on the patent's page, checking what each reference teaches.

Now let me examine the remaining cited references, especially the ones most relevant to server-side content generation.

Let me characterize the remaining cited references for completeness.

Obviousness Analysis — U.S. Patent No. 6,803,909 B2 under 35 U.S.C. § 103

Note on scope of this deliverable. This is a technical/defensive obviousness assessment based on (i) the authoritative full text of US 6,803,909 and (ii) the seven references listed in the "Patent Citations" / "Prior Art" section of that patent's Google Patents page, as retrieved and summarized in my searches above. I do not have the file wrapper, any examiner's reasons for allowance, or full claim texts of the references (I worked from abstracts and the quoted passages returned by search). Where I reach beyond what I verified, I flag it. This is not a legal opinion.

A. Analytical framework

Pre-AIA § 103 governs (priority 2000-02-23; US filing 2001-02-24). The four Graham v. John Deere factors apply, as reworked by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007): (1) scope/content of the prior art; (2) differences between the prior art and the claims; (3) level of ordinary skill; (4) secondary considerations. Under KSR, a combination may be obvious where the elements were "known in the field," the combination is of familiar elements according to known methods yielding predictable results, or where there was a "design need or market pressure" and a finite number of identified, predictable solutions. Critically for this patent, KSR also permits an "obvious to try" rationale where the field presents a recognized problem with a finite set of solutions.

Claim construction matters here. The whole invention pivots on the distinction between (a) three-dimensional virtual reality spatial data (e.g., a VRML file) and (b) pixel data ("rendered raster image data"). The specification supplies this construction expressly: "If the received picture image data is identified as pixel data, the Community Place Browser causes the VRAM 44 … to store the received pixel data as it is." So "pixel data" = post-render framebuffer content, and "generating pixel data … from three-dimensional virtual reality spatial data" = performing the rendering pipeline at the server. Any § 103 case must therefore show the server doing the rendering, not merely serving the VRML file.

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

A POSITA here would have a bachelor's degree in CS/EE (or equivalent) plus ~2–4 years' experience in networked graphics / distributed computing, with working knowledge of VRML 2.0, the client-server model, WWW/HTTP, and thin-client architectures (NC/X-terminal/remote-display). This is a low-to-moderate skill level for the 2000 timeframe — a fact that increases the obviousness exposure, since server/thin-client trade-offs were well-trodden ground.

C. What the cited prior art actually teaches (and where each is weak)

The key observation: the cited set contains a strong server-side-generation reference (Casio '557), a strong client-adaptation reference (Donohue '480), and a strong VRML/multi-user-VR reference (IBM '372) — but no single reference renders 3D VR data into pixels at the server. That is precisely the gap the patent occupies, and it is the only real § 103 battleground.

D. Element mapping of the independent claims

E. Proposed combinations

Combination I — Primary: Casio '557 + IBM '372 (+ Donohue '480)

Yields obviousness of claims 1, 5, 6, 7, 9, 10, 11, 13, 14 (and, with Donohue, 15–17, 19–21, 23–25, 27, 28).

• Casio '557 supplies every structural/functional element except the 3D-VR character of the image: a server intermediary that generates the display artifact at the server and sends it to the client for display, with generation conditioned on the client terminal's performance.
• IBM '372 supplies the 3D VR/VRML virtual world, avatars, viewpoints, and server-hosted multi-participant world.
• Donohue '480 supplies the "identify the client, then choose what to send" logic (its claim 14 is essentially the identifying part: identify client capability, select the compatible artifact).

Motivation to combine (KSR-satisfying):

1. Express problem recognition, admitted in the patent itself. The Background concedes that generating pixel data from VRML "is dependent on … a CPU processing capacity and a three-dimensional graphics processing circuit," and that with insufficient capacity "the pixel data may not be properly generated at real-time conditions." Casio '557 identifies the identical problem in the web context ("performance of the client terminal") and applies the identical remedy. KSR holds the patentee's own admission of the problem is evidence of the design need.
2. Known technique applied to a known structure. Server-side generation + delivery of rendered output is Casio '557's stated technique; applying it to IBM '372's VRML world is the substitution of one well-known content type for another in a known pipeline, yielding the predictable result of a rendered frame reaching the client.
3. Finite, predictable solutions. Given a low-capability client that cannot render in real time, the field offered essentially two options: (a) ship the VRML and hope, or (b) render server-side and ship pixels. Two choices = "obvious to try" under KSR.
4. Articulated rationale + reasonable expectation of success (the In re Kahn / KSR "articulated reasoning" requirement is met by the above).
5. Adaptation incentive. Donohue's "different browsers get different documents" reasoning maps directly onto "different clients get different content formats" — a design incentive, not a leap.

Combination II — Add Sony '001 (with a § 103(c) caveat)

Sony '001 is the closest architectural match (Community Place server, VRML2.0, Java scripts, avatars) but, like IBM '372, it client-renders. Combining it with Casio '557 gives the exact patent architecture. But see § H — same-assignee status may make Sony '001 (and '249) unusable as § 103 art.

Combination III — Donohue '480 + IBM '372 + Microsoft '414,677

For the 3D-UI/rendering context; weaker, since '677 has no server-rendering teaching. Best used only as a secondary "3D-on-the-desktop was well known" reference.

Combination IV — The "identify" claims (15, 17, 19, 20, 25, 27, 28)

Casio '557 alone reads strongly on the identifying part: its server "create[s] display data according to … the performance of the client terminal," which necessarily requires the server to determine client capability and branch. Add Donohue '480 for the "or … a request from the another data processing apparatus" prong of claim 17 (Donohue's server acts on the client's request and reads a browser signature). The residual difference — transmitting either VRML or pixels — is bridged by Sony '001 (VRML delivery to capable clients) in combination with Casio '557 (pixel delivery to incapable ones).

Dependent claims

• Claim 2 (avatar controlled by the other device) — squarely taught by IBM '372 (avatars are the user's in-world representative). Motivation: multi-participant VR. Strong obviousness case.
• Claim 3 (fixed viewpoint) — viewpoints are expressly disclosed in IBM '372; rendering from a fixed camera is the simplest mode of server-side rendering and is confirmed by the '909 specification's own statement that server-generated pixel data "corresponds to such picture image as viewed from a specific visual (fixed) point." Strong.
• Claims 16–17 (storage of client data; identification based on stored data or client request) — Casio '557 (history/ID table, client performance) + Donohue '480 (signature table, client request). Strong.
• Claims 4, 8, 12, 18, 22, 26 (satellite communication network) — Weakest point. None of the seven cited references discloses satellite delivery. A § 103 rejection of these claims would require an additional reference or an official-notice-type argument that satellite/broadcast delivery of rendered video was a known high-bandwidth channel. Note the specification itself concedes the point: "although the above description refers to the satellite communication network 25 … it is allowable to utilize any high-speed communication facility such as a CATV service line" — a helpful admission of breadth but not itself prior art. Flag: these dependent claims are the hardest to invalidate on this record.

F. The KSR "articulated reasoning" summary

A rejection would rest on: (1) identifying the inventor's own admitted problem (real-time rendering infeasible on low-capability clients); (2) finding a prior-art technique that solves that exact problem in the same client-server paradigm (Casio '557's server-side, client-performance-conditioned generation); (3) finding the VRML/multi-user-3D framework to which the technique is applied (IBM '372); (4) a reasonable expectation of success because the only new step — running the rendering pipeline on the server — was a routine, well-understood operation; and (5) the predictable result that the client simply displays what it receives. Claims 15–20, 21–24, and 25–28 add negotiation/format-selection steps that Donohue '480's "identify the client, pick a compatible artifact" teaching renders obvious.

G. Where the analysis is thin — and why the examiner likely allowed

1. No cited reference renders 3D VR to pixels at the server. Casio renders 2D display data; IBM and Sony ship VRML and render at the client. A challenger must bridge that gap by argument, not by a single reference's express disclosure. That is a real, non-trivial gap.
2. All of the seven references are in adjacent art (web personalization, VRML worlds, 3D UI). Analogous-art arguments are available but the § 103 motivation must be spelled out.
3. The patent's dependence on Community Place server + browser specifics (a Sony product) may have supported a commercial-implementation narrative.
4. No secondary-considerations record (no license, no industry praise, no copy) is visible in the patent page to rebut obviousness — which, if anything, favors a challenger.

H. Two structural caveats a challenger must confront

(1) Pre-AIA § 103(c) common ownership. Both US 6,493,001 and US 6,405,249 are Sony-owned, as is US 6,803,909. Because both issued after the '909 filing date, they qualify as prior art only under § 102(e) (application filings before the invention). Pre-AIA § 103(c) provides that subject matter qualifying as prior art only under § 102(e), (f), or (g) shall not preclude patentability under § 103 where the subject matter and the claimed invention were commonly owned (or subject to an obligation of assignment) at the time the invention was made. For a 2000-02-23 invention date, that is very likely the case for the Sony references. Conclusion: the Sony pair is probably disqualified for § 103 combination purposes (they remain available for § 102 novelty, which they do not defeat). This is why Combination I — built on third-party art (Casio, Donohue, IBM) — is the more robust theory, not Combination II.

(2) Donohue and Casio carry the weight. Donohue US 5,987,480 (issued 1999-11-16) and IBM US 5,982,372 (1999-11-09) and Netscape US 5,877,759 (1999-03-02) are all more than one year before the 2001-02-24 US filing date, so they are solid § 102(b) art with no common-ownership escape. Casio US 6,487,557 (issued 2002-11-26) is § 102(e) art but third-party, so § 103(c) does not apply. This trio is the backbone of any viable § 103 challenge.

I. Bottom line

• Best § 103 theory: US 6,487,557 (Casio) + US 5,982,372 (IBM), optionally with US 5,987,480 (Donohue) — motivated by (a) the '909 patent's own admission that low-capability clients cannot render VRML in real time, (b) Casio's express server-side, client-performance-conditioned generation of the artifact the client displays, (c) IBM's VRML multi-participant world with avatars and viewpoints, and (d) Donohue's identify-the-client-then-select-the-artifact logic. This combination renders claims 1, 5, 6, 7, 9, 10, 11, 13, 14, 15, 16, 17, 19, 20, 21, 23, 24, 25, 27, 28 and dependent claims 2, 3 arguably obvious.
• Weakest claims to attack: the satellite-recitation claims 4, 8, 12, 18, 22, 26 — no cited reference discloses satellite delivery.
• Use with caution: US 6,493,001 and US 6,405,249 (Sony) — same-assignee § 102(e) art likely disqualified for § 103 under pre-AIA § 103(c).
• Overall confidence: moderate-to-high that the core independent claims are vulnerable on the Casio + IBM + Donohue combination, but only if the challenger supplies a persuasive reason why a POSITA would run the 3D rendering pipeline at the server rather than ship VRML — i.e., the motivation must be argued from the admitted problem plus Casio's disclosure, because no cited reference expressly renders 3D VR data to pixels at the server.

Caveats I cannot resolve with available tools: (i) I did not retrieve the file wrapper/examiner's reasons for allowance; a genuine § 103 analysis should verify what the examiner considered and whether any reference was distinguished on the "pixel data vs. VRML file" line. (ii) I worked from reference abstracts and quoted passages, not complete claim texts; the element-by-element mapping in § D should be verified against the full specifications, particularly the Casio '557 disclosure of what "display data" comprises. (iii) I found no litigation or PTAB record for US 6,803,909 (consistent with the earlier sections), so there is no court's or Board's construction of "pixel data" to rely on — the construction in § A is my own, drawn from the specification.