US 8275897

I'll search for authoritative records on this patent and its litigation status.

Let me dig into the litigation dockets and any 2026-specific activity.

Now let me check the specific CAFC docket numbers associated with this patent and any 2026 activity.

I hit my search-step limit while probing the 2026 docket entries directly, so I'll give you what is authoritatively grounded (the full patent document itself, fetched from Google Patents/USPTO data) and flag clearly where I could not verify.

US 8,275,897 B2 — Patent Summary

Source (authoritative): https://patents.google.com/patent/US8275897/en (fetched 2026-09-28)

Bibliographic data (from the patent record)

Note: these bibliographic items come from the patent page itself; I interpreted the numbers literally and did not correct anything.

Continuity / family chain

The record states this application is a continuation of:

• US 11/230,953 (filed 2005-09-19, now abandoned), which is a continuation of
• US 10/628,801 (filed 2003-07-28, abandoned), which is a continuation of
• US 09/481,243 (filed 2000-01-11, now US 6,604,158), which is a continuation-in-part of
• US 09/266,394 (filed 1999-03-11, now US 6,601,104).

So US 8,275,897 shares its specification with the Realtime "accelerated data storage and retrieval" family (e.g., 6,601,104, 6,604,158, 7,415,530).

Abstract (verbatim)

"Systems and methods for providing accelerated data storage and retrieval utilizing lossless and/or lossy data compression and decompression. A data storage accelerator includes one or a plurality of high speed data compression encoders that are configured to simultaneously or sequentially losslessly or lossy compress data at a rate equivalent to or faster than the transmission rate of an input data stream. The compressed data is subsequently stored in a target memory or other storage device whose input data storage bandwidth is lower than the original input data stream bandwidth. Similarly, a data retrieval accelerator includes one or a plurality of high speed data decompression decoders that are configured to simultaneously or sequentially losslessly or lossy decompress data at a rate equivalent to or faster than the input data stream from the target memory or storage device. The decompressed data is then output at rate data that is greater than the output rate from the target memory or data storage device. The data storage and retrieval accelerator method and system may employed: in a disk storage adapter …; in conjunction with random access memory …; in a display controller …; and/or in an input/output controller…."

Technology in plain language (from the specification)

The patent addresses a practical bottleneck: mass-storage devices (e.g., ~22 MB/s disk drives) were far slower than the PCI bus and processor/memory bandwidth available in 1999–2000. The idea is to place a real-time compression/decompression accelerator in line between a fast data source/sink and a slow storage device, so compression makes the device appear faster ("effective" storage/retrieval rate increase).

Key disclosed elements (as claimed-in-spec / described):

• A data storage accelerator (10) that compresses input data blocks faster than they arrive, then stores them on a slower target storage device (45).
• A data retrieval accelerator (80) that decompresses blocks read from storage and outputs them at a higher rate.
• Lossless, lossy, or hybrid compression selectable via parameters (e.g., a first parameter indicating compression type; a second parameter indicating permissible information loss).
• Multiple encoders/decoders (E1…En / D1…Dn) operating in parallel or sequentially, with a compression-ratio threshold and compression-type descriptor used to tag and later decode each block; encoders that fail to beat a threshold are bypassed (null descriptor).
• Pipeline/"Method 1 vs. Method 2" timing diagrams and dynamic bandwidth adjustment (buffering, throttling input/output, or signaling the other end for a different rate).
• Application embodiments: disk storage adapter, RAM, video storage (FIG. 10), display controller (FIG. 11), and an I/O controller for analog/parallel/serial data (FIGS. 12–15).

Independent claims — ⚠️ important limitation

I do not have the authoritative claim text for US 8,275,897. The document I retrieved is truncated: the description cuts off mid-sentence in the discussion of FIG. 3, and the "Claims" section was not included in the fetched text. I therefore cannot give you a verbatim or reliable plain-language mapping of "each independent claim." Fabricating claim language here would be exactly the kind of error I'm instructed to avoid.

What I can say with reasonable confidence, based on the abstract and summary-of-invention:

• The independent claims in this family are generally directed to a method and/or system for accelerated data storage and a method and/or system for accelerated data retrieval in which data is compressed/decompressed "at a rate equal to or faster than" the input/output rate, so the effective bandwidth of the target storage device is increased.
• Claim content likely includes limitations around lossless and/or lossy compression selection, parameters indicating compression type and permissible information loss, selection of allowable encoders/decoders, and the relation between compression/decompression ratio and the storage-device access rate.
• This is a strong inference from the specification and related family patents — not a reading of the '897 claims themselves. To state the exact number of independent claims and their precise scope, you need the USPTO Patent Center "Claims" document or the printed patent's claim column.

Litigation status

Google Patents' litigation panel (populated from Unified Patents and Darts-IP feeds) lists "Family has litigation" for this patent, with these specific items:

• CAFC case 25-1529
• CAFC case 21-2268
• CAFC case 23-1035
• E.D. Texas, case 6:17-cv-00421

Links as listed: https://portal.unifiedpatents.com/litigation/Court%20of%20Appeals%20for%20the%20Federal%20Circuit/case/25-1529 (and parallel links for 21-2268, 23-1035, and 6:17-cv-00421). The page labels these as family-level litigation, not necessarily litigation over the '897 patent specifically.

Background from my searches on the broader Realtime Data campaigns:

• Realtime's zero-claim-compression assertions against Array Networks, Fortinet, Reduxio, Panzura, Aryaka, MongoDB and others were affirmed invalid under 35 U.S.C. § 101 by the CAFC in Realtime Data LLC v. Array Networks Inc. (nonprecedential), covering the '728, '825, '203, '908, '530, '458, and '751 patents — note that US 8,275,897 is NOT among the patents in that Federal Circuit decision, so that § 101 holding is not directly about the '897 patent.
• Realtime filed a Supreme Court petition (No. 23-498) seeking review of that § 101 affirmance; I did not verify its disposition.
• The family also faced substantial IPR and reexamination activity (e.g., IPR2017-01196 on the '530 patent, institution denied; numerous IPRs on related Realtime patents).

Uncertainty notes (please read)

1. I could not independently verify any specific 2026 CAFC activity on the '897 patent. My direct searches for CAFC dockets 25-1529, 23-1035, and 21-2268 returned no usable results before I exhausted my search steps. The existence of a 25-1529 docket link suggests a 2025–2026 appeal tied to this family, but I cannot confirm what it concerns, its parties, or its outcome. Treat the 2026 docket status as unverified.
2. The '897 patent is recorded as "Expired – Fee Related" with anticipated expiration 2019-03-11. Google Patents explicitly labels legal status as an assumption, not a legal conclusion — confirm with USPTO maintenance-fee records before relying on it.
3. Claim-level scope is unavailable in my source. Do not rely on the inferred claim summary above for validity, infringement, or FTO analysis.
4. I did not auto-correct any identifier; 8,275,897, 11/400,008, and the related numbers are reproduced exactly as they appear in the sources.

If you can supply the patent's claim column (or let me run additional targeted searches on the CAFC 25-1529 docket and the E.D. Tex. 6:17-cv-00421 record), I can give you a precise, claim-by-claim independent-claim breakdown and a verified 2026 litigation status.

I'll search for litigation involving US patent 8275897 using multiple sources.

Let me search more specifically for this patent number in litigation contexts.

Let me dig deeper into the specific cases asserting this patent.

Litigation Involving U.S. Patent No. 8,275,897

Short answer: Yes — litigation is known. U.S. Patent No. 8,275,897 ("the '897 patent," titled "System and methods for accelerated data storage and retrieval," Realtime Data LLC / Realtime Adaptive Streaming LLC, expired–fee related as of its 2019-03-11 anticipated expiration) was asserted in the Realtime Data / Realtime Adaptive Streaming enforcement campaign against DISH/Sling and related entities around 2017. Below is what I could confirm, with explicit flags where the record is uncertain.

Identified cases

1. Realtime Data LLC d/b/a IXO v. DISH Network Corporation et al. — E.D. Texas (the "Second Texas Action")

Sources: Google Patents litigation listing (T.E.D. Tex. case 6:17-cv-00421); EchoStar/DISH SEC Form 10-Q disclosures ("Second Texas Action"); Stanford NPE Litigation Database; Justia/UniCourt dockets (EchoStar Technologies, Dish Network Corporation party pages). Note: the case number is confirmed; the exact division/venue characterization varies slightly across sources.

2. Realtime Data LLC (later Realtime Adaptive Streaming LLC) v. Sling TV, L.L.C., Sling Media Inc., Sling Media L.L.C. et al. — D. Colorado (the "Colorado Action")

Sources: SLWIP case note; National Law Review case note; VitalLaw (IP Law Daily) report; EchoStar/DISH 10-Q disclosures. The Federal Circuit appeal of the fee/exceptionality ruling is Realtime Adaptive Streaming L.L.C. v. Sling TV, L.L.C., No. 23-1035 (Fed. Cir. Aug. 23, 2024) (vacated/remanded).

Important caveat: Although the '897 patent was pleaded in the Colorado Action, no substantive merits ruling on the '897 patent was rendered in that case — the asserted patents that were actually adjudicated were the '610 and '535 patents. I found no judgment of infringement or invalidity specifically on the '897 patent's claims.

Related/family litigation listings (may not be '897-specific)

The Google Patents "family has litigation" flags for this patent family (family 32072870) list:

• Federal Circuit case 25-1529
• Federal Circuit case 21-2268
• Federal Circuit case 23-1035
• E.D. Texas case 6:17-cv-00421

These are family-level links; I cannot confirm from the available material that 25-1529 or 21-2268 directly involve the '897 patent rather than sibling patents in the portfolio.

What I could NOT confirm

• The exact docket number of the Colorado Action.
• Any IPR/PGR challenging the '897 patent itself (the Netflix/Amazon/Hulu IPRs — e.g., IPR2018-01187, IPR2018-01630, IPR2018-01817 — concerned other Realtime patents such as 9,769,477 and 8,934,535, not the '897 patent). The '897 number appears in some PTAB exhibit lists only as a citation/prior-art-style listing, not as the challenged patent.
• Any outcome or judgment resolving the '897 patent on the merits.
• Whether the '897 patent was asserted in the earlier 2008–2010 Realtime Data actions (Packeteer, CME Group, etc.) — those suits predate the '897 patent's Sept. 25, 2012 issuance, so it could not have been asserted there.

Verification note

The two cases above rest on company SEC filings (EchoStar/DISH 10-K/10-Q language) and reliable secondary case reports, and the '897 assignment as U.S. 8,275,897 is consistent across sources. If you need litigation-grade certainty (exact Colorado case number, docket entries), the next step would be a PACER pull for D. Colo. and a check of the Federal Circuit docket for 23-1035, 21-2268, and 25-1529. I did not have PACER/Unified Patents Portal access in these searches, so I am flagging those gaps rather than fabricating numbers.

I'll start by verifying the canonical "no proceedings" signal and probing for any proceedings the ODP may not have indexed.

The general Realtime Data portfolio results are surfacing other family members (the '530, '104, '158 patents share this title). Let me zero in on the '897 patent and its application number specifically.

Key lead: Realtime Adaptive Streaming asserted the '897 patent against DISH/Sling, and DISH filed IPRs on the asserted patents in 2018. Let me pin down the specific IPR numbers and outcomes for the '897 patent.

PTAB / AIA Trial Report — US 8,275,897 B2 ("System and methods for accelerated data storage and retrieval")

Inventor: James J. Fallon · Original/Current Assignee: Realtime Data LLC (d/b/a IXO; later asserted by affiliate Realtime Adaptive Streaming LLC) · Priority: 1999-03-11 · Granted: 2012-09-25 · Application: 11/400,008 (continuation of 11/230,953 ← 10/628,801 ← 09/481,243 [now US 6,604,158] ← CIP of 09/266,394 [now US 6,601,104]) · Status: Expired – Fee Related; anticipated expiration 2019-03-11.

Proceedings overview

The canonical USPTO ODP "PTAB proceedings on file" block for US 8,275,897 returns ZERO AIA trial proceedings — no IPR, PGR, or CBM. Web research corroborates that the FWDs I can positively identify in this family are directed to sibling patents ('535, '610, and the '530/'506/'728-line), not to the '897 patent. There is one caveat worth flagging: in the Colorado litigation, Realtime originally asserted the '897 patent and DISH/Sling stated in SEC filings that it filed IPR "petitions … challenging the validity of each of the asserted patents" on 2018-07-03 — but I could not confirm a distinct IPR number, an institution decision, or a Final Written Decision specifically on the '897 patent, so I am not reporting a fabricated proceeding number. Bottom line for a defendant: the '897 patent carries no confirmed AIA invalidation, but it is (a) expired, and (b) squarely exposed to the § 101 line of cases that killed its siblings. It is a paper tiger, not a hardened patent.

Confidence note: Because I could not verify a specific PTAB docket for the '897 patent, treat the "possible 2018 DISH petition" as an open item to check on PTAB E2E / Docket Alarm rather than a finding.

Proceedings I could positively verify in the family (context, NOT on the '897 patent)

These are not proceedings on US 8,275,897 and are included only so you understand the lay of the land. Do not cite these as FWDs on your patent.

IPR2018-01169 — Netflix et al. v. Realtime Adaptive Streaming ('535 patent)

• Type: IPR
• Status: FWD 2020-01-10 — all challenged claims 1-14 held unpatentable (obviousness). Realtime appealed 2020-03-10.
• Defensive value: Shows PTAB willingness to invalidate this family's core compression claims when timely petitions are filed.

IPR2018-01342 — Google/Comcast (DISH terminated) v. Realtime ('535 patent)

• Type: IPR
• Status: DISH terminated 2020-01-17 as time-barred; proceeding continued with Google/Comcast; FWD 2020-02-27 finding claims 1-6, 8-12, 14 unpatentable.
• Note: Realtime pre-emptively disclaimed claims 15-30 on 2019-01-18.

IPR2018-01331 — DISH/ARRIS v. Realtime ('610 patent)

• Type: IPR
• Status: Instituted 2019-01-31; terminated as time-barred under § 315(b) on 2020-01-17; appeal dismissed for lack of jurisdiction 2021-03-16; rehearing denied 2021-06-28 (see Sling TV, L.L.C. v. Realtime Adaptive Streaming LLC, 840 F. App'x 598 (Fed. Cir. 2021)). Institution decision: https://www.ptablitigationblog.com/wp-content/uploads/2019/03/IPR2018-01331_DocketEntry_01-31-2019_9.pdf

The '897 patent's actual fight was in district court, not the PTAB

• Realtime Adaptive Streaming LLC v. Sling TV L.L.C. et al., No. 1:17-cv-02097-RBJ (D. Colo.). Suit filed 2017-08-31 asserting US 8,275,897; 8,867,610; and 8,934,535.
• Colorado joint status report summarizing the IPR/appeal picture for the five asserted patents: https://www.docketalarm.com/cases/Colorado_District_Court/1--17-cv-02097/Realtime_Adaptive_Streaming_LLC_v._Sling_TV_L.L.C._et_al/docs/172.pdf
• Handled to its conclusion on the '610 patent (not '897): SJ of § 101 invalidity 2021-07-31; affirmed Realtime Adaptive Streaming LLC v. Sling TV, LLC, No. 21-2268 (Fed. Cir. 2023-05-11); exceptional-case/fee appeal Id., No. 23-1035 (Fed. Cir., vacated 2024-08-23); related still-open appeal No. 25-1529 (per the patent page's litigation links).

Strategic summary

Canceled vs. sustained vs. untested on '897 itself. On the record available, no claim of US 8,275,897 has been canceled in an AIA trial, and no claim has been confirmed patentable in an AIA trial either — the claims are effectively untested at the PTAB. That is not good news for the patent owner's negotiating posture, because the "untested" label exists here only because the patent expired (2019-03-11) and stopped being worth a full IPR campaign, not because it survived scrutiny. Confirmed-surviving claims: none established. Confirmed-canceled claims: none established.

Estoppel landscape. With no '897 FWD, there is no § 315(e)(2) estoppel running against any petitioner on the '897 claims — a real tactical advantage. Any defendant not already time-barred under § 315(b) can raise § 102/§ 103 art on the '897 claims, or run a § 101 motion, without the "raised or reasonably could have raised" constraint. The family's § 315(b) history (DISH terminated as time-barred) is a cautionary tale: file within one year of service of the first infringement complaint, and be careful about complaints filed by an affiliate that did not yet own the patent.

Pattern signals. Same-patent-owner family, many petitioners (Oracle, Dell, Riverbed, Apple, SAP/HP, Teradata, NetApp, Rackspace, Veritas, Unified Patents). No evidence the '897 patent drew a Unified Patents petition (Unified's IPR2017-02129 on a Realtime patent was institution-denied). Realtime/Realtime Adaptive Streaming litigated aggressively and pursued appeals, but took adverse § 101 rulings and fee exposure on the sibling '610; note the Feb 2025 remand denying the exceptional-case motion and the pending appeal (No. 25-1529).

The real '897 risk is § 101, not § 102/§ 103. Across this family the Federal Circuit affirmed Alice-based invalidation (Realtime Data LLC v. Array Networks Inc. line), and district courts invalidated sibling claims on the pleadings. Any assertion of the '897 patent today runs through Alice step one with a hostile record of sibling outcomes.

Recommended next steps

1. Verify the '897 PTAB docket directly before relying on the negative. Check PTAB E2E and Docket Alarm / CourtListener for any IPR filed ~2018-07-03 by Sling/DISH (likely an IPR2018-0133x–0134x number) against US 8,275,897, and confirm whether it was instituted and/or terminated as time-barred like IPR2018-01331. I could not confirm such a number and will not invent one.
2. Lead with § 101, not the PTAB. The strongest, fastest kill for the '897 patent mirrors the winning play against the '610: a Rule 12(b)(6)/12(c) or early summary-judgment Alice challenge, citing the FWDs invalidating the '535 claims, Realtime Data v. Array Networks, and the affirmed '610 SJ ruling (Fed. Cir. No. 21-2268). Link the SJ affirmance disposition and the '535 FWD explicitly in your motion.
3. Because the patent expired 2019-03-11, damages exposure is time-limited; frame any demand response around the expiration date and the expired-fee status.
4. If you do petition the PTAB and are not time-barred, you face no estoppel on the '897 claims — a rare clean slate. Ground your § 103 art on the compression-primitive references already run against the family (Hsu, Franaszek, Langdon, etc., per the Teradata v. Realtime record), but expect the owner to argue the specific multi-encoder-selection limitations.
5. If you were served with a complaint asserting the '897 patent, calendar the § 315(b) one-year bar immediately, and check whether any related complaint by a Realtime affiliate (even a non-owner) could be argued to start the clock — the PTAB and Federal Circuit have split on that point and the family's history shows it is outcome-determinative (DISH's institution was wiped out on this exact issue).

I'll research the assignment record and litigation history for US 8,275,897 before drafting.

Let me dig into the specific assignment records and any security agreements in the Realtime Data chain.

Verification note up front (read before relying on the timeline): I was unable to pull a US 8,275,897‑specific record from the USPTO Assignment Center (https://assignmentcenter.uspto.gov/, search by patent number 8275897). The reel/frame entries below are real, citable records from sibling patents in the same Realtime Data continuity family (same inventor, same priority application 09/266,394, same assignee), which is why they are useful for reconstructing the ownership pattern — but I have not confirmed that any one of them is the record that covers application 11/400,008 / patent 8,275,897. I flag each entry accordingly rather than presenting family-level records as if they were patent-specific. Nothing below is invented; where a number is unknown I say so.

Inventors

Unusual pattern: this is an individual-inventor-started portfolio, not a corporate R&D spin-out. Every link in the chain traces back to one man assigning his own applications into his own LLC. The risk profile is the inverse of the "inventors bolt within 12 months" fire-sale tell — here the inventor and the assignee are effectively the same economic actor, which is the classic NPE-founder signature (cf. Stanford NPE Litigation Database tag "Individual-inventor-started").

Original assignee

Realtime Data LLC (New York, NY; later listed at Armonk, NY), also doing business as IXO. Corporate predecessor of record: Realtime Data Compression Systems, Inc. (the entity named in the parent assignment, Reel 009816/0037).

• Primary line of business: data‑compression technology development plus patent licensing/enforcement. There is no evidence in the record of a commercial product embodying the 8,275,897 claims; the CAFC opinion in the family appeal (13‑1098) describes Fallon circulating a white paper in 2002 and demonstrating the concept to potential licensees — i.e., licensing outreach, not a shipping product.
• Products shipped: none surfaced. Enforcement, by contrast, is documented extensively (see NPE signals below).
• Current status: operating as a licensing/assertion entity and still active as a plaintiff as recently as the 2025 CAFC appeal flagged on the Google Patents page (case 25‑1529). The 8,275,897 patent itself is "Expired – Fee Related" — term ran to the 2019‑03‑11 anticipated expiration and maintenance fees were not maintained, so this particular asset was abandoned even as the family vehicle remained active.

Assignment timeline

Family-level recorded chain (Realtime Data continuity: 09/266,394 → 09/481,243 → 10/628,801 → 11/230,953 → 11/400,008):

• c. 1999 (execution date not retrieved) / recorded date not retrieved — Reel 009816/0037 (frames 37–39)

  • Conveyance: Assignment
  • Assignor: James J. Fallon (inventor)
  • Assignee: Realtime Data Compression Systems, Incorporated
  • Correspondent: not retrieved in the source materials
  • Context: original inventor→company assignment of the parent application 09/266,394.
  • Scope caveat: cited in Realtime's own §3.73(b) chain-of-title statement; pertains to the parent, not confirmed for 11/400,008.

• Date not retrieved — Reel 018247/0405 (frames 405–410)

  • Conveyance: Assignment (described in-file as the "mesne assignment")
  • Assignor: Realtime Data Compression Systems, Incorporated
  • Assignee: Realtime Data LLC
  • Correspondent: not retrieved
  • Context: internal reorganization — clean-up transfer from the original compression-systems entity into the LLC that has been the named assignee on every Realtime patent since.
  • Scope caveat: family-level; not confirmed against app 11/400,008.

• Date not retrieved — Reel 011797/0735 (frames 735–736)

  • Conveyance: Assignment
  • Assignor: Fallon et al. (inventor)
  • Assignee: Realtime Data, LLC
  • Correspondent of record on the ancillary power of attorney: Michael V. Messinger, F. Chau & Associates, LLC, 130 Woodbury Road, Woodbury, NY 11797 (Customer No. 1473). Flag: F. Chau & Associates / Messinger recurs as the correspondent handling Realtime's POA filings across the family — a repeat-player recording counsel, though a single firm handling prosecution is weaker evidence than a recording agent on shell transfers.
  • Context: inventor→Realtime assignment for a family member (7,181,608).
  • Scope caveat: confirmed for 7,181,608, not for 11/400,008.

• Executed 2002‑08‑02 to 2002‑08‑07 / recorded date not retrieved — Reel 030877/0211

  • Conveyance: Assignment of assignors' interest
  • Assignors: Fallon, James J.; McErlain, Stephen J.
  • Assignee: Realtime Data, LLC
  • Correspondent: not retrieved
  • Context: routine inventor→company assignment for family member 8,553,759.
  • Scope caveat: confirmed for 8,553,759, not for 11/400,008.

• Effective 2017‑03‑07 / recorded 2017‑12‑20 — Reel 041573/0683

  • Conveyance: Assignment
  • Assignor: Realtime Data LLC
  • Assignee: Realtime Adaptive Streaming LLC (Texas)
  • Correspondent: not retrieved — this is the single most valuable correspondent to obtain; it is the recording agent for the entity that now holds the streaming/video half of the portfolio.
  • Context: transfer-to-asserter / intra-family split-out. Realtime moved a block of family assets into a separate Texas LLC shortly before a fresh wave of 1:17‑cv…/6:18‑cv… complaints (Realtime Adaptive Streaming v. Adobe, Polycom, Samsung, Amazon, Netflix).
  • *Scope caveat: confirmed as a Realtime Data LLC → Realtime Adaptive Streaming LLC transfer on the family (e.g. 8,934,535/8,553,759), but Google Patents still lists Realtime Data LLC as current assignee of 8,275,897, so this patent was likely excluded from the 041573/0683 block. Do not read this entry as a 8,275,897 transfer without confirming the frame.*

If the Assignment Center returns only these family records and no 11/400,008-specific entry, the correct reading is that 8,275,897 remained with Realtime Data LLC from issuance until its 2019 fee lapse — which is itself the finding.

Timeline diagram

timeline
    title Ownership of US 8275897
    1999 : Priority application filed by Fallon
         : Assigned to Realtime Data Compression
    2000 : Continuation in part filed
    2006 : Continuation application 11 400 008 filed
    2012 : Patent US 8275897 issued
    2017 : Family block to Realtime Adaptive Streaming
    2019 : Term expired for fee non payment

NPE / troll-pattern signals

1. Shell-entity transfer — Present (family level). Realtime Data LLC is an individual-inventor-started licensing entity with no product in commerce, and its sister entity Realtime Adaptive Streaming LLC received a portfolio block by assignment at Reel 041573/0683 (effective 2017‑03‑07). For this patent the transfer is unclear — 8,275,897 still shows Realtime Data LLC as assignee, so it likely stayed behind.

2. Known asserter in the chain — Present. Realtime Data LLC d/b/a IXO appears as a patent asserter in the Stanford NPE Litigation Database (party type "Individual-inventor-started," e.g. cases 1:11‑cv‑06698, 1:11‑cv‑06700, 6:17‑cv‑00421) and is typed "NPE (Individual)" in Unified Patents' litigation portal (plaintiff filter "realtime data," cases from 2019–2020 including 1:20‑cv‑01498, 2:19‑cv‑06160). The assignee is not merely NPE-like; it is catalogued as one by both major defensives.

3. Repeat correspondent across the chain — Unclear / partially present. The recording counsel on Realtime's power-of-attorney filings is consistently F. Chau & Associates, LLC (Michael V. Messinger) and the prosecution firms of record on family patents are Sterne, Kessler, Goldstein & Fox P.L.L.C. and Ropes & Gray LLP (with Jeffrey H. Ingerman named on one POA). I could not retrieve the assignment-record correspondents for the reel/frame entries above, so I cannot yet show recurrence specifically on the recording side. This is the highest-value gap to close.

4. Cascading transfers — Present but slow-burn, not the classic <24-month cascade. The chain runs Realtime Data Compression Systems, Inc. → Realtime Data LLC → (portfolio split to) Realtime Adaptive Streaming LLC across 1999→2017 — a long-duration intrafamily restructuring rather than a rapid chained-LLC flip. There is, however, a stable of sibling "Realtime" LLCs (Realtime Data LLC, Realtime Adaptive Streaming LLC, Realtime Array LLC) asserting different family blocks, which is the recognizable NPE portfolio-splitting pattern.

5. Pre-litigation transfer — Not present for 8,275,897. Realtime sued on this family as Realtime Data LLC itself years before any assignment, beginning with Realtime Data v. Packeteer, 6:08‑cv‑00144 (E.D. Tex., 2008), and continuing through the Morgan Stanley / CME / Thomson Reuters / CommVault actions. No upstream LLC transfer was staged within six months of the first suit; the original assignee was the plaintiff.

6. Bankruptcy fire-sale — Not present. No Chapter 7/11 or 363 sale of a Realtime entity appears in the record; the assets moved by voluntary assignment, not court sale.

7. Privateering — Not present. The patents were not spun out of an operating company; the inventor himself was the original holder. There is no operating-company parent left behind to benefit from the assertion.

8. Defensive aggregator — Not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. To the contrary, Realtime is a frequent target of Unified Patents IPRs (e.g. Unified Patents Inc. v. Realtime Data LLC d/b/a IXO, IPR2017‑02129, U.S. Pat. No. 8,717,204). The patent was neutralized only by expiration and fee non-payment (status "Expired – Fee Related"), not by defensive aggregation.

Verdict

NPE — high confidence.

The original and (apparently continuing) assignee, Realtime Data LLC, is a licensing-only entity that has filed dozens of patent-infringement actions as plaintiff (tracked as an NPE by both the Stanford NPE Litigation Database and Unified Patents), and the family shows an intra-family split-out at Reel 041573/0683 (effective 2017‑03‑07) into Realtime Adaptive Streaming LLC — the standard portfolio-segmentation move. Two strong signals (known asserter; licensing-only shell/segment entity) are independently documented, and the sole counter-indicator — that this specific patent appears to have stayed with Realtime Data LLC — does not change the character of the owner. Caveat to carry forward: I could not verify an 11/400,008-specific assignment record, so the exact reel/frame covering this patent, and whether it was swept into 041573/0683, should be confirmed directly.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search 8275897) · Google Patents legal events — https://patents.google.com/patent/US8275897/en · Stanford NPE DB — https://npe.law.stanford.edu/party/realtime-data-llc-dba-ixo · Unified Patents — https://portal.unifiedpatents.com/litigation/caselist?plaintiff=realtime+data

I'll research US patent 8275897 and its cited prior art references.

Let me get more specific citation data from the patent's Google Patents page and related records.

Let me search for the specific "References Cited" section of the patent and its prosecution history.

Prior-Art Analysis — US 8,275,897 B2

1. Patent identified (verified against the authoritative text you supplied)

No other patent number was substituted — everything below is tied specifically to 8,275,897 / application 11/400,008.

2. Important limitation — read before using the tables

Two honest caveats, because the operating rules require me not to fabricate:

1. The authoritative full text you supplied does not contain the front-page "(56) References Cited" section. It runs from the abstract straight into the description (it ends mid-way through the FIG. 3 discussion). Therefore the examiner-cited list for 8,275,897 specifically cannot be read off the authoritative document.

2. My search budget was exhausted before I could open the live USPTO PatentCenter page for application 11/400,008 (the tool returned "maximum number of steps"). I therefore must label the lists below as reconstructed from the co-pending family members (US 6,601,104; US 6,604,158; US 7,321,937; US 7,415,530; US 9,116,908), which share the identical specification and inventor and whose IDS/References-Cited lists are what the '897 front page reproduces. Treat the dates and numbers as reliable; treat the exact 8,275,897 front-page set as not independently verified.

I flag confidence per row.

3. The cited references

Table A — Applicant-submitted IDS references (most relevant set; sourced from the 11/553,426 IDS in the same family)

These are the references the applicant itself put of record, with the examiner's relevance grouping in parentheses. I could not verify every title within budget, so titles are given only where I am confident; otherwise "title not verified."

Table B — Front-page (examiner-cited) U.S. references, as reproduced on the sibling US 7,321,937 (identical spec)

Confirmed verbatim from the sibling's "(56) References Cited": US 4,302,775 (Widergren, 11/1981); US 4,394,774 (Widergren, 7/1983); US 4,574,351 (Dang, 3/1986); US 4,593,324 (Ohkubo, 6/1986); US 4,682,150 (Mathes, 7/1987); US 4,730,348 (MacCrisken, 3/1988); US 4,804,959 (Makansi, 2/1989); US 4,870,415 (Van Maren, 9/1989); US 4,872,009 (Tsukiyama, 10/1989); US 4,876,541 (Storer, 10/1989); US 4,888,812 (Dinan, 12/1989); US 4,906,995 (Swanson, 3/1990); US 4,929,946 (O'Brien, 5/1990); US 4,965,675 (Hori, 10/1990); US 4,988,998 (O'Brien, 1/1991); US 5,028,922; US 5,045,848; US 5,045,852; US 5,046,027; US 5,049,881; US 5,091,782; US 5,097,261; US 5,111,522; US 5,121,442; US 5,150,430; US 5,159,336; US 5,175,543; US 5,179,654; US 5,187,793; US 5,191,431; US 5,204,756; US 5,209,220; US 5,212,742; US 5,226,176; US 5,227,893 …

The sibling US 6,601,104 (the priority patent, same spec) adds: US 5,715,477; 5,721,958; 5,724,475; 5,778,411; 5,787,487; 5,808,660; 5,809,337; 5,812,789; 5,838,996; 5,841,979; 5,847,762; 5,870,087; 5,889,961; 5,915,079; 5,936,616; 5,960,465; 5,968,149; 5,974,471; 5,996,033; 6,000,009; 6,002,411; 6,011,901; 6,014,694; 6,026,217; 6,028,725; 6,032,148.

Table C — Foreign patents and non-patent literature (front-page)

4. Most relevant prior art, ranked, with § 102 mapping

Because the authoritative text cuts off before the claim set, the mapping below is keyed to the independent claims as disclosed in the Summary (a storage-acceleration method; a retrieval-acceleration method; and the "read a first parameter → select an allowable encoder/decoder" feature). I could not verify the verbatim claim text of 8,275,897 within budget, so anticipate vs. obvious is stated as a likelihood, not a conclusion.

1. US 4,988,998 (O'Brien, 1991) — data-compression system that successively applies at least two compression methods to an input stream. Most relevant to any claim reciting selecting among encoders based on a parameter. Potentially anticipates that dependent/sub-feature claim. § 102(a)/(b).

2. US 4,929,946 (O'Brien, 1990) — adaptive data compression including run-length encoding for a tape drive. Relevant to compression-for-storage claims; § 102(b) (issued >1 yr before priority).

3. US 6,008,743 (Jaquette, 1999-12-28) — data compression method/apparatus. Candidate § 102(e) reference (U.S. filing predating the 1999-03-11 priority) against the independent storage-acceleration claim if it discloses compressing data before writing to a storage device whose native rate is lower.

4. US 5,825,830 (Kopf, 1998-10-20) — compressed-data storage. Same theory as Jaquette; § 102(b).

5. US 5,974,235 (Nunally et al., 1999-10-26) — data access/storage. Candidate against storage and/or retrieval independent claims; § 102(e) possible depending on filing date.

6. US 4,876,541 (Storer, 1989) and Storer 1988 book — foundational compression; § 102(b) for the general compression steps but weak on the "input-rate > storage-rate / effective-bandwidth-increase" limitation.

7. US 6,091,777 (Guetz et al., 2000-07-18) and US 6,731,814 (Zeck et al., 2004-05-04) — compression-selection/bandwidth references; may anticipate the parameter/selection feature but their dates must be checked against the CIP claim date.

Key point for the § 102 analysis. The narrow limitations that distinguish 8,275,897 — (a) input data transmission rate greater than the target storage device's storage rate, (b) a compression rate that increases the effective storage/access rate, and (c) reading a first parameter indicative of compression type and selecting at least one allowable encoder — are the limitations any anticipating reference must meet in a single disclosure. The broad "compress then store" ideas are abundantly anticipated; the rate-mismatch + parameter-driven encoder selection combination is where many of these references fall into § 103 (obviousness) territory rather than § 102 anticipation.

5. Prior art actually tested in litigation (highest practical relevance)

8,275,897 is a Realtime Data family member that has been litigated heavily. The grounds actually run against the family (notably Oracle IPR2016-00373/-00374/-00375/-00376/-00377 and the Dell/Riverbed/IPR2016-01671 line on sibling US 7,415,530) relied on: O'Brien (US 4,929,946 / US 4,988,998), Nelson's Data Compression Book, Welch (US 4,558,302), Heath (US 5,973,630), and the Ziv-Lempel/Huffman papers. Notably, IPR2016-00375, -00376 and -00377 were denied institution, while -00373 and -00374 were instituted — i.e., the family survived some § 103 challenges.

6. Bottom line and residual uncertainty

• All citations above are family-level (the references the applicants/examiner put of record for the shared specification); I could not independently pull the dedicated (56) list for 8,275,897 because the authoritative text omits it and my search budget expired before reaching USPTO PatentCenter for 11/400,008.
• The strongest single-reference anticipation candidates are US 4,988,998 (O'Brien) for the encoder-selection feature and US 6,008,743 (Jaquette) / US 5,825,830 (Kopf) for the compress-before-store concept.
• I did not fabricate any patent titles or dates. Rows marked "title not verified" are exact number + date as reproduced in the family IDS/front pages; I declined to invent titles. If you can supply the (56) block from the 8,275,897 front page (or open USPTO PatentCenter for 11/400,008), I can finalize the per-claim § 102 table precisely against the actual issued claims.

I'll research the patent's claims and the prior-art record before analyzing obviousness.

I need the actual claim text and cited references. Let me dig deeper.

Let me get the exact claim language of US 8,275,897 and its cited references.

Obviousness Analysis — U.S. Patent 8,275,897 B2 (Fallon / Realtime Data LLC)

Under 35 U.S.C. § 103 (pre-AIA, given the 1999 priority chain)

0. Sourcing note and confidence flags

Before the analysis, three transparency points required by your rules:

1. The page text you supplied is truncated. The Google Patents rendering in the prompt stops inside the "DETAILED DESCRIPTION" and never reaches the "Claims" section, and it contains no "References Cited" list. The only prior-art material actually on that page is the "Prior art keywords" block (data; compression; data block; output; compressed), the priority/continuation chain, the re-examination note (claim 1 priority claimed from US 09/266,394), and the classification list. So I supplemented with (a) the applicant's own IDS filed in the sibling application 11/553,426 — same title, same family — and (b) the IPR/reexam record for sibling patents in this family. Citations below.
2. Claim 1 text is taken from the RPX litigation record for US 8,275,897 B2, not from the supplied page. I could not independently verify it against the USPTO claim listing because the supplied text is truncated. Treat claim-1 scope below as reported, not confirmed: https://litigation.rpxcorp.com/patent/[US8275897B2](/patent/US8275897B2)
3. I did not verify the internal disclosures of Clark (5,319,682), Crawford (5,771,354), Assar (5,479,638), Fall (5,991,515), or Rynderman (5,563,961) beyond their identification in the record. I flag where I am relying on a third party's characterization rather than my own reading.

Patent status (per the supplied page): priority 1999-03-11 (claiming benefit of US 09/266,394); application 11/400,008 filed 2006-04-08; granted 2012-09-25; anticipated expiration 2019-03-11; status "Expired – Fee Related." Family litigation includes Fed. Cir. appeals 21-2268, 23-1035, and 25-1529, plus E.D. Tex. 6:17-cv-00421.

1. Critical date and governing law

Continuation chain per the '897 specification: 11/230,953 (2005) ← 10/628,801 (2003) ← 09/481,243 (filed 2000-01-11, now US 6,604,158) ← CIP of 09/266,394 (filed 1999-03-11, now US 6,601,104).

• Because the application has an effective filing date before 2013-03-16, pre-AIA § 103(a) governs, with the 1999-03-11 date for subject matter carried in the 09/266,394 disclosure, and a 2000-01-11 date for subject matter that first appears in the CIP (relevant to bandwidth-adjustment / "modify a system parameter" concepts, which the specification introduces in the Fig. 6a–6b / 7a–7b flow).
• The incorporated U.S. Ser. No. 09/210,491 ("Content Independent Data Compression Method and System," filed 1998-12-11, commonly assigned) describes the encoder-module / descriptor architecture. It is commonly owned and shares the inventor, so pre-AIA § 103(c) offers no safe harbor against it if it is "by another"; and because the same architecture is described in the '897 itself, the Applicant-Admitted Prior Art in the '897's own Background is independently usable.

POSITA: an engineer with a B.S. in EE/CS and ~2–3 years' experience in data compression or storage systems design as of March 1999. The field is highly active and incremental (G06F 3/06; G11B 20/00007 "time or data compression"; H03M 7/30).

2. The claim at issue (as reported)

Claim 1 (RPX record): a method comprising — receiving a data stream having a plurality of data blocks; selecting, for a data block, an encoder from a plurality of encoders based at least in part on a data parameter associated with the data stream, the data parameter indicating an amount of information loss permissible; compressing with the selected encoder to produce a compressed data block, the compressed stream including data compression type descriptors indicative of the compression type used for each compressed block; and transmitting the compressed stream over an output, wherein the combined time to compress and transmit is less than the time to transmit the received stream uncompressed.

Elements to be met: (a) encoder selection from a plurality; (b) selection keyed to a permissible-information-loss parameter; (c) per-block type descriptors in the compressed stream; (d) the time-superiority limitation (compression + transmission < transmission alone).

The confirmed-independent claims in the sibling '530 patent (reexam certificate US 7,415,530 C1, citing claim 1 confirmation and new claims 24–26) recite the storage-side analogue: a data accelerator + memory device, multi-technique compression, bandwidth/rate adjustment by modifying a system parameter, descriptors, and the "compression and storage occurs faster than … in said received form" limitation. That tells us where the family's non-obviousness defense has historically rested — on the time-superiority / bandwidth-compatibility limitation, not on the multi-encoder selection.

3. The prior art of record

3.1 References cited by the applicant (IDS in sibling app. 11/553,426, filed 2008-01-14)

Key entries: 4,988,998 (O'Brien); 5,091,782 (Krause); 5,434,983 (Yaso); 5,483,470 (Alur); 5,563,961 (Rynderman); 5,574,953 (Rust); 5,611,024 (Campbell); 5,638,498 (Tyler); 5,640,158 (Okayama); 5,686,916 (Bakhmutsky); 5,825,830 (Kopf); 5,864,342 (Kajiya); 5,867,602 (Zandi); 5,974,235 (Nunally); 5,978,483 (Thompson, Jr.); 6,008,743 (Jaquette); 6,091,777 (Guetz); 6,195,465 (Zandi); 6,225,922 (Norton); 6,272,178 (Nieweglowski); 6,392,567 (Satoh); 6,404,931 (Chen); 6,577,254 (Rasmussen); 6,604,040 (Kawasaki); 6,661,845 (Herath); plus 5,832,126 (Tanaka); 5,836,003 (Sadeh); 5,838,996 (deCarmo); 5,839,100 (Wegener); 5,841,979 (Schulhof); 5,847,762 (Canfield); 5,861,824 (Ryu); 5,861,920 (Mead); 5,867,167 (Deering); 5,870,036 (Franaszek); 5,870,087 (Chau); 5,872,530 (Domyo); 5,883,975 (Narita); 5,886,655 (Rust); 5,889,961 (Dobbek); 6,330,622 (Schaefer); 6,345,307 (Booth); 6,421,387 (Rhee); 6,434,168 (Kari); 6,434,695 (Esfahani); 6,442,659 (Blumenau); 6,449,682 (Toorians); 6,452,602 (Morein); 6,463,509 (Teoman); 6,487,640 (Lipasti); 6,489,902 (Heath); 6,513,113 (Kobayashi); 6,529,633 (Easwar); 6,532,121 (Rust); 6,539,456 (Stewart).
Source: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1496876](/patent/1496876)/download-documents?artifactId=KjxR3AyyK0-Bgmm4ImQ9mh6B5AAzGMmoqQXbuPQzgBOGKzjWyPAUJj0

3.2 References relied on in the family's IPR record ('530 patent)

Petitioner's expert identified: Franaszek, US 5,870,036; Osterlund, US 5,247,646; Rynderman, US 5,563,961; Fall, US 5,991,515; Clark, US 5,319,682; Crawford, US 5,771,354; Assar, US 5,479,638; and Kawashima (JP application/translation).
Source: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1461658](/patent/1461658)/download-documents?artifactId=oxNvQMntH-xmy2CmCEeXit51rr-09SonUntKQmuBKr-TxDK7mf-maj4

3.3 What these teach (with quotations)

Franaszek (US 5,870,036) — "A system and method for compressing and decompressing data using a plurality of data compression mechanisms. Representative samples of each block of data are tested to select an appropriate one of the data compression mechanisms to apply to the block. The block is then compressed using the selected one of the mechanisms and the compressed block is provided with an identifier of the selected mechanism." It uses run-length, arithmetic, and dictionary/LZ1 coding; uncompressed blocks 210 are compressed by data compressor 220 and stored as compressed blocks 230; a CMD field 235 encodes the selected method (and dictionary). This maps almost verbatim onto claim 1 elements (a) and (c) and to the storage claims of the family.

Osterlund (US 5,247,646) — a data compression device "is interposed between a host computer and an optical disk controller to permit data storage and retrieval operations on an optical disk to occur at a faster rate than would otherwise be possible." This is the acceleration rationale itself.

Kawashima — a compression/transmission system that selects between compressed and uncompressed transmission, with the petitioner mapping the time-superiority limitation to Kawashima's explicit accounting of compression time (t_B1) plus transmit/store time (t_A1) versus uncompressed time (t_A1), i.e., t_AC1 < t_A1. https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1461666](/patent/1461666)/download-documents?artifactId=jtAvkmVsa3gmav0cGfkDo7IZRZwLgGoyaCcuiWdFQ8mA-7zEerQtwDI

Applicant-Admitted Prior Art in the '897 Background — lossy vs. lossless compression, negentropy limits, and the practice of lossy encoders "delet[ing] information content in excess of the display resolution or contrast ratio of the target display device." This is the specification's own admission that lossy/lossless selection based on permissible information loss was known, which is precisely claim 1 element (b).

Claim 1 priority / reexam note on the supplied page: priority claimed from US 09/266,394; and the '530 reexam determined that the requester's references did not disclose "said compression and storage occurs faster than said data stream is able to be stored on said memory device in said received form" — the PTO's own identification of the crux.

4. § 103 combinations

Combination A — Franaszek + Osterlund (primary; anticipates/renders obvious most storage-flavored claims)

• Franaszek supplies: a data stream of blocks; a plurality of compression mechanisms; per-block selection of an encoder; a type identifier (CMD) stored with each compressed block; storage of the compressed stream.
• Osterlund supplies: the placement of the compressor between host and storage controller for the express purpose of making storage/retrieval faster, i.e., the acceleration insight.
• Motivation (KSR, design incentives/market forces): both references address the same problem the '897 Background itself identifies — storage media are far slower than the host bus (disk ≈ 22 MB/s vs. PCI ≈ 528 MB/s). Franaszek optimizes which compressor to use per block; Osterlund optimizes where to put the compressor. A POSITA seeking faster effective storage would naturally pair a per-block best-encoder selector with a between-host-and-device compressor. Both are in the same field (data compression for storage), the combination is a mere arrangement of known elements, and the result (higher effective storage bandwidth) is predictable.
• This combination squarely meets reported claim 1 elements (a) and (c), and (d) via straightforward time arithmetic: if you compress before store and the compression ratio exceeds the ratio of input rate to storage rate, total time shrinks — arithmetic the '897 itself performs ("if the storage device … is capable of storing 30 MB/s and the [accelerator] is capable of providing an average compression ratio of 3:1, then 90 MB/s may be accepted as input").

Combination B — Franaszek + Osterlund + Kawashima (targets the time-superiority limitation directly)

Kawashima explicitly compares compress-plus-transmit time to transmit-uncompressed time and chooses the compressed path only when it wins. Adding Kawashima to A supplies the express time-superiority teaching that the PTO treated as the distinguishing feature in the '530 reexam. Motivation: Kawashima addresses the identical design tension ("reduction in data if the time to compress the data is significant and time is an important consideration"), so a POSITA optimizing end-to-end latency would consult it. Result again predictable.

Combination C — Franaszek + Osterlund + Rynderman / Zandi / Kawasaki (targets the permissible-information-loss parameter, element (b))

Claim 1's "data parameter … indicating an amount of information loss permissible" is, in substance, rate/quality control: choosing a lossy vs. lossless path and a quality target. The '897 Background admits this; Rynderman, Zandi (5,867,602 / 6,195,465), and Kawasaki (6,604,040) — all in the applicant's own IDS — are lossy/lossless imaging and video coding references in which a quality/loss parameter governs encoder choice. Motivation: the '897 spec's own worked example (an incoming video stream with information value 7FFF invoking an encoder that reads display resolution) is just this known rate-control paradigm applied to a storage path. Predictable result; the parameter is a design choice with an articulated rationale.

Combination D — adding a timing module (dependent claims)

The spec's "timer … to measure the time elapsed during the encoding process against an a priori-specified time limit" is a known real-time scheduling technique, and Franaszek already contemplates sampling to bound per-block work. Applying a deadline to encoder selection to preserve real-time behavior is routine optimization (KSR: "a court must ask whether the improvement is more than the predictable use of prior art elements according to their established functions").

5. Why a POSITA would have been motivated (KSR factors)

1. Identical field and problem. Every reference is directed to compression for storage/transmission bandwidth; the '897's stated problem (media bandwidth far below host-bus bandwidth) is the same problem Osterlund and Kawashima address.
2. Predictable combination of known elements. Multi-encoder selection (Franaszek), compressor-before-storage (Osterlund), and descriptor-tagged blocks (Franaszek CMD field) each perform their known function; the aggregate performs no more than the sum.
3. Market/design pressure. The Background section documents an industry-wide race for effective storage bandwidth (SCSI-3, Fibre Channel, striping). KSR holds that such pressure is itself a motivation.
4. No teaching away. Nothing requires a single fixed codec or forbids descriptor tagging; Franaszek affirmatively teaches tagging.
5. "First/second" limitations are not limiting. In the family's IPR briefing, the petitioner correctly invoked Traffix Devices: "first" and "second" merely distinguish repeated instances and do not impose seriality.

6. Weak points in the obviousness case / counterarguments

• The time-superiority limitation is the family's historical redoubt. In the '530 reexam, that exact limitation was found not to be taught by the requester's references, and claims 1, 2, 16–21 and 23 were confirmed, with new claims 24–26 added. Any § 103 attack must present an express (not inferential) teaching that compress-time + store-time < store-uncompressed time. Kawashima is the best such reference; Franaszek and Osterlund alone may be attacked as only implying it through arithmetic.
• Critical-date exposure. Franaszek issued 1999-02-09 — one month before the 1999-03-11 priority date. Under pre-AIA § 102(e) it is prior art based on its 1995 filing date, so it qualifies. But if any '897 claim depends on subject matter first disclosed only in the 2000-01-11 CIP (e.g., "adjusting a data rate by modifying a system parameter"), the relevant art set broadens.
• Antedating. For a § 102(a)/(e) reference, Realtime could attempt a 37 C.F.R. § 1.131 or § 1.132 antedating showing. That is not available for the on-sale/printed-publication art.
• Secondary considerations. The family has been litigated extensively (E.D. Tex. 6:17-cv-00421; Fed. Cir. 21-2268, 23-1035, 25-1529) and § 101 rulings in the Carbonite case found the claims non-abstract, describing an "improvement in computer capabilities." Commercial success, copying, and industry praise in the deduplication/backup-appliance market could be asserted — but the nexus between the claim-1 elements (multi-encoder selection + descriptors) and any such success is weak, because those features are straightforwardly Franaszek.
• Independent § 112 issues (out of scope but relevant to overall vulnerability): the "data parameter … indicating an amount of information loss permissible" is described in the spec only in descriptor terms (binary flag; 16-bit FFFF/FFFE–0000 values). Claim 1 recites the parameter as "associated with the data stream," which invites a written-description/enablement challenge.

7. Bottom line

On the record above, claim 1 and its storage-/selection-oriented dependents are, in my assessment, more likely than not obvious under § 103 over Franaszek (US 5,870,036) in view of Osterlund (US 5,247,646), and — for the time-superiority limitation specifically — further in view of Kawashima. Franaszek supplies the plurality of encoders, the per-block selection, and the per-block compression-type identifier stored with the data; Osterlund supplies the express motivation to compress-in-line-to-store-faster; Kawashima supplies the express accounting showing compress-plus-transmit time less than uncompressed transmit time. The '897's own Background supplies the admitted lossy/lossless and permissible-loss paradigm that meets the "information loss" parameter element. Under KSR, the combination is a predictable use of known elements with an articulated, market-driven rationale.

The strongest non-obviousness position for the patentee remains the time-superiority/bandwidth-compatibility limitation, which the PTO upheld in the '530 reexamination; the weakest is the multi-encoder-selection-plus-descriptor subject matter, which Franaszek discloses nearly verbatim.

Confidence: High on Franaszek/Osterlund content (direct quotations from the IPR record). Medium on the exact scope and numbering of the '897 claims, because the supplied page was truncated before the claims and I relied on the RPX litigation record. Low-to-none on the internal disclosures of Fall (5,991,515), Clark (5,319,682), Crawford (5,771,354), Assar (5,479,638), and Rynderman (5,563,961) — I did not read those references and am only reporting that the petitioner's expert relied on them.