About Patent Case Watch

Patent Case Watch is a free, public research starting point: a browsable, searchable index of patent-case opinions and orders from the U.S. District Court for the Eastern District of Texas and the U.S. Court of Appeals for the Federal Circuit. Every document comes from an official government publication. Most come from the govinfo.gov USCOURTS collection, which the U.S. Government Publishing Office maintains. Federal Circuit decisions from the years that collection does not reach come from the Federal Circuit’s own opinions and orders archive. The PDFs stay on those government sites — this site does not re-host them.

This site last refreshed the full index on September 10, 2026. Each detail page carries its own retrieval date, which is the day this archive last read that particular document, not the day the index was last refreshed. The archive re-checks both sources on a rolling basis, and new decisions appear at the top of the feed.

Patent Case Watch is not legal advice, not a complete legal database, and not a substitute for reading the opinions themselves. The Terms of Use set out what that means in full.

What Our Archive Does Not Cover

Neither source publishes complete dockets. Both carry the opinions and orders a court chose to transmit, and both omit briefs, exhibits, hearing transcripts, and some orders. A case’s absence says nothing about whether it exists.

Eastern District of Texas coverage runs from roughly 2005 onward and follows what that court sent to govinfo. Federal Circuit coverage is split: govinfo holds almost nothing from that court before 2013, so this archive takes the earlier years from the court’s own published record instead. Each page names the source it came from.

How We Classify Records

A program reads the opinion text and, where the source supplies one, the court’s docket-entry line, and extracts the bibliographic fields: decision date (from the signature block, the “Decided:” line, or the filing stamp), judge, and document type. Documents cross-listed under several consolidated appeal numbers appear once, and the record notes the other docket numbers.

The two sources describe a document differently. govinfo carries the clerk’s docket-entry line, which states the disposition outright and identifies Federal Circuit Rule 36 summary affirmances. The court’s own archive gives the appeal number, the court appealed from, and whether the decision is precedential, but no disposition line. Records from that source are therefore classified from the opinion text alone.

An AI language model assigns the holding and the subject-matter tags. It reads each opinion, alongside the docket-entry line where one exists, and chooses from a fixed list — it cannot invent a category. Where it cannot determine the disposition, the record says “See Opinion” rather than guessing. A small number of records also carry an AI-assisted editorial summary, labeled as such on the page. No lawyer reviews any of it. These are research filters, not legal characterizations, and they can be over- or under-inclusive — always confirm against the opinion itself.

Known Limitations

Some documents were published as scanned images with no text layer. Those pages say so and offer the PDF as it is. Where a source does not serve a document at all, the record shows “source document unavailable” and offers no download. Case names appear as they do in court records, lightly normalized for capitalization.

Corrections

Found something wrong in a record? The Contact page explains how to report it.