Laboratory notebooks carry a printed line at the bottom of each page: "Disclosed to and Understood by." A colleague — technically qualified but uninvolved in the work — reads the entry, and if they can follow it, signs and dates.
The interesting word is "understood." The system didn't ask for acknowledgment of receipt. It demanded evidence that a second qualified person actually grasped the content. And under U.S. patent law, the witness's date established the date of invention. Your entry proved you'd written something down. The countersignature proved someone else could parse what you meant, on a specific day, and could later testify to that fact.
We now generate enormous, tamper-evident records of what happened and when. What we almost never capture is whether anyone understood what it meant at the time it was recorded. The witnessing practice was a small, deliberate fix for exactly that gap. As of 2014, there was still no well-accepted method for replicating it electronically.
The wording: "Disclosed to and Understood by:" or "Read and Understood by:" — printed at the bottom of each notebook page.
Who qualifies: A non-inventor with enough technical background to follow the science. If you can't parse the work, you can't serve as witness.
Timing: Same day preferred; within one week at most.
Blank-space rule: Any empty area on the page must be crossed out before the witness signs, blocking later additions.
Conception vs. practice: Attesting to an idea required the witness to have read and understood it. Attesting that work was performed required only that they observed it.
After 2013: The U.S. shifted to "first to file," reducing the legal stakes. The practice persists for validity challenges and derivation disputes.
The digital gap: A 2014 study found it "virtually impossible" for a witness to prove their electronic notation wasn't inserted by someone else.