Patent Information Users Group, Inc.

This was posted on a listserv this morning and I thought everyone would be interested in it.

I recently found that a WIPO A1 document downloaded from Google Patents (patents.google.com) was not the same as the A1 document available directly from PatentScope. That may be old news to you, but it was news to me.

The document available from Google Patents (via the “Download PDF” link) was already OCR’d — how convenient! — except that the OCR step introduced typographical errors throughout the document.

The same OCR errors were evident under the “Description” and “Claims” tabs on the PatentScope page; the “official” version, under the “Documents” tab, was pristine.

It seems that the International Bureau provides OCR’d data to Google, and then Google Patents uses that to make a downloadable version that looks like a true copy (but is not).

Caveat emptor.