Microsoft's Acquisition of Powerset

[Miles P. Jennings, Jr.]

Charlie Hoffman's first video, "How XBRL Works," is a clear presentation of converting unstructured text into computer-readable structured text or XBRL. (See Video) In a real sense, Powerset is making it possible to, as they say, "search for meaning." For a very clear statement on Powerset's service, check out the article by Juan Carlos Perez on CIO.

Microsoft has had a longtime interest in XBRL, an interest in structuring text and making it computer-readable. Microsoft's partnerships with XBRL companies make it possible to perform deep searches into the financial reporting of SEC-filing companies.

Here is a statement from Perez's article:
"'We're changing the way information is searched by doing a much deeper analysis of the pages we index,' said Scott Prevost, Powerset's product director.

'Keyword engines treat pages as word bags, indexing their content without grasping its meaning,' he said. Meanwhile, Powerset's engine, applying technology developed in-house as well as licensed from Xerox's PARC subsidiary, creates a semantic representation by parsing each sentence and extracting its meaning. 'Meaning is what we index,' he said."

The public SEC database of XBRL-ized financial statements of SEC-filing companies is just now starting to be built. It has taken almost a decade of hard work to build the taxonomies and design the operating software. Over time, smart searching of that database will allow better understanding of the companies' financial reports.

The recent acquisition of Powerset by Microsoft seems like a smart move to challenge and improve on the keyword search engines. In the same way, structured XBRL text will allow better and deeper searching of financial reports.

Although I have a sense that there is a clear similarity with Powerset's goals and that of XBRL developers and users, I hope some of our members will provide comments on the semantic value of XBRL in understanding financial reporting.