HP focuses on patent quality, IBM on quantity

Microsoft may get a lot of grief for its patent strategy, but IBM carries the biggest patent portfolio by far. In fact, Big Blue has acquired more patents than any other company on the planet for the last 16 years running.

This is particularly interesting when you discover, as The Wall Street Journal recently did, that Hewlett-Packard, one of IBM's fiercest competitors, is actually slowing its patent applications. In HP's words, its emphasis now is on patent quality, not quantity:

HP was once trying to compete with IBM as the most prolific patent producer...But after Chief Executive Mark Hurd arrived at HP in 2005, the Palo Alto, Calif., company spent less time and money filing new patents as the CEO strove to make HP more efficient, said Kevin Light, a deputy general counsel and vice president for intellectual property at HP.

HP has been focusing on "the quality of the patents that we seek, as opposed to the quantity," Mr. Light said. As a result, he added, HP is seeking broad patents that relate directly to its main businesses, avoiding the costs of filing patents that may relate to more specific processes....HP seems to be keeping secret certain innovations, especially in its services division, rather than filing patents.

This strategy is a far cry from IBM's position, which has been to seek a wide array of patents on just about everything, some of which have been downright silly, like its patent on, wait for it, patent trolling. Or how about the patent on offshoring?

That said, IBM has also been generous in granting patent protections to open-source projects. I just wish that more companies would follow HP's lead: either limit your patent ambitions or stop seeking them altogether. However, in our world of patent lawsuits, that's probably not going to happen, as companies feel the need to build up a defensive patent portfolio.

[drosenthal7757]

Or how about a patent on grammar checking?

"Or how ABOUT the patent on offshoring?"

[MadLyb]

This is a great and noble ideal, but how do you measure the 'quality' of a patent?

For patents that have direct product impact, it should be easy, but most patents are for future products and have little concrete ROI on day one.

[Commander_Spock]

How about developing a parent to be used to help solve the problems in the U.S. Housing and Financial markets!!!

[kwhsy82]

Having worked at a large, WA-based software company:
1. Better patents including focusing on actual areas of interest to your company; the likelihood they will stand up (you can scan databases for relevant ones); relevance to important computer trends; sophistication of technology that they represent.
2. It takes time/money to patent things. It can take a developer a week to write up and document a patent. HP and IBM's decisions may simply reflect how they want to allocate resources. I don't think there is a 100% correlation between "how many patents you apply for" and "what you do with them". Five crucial patents in a hot area like cloud computing and you can be quite aggressive.

[Harrity]

Re:"HP Focuses on patent quality..." In an uncertain economic climate like we are experiencing today, HP's decision to severely curtail its research and development (HP has gone from #3 to #10 in the past several years in patent development) should be alarming to investors and others. Now more than ever, innovation (via research, etc.) is needed more than ever to deal with some of the globe's greatest challenges.

It is clear that IBM is committed to innovation through its technology leadership (it dwarfs the competition) and, equally important, is sharing the results of its research (ie patents...) with thousands of other researchers worldwide. HP seems to be circling the wagons in search of short-term gains instead of looking to the future.