patents and open source

[tmetro-blu-5a1Jt6qxUNc at public.gmane.org (Tom Metro)]

A Linux Magazine article on how software patents are used (in this case, 
specifically by Microsoft) to tamp down open source competition.

Trimming the FAT: Linux and Patents
http://www.linux-mag.com/id/7325/

   Microsoft has long been claiming that there are hundreds of patent
   infringements in the Linux kernel, but has never come to the party
   to tell the free software world what they are. The reason Microsoft
   does this is because they feel that it keeps the balance of power in
   their hands and enables them to use the threat of patents to fight
   the adoption of Linux.

   The truth of the matter is that if the patent issues were made
   known, the free software world would work around them, solving the
   problem. Unfortunately this is a problem Microsoft wants to have and
   so to maintain leverage they aren't publicly disclosing their
   claims. Microsoft has been approaching companies to pay royalties
   over patents, and forcing them to sign non-disclosure agreements so
   as to not release the details of the deals. But this backfired in
   late 2008 when TomTom refused to accept the terms Microsoft was
   presenting and the patents were then thrust into the public
   spotlight.
   ...
   TomTom uses the Linux kernel's implementation of the VFAT file
   system in its devices and when the company refused to meet
   Microsoft's demands for royalty payments, Microsoft in turn filed a
   patent infringement action against them in February 2009.
   ...
   Andrew Tridgell recently submitted a patch to the Linux kernel which
   bypasses the VFAT patent issue in question. The fix addresses the
   core of the long filename issue, creating files on a VFAT partition
   with names longer than the 8.3 standard, that is; eight characters
   for the filename and three for the extension. In hindsight the fix
   is very simple; just don't allow the creation of filenames with more
   characters than the 8.3 standard permits. The patch still allows the
   system to read pre-existing longname files and supports case
   sensitivity. By default however, new files are created in the 8.3
   standard, all in uppercase.
   ...
   Microsoft wants the world to believe that there are major patent
   issues in the Linux kernel, without telling what they are. By doing
   this they are trying to create a cloud of uncertainty around the use
   of free software and as a result are hoping to deter companies from
   taking up the technology over their own. By moving any patent issues
   into the public spotlight however, those in the free software world
   can help break down the cloud of fear, uncertainty and doubt that
   Microsoft is spreading. The future problem will be, are companies
   willing to fight back and make details of the patent claims public?


I'm not sure why the above article, dated May 6th, 2009, didn't mention 
that the patent suit against TomTom was settled back in April:

http://www.sdtimes.com/link/33382

   Star-crossed patent litigants Microsoft and TomTom have settled
   their differences out of court; financial terms were not disclosed.

   The settlement requires TomTom to pay Microsoft for coverage under
   eight car navigation and file-system management patents for all past
   and future sales of relevant products. TomTom has also agreed to
   phase out functionality related to Microsoft's FAT LFN (Long File
   Name) file-system patents within two years.
   ...
   TomTom responded to Microsoft's legal actions with a countersuit,
   alleging that Microsoft's Streets & Trips navigation software
   product infringes on TomTom's patents. Under the terms of the
   settlement, Microsoft will receive coverage for four patents from
   TomTom.


So it seems TomTom caved.

On a more encouraging note...

Experts mull changes to software patent process
http://www.sdtimes.com/link/33433

   In response to the controversy surrounding software patents, the
   United States Patent and Trademark Office (USPTO) has tightened up
   considerably on granting them, and they are now harder to get than
   any other type of patent (except for business methods). That's the
   view of Mark A. Lemley, a professor of law at Stanford Law School
   and director of Stanford's program in law, science and technology.
   ...
   Lemley suggested that one of the easiest reforms for Congress to
   enact would be a system that would give stronger patents to those
   who are willing to go through, "a more searching review in the
   USPTO."
   ...
   James Grimmelmann, an associate professor at New York Law
   School...said that industry peer review would address some of the
   most recurring and serious problems with bad software patents, such
   as when patents are too abstract or in cases where prior art exists.
   "Programmers know the tricks of the trade that are not in the
   universe of documents that USPTO reviewers look at," he said.
   ...
   The Supreme Court also raised the threshold for obviousness in 2007
   when it ruled in favor of KRS, a company that refused to pay
   royalties on a rival's patents on the grounds that the patent
   combined preexisting elements in a predictable manner, said
   Grimmelmann.
   ...
   While patent reform in Congress has proven a long and difficult
   process, Lemley said, he believes that some sort of reform bill will
   pass in this Congressional session.