[plug] NEW M$ Patent?

[Harry]

On Fri, 27 Feb 2004 14:20:16 +0800 "Bennett, Phillip" <PBennett at arg.net.au>
wrote:

> Has anyone seen this today?
> 
> http://www.theinquirer.net/?article=14372

> Does anyone have any more info / thoughts?

One /. comment explains:

"
MS clearly states that there is prior art, which makes me think that they
aren't patenting desktop pagers but some kind of enhancement to it. (I only
had a quick glance at the patent)
"

Often companies (or their competitors) will seek to surround an existing
patent with patents which extend the idea (that is, define a new feature
or innovation that extends but falls outside of the existing patent.
I think this is called patent mosaic-ing.

If the new patent applicant is a competitor, this can serve a few purposes:

1. To prevent the original patent holder from making improvements which
they can use in the market against that competitor,

2. Another option is that the competitor negotiates for an exchange of
patent licences so that they can join the original patent holder in a
product market,  or

3. To prevent the original patent holder from creating a new patent which
overlaps the old one when it expires and so extending the exclusive income
horizon of the invention.  

The original patent holder creates a mosaic of surrounding patents for
similar competitive advantage. Sarich Motor Corp has a huge patent
portfolio for these sorts of reasons.

Since Microsoft's threat is innovation that it cannot thwart in the courts
an obvious approach when you have plenty of funds is to apply this mosaic
to existing ideas; that is, the prior art in the public domain. This hems
in the existing published approach to prevent it from being developed
further.

Also, patents sometimes contains documented examples of prior art to also
clearly demonstrate what the patent is _not_ defining in order to prevent
the patent being challenged with the usual prior art arguments.

I'll let you figure out what Microsoft is doing :-)

To extend the discussion a bit ..

I jumped up at the LCA2003 Forum and asked the first question. BDale had
given a talk which included the softwware radio project (signal processing
to recieve radio by digitising the incoming RF without downconverters etc).
The people doing this project were aware that it was a breaking technology
and, like the Genome project, wanted to do work to try to force the core
concepts, knowledge, and techniques into the public domain.

I noted this and asked why the community doesn't just go crazy with rough
code for new ideas (I referred to kernel techniques) to thwart the
accelerating lodgement of patents as a commercial exploitative tool.

I knew nothing about SCO and the impending nuiscence that they became but
Andrew Tridgell explained that arguments about kernel code was more to do
with existing code rather than patents. I now (!) see his point but his
response didn't tackle the patent question.

But Andrew does something that is actually aligned with this idea of making
any utilities and ideas public and that is his junk code directory which
he went through at LCA2004 I believe (I wasn't there). This has test
utilities and other useful stuff that he has hacked up that may be 
reused for his development work but it also a repository of prior art. 

If you are a coder, particularly of GUI or MMI code of any sort, then I'd
strongly recommend that you too maintain a junk code directory for all the
little scrappy ideas that you try or tests that you make. Make sure that
mm.c (or mickey_mouse.pl or whatever) is dated and copyrighted for when
it's going to needed :-)

cu
Harry

-- 
Are you a computer angel?	http://www.computerangels.org.au/