Re: New MS patent: sounds like PG db rules

Justin wrote:

> Only objective is to protect everyone from stupid and ridiculous
> lawsuits. The entire blackberry lawsuit is example of things to come.
> Where another company had a patented that process of moving email to a
> phone for years but never used it. This company waited in the
> background for years for the service to become popular then sued
> blackberry. It cost millions of dollars to defend and it was nothing
> more than legal stealing.

Such behaviour often borders on extortion IMO, with the use of
preliminary injunctions against product sales or other before-proof
punitive measures as threats to encourage an early settlement. Why go to
all the trouble and expense of proving your patent claim's validity when
you can so severely disrupt someone's business without doing so that
they're forced to settle? If you can force a few small companies to roll
over before going after anyone big, so your patent appears to have more
validity, then all the better.

Regarding the blackberry lawsuit:
http://www.infoworld.com/article/06/03/03/76106_030306HNblackberrydisputeover_1.html

"Research In Motion (RIM) and NTP have settled their long-standing legal
battle, with RIM paying NTP US$612.5 million to settle all of NTP's
patent claims against it.

"NTP has granted RIM an unfettered right to continue its business..."

Wow. They granted RIM the right to continue its business. How lovely of
them. Positively generous, even.

This is a very real concern. As the USA has convinced much of the world
to sign patent reciprocity agreements (well, to honour US patents
anyway) as part of free trade treaties, it's not just a worry if you
live or work in the US either. Given the absurd generality of some
patents I'm sure I infringe patents practically any time I write any
code - though I wouldn't know which ones nor would it be wise for me to
find out.

> I can see MS or other company patenting a process that Postgresql has
> used for some time or independently invented it gets sued over as a
> means to extract money from companies and others that used the tool.

If PostgreSQL has been using the process for some time it should ideally
not be a problem. Unfortunately, it still may be, as the USA permits
patents to lurk in pre-approval almost indefinitely (adding claims as
needed) so the patent might apply to fairly ancient ideas despite recent
approval. There's also the whole first to invent vs first to file drama.

Given PostgreSQL's impressive rate of development I'll be surprised if
it *doesn't* run across patent issues. Databases are prime patent
territory, and Pg is picking up impressive modern capabilities with
every release. It's hard to argue that they're obvious, either, even to
someone who's a professional in the field.

PostgreSQL is also becoming powerful enough to be a noticeable threat to
big commercial database vendors' low/mid end markets.

I'm not too worried, personally, as I'd say most patent aggressors will
probably go for MySQL first, at least if their claims apply to MySQL as
well. They, after all, have revenue and a reason to feat patent scare
tactics.

There's also not much to be done about it; it's really up to the
glacially-slow-but-promising patent reform efforts and to an extend the
companies that extend and build on Pg.

--
Craig Ringer