[ILUG] Enda Kenny and Patents

[Justin Mason]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


Enda writes:
> > the argument that pro-software patent people make is that software
> > patents are required for innovation.  that argument is demostrably
> > false.
> 
> You've picked up the argument wrong. The argument is that software patents
> are required to protect the financial investment made in an R&D process. 10
> man team working for 3 years on an R&D project and they turn out a solution
> at a cost of 30 man years development work just so that anyone else can reap
> the benifits at their expense? I don't think so.... thats exactly whats
> hampering true inovation in the Irish software industry (as was aptly
> described by Chris Horn circa '97 in his famous speach to the Irish software
> industry) which leaves an industry of implementors rather than innovators.

Got a link for that speech?

That doesn't match up with my experience, either during my time developing
software at Iona, or with my own software development since.

In my experience, the *research* for computer software engineering is very
minor, compared to that required upfront in other industries.  Then
*implementation* takes a lot of hard work and time.

Also, can you define what you mean by "innovation" that makes it different
from "implementation"?

Often, in my experience, a very good, very useful, strong, innovative
software product isn't innovative at the level that software patents
cover.  Instead, it's innovative as a greater whole, and half of what
makes it great is implementation, API and user interface, and their
cohesiveness. (e.g. Firefox vs. MSIE -- nothing strictly new there, just
put together a lot better and more efficiently.)

And OTOH, products that include innovative stuff at the
software-patentable level, can be gigantic, unusable turds.   (e.g.
Clearcase, which uses some great technology under the covers, and
then wraps it in a horrible slow inefficient unreliable mess.)

So in other words, they're not really "innovative" at a core level.
They're instead a very nicely designed *implementation*.

Maybe the Irish software industry has difficulties producing these,
or did at least, but it's not the kind of innovation that software
patents protect.

> > uh, you're not getting the point.  yes, a patent is a gov't granted
> > monopoly.  however it is granted with the goal of encouraging innovation.
> > however the software field was already generating a huge amount of
> > innovation before software patents even existed.
> 
> And the resulting trends where the innovators were swallowed by the fat cats
> without due recompense has left the industry in a situation where we'd
> rather implement than innovate.

Who?  which innovators are these that we've lost?  Damn, wish I'd heard
about that terrible event! ;)  Did the innovators who designed Orbix for
Iona get "swallowed"?  Nope, they went on to start new companies,
including Cape Clear and others, in turn creating new innovation.
Somehow, they did this without patent protection in Ireland ;)

That's just from my experience; I'm sure there's lots of other stories
from other companies.

> > uh, no.  the link you point out has a few flaws.  first, the credited
> > inventor of the spreadsheet is not the person who got the patent - all
> > the way back in 1961.  second, visicalc came a while after that patent
> > expired.  i doubt visicalc's creator even knew about the patent.
> 
> Thats the biggest flaw I see in the patent system, where someone could
> accidentally and unknowingly infringe somone else's IPR and then get
> strangled with a loss of revenue claim that doesn't even match nor
> proportionatly match the associated revenue gains the IPR infringement gave
> the infringer. This is an issue that could be resolved properly with policy,
> and is an issue that is not confined to the software industry. I imagine if
> we spoke to a patent atty they have plenty of precident to quote that deals
> with the issue.

Don't forget the cost of litigating a patent case -- multiple $millions.

> >> > patents are a useful tool in other industries.
> >> Why wouldnt they be useful in the software industry then?

see the FTC report, (Chapter 3, V.G. pages 164-165) as quoted on
http://www.ffii.org.uk/ftc/ftc.html#software :

  The software and Internet industries generally are characterized by five
  factors: (1) innovation occurs on a cumulative basis; (2) capital costs
  are low, particularly relative to the pharmaceutical, biotechnology and
  hardware industries; (3) the rate of technological change is rapid, and
  product life cycles are short; (4) alternative means of fostering
  innovation exist, including copyright protection and open source software
  ...

The FTC report, or at least this summary page, is, as always, well worth a
read -- especially where comparisons between the software industry and
other fields, and the US industry's feelings about patenting, are
relevant.

>> they demonstrably have not been.
>You have examples to back that up?

Let's see what the US software industry had to say:
http://www.ffii.org.uk/ftc/ftc.html#questionable -- 'Effect on the
Software Industry: Questionable Patents Create Uncertainty and Hinder
Innovation'.

  Questionable patents may have a disproportionally adverse impact on
  entry by small firms and individuals who lack the resources to challenge
  such patents. As one software programmer commented, "the ease with which
  the US Patent Office has been granting patents in the last few years has
  already dampened my plans to write software as a primary business."
  [358] In contrast, a panelist from a larger firm suggested that
  incentives to innovate are not undermined by questionable patents. [359]
  The panelist observed that it is "a fairly straightforward exercise for
  our research department to investigate the relevant prior art [for an
  overly broad patent] and therefore obviate any further discussion on the
  matter." [360]

There's a few others, but the "small guy vs big guy" is most relevant
to the Irish homegrown software industry IMO.


PS: the one thing that Ireland *is* missing, compared to the US -- in
fact, more specifically, compared to *Silicon Valley*, which is the only
place where small innovative companies are really being allowed to become
established -- is a really good angel investor community who understand
technology and are prepared to take risks.

Venture capital deals are risky -- the VC firms often seem to get
jittery too early, pull out, and the company folds.

OTOH, angel investors who have insufficient experience in the technology
field often wind up funding the same old crap -- business/banking-oriented
boring stuff, which really is not innovating anything, instead of the
next Google.

(BTW this isn't just Ireland's problem.  Same thing happens in the
software business in India, in the UK, in Orange County, Chicago,
etc. etc. etc.)

- --j.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)
Comment: Exmh CVS

iD8DBQFA8hEfQTcbUG5Y7woRAnM+AJ4+R/tx9GbzSopkp03xjlTSFz5x/wCg50uU
wXFZ3xJQXQzSVxvG1VjKnCw=
=tEuT
-----END PGP SIGNATURE-----