The BSA (Business Software Alliance, or
Bad Statistics Alliance, depending on who you talk to) have released
yet
another one of their comical studies. I have been very
critical of these studies (See:
Lies,
Damned lies, and IIPA/BSA/etc statistics). What I
recommend people do is skip to the methodology section and see what
they are measuring, and decide for themselves whether what they are
measuring is harmful or beneficial for the Canadian economy.
Brian
Jackson wrote an article that quotes Michael
Geist indicating the study was "shockingly misleading". I will go further and suggest that what the BSA is really asking for
is a time machine, not copyright reform.
Twenty years ago the software
marketplace was different than it is today. A tiny few companies,
primarily those who make up BSA today, dominated the software
industry as people perceived it. There was "one true"
business model: software was sold the same as computer hardware,
distributed via brick-and-mortar retail channels, charging royalty
for boxes or software pre-installed on a computer.
Advancements in communications
technology have brought about change, and the BSA study is a
measurement of that change, not some alleged benefit of turning back
the clock 20 years. If you take a closer look at the methodology
section of the BSA study you will find that many of the things that
the BSA is claiming is harmful and in need of a legislative fix are
are actually positive developments in the software marketplace.
More and more software is
distributed in ways that do not involve brick-and-mortar retail
channels, being pre-installed on a computer, or even charging royalty
fees. While the BSA calculates this as a loss due to "piracy"
in their studies, this should be recognized as a positive development
that reduces software costs and benefits the economy. We do not want
to go back in time, and most reasonable people would consider the
growth of online software distribution to be a good thing. Many
people are predicting that the "app stores" that have been
part of the Linux software distribution system for decades
(centralized repositories with one-click download/installation), and
have recently been added to mobile computing for Android and Apple
iOS, will become the norm for all computing platforms.
We are slowly migrating away
from desktops and laptops to mobile devices from the cell-phone to
tablet and other form factors (embedded in televisions). IDC
competitor Gartner expects that Symbian and Android together will
account for 59.8 percent of the total worldwide mobile OS market by
2014, Symbian with 30.2 percent, and Android with 29.6 percent. Both
Symbian
and Android
are open source mobile operating systems, with open source being a
competitor to the software production, distribution and funding
models promoted by BSA. It is far more likely that the trend that we
see in mobile computing will migrate its way back to older
form-factor computers than the very outdated methodologies used for
desktop software trying to be mangled into the mobile space.
Without the complexity of the
legacy retail system, it has allowed for the growth of the
independent software marketplace. Being independent means separate
from the incumbent software vendors that make up the BSA. This
growth in the competing software marketplace will quite appropriately
register as a loss for BSA members. This includes software
professionals that participate in the growing commercial Free/Libre
and Open Source Software (FLOSS) marketplace, as well as the smaller
software vendors that make a great living through AppStores without
having the overhead of maintaining a retail distribution system
themselves.
Here are a few of the factors
in their study.
- Estimating losses: Their
methodology doesn't adequately differentiate between infringing
software and people using legal alternatives. IDC, the company they
hired to create this study, is notorious for under-estimating the
usage of FLOSS and other non-BSA software. When the study claims
that Canada's infringement rate was 29% in 2009, that could be 9%
infringement and an under-estimation by 20% usage of competing
software for all we know.
- Multiplying by
software-related services market: The methodology is focused on
services provided by or on top of BSA members, including outdated
overhead such as pre-Internet retail. This also incorrectly suggests
that someone not paying royalty fees for software is not going to
hire value-add services from competitors, something that makes
absolutely no sense to suggest. The only thing lost when someone
doesn't pay a royalty to a BSA member is the royalty to the BSA
member. Nothing else can be extrapolated from that, so this
multiplier has no legitimacy in the methodology.
In fact, the opposite is likely true. Money spent on software royalty payments is money that is then not
available for value-add software related services. This means that
increasing royalty payments to BSA members actually subtracts that
value-add to the software user, reducing their productivity and thus
their profitability.
- New jobs: This is linked to
the false claims from the services market, but also linked to the
fact that royalty payments leaving the country may theoretically
create jobs in some other country, but won't benefit Canada. In
fact, the more money that is leaving the country in the form of
royalties, the less money that is available to pay Canadians.
Note that I am not excusing
infringement, just putting this "study" into a more valid
perspective.
I will agree with the BSA that
reducing software copyright infringement would be a good thing. The
easiest and most long-term beneficial way to do this is to encourage
people to switch from illegally free software to legally free
software. While twenty years ago it was valid to claim that what the
BSA members offered didn't have viable alternatives, that is no
longer true. People who aren't evaluating competitive software are
doing so based on nostalgia, not valid technological or business
reasons.
Some representatives of BSA
members have admitted that while their first choice is people payed
them money, their second choice is people infringe since they least
want people to switch to competing independent software. If you look
closely at their legislative suggestions they are also primarily
aimed at reducing competition, not reducing infringement.
The IDC study makes some
pretty classic statistical mistakes. For instance, they suggest that
it has been found that countries that they have received higher
amounts of royalty payments from are countries that are richer. This is a correlation, not a causal relationship. Most reasonable
people would believe that being a richer country is the cause, and
royalty payments is the effect, but the study makes the ludicrous
jump to suggest the opposite : that higher royalty payments would
make the country richer.
The study claims that "IDC
has concluded that software revenues lost to software piracy would be
recouped dollar for dollar if piracy rates were lowered". This
is not a conclusion, but the assumption upon which their entire
methodology was built. IDC can't tell a correlation from a causal
relationship, and they can't tell an unfounded assumption from a
conclusion. I hope that the politicians that these "studies"
are aimed at will take the time to look at the methodology, and
understand just how wrong the claims made really are. We need
policies which promote these advancements in the software
marketplace, not policies that seek to turn back the clock to the
days when the tiny number of companies represented by the BSA
practically owned the marketplace.
---
Russell McOrmond is a self employed consultant,
policy coordinator for CLUE:
Canada's Association for Free/Libre and Open Source Software,
co-coordinator for Getting Open Source Logic INto Governments (GOSLING),
and host for Digital
Copyright Canada.