[Linux-aus] RE: Why short sightedness is bad for Australia
brendon.chase at zdnet.com.au
Fri Jun 25 15:03:03 UTC 2004
Just to clarify this early:
This is a commentary written by Tony Healy, a software developer who does
NOT work for ZDNet, Builder, or any part of CNET Networks. It is not the
personal opinions of this Editor, or the company who published it.
It also had nothing to do with promotion. It's an issue that is being put
forward at the highest political level in this country and something I have
personally covered earlier this week. This legislation will affect the
audience it was written for, Software Developers. As always, there are two
sides to this argument, hence the commentary.
While there are many valid points in the talkback and on this list in
response to this commentary, it's only fair that I've invited a Linux
Australia spokesperson to write a reply for publication.
If anyone has any further questions feel free to contact me off list.
If you posted that article to create controversy for the promotion of
your website, all I can say is that I'm seriously unimpressed.
Regardless of your opinions on Open Source, most of Asia including at
nearly ten times the population of the entire USA is moving en bloc
towards an Open Source software model. What this means in practice is
that if Australia doesn't address it sensibly, we will simply be
backwatered and ignored. The head-in-the-sand option won't work here.
Your comments on copyright term extensions would be reasonable in a
world of altruists, but I don't know where tyou're going to find one of
those. In real life, copyrights cause useful extensions of existing
technologies to be suppressed, both by dog-in-the-manger tactics and
almost inadvertantly as hordes of would-be inventors silently give up
when faced with a tangled morass of overlapping patents. Extending
their domain will simply make those problems worse.
The DMCA is a legislative patch for broken and failed business models,
and there is already a long list of problems with it which US
legislators are hoping to reduce (but not eliminate) with a patch on
this patch. Its nett effect for the Australian developer community if
applied here will be to drive development offshore to countries not
encumbered by it.
> Open source activists don't represent the Australian software
That's just flat-out wrong. We do represent "the Australian software
industry" just as much as, say, MYOB does. It's insulting for a
self-proclaimed expert on the topic to make strong and unsupported
assertions to the contrary.
> The software industry generally means firms that develop software
> and charge for it.
You have completely bypassed at least one entire segment of our
Not all software is developed to be sold en masse. Quite a few
developers turn a penny doing custom and one-off work. Often, they have
neither the time for nor interest in taking these one-off solutions any
further, yet by publishing them in an accessible fashion (that is,
under an Open Source licence) they make them available for other
devlopers - including, of course, Australian developers writing for a
shrink-wrap market - to use as building blocks or stepping stones for
their own creations.
> These are high value firms that create valuable IP
These companies do not *create* IP, they *realise* it, or as the lawyers
would say, they "engross" it, drag it out of people's heads and make
working implementations of it.
Whether or not they charge others for the use of this concrete property
or not is their own business, but when they use "intellectual property"
laws to prevent others from thinking in the same space, that damages
the entire development community and shrivels their market as a whole.
I can sit down with a bare machine and a set of Mandrake Linux CDs, and
within an hour have a working server supplying name, web, mail
(including viruses and anti-spam), FTP, login, file, firewalling and
statistical services to the world and/or an intranet. Give me another
half hour and I can turn a vaguely worded press release into a fully
operational, dynamic and interactive web site and CMS using Plone and
ZWiki. Including time spent lugging machines around and plugging cables
in, it is reasonable to expect to turn AUD$2000 into a vibrant web
presence and Internet gateway in a couple of hours. It's quite a
profitable business for me, pleases the customers no end, and I make
extra income from writing or customising software, "tailoring" the
system to the customer's needs.
If I had to IP-licence and litigatively cross-check all of this before I
used it, lots of businesses including mine would suddenly lose all hope
of viability. The difference between the solution above and one from
our favourite convicted monopolist is at least a fourfold price
increase, plus all manner of arbitrary and complicated limitations on
what the owner of the system can and can't do with it.
As things stand, I can operate with a presumption of innocence and
simply keep an eye on the news for signs that any of my components are
in "intellectual property" trouble - and if they are, I or anyone else
can reasonably expect to be able to rewrite the offending subsystem to
avoid the issue. The onus is on the entity alleging infringement to
demonstrate a valid claim, not on me to demonstrate the invalidity of
all possible claims.
The DMCA reverses this. One American was thrown off an airplane for
wearing a "suspected terrorist" badge, but the point he was making -
and which the incident amply illustrated - is that all airline
passengers have now to some degree become terrorists by default, and
must prove otherwise. The terrorists have won.
The DMCA is doing the same thing to American software developers. Why do
we want to do it to our own developers? If we do, expect to see me
start wearing a "suspected software pirate" badge.
> Open source advocate Russell, who appeared before the committee
> is actually an IBM employee.
Rusty is an IBM employee because he wrote a number of Open Source
packages (ever heard of iptables?) long before meeting IBM, and anyone
who knows him would expect to have to tie him down to *prevent* him
from testifying, regardless of who he's employed by, or not.
IBM is one of the biggest software developers in the world, and far from
dumb, and not at all shy about following up on their pound of flesh. If
there was a genuine issue between developers and FOSS, IBM would be
aware of it, and wouldn't be supporting FOSS so avidly.
To try to imply that IBM would be working against the Australian
developer community is just plain silly. They'd be working against
themselves. Directly against themselves, in the case of AusLabs, the
unit which employs Rusty Russell.
> Most members of the recently formed Open Source Industry Association
> are commodity firms and include booksellers, lawyers and casual
And the rest. Google for software projects I've contributed to, and
remember that I'm far from unique in this respect, many OSIA members
have contributed far more.
What you'd have said if you were truly interested in a complete
presentation is that OSIA members come from all walks of life. What
we'd have said in response is that we're delighted with that
assessment, it's what we aim for. Including those of our membership who
develop software for a living.
> Open source activists nowadays stress that they love and support
> copyright. Technically they are correct; they use copyright to
> force restrictions on downstream users.
The key phrase which you left out was "of the software they wrote."
It never ceases to amaze me that people who can't even see source from
MYOB or Pretzel Logic are complaining about how they can't paint
*someone*else's* software with their own livery and abuse the source to
it however they see fit.
Compare the restrictions in a typical EULA with those of the GPL. The
GPL - just about the most restrictive Open Source licence available -
is freedom and gentleness incarnate, relatively speaking.
Quoting GPL clause 5:
You are not required to accept this License, since you have
not signed it. However, nothing else grants you permission
to modify or distribute the Program or its derivative works.
And what happens when a developer playing dog-in-the-manger closes down?
Nobody dares use their "IP" because there is no company to do an
intelligent deal with, typically just administrators who will hold out
for as much money as they think possible and sue if you go ahead
anyway. Until the "IP" expires, that pocket of knowledge, experiment
and experience is effectively denied to the rest of the human race...
where they respect "IP" laws, which the majority don't.
Now here come people wanting to make that situation worse, yet who would
be enraged if development in the field moved to India or Romania
because of it.
> Problems arise where government funded research is released under
> licences like the GPL. That work is available to everyone, but the
> best developers are barred from commercialising it.
Flat wrong again. Quoting the GPL preamble:
When we speak of free software, we are referring to freedom,
not price. [...] if you distribute copies of such a program,
whether gratis or for a fee, you must give the recipients
all the rights that you have.
There's nothing wrong with charging for GPLed software, and many
companies do (MandrakeSoft, SuSE and Red Hat, for three examples). The
only restriction is that acquiring complete source code
*in*addition*to* the binaries you've already purchased not be
> Computer sellers such as HP and Sun use open source as a way to
> depress software prices and thus expand the markets for their
> expensive computers.
This is a highly illogical assertion.
HP and Sun both *sell* software, and Sun also give away (for $0) certain
other software, with and without source. If they just wanted market
penetration, they'd give it all away for free.
In point of fact, Sun is to some extent even giving up on making their
own hardware, and using AMD gear instead. HP is getting into
ever-cheaper computers, competing with the likes of Dell. All of which
makes a non sequitur out of your arguments.
> For example, scientists and academics share source code because
> it's not central to their jobs or work.
I call bulldust.
Open Source programs like Terra are absolutely central to the work of
many earth scientists, and they get shared around enthusiastically.
Cinematographers also share source code (cinepaint, for exampler, and
many Maya converters and plugins), and that's absolutely central to
> That type of software is technical and rough. It's quite
> different from the highly engineered products that business and
> home users demand.
You mean software like OpenOffice, Mozilla, the KDE suite, The GIMP,
QCAD and Evolution?
> There are no demands for Intel and other chip makers to provide
> the diagrams for their circuits, or for car makers to provide
> the engineering drawings for their cars.
You need to get out more. There most certainly is such demand. There's
an entire industry revolving around "chipping" cars, and legislation
slated to go through the US Congress and designed to do exactly that.
> I discuss these issues further in a paper for the US public
> policy research institution, the Institute for Policy
> Innovation. (Healy, 2004)
Ah, yes. The Open Source community has had exposure that that particular
group of paid mouthpieces before. You need to learn a bit more about
their history and sponsors before writing anything for them.
> One is for people and government to use public software such as
> Linux in preference to commercial software. This doesn't directly
> effect the Australian software industry one way or the other,
> although Linux is a less polished platform for commercial
> developers to target, and widespread usage would probably reduce
> the amount of third party software development. This directly
> contradicts the claims and beliefs of political backers of open
I call bulldust again.
2/3 of all internet-visible web servers, over 3/4 of all email servers,
the vast majority of name servers and so on. This is not the hallmark
of "less polished" software.
Tony, when did you last install or use a Linux distribution? I have
customers' staff who have used a Linux box for a full day without
noticing that it was Linux. Just how much polish are you looking for?
> The second meaning is that commercial firms, including
> successful Australian firms, should disclose their technology
> to rivals by exposing their source code.
Give me a single example of a successful Australian softweare
development firm's source code being demanded from them by FOSS
advocates. I'm having trouble believing that such a beast exists.
> To compete in the world economy we need strong copyright
> protections so our smaller firms can protect their innovation
> against the larger marketing and distribution engines of
This statement is also staggeringly ignorant of real-world events.
We already have strong copyright protection, to start with. The DMCA and
other provisions entailed in the FTA have nothing to do with copyright.
For seconds, large corporations ("multinationals") have repeatedly
demonstrated both the capability and will to crush "smaller firms"
whose "intellectual property" stands in their way. All that the
"intellectual property" rules effectively do in these cases is provide
a central repository in which the crusher can conveniently fish for
ideas. Making the rules more stringent advantages only those able and
willing to wield the most legal muscle.
The only "small" firms to benefit so far have been what might be fairly
termed "intellectual property pirates" or perhaps more accurately
"highwaymen", and more colloquially "patent squatters". These companies
acquire patents solely for the purpose of extorting royalties and
compensation with them.
There is no development, no social or national advantage here, only a
crime - hoarding - for which people are shot during time of war. And
"smaller firms" are ill-equipped to deal with this now-flourishing
(thanks to the DMCA) parasite industry.
Shall we sum up? The best points in the article are built on very shaky
ground, and it goes downhill from there. In terms of convincing power,
it was very well written, but in terms of technical and factual merit
it's hopelessly lost. Many of the basic facts and assumptions fly in
the face of observation, and the processes therein outlined contradict
American experience - which is based on concrete reality, rather than
theory or rhetoric.
You should probably submit an article to the Alexis de Tocqueville
Institution as well. Their "research" techniques are similar in many
ways to IPIs and to what I'm reading here. I was particularly impressed
with Brown's reinterpretation of the information available to him, see
here for a good collation, and follow the links:
http://cyberknights.com.au/ Modern tools; traditional dedication
http://plug.linux.org.au/ Vice President, Perth Linux User Group
http://slpwa.asn.au/ Committee Member, Linux Professionals WA
http://linux.org.au/ Past Committee Member, Linux Australia
http://osia.net.au/ Member, Open Source Industry Association
More information about the linux-aus