Australian Open Source?
Australian Open Source?
Aside from Linux, Open Source is more often than not conceived as US-based, anti-Microsoft act of anarchy rather than a conventional form of software making. Tim Smith spoke with three major Australian Open Source influences.
Doctor Silvia Pfeiffer is works at the CSIRO. She writes code. She writes the kind of code that has been changing the Internet. She also evangelises on behalf not only of that organisation but also Australia. With a distinguished track record in ‘Closed Source’ as well as Open Source applications, she is a key voice to explain just how Open Source works in our region.
The CSIRO - Dr Silvia Pfeiffer
IDM: Can you tell us about your role at the CSIRO?
SP: I am foremost Project Manager for the Structured Media project, which is developing the Open Source Annodex technology and researching into novel ways of authoring media annotations and novel applications of interactive media. My team and I deliver milestones towards the CeNTIE-2 (Centre of Networking Technologies for the Information Economy) project, which is one of three advanced network projects funded by the Australian Government’s Advanced Networks Program.
I am also the Media Industry Focus Group convener for CeNTIE Last but not least, I am the Science Leader for the Multimedia Science Area. As such I bring the group of scientists and engineers in the CSIRO ICT Centre together that work in this science area, organise regular presentations and opportunities for discussion.
Our work also involves the commercialisation of the outcome of our R&D. We have a set of commercialisation and business development managers that help us in this respect. Dr. Gautam Tendulkar is the General Manager of the Commercialisation area.
IDM: What is the biggest confusion/obstacle you’ve discovered when educating commercially-focussed people in Open Source.
SP: Commercially-focused people sometimes regard Open Source software as a threat. Perceived threats are that it involves “giving IP away for free” without the possibilities of extracting money. Giving away IP for free in an organisation whose sole purpose is the creation of IP and the extraction of value from IP is indeed a difficult notion to tackle. However, it must be understood that for some IP and in some situations it is better to give it away - or rather to contribute it back to a community that is contributing much to the advancement of research.
Open Source actually helps solve many problems where collaboration between different research institutions is envisaged. Rather than having to manage a complicated network of closed source libraries that institutions would cross-license to do their collaborations, Open Source encourages and enhances possibilities for collaboration between researchers, since they can look at and enhance their colleague’s code freely, thus creating more value.
IDM: What are the biggest drivers towards the recent take-up of Open Source among previously commercially-focussed organisations?
SP: First of all, there is no contradiction between commercial focus and Open Source: you can be commercially-focused and still work with Open Source Software.
For Australian companies and for the Australian economy I can see several advantages to going towards Open Source technology.I think the foremost reason at the moment is quality. Many companies have realised that some Open Source software is actually better than their commercial Closed Source equivalents. This is not surprising since Open Source code is subject to the review of many eyes and bad quality code will not pass easily to increase the reputation of an Open Source programmer. Developing good code is a matter of pride.
The largest drawback for adoption of Open Source code by commercial companies is the perceived lack of support. A group of Open Source developers has no legal contract or obligation to provide support - however experience shows that in a vibrant developer community, the effective support you can receive for an Open Source system is much more timely and accurate than for commercial products. And with an increasing number of local companies that provide support for specific Open Source systems, it become more and more easy to make the decision for running an Open Source system rather than a commercial one.
Another reason to go for Open Source software rather than for a proprietary system is that you make yourself independent from the software provider and their company. You are not dependent on waiting for bug fixes from them, you are not dependent on waiting for upgrades or being forced into upgrading when you’re not ready for it. And: you are not broken when their company goes broke. It thus helps reduce commercial risk.
As a country, Australia has a large number of highly qualified software engineers coming out of its universities. However, the number of Australian software development companies that sell their own software on a large scale is rather small. Many of the core software used in corporations is still imported from abroad - particularly the US.
Australia is too small a country to build its own standing with proprietary software in this market. However, small companies can have a large impact with Open Source software that is adapted to their or their clients’ needs. The ever-improving quality of Open Source software is a good reason to go down that track.
IDM: What are the biggest challenges faced by Open Source developers in terms of making money from their work?
SP: Open source developers are mostly motivated through the challenging technical work they do, through the reputation they get from it and the sense of achievement when solving a problem that hasn’t been solved before. This goes for academic as well as hobbyist Open Source developers. Some are also strong believers in the Open Source way and would not develop any other type of code.
Open source developers hardly ever make money from their work directly. However, they increase their knowledge and reputation through it and thus become highly valued individuals, who sometimes find well paid jobs that allow them to do what they enjoy most: develop high quality Open Source software.
Commercial motivation around Open Source software mostly comes from the companies working in the Open Source space. Such companies can make money directly from the Open Source software (e.g. distributions), from selling hardware that the Open Source software runs on, from selling proprietary software that complements the Open Source software, or from service contracts.
IDM: How would you estimate Australian Open Source development has progressed in the last decade?
SP: Australia has a very vibrant Open Source community. The Boston Consulting Group does a regular “Hacker Survey” (see http://www.ostg.com/bcg/). If the numbers can be taken as indicative for the distribution of the Open Source development community around the globe, it shows that in absolute numbers of Open Source developers Australia is amongst the top 10 and taken relative to the number of inhabitants Australia seems to be among the countries of top density of Open Source developers. This is an amazing advantage for a country that is so strongly dependent upon software imports. I believe that Australian companies are slowly starting to realise this advantage and are starting to tap into this potential of a highly qualified technical workforce that can help stand up in the global market place.
IDM: According to Annodex, the reason for its formation is that: “...the CSIRO wanted to focus its work more on closed-source research and products built on top of the technology.”
SP: This statement is only targeted towards the Annodex technology and must not be taken too literally. We will continue to do R&D with Open Source software and contribute to Open Source software. We are even currently in the process of developing a more formal strategy for the Open Source publication of software resulting from our research.
The CSIRO has been working on Annodex for 5 years now and published much Open Source software during that time. We will continue to contribute to that base. However, the CSIRO also has to consider commercialisation of technology. With Annodex, we have reached a situation where we can build ‘Closed Source’ applications to complement the Open Source and Open Standards. This is similar to the World Wide Web: while the standards and base technologies that run the World Wide Web are open and free, there are plenty of Closed Source applications that live on top of it, such as Web Page authoring systems (e.g. MS Frontpage) or Web Site management systems (e.g. Vignette).
AUUG
The Australian UNIX and Open Systems Users Group is a professional association for end users, corporations, developers and vendors. AUUG is a national body with chapters that organise local activities in most capital cities. It is also the second oldest Open Platform user group in the world. We spoke with AUUG President, David Purdue and Vice President, Steve Landers about the professional’s view of Open Source.
IDM: ‘Open Platform’ or ‘Open Source’?
DP: We also talk about Open Standards, Open Protocols and Open File Formats. From AUUG’s point of view what is important is the interoperability - making all my ICT components play nice together. It is important that if I have a Brand X computer in my data centre then I can add a Brand Y computer and have them share data - ideally, have them share disks. It’s also important that if a colleague sends me a Brand M word processing document that I can edit it with my Brand N word processing program.
SL: And it is important that the documents I archive today can still be read in 10 years time when the vendor of my WP and/or database software no longer supports the version of the software used to produce them.
DP: Why is interoperability important? Because it levels the playing field - allowing everyone to play in the same space, leading to efficiencies and innovation. And competition, which is where the economic arguments come in.
SL: It also avoids lock-in - I can focus on the best solution for the given problem without having to rely on the vendor “doing the right thing”. Note that this doesn’t preclude commercial software - the Internet is a good example where there are many commercial products that use standard means of communicating with each other and with Open Source software.
DP: But interoperability can only be achieved if everyone knows the rules - hence the “Open” word, which implies that the rules for an API, or a protocol, or a file format, are agreed upon by all the relevant parties (a standards body, a consortium, or some other keeper of the flame), and easily accessible to anyone who wants to implement them.
Agreed upon and adhered to. One of the bugbears is organisations who agree to a standard, and then implement their own “extensions”, effectively making their product “non-standard” Open Source is the ultimate extension of this openness, because in the end APIs, protocols and file formats are defined by the software that implements and uses them.
IDM: Open Source is just Linux and FireFox, right?
DP: Well, the FSF/UNESCO Free Software Directory currently lists 4,454 software packages, and it is not a complete list. There are open source equivalents for almost all software applications out there - operating systems, web browsers, email clients, email servers, photo editing, games, database systems, even accounting systems, CRM systems and ERP systems. This software is not restricted to the Unix space, either. Many of these Open Source applications will run on Windows systems, and there is a significant body of Windows-only Open Source apps.
IDM: Open Source/Platform is just anti-Microsoft, right?
DP: There is a fair degree of anti-Microsoft sentiment with the Open Source/interoperable system community, but being anti-Microsoft is not the driving force behind the movement. Open Source often appears anti-Microsoft because Microsoft is (or at least has been) anti-Open Source. For much of Microsoft’s history interoperability was a threat to them - it did not fit their business model. For example, it suited their interests for everyone to be exchanging Word 97 documents when the only way to view and edit those documents was to purchase Word 97. However, customer demands are driving Microsoft to be more open. It drove Microsoft to put an Internet Protocol stack into Windows, for example, and is driving them to adopt a more open XML format for office documents.
SP: It’s only fair to mention that AUUG had a speaker from Microsoft at 2005’s annual conference. He gave a presentation on Microsoft’s Unix interoperability products. From AUUG’s perspective, if organisations are supporting standards and interoperability then we want to hear their story.
IDM: How does anybody make money from Open Source if it’s free?
DP: First point - making money is not always the aim. Sometimes the aim is to build something you need, and then share the result. If other people need it, they may well contribute improvements to it. Result - you end up with a better piece of software. I can think of five ways to make money with open source.
(1) Ask for donations.
(2) Consulting. Find an Open Source solution for someone, implement it, maintain it, and charge them for your time.
(3) Sell CD-ROMs of it. If it is popular you can also sell T-shirts, posters and other paraphernalia.
(4) Write the definitive book about a piece of Open Source.
(5) Least common is a dual license model - if you just want to use this software for yourself, then it is free. If you want to build this software into a product you sell, then you have to pay us a license fee. This is the model used for the Berkeley database maintained by Sleepycat Software.
Related Article: