An Interview with the OpenOffice.org Team
This article was first published on :www.newsforge.com.
I asked the OpenOffice.org team a few questions about OpenOffice.org in general, and their upcoming release. Questions 1-3 were answered by Louis Suarez-Potts, while questions 4-6 were answered by Colm Smyth of Sun Microsystems. Louis is OpenOffice.org’s Community Manager, member and chair of the Community Council, and lead of many OpenOffice.org projects including the Native Language Confederation.
Colm is a StarOffice Architect, and was responsible for defining the product concept for OpenOffice.org 3.0 (or StarOffice 9).
Ques.1: What are the core strengths of OpenOffice? What do you think are the features that will help you replace the dominent office suite?
Ans. 1 : OpenOffice.org (as we must refer to it) is to begin with open: that’s its key strength, as it that element allows a vast community to inspect, improve, and work on the source. In terms of the application, well, it depends on your audience. A so-called knowledge worker would key features such as the built-in PDF export or Shockwave Flash (for presentations) to be very important, as it makes it easy to share non-editable files. They’d also like the new component (new with 2.0), Base, which is our Access equivalent. Only it works better and is a real relational database. Other things that make the application important include, in 2.0, a very good Calc, which is the equivalent of Excel, and what is probably the best presentation software.But that’s just for starters. OpenOffice.org is, with 2.0, using an open standard file format, the OpenDocument. By using this file format, which is open and a standard, we thus defy proprietary exclusion and dependency. There is no possibility of vendor lock in with an open standard. Furthermore, because it is sophisticated XML, it can work with other equipped applications, meaning that developers, companies, can employ OpenOffice.org in heretofore unknown ways; innovation is engendered in this manner.
Finally, what is the key feature in OpenOffice.org 2.0 that will replace the status quo? Offhand, I’d say it is simply that it does what people want easily and simply and does not also drag them into stuff they never thought they had to do. Good software is not something you are saddled with for the rest of your life, it’s what gets the job done.
Ques. 2: Do you believe the day of proprietary formats will ever come to an end?
Ans. 2: Not entirely, no. But for things like office suites, yes. What would determine a move to open standards is the size of the market. I can see scientific arenas, say, using proprietary formats and software because of limited markets.
Ques. 3: One major advantage of OOo is the internationalization. Do you see this opening up new markets in developing countries?
Ans. 3: Yes! Very much so. And we’ve been the leader here. OOo has provided the nucleus for any number of local efforts to gain control over software production and distribution. Our native-language confederation, which I help lead and helped create, has enabled millions to use office software in their language and to get support for it. It’s really a fascinating and important point. Open source is not happening only in the US. In fact, much of OSS is happening in India, China, Brazil, Europe, where governments are trending to using FOSS (free/open source software) and away from proprietary and monopolistic software for reasons that are as much political as pragmatic and economic. Political, because FOSS offers a way out of the hegemony of US software products and English-only applications and pragmatic because not only is it cheaper but FOSS encourages the formation of local software production and distribution economies. An engineer learning and working with Linux contributes more to local wealth not to the wealth of a remote company, even if there are local Microsoft training companies. He will still end up supporting Microsoft’s business. And there is the language issue. Since WWII, the lingua franca of business has been English, and software used reflects that. To be sure, there are Japanese, Chinese, French, German, etc., versions of proprietary software, but these are usually inadquate and imperfect; and the user has no choice but to use it. Open source has changed that. By giving the user–the native speaker–access to the localization process (translation, customization), the application can evolve to meet the speaker’s real needs. Thus, our Welsh localization, our Indic language localizations, our Japanese, Czech, Serbian, Swahili, Basque, etc., localizations, all these are superior to proprietary versions, for they were created by the communty actually using the application according to an open model of development. Errors are corrected, innovations incorporated.
Ques. 4:The upcoming version 2 release must have the development team working overtime. Have there been any particularily difficult problems you are facing/have had to face?
Ans. 4: Just to set the context – OpenOffice.org/StarOffice is a huge code-base comprising millions of lines of code developed over a period of more than 15 years. It is a much larger challenge to work on a single native code-base of this size because logistically it takes a long time to build and test, which impacts not just on releases but on daily and developer builds. Techniques like agile programming and test-first development simply don’t scale this far, and developer support tools that check stack and memory access are unable to cope with the code size and number of API symbols. The code-base is very well layered and somewhat modular, but not componentised.
In this particular release, we set ourselves (and almost entirely completed) a challenging set of goals in terms of new features (in the 100s), along with bug fixes and many minor enhancements numbering in the 1000s. Due to development scheduling, we had a large number of projects integrating almost simultaneously. This put a strain on our quality and build processes, as well as requiring developer effort to resolve code and feature merge conflicts.
In the next release, we will avoid these issues in 3 ways: 1) the plan for the release will specify fewer features, allowing the team to respond to customer requirements during development; 2) the development schedule will use clear time-boxing, with individual projects aiming to complete within a specific part of a development phase; 3) we will improve planning and inter-team communication so that dependencies are detected and resolved earlier.
Ques. 5: Can you give us an idea of the new features in OOo ver 2.0?
Ans. 5: The main focus of our efforts and the most important benefits that customers will see is improved usability and significantly improved interoperability with Microsoft Office formats. This addresses the day-to-day needs of many more end users and makes OpenOffice.org/StarOffice a real alternative.
Ques. 6: Do you decide new features from industry feedback? Specifically, do you have any corporate/industry partners who can give you an idea of the needs of a large corporation so that you can incorporate them?
Ans. 6: We use a broad range of sources for feature requirements ranging from customer interviews and surveys, market and competitive analysis, input from Sun’s service and support teams, direct contacts by our engineers with larger customers and global partners (ISVs, ASPs, OEMs) and feedback from local service partners performing customer migration and deployment. This requirements process has been especially broad and intense in planning OpenOffice.org 3.0; while 2.0 focusses on offering a compelling alternative to Microsoft Office, in 3.0 our focus will extend to several unique and valuable features, especially around collaboration and workgroup productivity.