LibreOffice cash-for-code strategy tests open source ethic

The Document Foundation’s tender for the development of an Android implementation of LibreOffice begs serious questions, namely: Can an influx of cash into open source code creation succeed, and how do pay-for-code plays from nonprofit foundations affect the ethics and work ethic of today’s open source community?

For those who haven’t heard, the German nonprofit behind the successful open source LibreOffice productivity suite issued a tender document last week at the LibreOffice Conference seeking a bid for the one-off task of extending the LibreOffice document viewer prototype into a basic document editor on Android. It covers all the core program modules: Writer, Calc, Impress, Draw, and Math.

[ Learn more about the open source profit motive with our 9 open source secrets to making money. | Track trends in open source with InfoWorld’s Technology: Open Source newsletter. ]

The community has already experimented with bringing a LibreOffice editor to Android, first in the form of an early proof-of-concept, then with a LibreOffice document viewer built on the Mozilla for Android framework by Collabora developers, among others. But the number of volunteers turning up to work on the project has clearly been disappointing. That’s not surprising, given the scale of the task facing any would-be developer; clearly the enthusiasm of casual volunteers has been chilled.

The truth about open source and money

Contrary to popular sentiment, open source software is rarely written as an act of philanthropy. Rather, it represents the overlap on a giant Venn diagram of the vested interests of the participants. But with the proposed implementation of LibreOffice on Android, commercial motivations for engagement will be hard to devise.

Developers interested in coding the Android port won’t be able to generate future income from their work, as the Documentation Foundation policy (rightly) keeps all the work open for the community. As a result, it’s hard to imagine a business model that would fund the substantial up-front investment needed to write the Android port. So far the only commercial activities have been Web-hosted LibreOffice solutions such as the one provided by RollApp.

Technically, one could create a proprietary Android implementation of LibreOffice following a quiet change of licensing that TDF executed in the past year. This would allow the implementation to be monetized by traditional software industry tactics based on artificial scarcity. TDF changed the license from the LGPL v3 to a duality of LGPLv3 and MPLv2. The new Mozilla license is explicitly compatible with the (L)GPLv3, but allows new work under any license — including proprietary — as long as its source is in a separate file.

To make matters more interesting, TDF is relatively cash rich. Donations pour in from the millions of grateful users of LibreOffice around the world, resulting in a significant cash surplus for the foundation. As I explained in my session at OSCON this year, money is not always a blessing to open source projects. Open source projects are not “about” money. They arise from the overlap of the extrinsic interests of their participants — some commercial, some not — and not from any intrinsic motivation to monetize the code. That’s why so many become nonprofit entities.

Root of all evil?

Communities with a business model go by another name: businesses. While there are costs for a community — hosting, domain names, and so on — they are limited and should be easy to cover from small contributions by the project participants and users. (Of course, it’s always smart to maintain a cash reserve to avoid unexpected problems.)

At issue here is the fact that it is very unusual for open source foundations to spend money on the development of code. In fact, I recommend against it, as there is a great risk of chilling the contributions of people who are not being paid, as research has shown.

As it stands, TDF is faced with a mixed blessing: It has substantial funds, but as a charity, it is legally obligated to spend those funds over the next year or so. To this end, TDF has thus far been acting intelligently. It has invested in development infrastructure, especially for testing. It has hired sys admin and administrative staff. It has backed community activities. But there is still a big pot of cash left over, with the clock ticking.

Knowing it’s unwise to spend money where the community has extrinsic motivations, TDF has turned to the Android port. As outlined above, extrinsic motivations are few, and demand for an Android version of LibreOffice, already high, will surely grow as governments standardize on Open Document Format for communications.

Thus, TDF is spending money in hopes of bootstrapping a necessary new community within its purview. Florian Effenberger, executive director at TDF, told me:

In order to accommodate [the need for an Android app] within a reasonable timeframe, we must bootstrap a good-enough version to attract skills, which at the moment do not seem to be available inside the LibreOffice community. We see the tender as a first step toward a more diverse developer’s community, capable of ensuring the future of LibreOffice both on the desktop and on mobile platforms. The Board of Directors will oversee the tender process with extreme care, to avoid unwanted side effects.

Having tried to get active participation in an Android port for quite a while with little effect, TDF seem to have confirmed there are few developers with an extrinsic motivation who will be impacted by a tender. The tender is for a one-off work item; while the developer involved will be the early and acknowledged expert in the code, they are not guaranteed long-term income from TDF. Effenberger told me:

The scope of the tender is large enough for a group of developers, which reduces the risk that the community perceives the opportunity as limited to just a couple of individuals. … This does not mean that a company already investing in LibreOffice development cannot hire hackers with the appropriate skills and participate in the tender.

That means other developers will still be able to bring their own extrinsic interests to the project and collaborate. Spending a significant sum of money thus seems a smart move. The whole activity is a gamble, but if it works, it could at last bring a viable open source document solution to the burgeoning tablet market. The biggest question seems to be: Will anyone bid to do the work?

This article, “LibreOffice cash-for-code strategy tests open source ethic,” was originally published at InfoWorld.com. Read more of the Open Sources blog and follow the latest developments in open source at InfoWorld.com. For the latest business technology news, follow InfoWorld.com on Twitter.