I’ve been writing about free software for nearly 20 years, and about Microsoft for over 30 years. Observing the latter deal with the former has been fascinating. At first, the US software giant simply dismissed free software as unworthy even of its attention, but by the early years of this millennium, that was clearly no longer a viable position.
As I’ve charted elsewhere in my "Brief History of Microsoft FUD", it made various attempts to discredit open source, all of which were dismal failures. As it became clear that this strategy would not work, it adopted another, somewhat more sophisticated. This involved trying to match aspects of open source without actually embracing it. The first manifestation of this was "shared source":
Microsoft Corp. announced this week it is making the programming code for its Office 2003 software suite available to government agencies around the globe, a move partly aimed at allowing them to inspect the product for flaws and security problems.
Though Microsoft usually guards such software coding tightly, the step is an extension of an initiative the company began in January 2003 giving about 60 governments access to the inner workings of the Windows operating system. This is the first time the software giant has shared the source code for Office, which includes the Word text processing, Excel spreadsheet, and PowerPoint presentation programs.
In the 2004 Washington Post article quoted above, the company was quite frank about what it was doing:
While government clients will be allowed to examine the source coding for Office, they will not be allowed to modify it. “Shared source is all about learning from the open-source movement and applying it to our business model,” [director of Microsoft’s Shared Source Initiative] Matusow said.
Despite that attempt to match open source with shared source, Microsoft had as little success in heading off the growing interest in free software here as elsewhere. So it then dipped its toe in the open waters a little deeper, by introducing its own shared source licences, some of which met the Open Source Definition:
The board of the Open Source Initiative (OSI) has approved two Microsoft Corp. licenses that allow proprietary source code to be shared, a move that is likely to inspire protest and spur controversy for die-hard open-source proponents.
The Microsoft Public License (MPL) and the Microsoft Reciprocal License (MRL), two of Microsoft’s so-called “shared source” licenses, are now viable OSI licenses for distributing open-source code alongside more widely used community licenses such as the GNU General Public License and the Mozilla Public License.
“Today's approval by the OSI concludes a tremendous learning experience for Microsoft and I look forward to our continued participation in the open-source community,” said Microsoft General Manager of Windows Server Marketing and Platform Strategy Bill Hilf in a press statement.
Well, maybe. Or perhaps it rather hoped that by adding to licence proliferation it would sow a little confusion in the open source world, and slow things down a little while it used its “learning experience” to come up with something more deadly.
Of course, that didn’t work either, but that hasn’t stopped people trying the same approach in another sphere where openness is displacing traditional closed approaches: academic publishing. As I noted just recently, open access is doing to high-priced academic publishing what open source did to Microsoft. And it seems that the academic publishers are going through exactly the same stages as Microsoft in their response. Here, for example, is a fine example of the FUD that publishers were pushing out about open access back in 2007:
Recently, there have been legislative and regulatory efforts to compel not-for-profit and commercial journals to surrender to the Federal government a large number of published articles that scholarly journals have paid to peer review, publish, promote, archive and distribute. Mrs. Schroeder stressed that government interference in scientific publishing would force journals to give away their intellectual property and weaken the copyright protections that motivate journal publishers to make the enormous investments in content and infrastructure needed to ensure widespread access to journal articles. It would jeopardize the financial viability of the journals that conduct peer review, placing the entire scholarly communication process at risk.
That backfired spectacularly – especially another claim that government policies mandating open access for publicly-funded research were:
opening the door to scientific censorship [emphasis in original] in the form of selective additions to or omissions from the scientific record.
Like Microsoft, the academic publishers have now moved on, and adopted exactly the same approach: coming up with their own licences that are confusingly-similar but not equivalent to existing Creative Commons licences, widely used in the open access world:
the [International Association of Scientific, Technical & Medical Publishers – STM] has produced sample licences for a variety of uses within open access publishing. The licences on this page have been designed to provide easy to use, ready-made terms and conditions which publishers can adopt and/or adapt to the needs of their users. The “full” licenses can be used as stand-alone options, while the “supplementary” license clauses can be used to supplement other existing standardised or bespoke licences.
Understandably, the open access community is keen that people understand why the new licences are both unnecessary and unsatisfactory, and have launched a campaign asking the STM Association to withdraw its licences:
In their current formulation, these licenses would limit the use, reuse and exploitation of research. They would make it difficult, confusing or impossible to combine these research outputs with other public resources and sources of knowledge to the benefit of both science and society. There are many issues with these licenses, but the most important is that they are not compatible with any of the globally used Creative Commons licenses. For this reason, we call on the STM Association to withdraw them and commit to working within the Creative Commons framework.
The Creative Commons licenses are the de facto global standard for providing users with legal confidence of their rights to reuse content. They are not perfect, but they have been applied to over a billion resources by millions of authors. Creative Commons licenses are the preferred option supported by major content platforms and Open Access publishers. They are recommended by governments in Australia, Europe, the United States and elsewhere. If research outputs are to be a first class citizen of the web then they should use the same licenses.
Using the STM model licenses would make the research literature legally incompatible with hundreds of millions of Creative Commons licensed pictures on Flickr, videos on YouTube, articles on Wikipedia and across the web. Not all Creative Commons licenses allow all forms of use, and not all are compatible with Wikipedia but all Creative Commons licenses use common terms and a common and established legal framework. By contrast, the STM model licenses will increase costs for all stakeholders by creating legal uncertainty that can only be resolved by legal action, probably in multiple jurisdictions. Confusion and inconsistency are not in the long term interests of any stakeholder.
As well as highlighting the interesting way in which history repeats itself in the open world, this latest development also emphasises once more the centrality of licensing. Just as the creation of the GNU General Public Licence by Richard Stallman effectively gave the nascent free software community its constitution, so the Creative Commons licences have played a key role in defining the open access world.