Microsoft's quest for shared-source approval

Microsoft has submitted two of its "shared source" licences for approval by the Open Source Initiative (OSI). The significance of this can be read in different ways. Some see it as a victory for open source, a capitulation on the part of Microsoft, and an admission that open source will be the way of the future. Others see it as a divisive move, designed to emphasise existing splits within the free and open source community. In reality, it is probably a bit of both.

Either way, Microsoft's submission of licences to the OSI illuminates some problems, not all of which are of Microsoft's making. Microsoft has always had a problem with free and open source software. The seeds of this animosity can be traced as far back as the famous Open Letter to Hobbyists written by Bill Gates, "General Partner, Micro-Soft", on February 3, 1976, in which Gates declared: "As the majority of hobbyists must be aware, most of you steal your software. Hardware must be paid for, but software is something to share. Who cares if the people who worked on it get paid?"

"Who can afford to do professional work for nothing?" he asked. "What hobbyist can put three man-years into programming, finding all bugs, documenting his product and distribute for free?" This is a question that has been answered many times by free software developers during the intervening thirty years.

The merry band

However, by adopting the EULA and software in a box Microsoft grew from being a mere "break-even operation", as Gates claimed it was, to becoming the richest corporation in the world. Free and open source software didn't become a serious problem to Microsoft until the late nineties when Linux and Apache began to make serious inroads into the web server market. A leaked report, written in 1998 by a Microsoft engineer, Vinod Valloppillil, praised "the ability of the [free software] process to collect and harness the collective IQ of thousands of individuals across the internet," and suggested that the resulting software was of such a high quality as to constitute a "direct, short-term revenue and platform threat to Microsoft." Nonetheless, a year later, Ed Muth, a Microsoft group marketing manager, felt able to compare free software developers to Robin Hood and his merry men, declaring that "complex future projects to add such functions as automatic translation of email require big teams and big capital. These are things that Robin Hood and his merry band in Sherwood Forest aren't well attuned to do."

In July 2000, Steve Ballmer declared: "Linux is a tough competitor. There's no company called Linux, there's barely a Linux road map. Yet Linux sort of springs organically from the earth. And it had, you know, the characteristics of communism that people love so very, very much about it. That is, it's free."

A PacMan and a cancer

By 2001, it had become obvious that free software developers were well attuned to the creation of large projects, and had found working business models to distribute the software. Free is hard to compete with, especially when the methodology tends to result in better quality software. During this time, Microsoft was on the attack, worried in particular by the impact that free software was having on government computing projects, with their emphasis on cost and accountability. Within a matter of months Microsoft executives Steve Ballmer, Bill Gates, Craig Mundie and Jim Allchin all made statements about the dangers of free software.

In July Ballmer declared that "Linux is a cancer that attaches itself, in an intellectual property sense, to everything it touches." Later the same month, Bill Gates compared the GPL to PacMan. "If you say to people, 'Do you understand the GPL?'", he said. "They're pretty stunned when the PacMan-like nature of it is described to them," as well they might be, because it doesn't quite work like that.

Craig Mundie told a journalist that the "viral aspect of the GPL poses a threat to the intellectual property of any organisation making use of it. I would challenge you to find a company, who is a large established enterprise, who at the end of the day would throw all of its intellectual property into the open source category." Irving Wladawsky-Berger, a senior vice president at IBM, which had dedicated a large part of its resources to promoting Linux, responded: "If we thought this was a trap, we wouldn't be doing it, and as you know, we have a lot of lawyers."

The competitive threat of free and open source software has been difficult for Microsoft to respond to. The attacks on the GPL have backfired, and nobody has believed them. The SCO case against IBM failed to dent the progress of Linux, and the regular patent threats made by Microsoft are just noises to distract potential Linux users, so kick-boxing isn't going to work. Microsoft still monopolises large sectors of the computer industry, but the rise and rise of open source looms bigger every day. Slowly, inexorably, Microsoft is coming to the realisation that it will have to engage this enemy on a more level playing field.

A desirable objective

From Microsoft's point of view, the most galling thing about open source is the appeal it has for developers. In order to spread the popularity of its single platform proprietary technologies among developers it has to retain and extend the appeal of technologies such as .NET and the proprietary languages it uses - and this means catching the imagination of developers, and locking them in to Microsoft technologies. This is the attraction of open source to Microsoft. Each of the "shared source" technologies it has sponsored is locked into its proprietary version of userland. This inevitably limits its appeal to open source developers. Nevertheless, it is taking an adventurous approach towards the newer programming languages and is being pragmatic in its approach to developer communities, in the hope that by submitting its licences for approval it will gain an aura of respectability among a larger community who will take their expertise onto .NET and Windows in the workplace.

A secondary benefit, from Microsoft's point of view, is that it may serve to accentuate the fault line that has long existed between the "open source" and "free software" wings of the user and developer communities. The Open Source Initiative (OSI) is a relatively recent creation, having originated in 1998 as a split from the free software movement, although not all the protagonists would have recognised it as such. The OSI was the creation of some of the better known figures who had emerged out of the noise surrounding Linux, primarily Eric Raymond, who was best known for the book, "The Cathedral and the Bazaar", and Bruce Perens, a free software advocate and one time project leader of the Debian GNU/Linux development community.

Some of the protagonists convinced themselves that the GPL was too "extreme" and would frighten business. Others felt that "open source" was a less ambiguous and more comprehensible description than "free software". Their answer was the encouragement of a more liberal licensing regime, based on the premise that "open source" methodologies produced better software, rather than the notion that software should be "free". The main public role of the OSI, which is now led by Michael Tiemman of Red Hat, is to arbitrate on the validity of "open source" licences.

The term "open source" has caught on as a universal shorthand to describe free software, but open source and free software are not the same thing. Free software has an ideology, a philosophy and a licence that were developed over a number of years in response to the problems encountered in furthering the idea that software should be free. Open source, on the other hand, has no connotations beyond the visibility of the source code, and claims, for this reason, to be friendlier to business than "free software". Strangely enough, the GPL has remained by far the most popular "open source" licence. The great majority of the better-known and more successful "open source" projects are released under the GPL, and can be described accurately as free software. JBoss (now owned by Red Hat), for instance, has always described itself as "professional open source" but releases its software under Free Software Foundation (FSF) licences.

We're the same as you

The greatest problem that emanates from the role of the OSI is the feeling that it has weakened rather than strengthened the public front of the free and open source software movement. There has been a proliferation of licences, many of which conflict with one another, are not compatible, and serve little useful purpose, other than satisfying the doubts of the participating companies. There has also been a proliferation of companies that describe themselves as "open source", some of which are a long way from fulfilling the objectives of either the FSF or the OSI.

The main argument against the OSI giving approval to the Microsoft "shared source" licences is not antipathy to Microsoft, nor as a response to its historic hostility to free software, but that there are too many "open source" licences already, with very little clear distinction between them.

If Microsoft, or any other company, wants to become a member of the free and open source software community it should be asked to adopt an existing licence, or to help in the long term process of rationalising the existing licences, and play with the community rather than against it. Tiemann has already hinted that Microsoft's primary offering, the Microsoft Permissive Licence (MPL), will fail the procedure because it is not "permissive" as the term is understood by the open source community, and because "the specific innovation of maximum incompatibility of the MPL is not what we were looking for, so I think what we have is a submission that has two fairly major strikes against it."

Some observers have felt that the OSI has been more keen on enrolling companies into the open source camp than ensuring the long term integrity of the licences in the past, which has resulted in an unwelcome proliferation of incompatible licences. If the trend continues, there will be an inevitable weakening in what is understood by "open source". A sensible long-term strategy would be to work with all participants to reduce the number of licences to a handful, and preferably to urge all the participants to adopt one of the GPL variants, or the BSD licence, which between them cover most of the sensible demands of an open source licence. The community is in a position of strength. Microsoft needs developer participation more than the community needs a company like Microsoft, and it certainly doesn't need another set of potentially incompatible licences.

Email to a friend

Print this page