December 6, 2007

GNOME/OOXML podcast shows two sides closer than appears

Author: Bruce Byfield

Despite technical difficulties with the phone lines,'s live podcast with Jeff Waugh of the GNOME Foundation and Roy Schestowitz, cofounder of the Boycott Novell site, attracted a large audience eager to discuss GNOME's involvement with the efforts to make the Microsoft Office Open XML (OOXML) document format an ECMA standard. Hosted by Rod Amis on his Lightning Strikes show at BlogTalkRadio, and with questions from's Editor in Chief Robin Miller and me, the discussion revealed that the two sides of the issue are closer than they have appeared in the past.

The discussion came about as a result of GNOME's support for the efforts of Jody Goldberg, the lead developer of the Gnumeric spreadsheet, to use the ECMA process to force Microsoft to reveal more about OOXML and, by extension, its previous binary formats. Critics charge that this move lends OOXML credibility, and worry that it might come at the expense of support for the rival, community-based OpenDocument Format (ODF). Complicating matters, because Goldberg's efforts had previously been supported by his employer Novell -- a company that many free software advocates regard as a Microsoft collaborator -- some critics began scrutinizing the remarks of everyone remotely involved for signs that the GNOME Foundation was covertly supporting Microsoft.

As reported earlier, the GNOME Foundation issued a statement explaining its actions. When the statement did little to quiet criticisms, Jeff Waugh, a member of the foundation's board, suggested a public debate.

The discussion

The discussion was supposed to start with an opening statement from Waugh. Instead, Waugh had trouble phoning in to the show, and Schestowitz began a rambling explanation of his viewpoint. Schestowitz said that he was not prejudiced against GNOME, and acknowledged that the project might have "to interact with somewhat more questionable groups." However, he also seemed to express concern that efforts by projects like Mono (a GNU/Linux implementation of .Net) might expose users to patent infringement suits. In fact, Schestowitz claimed that "Microsoft actually goes into various large companies now and is collecting fees from people who use GNU/Linux desktops and servers without actually specifying the infringements."

At this point, Waugh joined the conversation. According to Waugh, GNOME's involvement is limited entirely to support for Goldberg, "to ensure that Microsoft provide as much documentation as possible to make it easy for him to implement OOXML in Gnumeric specifically. And if he did not continue his participation, he would not be able to hold [Microsoft's] feet to the fire and make sure they came through on the various bits and pieces of documentation" needed for the OOXML standard. The advantage of supporting Goldberg's efforts, Waugh said, is that it helps free software support not only OOXML, but Microsoft's previous binary formats as well. He added that, far from being a Microsoft collaborator, Goldberg has "been a thorn in their side going through this process, because he's forced them to do work that they otherwise wouldn't have bothered doing."

Schestowitz responded that the OOXML standard was "quite poor" in its design because it did not fully support non-European languages and did not make use of existing international standards in its specifications. He also questioned the value of the documentation, given that Microsoft would undoubtedly extend and alter the format in future releases of Microsoft Office. His implication seemed to be that participating in the OOXML standard meetings served no useful purpose.

Waugh replied that the quality of the specification mattered less than the opportunity to learn more about the Microsoft document formats, and help users to live with them. "We do that to make sure that users can interoperate with their friends and colleagues and are able to choose free software even though people around them may choose not to," Waugh said.

Miguel de Icaza then joined the discussion. After answering a few brief questions, de Icaza let Waugh do most of the talking for GNOME. Waugh acknowledged that people were angry with GNOME's actions "because anything that involves Microsoft winds up looking a bit like a threat," and even that, as critics had claimed, GNOME's actions might be misinterpreted as support for Microsoft. However, Waugh denied that the involvement with OOXML precluded support for ODF, which he called the standard "that provides the greatest opportunity for collaboration across the community." He admitted that using OOXML might risk patent threats, but suggested that, these days, any software might face similar threats.

While some groups, like GNOME's rival KDE, might choose not to become involved with OOXML, Waugh said, "I don't think disengagement is the answer to our problems here. It's much more important that we give people the interoperability to use free software. I think we have to fight them at the gate, rather than sit back in our own little community and feel comfortable."

In answer to a question Miller relayed from the IRC channel that was running concurrently, Schestowitz tried to articulate his views about how free software should deal with OOXML. He suggested that GNOME should not continue to participate in defining the standard "because any participation, no matter how passive, will seem like support," and seemed to suggest that free software could not coexist with OOXML. At any rate, he thought Waugh exaggerated the immediate need, arguing that, contrary to Waugh's firsthand experience as a consultant, "not many people, if anyone at all, has received these files. In circumstances where you do, it is very simple to request the file in a binary format if necessary."

Waugh was in the middle of explaining that Microsoft Office 2007, which uses the OOXML format by default, already had a larger user base than, its free software rival, when the hour allotted for the show ran out.

The aftermath

In terms of the audience, the podcast was undoubtedly a success. According to Amis, more than 600 listened via the Internet, and another 50 via the phone. About 45 logged on to the IRC channel during the debate, although presumably almost all of them were also listening in some way. According to Amis, these are at least six times the number who usually listen to a live daytime podcast.

Moreover, if the usual ratio holds true, 10 times the live audience can be expected to download the show over the next week. Three hours after the show, it had already been downloaded nearly 300 times. You can also download a transcript of the IRC discussion logged during the live program.

But was anything resolved as a result? Neither Schestowitz nor Waugh converted the other, although Waugh perhaps persuaded more people because his comments were more articulate and organized. However, both were noticeably more restrained than in their online exchanges, and the lack of accusation and counter-accusation -- to say nothing of the wilder conspiracy theories -- may help listeners to realize that the two sides are closer than they appear.

Both sides are clearly concerned with what's best for free software, differing only in how they define their objective. That is a small point, but perhaps this realization can finally start to put the issue in perspective. is already considering future podcasts on issues in the free software community (with luck, without the glitches of this first effort). If you have any thoughts about future podcasts, or any possible topics, please contact us at


  • Office Software
  • Community
Click Here!