On document formats, Microsoft slowly bends to the customer's will
The hi-tech world woke up to news this morning that Microsoft has finally decided to put some wood behind the OpenDocument Format (ODF). I deliberately chose the word wood instead of support even though most reports are construing this as some form of support. It is, but not in the way you'd typically expect.
ODF is, as its name suggests, an open format for the storage and retrieval of documents that are created and accessed by productivity applications including word processors, spreadsheets, and presentation software. Although Microsoft disagrees with the positioning, ODF is viewed by most as a substitute for the document formats long-employed by Microsoft Office: formats that, due to their historically proprietary nature (in other words, not openly deployable by competing products), have afforded Microsoft with its Office suite uncontested domination of the desktop productivity application marketplace.
Although the specification was published last year by OASIS, it was primarily the work and politicking of Microsoft's rivals such as IBM and Sun (companies with an interest in breaking that domination) that brought the ODF format to fruition, established its adoption as an agenda item for organizations to consider, and saw to it that ODF was blessed with the imprimatur of the International Organisation of Standardization (the ISO). That work, particularly on the adoption front where governments are seeking more control over their IT infrastructures (cost, choice, etc.) than the addiction to proprietary technologies such as Microsoft Office typically permits, has paid off.
Over the last 18 months, amidst quite a bit of high drama (for the technology business) and despite the Redmond,WA-based company's continued derision of ODF and everything it stands for, Microsoft has very incrementally been capitulating to ODF's tour de force. That capitulation began early last year (2005), when, in an effort to preserve its foothold in the Commonwealth of Massachusetts (and the appeal of its formats), Microsoft relaxed the licensing terms to its XML schemas for Office so as to freely permit the sort of third-party support that was important to that state as well as other governments. But, when, after exploring the definition of "open" more deeply, the state's IT department still judged Microsoft's formats to be too-closed on the basis of their single-party stewardship and lack of access to open source developers, Microsoft again tried to assuage Massachusetts and others by not only coming up with a license to its XML formats for the yet-to-be released Office 12 (now, officially, Office 2007) that was more open source friendly, but by essentially putting those formats on the same type of course that ODF followed to becoming an ISO-ratified standard.
Despite lingering questions about the hypocrisy by which two formats for the same thing both end up as ISO standards (hey, the greatest thing about standards is that there's so many of them), most people have a limited appreciation for the distance Microsoft covered in less than 18 months. Office is widely regarded as Microsoft's key to the company's domination of the desktop. So long as Office is the only product that natively supports most of the documents created over the last decade (and new ones being created as we speak), most PC users need Office and, with the exception of the handful of Mac OS users running the Mac OS version of Office, pretty much everyone who needs Office needs Windows too.
Windows and Office have been and continue to be the core cash cows in Microsoft's kingdom -- a kingdom whose biggest and most impregnable embattlement is probably its existing and oft-used proprietary Office formats (the Office 97 formats which have gone unchanged trhough Office 2000, Office XP, and Office 2003). Second to that is probably the proprietary connection between its Outlook e-mail clients and its Exchange e-mail servers. By beginning the deconstruction of what has historically been its biggest embattlement, at least with respect to its newest file format, Microsoft appears ready to give away some keys to the kingdom. As evidenced by the two steps it took to that openness rather than just one big one back in early 2005, opening up what traditionally has been kept closed was not an easy pill to swallow. But, in the face of a growing "open establishment" --- an establishment that has already taken a significant toll on the company's server business and that is now clearly poised for a desktop invasion (especially on the international front) -- the company had little choice.
Unfortunately for Microsoft, giving the kingdom key that it gave away may not have been enough which brings us to today's news. Historically, new versions of Microsoft's Office like Office 97 and productivity applications (prior to the existence of "office suites") have come with updated file formats and once those versions of those applications got some serious market traction through new purchases and upgrades, the new file formats that they supported have ended up as the new de facto file format market standards. The formats that originally shipped with Office 97 are in that position today. So, in a world where ODF didn't exist and where Microsoft Office 2007 was on track to usher-in a new file format, that format -- in this case, the Office Open XML file format -- would probably have ended up as the de facto market standard since Microsoft plans to make it Office 2007's default file format.
Had Microsoft ever in the past given the keys away to what would eventually become a market de facto standard, it would have been no small gesture to the open world and ODF would never have come into existence. But now that ODF does exist, and now that it's an ISO standard (even though I find that to be a dubious honor), Office Open XML's chances at becoming the de facto market standard are no longer what gamblers call "a gimmee" (the sure bet that it was with previous Office formats). Thusly, from the market's perspective, those keys aren't nearly as valuable. Especially since it's not in use yet. Today, the Office Open XML file formats are only being tested as a part of the Beta 2 distribution of Office 2007. Once Office 2007 ships (supposedly, at the end of this year), its anybody's guess as to how fast it will be adopted and, where it is adopted, whether or not users will accept Office Open XML as the default file format or if they'll switch it back to the old Office 97 formats.
According to a Microsoft spokesperson, approximately half the installed-base of Office users were on Office 2003 by the end of 2005, roughly 2 1/3 years after it first shipped in August 2003. Microsoft will be offering Open Office XML compatibility updates to users of Office going back to Office 2000. But if Office 2003 (which came with its own XML formats that have not been widely adopted) is any indicator, then my guess is that it's going to be at least two years (an eternity given the way things are changing) before we have any real sense of Open Office XML's market traction, and subsequently, whether Microsoft's efforts to make it open will have mattered.
Microsoft points to the 275 partners that are poised to ship their solutions with Office 2007 as a sign of significant traction. But I liken that to the growing number of applications that support ODF (that the ODF camp loves to remind us of). The real test of either will be the adoption by end-users and, right now, ODF may be ahead, but only so due to the sheer fact that Open Office XML isn't shipping yet Adoption of ODF also pales in comparision to the usage of Office 97's format and so, if ODF has a lead, it isn't by much. And Office 2007 won't need to get into too many hands for Open Office XML to pull even, if not pass ODF from an installed-base point of view.
That said, ODF is getting enough traction, especially on the international front, that Microsoft sees it as a threat to Office. Without support for ODF, Office automatically gets bumped off the procurement lists of any organizations (especially governments) that decide to go the ODF route. With more organizations going that route now that ODF is ISO ratified, Microsoft had a choice: say good riddance to those customers or find another ODF-compliant way to keep Office on the procurement list. Judging by yesterday's announcement which was another baby step towards a place that Microsoft probably never imagined itself in just a couple of years ago, it wisely chose the second path. It's a circuitous one. But it's a path nonetheless.
How circuitous? Well, let's put it this way: Microsoft has left itself some room for more baby steps.
In yesterday's announcement, Microsoft basically said that it will be the sponsor of an open source project, the net results of which will be a technology that can translate Open Office XML documents into ODF documents and that will be available in at least one form factor -- a plug-in for Office that allows users to open and save ODF-based documents through Office's menus (rather than having to run an external utility). The process, which apparently involves exporting to ODF even if the file you're saving was opened as an ODF file, may not be as smooth as working with Office's native formats. Nevertheless, compared to where the company was at the beginning of 2005 with respect to ODF, its position today is stunning. As I said earlier, the distance it has come, for such a large and historically proprietary company should not go underappreciated. For the last 18 months, we've watched in real-time how the world's most powerful software company has been forced to reconcile its culture with the open direction the world appears to be heading and, as slow and as incremental as the metamorphosis appears to be happening, it is indeed happening.
What's left to do? Well, this was, after all, just a baby step. For example, while the technology applies to the conversion of Open Office XML to ODF, it doesn't apply to the conversion of older Office file formats. So, since Open Office XML's acceptance in the marketplace is still a question mark, so too is a conversion tool that works with it but not any of Office's older file formats. According to a Microsoft spokesperson, unless end users turn to third party utilities with which Microsoft has no involvement, going from the old Office 97 binary file formats directly to ODF is a two-step process. The first step involves saving the files in the Open Office XML format and the second step involves converting it from that to ODF. The process can be done one file at a time through Office.
Alternatively, Microsoft will be offering a non-open source but freely downloadable bulk converter that converts files in bulk from the old binary formats to Open Office XML. Given the open source nature of the Open Office XML-to-ODF technology, it's entirely plausible that a bulk converter for the second step will come from the open source community. That said, giving that it's technically feasable to go from one (the binary) to the other (Open Office XML) and then to the third (ODF), it's also technically feasible to remove the unnecessary friction (the middle step) by collapsing the conversion process into one step (or, to something that appears to the end-user as one step).
Not only isn't Microsoft releasing such a one-stop tool (another nice baby step it could take), the bulk converter's non-open source nature means that open source developers won't be able to easily stitch the two technologies together to produce one. One problem? To open source the bulk converter, Microsoft would probably have to also change the license for its older binary formats to something open source friendly -- something it hasn't done it's apparently not yet prepared to do. But that could be another baby step. Or, at the very least, Microsoft could offer a proprietary bulk converter that converts to whatever XML format the customers wants (Open Office XML or ODF). Call that baby step 3.
Citing issues relating to the 275 software partners and "loss in fidelity" (a term that comes up a lot when Microsoft downplays the viability of ODF), Microsoft says there's plenty of technical justification for why bulk converting directly from its older binary file formats to ODF in one fell swoop is unnecessary. But, as an advocate for the millions of users of technology (and not of the 275 partners or any particular vendor), my position is to provide the functionality (if it's doable in two manual steps, it's doable in one) and leave it to the end-user to decide. If, as Microsoft says, ODF users will be denied a certain amount of fidelity that they would otherwise have with Open Office XML (a claim that the ODF establishment refutes), then Microsoft has nothing to fear. Most users are saavy enough to know when something doesn't meet their needs.
Yet a fourth baby step that Microsoft could take would be to just natively build ODF support right into Office so that a plug-in is not necessary. The open source nature of this announcment is a double-edged sword. On the one hand, open sourcing a certain body of code stimulates innovation and ensures that the relevant formats and technologies will be available in perpetuity (as opposed to proprietary, closed-source technologies to which long term access is never a guarantee).
But, on the other hand is the question of support. While Microsoft is sponsoring the development of this open source code, the code itself is coming from third parties and is not officially supported by Microsoft. According to a Microsoft spokesperson, Microsoft will take bug reports and it will be a Microsoft product manager's job to relay the bug reports to the open source developers working the code, but Microsoft will not support the code or taking responsbility for its malfunction. This may be OK for a certain segment of Microsoft's customers. But there are very large segments of customers -- particularly in the government and corporate spaces that explicit support matters to -- where any announcement that's not officially supported in any of the existing support plans registers as little more than lip service. By incorporating the technology directly into Office 2007, which Microsoft still has time to do, the technology is automatically included in whichever of the Office support programs an organization is signed up for.
In the big picture, it has always been the customers of promising and viable underdog technologies that force the hands of the status quo leaders. Not the technologies themselves. Customers of AMD's hybrid AMD64 technology come to mind. Intel came up with every reason in the book why AMD64 made no sense to customers. But as soon as it started offering the same technology, those same customers to which it theoretically made no sense started buying it. Microsoft has been at this crossroads before, it's here now, and it will be there again. In what it says, Microsoft may come across as thinking it knows what's best for it's customers. But, as evidenced by where it was at the beginning of 2005, and where it is now, in what it does, Microsoft is clearly recognizing that customers often know what's best for themselves. Said Microsoft's GM of interoperability and standards Tom Robertson of the company's sponsorship of the open source project:
"We're hearing that (customers) don't want homogeneity--they want diversity; they want translatability," Robertson said. "And some customers are saying they would like us to focus on this to a certain extent, to make sure the product is high quality."
If that's really true and the company has already come this far, expect more baby steps. The customer is in charge.
Disclosure: In the spirit of media transparency, I want to disclose that in addition to my day job at ZDNet, I'm also a co-organizer of Mashup Camp and Mashup University. Microsoft and Sun, both of which are mentioned in this story, are sponsors of both upcoming events. For more information on my involvement with these events, see the special disclosure page that I've prepared and published here on ZDNet.