Double standards for Microsoft and OOXML

As everybody knows by now, Microsoft has been rebuffed in its attempt to fast track Offce Open XML (OOXML) through the International Organisation for Standardisation (ISO) as a second international standard for office applications.

It takes little intelligence to comprehend that two standards are not better than one. The existing standard, the Open Document Format (ODF), is better conceived, has already gained acceptance with many governments, some US state governments and other bodies, and suffers none of the widely anticipated implementation problems of OOXML. The adoption of a second standard is unnecessary and unwarranted, will distort the market, and is an impediment to the adoption of either standard. In any other business we might describe the process as mischievous, or worse.

The purpose of a standard is commonality. A standard for office applications is badly needed. A standard document format means that an office document created now can be read 10 or 20 years in the future. A standard makes it possible for a document created by a user in one company on one piece of software to be readable by any user in any company on any piece of compliant software. A document can be stored and retrieved, and is not lost when today's whizz-kid office suite has become a thing of the past.

Although Microsoft Office currently rules the roost it hasn't always been so. Back in the 1980s, Lotus 1-2-3 had more than 90 per cent of the spreadsheet market and Excel was a clunky no-hoper. WordPerfect was a billion dollar giant, and Word had less than 10 per cent of the market. Some would say that the subsequent success of Word and Excel owed more to the PC monopoly of DOS and Windows than it did to the inherent virtues of the products themselves, but the once "de facto" standards of WordPerfect and Lotus 1-2-3 have long since flown out of the office window.

Warts and all

The objective of an international standard is to provide a clean intersection between different implementations of software and hardware, and to allow everybody to drive on the same side of the road. Some rules are an obstacle to progress. Some promote progress and innovation. A white line down the middle of the road encourages us all to move at our own speed, without fear or catastrophe, just as long as we all follow a simple set of rules. If there are no rules, there are barriers to entry, and accidents will happen. If there are two standards, some will drive to the left, and some to the right, accidents will happen, and progress atrophies.

It follows that an international standard is not dependent on proprietary interests, does not contain binary specifications, does not contain specifications that contradict existing standards, does not include undisclosed patents and incomplete licensing terms, does not exist to deal with one vendor's bugs, and is not culturally specific. Yet, Microsoft's detractors suggest that, by accident or design, OOXML suffers from all these failings and more. OOXML is not so much a specification, as an incomplete description of Microsoft's existing proprietary data formats, warts and all. One of the better known anomalies is the treatment of 1900 as a leap year to satisfy an ancient bug. The industry deserves better.

There is a standard specification for office application data formats. The ODF was created by a technical committee of the OASIS industry consortium, and has benefited from industry wide participation in its development. ODF has been welcomed by many governments because it is the elegant solution to a problem that has become intractable. How do we share our documents, and how can we retrieve them when Office 2012 has changed its formats yet again? The ODF specification is 600 pages in length. The OOXML specification already occupies 6,000 pages - enough to fill a library.

More relevant to our current discussion is the knowledge that a computing standard allows a user to be platform, vendor and software independent. Standards make networking possible. For this reason, standards have tended to be opposed by incumbent monopolies, who want to sustain their monopolies and create barriers to entry.

In some cases, the incumbent monopoly has subverted the standard to further its own interests. After all, the solution to competition from Linux and free software was stated clearly by a Microsoft engineer, Vinod Valloppillil, in a leaked Microsoft internal report as far back as 1998: "OSS [open source software] projects have been able to gain a foothold in many server applications because of the wide utility of highly commoditised, simple protocols," he wrote. "By extending these protocols and developing new protocols, we [Microsoft] can deny OSS projects entry into the market."

Valloppillil went on to describe areas in which Microsoft has "de-commoditised" protocols, and can do so in the future, with the objective of sabotaging interoperability, namely: "DNS integration with Directory. Leveraging the directory service to add value (sic) to DNS via dynamic updates, security, authentication," which translates as using market dominance in one area as a means of stretching and proprietising standards in another, suffocating innovation in the process. We do not need to ascribe such motives to Microsoft to realise that a second standard for office applications of such length and complexity is a recipe for confusion.

All in the game

The absence of standards for any segment of computing is bad for users, choice is removed, and prices escalate. The problem for Microsoft is that, for the first time in years, it has a realistic competitor in the office space. Whether you run Windows, Linux or the Mac, it makes sense to use OpenOffice for one simple reason above all others. OpenOffice comes at zero cost.

Coincidently, OpenOffice also supports ODF as its default data format, which has made it increasingly attractive to governments. Governments have also been increasingly vocal in demanding open standards for office applications, which some may feel is the primary reason why Microsoft has been pushing for the adoption of a proprietary standard over which it can wield control. The problem remains that implementation of OOXML will not be easy, or possible, on other platforms, and there will be one broken standard for Microsoft, and one for the rest. In this eventuality many governments may well decide to go with the rest.

Microsoft has the ability to do the best by its users, and conform to the existing standard. Or it can continue the damaging political process that has jeopardised the reputation of the ISO amid allegations, justified or otherwise, of "ballot stuffing", "bribery" and "gaming" of the standards system. Standards belong to us all, and are not the property of the biggest and baddest players.

The next round in this battle will take place at a meeting in February 2008 in Geneva, Switzerland, which allows Microsoft the opportunity to propose modifications to satisfy the objections of the national standards bodies. If these fail the process will be abandoned, although the standard can be re-submitted for consideration at a later date. Microsoft owes its countless millions of Office users to play it straight, and follow the industry standard. But nobody is confident that this will happen.

Email to a friend

Print this page