ODF vs. OOXML: War of the Words Chapter 4

This is the fourth chapter in a real-time eBook writing project I launched and explained in late November.   Constructive comments, corrections and suggestions are welcome.  All Microsoft product names used below are registered trademarks of Microsoft.

Chapter  4 –  Eric Kriss, Peter Quinn and the ETRM

By the end of December 2005, I had been blogging on ODF developments in Massachusetts for about four months, providing interviews, legal analysis and news as it happened.  In those early days, not many bloggers were covering the ODF story, and email began to come my way from people that I had never met before, from as far away as Australia, and as near as the State House in Boston.  Some began with, "This seems really important – what can I do to help?"  Others contained important information that someone wanted to share, and that I was happy to receive.

One such email arrived just before Christmas in 2005.  In its entirety, it read:

Enjoy reading your consortiuminfo blog ... keep it up.
Happy New Year,
Eric Kriss

This was a pleasant and welcome surprise.  Until the end of September, Eric Kriss had been the Massachusetts Secretary of Administration and Finance, and therefore Peter Quinn's boss.  Together, they had conceived, architected and launched the ambitious IT upgrade roadmap that in due course incorporated ODF into the state's procurement guidelines.

Naturally I responded to Kriss’s email, and suggested that perhaps he might consider granting me an interview. The answer was "no."  But over the next several months Eric was nonetheless willing to help me out from time to time when I needed insight into current events as they evolved – as they began rapidly to do, beginning with the abrupt resignation of Peter Quinn less than a week later.

Eric usually responded to my questions with single sentence emails or pointers to material already available on line.  His reluctance to share more than might be deemed appropriate after serving in government was as frustrating as it was admirable.  Inevitably, the exchange reminded me of Bob Woodward’s erratic interactions with the gnomic source he called "Deep Throat" during the Watergate era.  But early in April I received an email from Eric that included not only several helpful links, but this message as well:

Also, I’ve been thinking about your offer of an interview and perhaps the time is right.

We settled on a time and place, and a few weeks later I was waiting for Eric in the lobby of a business hotel on Route 128, looking out over the offices of technology companies large and small.

When Eric arrived, my first reaction was to wonder if someone faking his email address had spoofed me.  I had never seen his picture, but knew that he was 57, had helped Mitt Romney in 1983 launch what grew to be a $50 billion private equity fund, and had also served in state government before, serving in Massachusetts Governor Weld’s administration from 1991 through early 1993.  That first time around, he had been both the Chief Financial Officer of Massachusetts as well as its Assistant Secretary of Administration and Finance.  In between, he had been CEO of several very successful technology based companies, two of which he had founded.  Others he had been brought in to turn around – and did.  But the person shaking my hand looked twenty years younger, and was wearing a T shirt, sneakers and jeans – hardly the stereotypical Bain Capital, Beacon Hill type.

Had I done my homework more thoroughly, I would not have been surprised, as Eric Kriss is a man of many parts, most of which aren’t usually found in the same package.  In addition to the stints in coat and tie, he had written many articles, reviews and anthologies, authored four books on music (three or which focused popular piano styles), produced and performed on a Grammy Award-nominated album, and recorded a CD of his own music.  Depending on the year, you could find him as either the CEO of a major Victoria’s Secret lingerie supplier, or the Managing Editor of Shuttle, Spindle & Dyepot, a glossy quarterly for craft weavers. In other words, not someone who would be likely to feel constrained by the usual boundaries of state government and finance, let alone document formats.

As we settled in, Kriss told me that he had never expected to reenter public service after leaving the Weld administration in 1993.  But some time after he merged a company he led called MediQual into a larger company, the phone rang.  It was his old friend Mitt Romney, telling him he was now running for governor.  Kriss agreed to help out, and in March of 2002 became Romney’s chief policy advisor, working up stands on topics like education reform, tax policy, and "smart growth."  After Romney was elected governor in November of 2002, Kriss led the transition team.  Perhaps inevitably, in January of 2003 he found himself part of the administration, sitting in the same offices he had expected never to return to a decade earlier.  Only this time he was not Assistant Secretary of Administration and Finance, but Secretary, responsible for $25 billion operating, and $2 billion capital, annual budgets.

As numbers and a title like that would suggest, the Secretary of Administration and Finance wore many hats.  The former entrepreneur, editor, turn around specialist, musicologist and fund manager was now responsible for overseeing an equally diverse range of functions, including tax collection, collective bargaining, purchasing,  construction and real estate – and the state’s information technology (IT) infrastructure.

Kriss wasn’t trained as an engineer himself, but he did have a strong background in technology-based, fast-growing companies, and most recently had founded a Web development he still operates today.  Consequently, he was comfortable with IT, and with what goes into purchasing, scaling and managing IT.  He also knew how important IT systems were to any enterprise, and particularly to those that dealt with large amounts of information – like MediQual, which had been a medical data management company.  Governments, with their millions of records containing data on everything from legislation to voter data to entitlement programs, were even more data intensive, and therefore IT dependent.  In his new role, Kriss knew that he would have ultimate responsibility to the governor and the voters for how well those systems functioned in the service of the electorate.

One of the people he quickly got to know was Peter Quinn.  When he met Quinn, Quinn’s title was "Commissioner of the Information Technology Division" (ITD), a vague and uninformative title.  Kriss recalled from his Weld administration days that whenever the then-Commissioner of the ITD was introduced to someone, the inevitable reaction was a blank stare.  The first time Kriss and Quinn entered a meeting, Kriss therefore decided to utilize what he described to me as, "the general ability of executive heads to make up the titles they wish."  As he and Quinn shook hands with their guests, Eric introduced Peter as "the state’s CIO."  In fact, the position did not then legally exist (it does now).  But life for both of them was easier from then on, because Quinn’s real role in the administration was immediately clear.

Peter Quinn was also a newcomer to state government, having come aboard during the last months of the previous administration.  Peter’s job, regardless of title, was managing the IT structure of the twenty agencies of the Executive Branch of state government.  The CIO title, however, implied greater control than was in fact the case.  In Massachusetts, each of these agencies not only created and managed its own budget, but also had the authority to purchase whatever IT goods and services it wished within its budget, once approved.  As a result, while Peter had responsibility for making sure that over 50,000 state employees could communicate and create and use information, the equipment that supported all of those desktops was an unbelievable hodge-podge of computers, software and other technology that had accumulated over the last twenty years.  Or, as Quinn, liked to say, "If any computer manufacturer that was ever in existence had ever made something – well, we had one of those.

All of which was hardly surprising, nor too different from what Kriss had expected.  He knew how purchasing was done in Massachusetts from his previous stint on Beacon Hill, and also done enough consulting to know that most governments spend only about 1% of their budgets on IT systems, as compared to the 5 – 7% large businesses dedicated to the same purpose.  And yet governments were far more information intensive than the great majority of private sector enterprises of comparable size.  Moreover, governments were mindful of the benefits of stability, and were typically much slower to migrate from existing systems to newer technologies.  Understandable or not, Kriss thought the IT infrastructure in Massachusetts was out of date and ripe for an overhaul.

Quinn thought so as well.  Like Kriss, he had spent most of his life in private industry, primarily in the financial sector.  Most recently, he had been CIO of a company that managed shareholder records for mutual funds, and was used to better-integrated systems.  Kriss thought Quinn was a "breath of fresh air" compared to what he had expected, and the two hit it off immediately.

Soon, they were talking about what they could do together to modernize the systems for which they were responsible.  Clearly, the state’s IT infrastructure needed upgrading and rationalizing.  With a new governor happy to cultivate a tech-savvy image and to let a trusted lieutenant have his head, why not do the job right?  Instead of simply replacing this computer and that, why not assume a clean slate, and leapfrog from the funky systems they had inherited to the type of infrastructure that an information age government should have?

As Kriss recounted the tale (and as I struggled to keep up with my note taking), it was clear that even three years later he was excited by the opportunity that he and Quinn had embraced.  What had captured their imaginations was not the mundane task of reconfiguring a vast, out of date patchwork system into something more state of the art, but of designing the type of architectural model upon which such a system could be built – in Massachusetts or anywhere else.  At the core of such an architecture would be two fundamental building blocks: "open source software," and ”open standards."  The rules for using those building blocks would be laid out in a master plan, which would function going forward as a living, periodically updated framework that they would call the Enterprise Technical Resource Model, or "ETRM."  And although neither was an expert in either open source or open standards, they were excited by the prospect that they believed these tools offered to achieve two crucial end states: "interoperability" and "vendor independence."

Those aims were important for achieving certain central goals that were becoming both popular in IT circles as well as feasible in the marketplace.  One of those goals was to make it easier for one agency to trade information with another.  Such a capability might seem obvious, until the mongrel nature of the state’s IT infrastructure was taken into account.  Nor was Massachusetts alone in that regard, as state and national governments around the world were struggling with the same mixed environments of systems that individual vendors like IBM, HP, Sun and others had sold them over the decades, each in an effort to lock in its customers.  Each of these systems was "proprietary" rather than "open," meaning that they were built to unique designs that made it difficult for the systems sold by one vendor to operate with the systems of others.  In fact, even systems within a single agency often couldn’t communicate with each other, either, because some equipment had been purchased from one vendor at one point in time to perform a particular function, while other vendors had fulfilled other discrete requirements.  None of these vendors, of course, had any reason to make communication between its wares and those of its competitors simple.

Open standards promised a way to ameliorate this situation, because if two systems used the right standards, they were each referred to as "open systems," and could communicate with each other – becoming, in IT-Speak, "interoperable."  In truth, it was rarely, if ever that simple, but using open standards as pervasively as possible could make it far easier for IT managers to enable their systems to "talk" to each other.  And that would save a great deal of time and money.

Kriss and Quinn believed that using open source software, instead of proprietary software, would also reap large rewards.  In this case, proprietary referred to software that had to be used as it was delivered from the vendor, because the vendor not only withhold the type of computer code ("source code") that would make changing the software comparatively easy, but also made the software available under a license agreement that prohibited the customer from making any changes.

Having access to source code would have benefits for the state beyond the freedom to make changes to software whenever it wanted to.  To begin with, Massachusetts could continue to maintain that software even if the original developer no longer supported it, or had gone out of business.  In short, the state would have become "vendor independent," because it no longer needed to worry about being abandoned by a vendor, or forced to continue to purchase upgrades it didn’t really want.  Moreover, once an enterprise had converted all its systems to open standards and open source software, it would have far more choices in the marketplace, and therefore could drive harder bargains between vendors, because it could fulfill the same need with more alternative products and services from more vendors.

That concept captured the visionary sides of Kriss and Quinn’s personalities.  But if they had spent more time in civil service, they might have been less optimistic about their ability to pursue such an ambitious goal in a political environment.  After all, not only would they need to invade the turf of multiple career civil servants (some of whom worked in departments that did not report to Kriss), but they would need the cooperation of legislators that had little or no acquaintance with technology at all.  Also, it was an open secret that their boss, Mitt Romney, viewed the State House simply as the launch pad for a run at the White House.  His appointees, as a result, weren’t likely to be long-term forces to be reckoned with.

Finally, and most fatefully (as they would learn), while overhauling the State’s IT systems would benefit some large and powerful IT vendors, by definition it would also take business away from others.  Those vendors not only had lobbyists, but plenty of money to spend on them.  As Peter Quinn in particular would later learn, rocking the government procurement boat was not something to be undertaken lightly.

But in 2003, that was a lesson yet to be learned in the future.  For now, Eric Kriss and Peter Quinn could enjoy planning the government enterprise of the future, meeting out of the limelight with their staffs on Beacon Hill in Boston.  When Kriss had accepted Mitt Romney’s invitation to join the new governor’s administration, he had promised himself that his stay in government would be short – just one year.  But before he knew it, one year had turned into three as the ETRM slowly took shape.

->Next Chapter:        Open Standards
In which the questions, "What are ‘open standards,’ and why do they matter?" are answered

<-Previous Chapter:   What a Difference a Decade Can Make
In which Microsoft captures – and locks in – the office suite marketplace

  <All Chapters>

sign up for a free subscription to Standards Today today!

Comments (12)

  1. Absolutely riveting, I can’t wait for the next chapter.

    BTW, missing a closing quote with ‘vendor independence’. In the paragraph that ends as follows:

    achieve two crucial end states: "interoperability" and "vendor independence.

  2. So what is Massachussetts to make of IBM’s Lotus Symphony offering ?

    As far as I can tell, IBM and Microsoft play the role of Coca-Cola and Pepsi; large vendors in the business. In that metaphor, Lotus Symphony is like a voucher for 5 cents off a can of Coke; Coca-Cola doesn’t sell the vouchers, it tries to distribute them to all of Pepsi’s best customers in the hope of persuading them to switch brands and indirectly to grow the business of what they do sell.

    The vendors are caught between the trio of commercial laws, namely Copyright, Patent, and Anti-Trust. Anti-trust means you cannot design a loss-making business, and that in effect puts a floor under the price of Microsoft Office and IBM Lotus Notes below which the vendors will not go. The salesmen get measured on how high above the floor they sell at.

    • So what is Massachussetts to make of IBM’s Lotus Symphony offering ?

      Interesting question but I am not sure what you are asking? Why should they make anything of it?

      As far as I can tell, IBM and Microsoft play the role of Coca-Cola and Pepsi;

      With all due respect, I’m not sure that is the best analogy. I for one have no ability to tell the difference between the taste of Coke and the taste of Pepsi and the taste of supermarket own brand. Maybe I don’t drink enough of it (but I want my teeth to last at least until I am 30).

      I can however easily tell the difference between ODF and MS’s proposed format (see for example this picture). I have tried to read and understand both specifications. I also managed to implement ODF support in a web application using an ODF library, it was fairly painless. I honestly have no idea how I would do that for MS’ format. Also the difference in provided library support is quite vast.

      large vendors in the business.

      Though I’m sure to some companies, knowing that some big companies Sun and IBM support a standard is more helpful than just open source alone, I think that didn’t really matter to Kriss and Quinn in the context of TFA. They were outlining an infrastructure that was in the best interests of the state first; getting companies to tender to help is a stage or two along.

      It might be shocking concept to us in the IT world but in other industries you have this concept that the customer is always right, and that supply will follow demand.

      So IBM’s offering works with ODF. You can save your data in ODF, and a few years down the line, you can switch to Sun’s product OpenOffice, or to KOffice, or to Google docs, or you can hire someone like me to build you a custom made browser-based system that can save and open ODF files.

      I’m sure IBM want you to then buy other IBM products, but you don’t have to, because your data is in a truly open format, a quite elegant format (well, for office software) that any programmer with a basic knowledge of XML and half a brain can implement using any language on any operating system.

      What Microsoft is offering is that you can save your data in their format by buying their tools, and in three years time you can buy their tools again, and again, and again. Your data is in their format which at the moment is supported by precisely no-one else.

      It is a really really complicated format and it is not clear at the moment if there will be any independent implementations or if that is even possible, given the low quality of the specification. So at the moment choosing OOXML is choosing lock-in.

      I think that is quite a clear difference.

      The vendors are caught between the trio of commercial laws, namely Copyright, Patent, and Anti-Trust.

      Well, I think Microsoft is slightly caught in Anti-trust, as regulators around the world queue up for their turn and their few hundred million. Fortunately, Microsoft are good for it, they have a huge pile of money with no original ideas about how to spend it (Zune anyone?). IBM are crafty when it comes to these issues, so they tend to avoid these problems.

      Anti-trust means you cannot design a loss-making business,

      Well again, I think Microsoft sees anti-trust fines as taxes, part of the cost of doing business in the real world, it feels no shame at being dragged into court and told off. IE and Windows Media Player were essentially bundled in as loss-leaders.

    • I am concerned *both* with document formats *and* RFID *and* with "Real ID."  All have the ability to limit freedoms and invade privacy in significant ways.   Your choice is a false one.

  3. I’m sorry, but I don’t know what "… and to let a trusted lieutenant have his head…" means. Is it a typo, or another colloquialism?

    Very interesting, anyway. Thanks.


    • Sorry, yes, a colloquialism, although I would not have thought that this one is unique to America.  To "give a horse its head" is to loosen the reins and let the horse run.  Hence, what I meant here was that they were being given the freedom to run with their ideas.

        –  Andy

      • Ahhh… There is an equivalent expression in British English – "To be given free rein" – I’ve never heard your American one, either in the UK or in Australia. I did guess it might be related to "Lend me your ears", but now I see it isn’t.

      • For what it’s worth, I have never left the United States.  While I have been familiar with "given free rein" for as long as I can remember (but never really thought about its literal origin before), I don’t remember hearing the phrase "giving a horse its head" before.  Although from the U.S., I was as confused about that phrase in the article as the first poster.  Maybe that "head" phrase is a regional thing, or something learned by those around horses.

      • A German newspaper will use a vocabulary that has thousands more words than an American paper, because the latter is dumbed down so that someone with grade school reading skills can read it with ease.  I don’t think that makes US papers superior to German ones – quite the opposite.  Cutting out idioms strikes me as the same thing – they are part of  what makes language and culture interesting.

          –  Andy

      • I like rich language, but idioms, like jargon, often require a cultural background in common between writer and reader – context alone is not always enough. In literature, it can add to the desired effect, but in factual writing it can obscure the message the writer desires to convey, at least without a glossary.


Comments are closed.