The Folly of the Short Game in IT

Primary tabs

In 2002, the then-powerhouse IT company Sun Microsystems was considering purchasing Microsoft Office for its 35,000 staff. The company worked out the per-seat cost (taking into accounts volume discounts) and instead decided it would be more sensible to acquire Microsoft Office competitor, German firm Star Division, for their Star Office product which offered reasonable MS Office compatibility (thanks to Star Division's developers reverse engineering MS Office's proprietary file formats). Sure, the US$73 million acquisition might be more than the first year cost of MS Office, but having bought it, Sun would never pay for MS Office in future years either. Plus, holding the code, Sun could tune Star Office to be precisely what the business needed. And there was another motivation.

With control of Star Office, Sun could attempt to undermine and perhaps break the MS Office proprietary monopoly which, at that time, had completely taken over most markets in the world. The way they did this was to advocate for an open standard for productivity software file formats, and, open up the Star Office source code to the global software community, not unlike Netscape did with their Netscape Navigator "Mozilla" code base, which subsequently resulted in Firefox and Thunderbird, the flagships among other Mozilla Foundation open source applications. Star Office, since open sourcing, has forked into OpenOffice and LibreOffice for good reasons.

Sadly, Sun Microsystem's quite rational decision did not help them prevail over the Microsoft Windows and Office juggernaut. The reason for this: most businesses, institutions, and governments choose perceived short term gains over long term strategic advantage. The market dominance of the proprietary Microsoft offerings shows that

  • people and organisations will almost always act against their best interest for some immediate gain,
  • marketing is very very effective, and
  • people can rationalise making short-sighted decisions if they perceive safety in numbers.

So, what does this mean for us here in New Zealand?

Buying the whipping stick

The biggest software purchaser in NZ is the government. Although tricky to measure (thanks to "commercial sensitivity" dodge clauses in the OIA), we think the government spends 25-40% of the country's annual outlay on IT services and software licenses. The vast majority of that spending is on proprietary or closed software which almost always preferentially uses proprietary file formats, data schema, and network protocols. In this context proprietary means entirely controlled by the supplying vendor, who can change them at their pleasure, unilaterally. Those include almost all Microsoft, Apple, Oracle, PeopleSoft, SAP, ESRI/ArcInfo, applications, along with thousands of smaller players. Even suppliers like Google, Facebook, Amazon, GitHub, Trello, Atlassian, Slack and others (many of them cloud suppliers) who trumpet their "openness" are similarly... closed.

By having this unilateral control of their software, and all the data and files users create with that software, these proprietary software vendors achieve an implicit monopoly over the user: the user depends on that vendor if they want to retain control of and access to their own data. It is effectively a "slow burn" hostage situation - the more the user uses the software, the more dependent they become on that specific software and beholden to the vendor. This is a monopoly because the user cannot simply adopt a competing product without the huge additional cost of recreating all historical data and files or somehow converting them to a new format, requiring the support of the proprietary vendor. Buying this proprietary software means that the user has not only handed the vendor an effective monopoly - a whipping stick, with which to exploit them - it has paid handsomely for the privilege.

This, however, is precisely the position that nearly every government, business, and institution in the western world finds itself today. It is the status quo. As a result, decision makers seem to be blind to the fundamental folly in their current and past strategic decisions. Everyone knows that monopolies are bad, but expedience has won over principle.

Invest in the future

Given that most of the world is in the grips of existing proprietary software monopolies - proprietary software is, if you think about it, not dissimilar from ransomeware because of the cost to users to get out of them - but what is the alternative? Sun Microsystem's approach gives us an example, and, because Sun Microsystems is no longer around (they were acquired by Oracle in 2010), shows us how important it is to circumvent the monopolies.

The Proprietary Imposition

The Sun Microsystems approach might well have prevailed - and the market would be more competitive and costs for software for everyone today would be far lower - if there had been a level playing field. Namely, if the proprietary formats, data schema, and network protocols - necessary to create the almost insurmountable monopolies software vendors try to create - were not allowed. But how could that be achieved?

The reason Sun Microsystems lost out is that they did not succeed in rallying enough of the market to their cause. The insidious thing about proprietary software is that - in order for people to effectively collaborate using it, they all need to be using the same software. Not "compatible" (there's not really any such thing, because that's not in the controlling vendor's interest) but "the same". If an influential entity, say, a national government, adopts a proprietary software package and (either out of ignorance or malevolence) requires people to use that software's proprietary file formats in order to, say, achieve required government compliance or do business with the government, then it becomes a de facto requirement that people use that proprietary software package.

Virtuous Customer Cartels

So far, we have seen few instances of customers organising themselves to combat the big software vendors - if they can get enough of the market to rebel against the vendors' terms, they can force the vendors (who want to stay in business) to change their practises. An coordinated, organised customer base could demand that proprietary vendors do something which they will fight tooth-and-nail rather not do: give up their monopolies... This is where governments come in. In New Zealand, the government could create a mandate that government software procurement was limited to software adhering to vendor-neutral open standards - that can't be controlled by the whim of a single vendor like the current status quo.

Can I get a mandate!

Such a move would not be without precedent: for all their recent policy blunders, one thing the UK have done right in the past few years is to mandate open standards for software in UK government. With the stroke of a pen, we could do the same in New Zealand, and all the software procured (and used to interact with the denizens or citizens or residents of NZ) by the government would need to use open standard file formats, data schema, and network protocols... overnight, the proprietary vendors would have to ensure good support for those open standards if they wanted to keep their contracts. Suddenly, our government would be able to consider a much broader range of competing products and services: an actual, competitive software marketplace - the "level playing field" of legend and lore! Plus, it would have the option to use open source software (which is typically very good at supporting open standards, because why wouldn't you if you don't have a proprietary lock-in motive), supported and tuned to NZ requirements by domestic software firms (bolstering their own businesses, increasing export opportunities), giving NZ Inc. far more control over its own software destiny. Win. win. win.

Oh, and another "win" occurs to me - with such a mandate, Hon Peter Dunne could trumpet this bold, tranformative move as the first decisive step towards the admirable goals of the D5 Charter to which he signed up NZ back in 2014 (but with very little to show for it since).

Add new comment

Note to commenters: due to problems with spam comments, your comment will only appear on this site after it's been deemed (by me) to be legitimate.