The-Source.com

Interesting reading from Chris Dixon on the tradeoff between open and closed, where Mr. Dixon takes a look at a Harvard Business School paper “Opening Platforms: How, When and Why?”  by Thomas R Eisenmann, et. al.

Information is Beautiful

The Mix of Open and Closed

This chart is from Mr. Eisenmann’s paper, where he is particular in what he considers “Open”:

A platform is “open” to the extent that: 1) no restrictions are placed on participation in its development, commercialization or use; or 2) any restrictions—for example, requirements to conform with technical standards or pay licensing fees—are reasonable and non-discriminatory, that is, they are applied uniformly to all potential platform participants.

I don’t think this definition of “open” holds up well under close examination. I don’t think any platform technically meets the first clause, and I think every platform could technically meet the second.

He does expand on this a bit in the text where he clarifies that the iPhone “is only open for some prospective demand-side users under onerous terms” – so the chart doesn’t capture such distinctions. 

(I contend applying such distinctions would lead to all boxes being “closed” for all non-Linux platforms and obscure the points the paper is trying to make.) 

Even given these caveats, I recommend reading the paper. The main value of the paper to me was seeing a more academic analysis of “open vs. closed”  that employed real-world examples.

Interoperability

Take for example, the analysis of interoperability:

As markets mature, the sponsors of rival platforms who previously have eschewed compatibility may find it attractive to make technical modifications that allow interoperability (Katz & Shapiro, 1985; Farrell & Saloner, 1992).

[...]

Interoperability is achieved through the use of converters, which are also known as adapters or gateways (David & Bunn, 1988).

After which, several properties of converters are listed:

• They are costly, and the cost is “typically borne by the weaker platform”
• They can be one or two way
• They can be developed unilaterally or bilaterally
• Cross-platform transactions might be of lesser quality (perhaps intentionally to “maintain differentiation”)

I would contend that I think all these properties can be  – and often are – intentionally skewed  in order to “maintain differentiation”.

Looking at Linux

The paper makes a very interesting point when talking about the open-ness of Linux:

For Linux and other platforms, openness at the sponsor level entails greater openness at the user level, as it implies not only nondiscrimination in platform access, but also in the process of defining platform standards.

Allow me to expand on this point:

• Open-ness at any “higher” level entails greater open-ness at any “lower” level, and
• Closed-ness at any “higher” level entails greater closed-ness at any “lower” level

The retention of acceptable accessibility is why more often than not, companies based on closed and proprietary practices are unable or unwilling to produce truly open products, protocols or platforms.

I’m not saying it is impossible, I’m just saying that if the core principle of a organization is that closed and proprietary is optimal, it will be difficult to permit open and free practices.

Conversely, if the core principle of an organization is that free and open is optimal, it will be difficult to permit closed and proprietary practices.

The Devil is in the Details Comments

One reason I tend to mention the comments when I talk about a blog posting is because I feel they reflect a larger and more popular interpretation and reaction to the point or position of the author.

The author has attempted to refine and strengthen his point, whereas comments contain a more immediate reaction springing from a current understanding of the problem domain.

For example, take a comment made on Mr. Dixons blog:

But the fact that we’re struggling with the definition of “open” here proves what a useless term it is. The fact that we’re using the term “open” to describe about four or five different points of friction in the process of getting a technology idea to market cripples our ability to move the discussion forward. We need terms that are more specific, and we should make a differentiation between 1) full-stack openness (where every bit used to create a solution is open-sourced from end to end), 2) the openness that is fostered by the presence of a platform, and 3) an open marketplace

I appreciate this comment’s point, but the problem is not so much with “open” being useless as it is the enthusisastic marketing misapplication of the term by companies desperate to appear “Open”, while only relaxing their closed and proprietary grip to the slightest degree possible to make the claim.

To a large degree, the fault lies with Open Source proponents, who – instead of calling out bogus claims – have encouraged them according to the thinking that “a little open is better than no open”. Of course, encouraged that their ludicrous claims to be open have been blessed, companies have more expanded the claim than expanded the open.