Saturday, February 13, 2010

Indian Openstandards, or double standards

The government of India decided that it was necessary to formulate a standards policy on egovernance. It released a draft policy on open standards for eGovernance-
http://egovstandards.gov.in/Policy_Open_Std_review.
The policy draft was excellent, in that it mandated open, unencumbered (not restricted by patents) standards.

Ofcourse such a policy would be inimical to certain business interests and NASSCOM stepped in with some weasel worded recommendations. The recommendations were submitted inspite of strong objections from many of it's members. In effect Nasscom recommends that
  1. Patented encumbered technologies be permitted for inclusion as part of standards, as it helps wealth creation and will stifle innovation if not permitted.
  2. Multiple standards be permitted , as it provides choice.
The upshot of such recommendation, particularly in the context of egovernance, is that you and i will have to pay for software for using egovernance services. One might note that many of these services are fundamental rights and many obligatory duties.

Firstly standards do not stifle innovation or the ability to earn. In fact it allows much wider participation and hence much more innovation and creation of wealth. Nothing in a standard prevents multiple implementations and thereby innovation, efficiency and consequent creation of wealth. The above is true only if the standard is available freely and without patent / copyright encumbrances, for use and implementation. Thus if the standards are not open and unencumbered, the standard will actually stifle innovation and wealth creation. Quite the opposite result of what the recommendation implied. NASSCOM is confusing wealth hoarding for a microscopic minority with wealth creation for the majority

Secondly multiple standards came into existence PRIOR to or in parallel with the standards process, primarily to carve out niche markets. In most cases geographically isolated regions had set a particular standard and subsequent globalisation forced the use of multiple standards. In every case of multiple standards, the end user bears the hughe costs of supporting interoperability ( by having to purchase additional goods and or services).

Some examples:

PAL /SECAM /NTSC tv broadcast systems.
Every television set had to incur additional costs to incorporate circuitry which most people rarely used.
Every broadcaster had to incur costs converting feeds from one format to the other.

The Indian Railways
A particularly useful example is The Indian railways. IR will be spending a whopping Rs.16500 Cr (and most likely a substantially higher amount) for converting 18000Km of narrow and meter gauge lines to broad gauge in the next few years.
This is excluding the costs of 7000 odd Km already converted since 1992. They need to do so to minmise their costs and provide seamless services. Most readers will be very familiar with the hassle of transiting from Broad gauge to Meter gauge in the course of their travels.

There are numerous examples where not following a standard (eg. roads - Indian road standards are followed more in the breach) results in everybody else incurring a penalty forever. Another example is the humble electrical socket. Everyone knows the hassle of using electronic equipment purchased in one country, when moved to another (UK - USA for example).

The existence of multiple standards should be viewed as a failure of the standardisation process and a burden on the public, instead of being justified on grounds of innovation, creation of wealth and other lame reasons.

There are several other technical reasons (software bitrot, vast volume of data, life times of data, authenticity, security etc), that are even more pressing in case of egovernance, which will absolutely mandate single, open and unencumbered standards.

Only an idiot will commit the folly of proposing and approving encumbered and multiple standards when there exists an opportunity of avoiding it.