Mast Kalandar

bandar's colander of random jamun aur aam

Thu, 09 May 2002

< Web server re-organisation | · | Djkstra passes away >

E-mail regarding open standards


floss [link] [comments ()] [raw]


Hello,

Apologies for this intrusive mail. I am writing to you as a citizen
who is concerned about the manner in which computers will be used in
the administration of our country.

You may have heard that the European Union, the United Kingdom, China,
Peru and Pakistan have recently (in various ways) announced initiatives to
embrace free (as in "azad" or "mukta") standards in the use of electronic
media. On the other hand not enough has been done by state administrations
in India to define, extend and adopt such standards. Meanwhile, some
state-run organisations feel that distribution of material in proprietary
formats amounts to "electronic availability". In turn the government
often insists that applications and submissions made to it be prepared
in proprietary formats.

I feel that it is important that there be some bye-law to our various
electronic communication acts or perhaps some additional legislation
that insists on:

1. The use of open standards to record, store and distribute
administrative documents and related material for our Government.

2. The common availability of the software required to handle these
formats.

3. Some distributed expert-help support structure to ensure that these
facilities can be used by all.

4. The long-term viability of the electronic records. (The formats or
the programmes that produce them or the expertise to handle them
should be recorded in sufficient detail that would allow
reconstruction if the need arises).

Since we are still in the early stages of the use of the electronic
media for adminstrative use, we do not have to invest much in the
"migration costs", which will be substantial if this change is required
a later date.

Below is an analogy of the exiting software scenario with a similar
scenario in the paper and ink world. A draft by Mr. Bruce Perens of
the SPI (Software in the Public Interest) on what constitutes an "Open
standard" in the realm of electronic media is available at

http://perens.com/OpenStandards/Definition.html

Regards,

Kapil.

An analogy:
==========

Suppose the government insisted that all official documents be recorded on
paper produced by one manufacturer on the basis of *that* manufacturer's
advertisement (rather than guaruntee) that the paper will last thousands
of years. Moreover, it is also mandated that the paper be written on
with ink made by the same manufacturer which is invisible unless seen
with glasses provided by the same manufacturer.

Also assume that the manufacturer distributes the paper, ink and glasses
very well so that these are available in all the shops and don't cost
too much. The manufacturer is also giving the material free to schools
in poor areas. The manufacturer also gives jobs to a few people who can
be trained to improve the paper, ink and glasses.

Finally assume that is not permissible to submit the paper or ink or
glasses to scientific analysis to check the manufacturer's claims.
This is in spite of the fact that there are known cases of important
documents where the paper crumbled or the ink vanished or the use of
certain glasses actually destroyed both ink and paper. In each case
the manufacturer comes out with a "new version" of the system where
the defects have been ironed out.

I think it will be foolhardy of any public body to place its trust in
one such manufacturer or even five or six of them. What one can trust
is a system where the government dictates certain technical criteria
("standards") that the paper and ink should satisfy. All manufacturers
can then compete to produce paper and ink that satisfy those criteria.
A small ("khadi") paper industry can also try to produce the relevant
material in the village if it wants or needs to do so.

End of Analogy.


Archives

< May 2002 >
SuMoTuWeThFrSa
    1 2 3 4
5 6 7 8 91011
12131415161718
19202122232425
262728293031 

2016, 2015, 2014, 2013, 2012, 2011, 2010, 2009, 2008, 2007, 2006, 2005, 2004, 2003, 2002, 2001, 2000, 1999, 1997, 1995,