[Date Prev][Date Next] [Thread Prev][Thread Next]
[Date Index] [Thread Index] [New search]

RE: The military, MS, and XML

It has been awhile since I have contributed to the Framers forums
because my current writing efforts have been for readers that are only
Word or HTML users. However, I am in a professional role as a "future
taster" for the company I work for, and I am also currently involved
with writing a hypertext book on the evolutionary and revolutionary
history and future of knowledge management tools and technologies.

Richard and Chris both make some comments here that I think would
benefit from further discussion.

The major early support for SGML and XML did come from the military for
two important reasons.

1. To facilitate management and training standardised document
structures were considered to be important. The SGML file output
specification established styles independently from the creation of
text, and the DTD enforced a consistent logical structure on documents
conforming to the DTD. Once the authoring environment was established
and understood, authoring was greatly facilitated because authors no
longer hand any responsibility or concerns about output formatting. By
contrast to MS Word's continual formatting issues, authors could achieve
up to 50% more productivity in the SGML environment.

2. Configuration management of elements of text within documents (where
particular bits of text relate to particular physical bits of the
product as delivered) is highly important in aerospace and defence
engineering projects where engineering changes are a fact of life, and
no two vehicles or facilities being documented are identical. SGML/XML
in conjunction with a content and/or product data management environment
makes this fairly easy to achieve. Given that the military are very
major customers for complex engineering projects, many of the defense
standard DTDs incorporate a number of configuration management
capabilities that greatly facilitate this kind of management. Properly
understood and used, the same advantages could be used for software

In this sense the defense industry leads the way because its incentives
were very strong. The result is not "bad", even if you consider anything
to do with the military as being inherently evil.


Microsoft Word achieved its market dominance through a phenomenon some
call "network externality". 

In short, when people exchanged information and knowledge using
physically ponderous paper documents, what tools were used to put the
words on paper was a relatively unimportant issue. Many different word
processing applications could survive in the market - although the rule
of three would probably still operate as the market matured - with one
main plus a couple of other important suppliers.

However, as soon as people started exchanging knowledge and information
using electronic formats that had to be readable and preferably also
editable by both sides of the exchange, it became critical for both
parties to have compatible applications. As soon as a few large
communicators adopted the same system all those communicating with them
needed to adopt compatible systems. In such an environment, once one
communication standard achieves dominance everyone else has to fall into

XML may (and probably will) change this for two reasons.

1. the medium of information/knowledge exchange would no longer depend
on specific proprietary formats - and Microsoft has generated enough
resentment by its monopolistic practices on top of the advantages it
gained more or less accidentally as a result of network externality that
many communicators have an incentive to look for other standards for

2. XML's mark-up provides (or at least can provide) a lot of valuable
semantic information about the content being communicated that is able
to be parsed and processed by appropriate knowledge management
applications. Word's format oriented mark-up is vastly more difficult to
parse semantically. Microsoft faces a major conundrum - do they adopt
XML as a medium of exchange? In this case the network externality effect
no longer reinforces their monopoly. Or, do they try to disable XML as a
non-proprietary standard, and risk losing everything if enough people
begin using XML for it to become the preferred medium of exchange?


Bill Hall
Documentation Systems Analyst
Strategy and Development, Tenix Defence
Williamstown, Vic. 3016 Australia
URL: http://www.tenix.com

Honorary Research Fellow
KM Lab, School of Information Management & Systems
Monash University
Caulfield East, Vic. 3145
URL: http://www.sims.monash.edu.au/research/km/

-----Original Message-----
From: Richard Combs [mailto:richard.combs@voyanttech.com]
Sent: Wednesday, 30 April 2003 1:55 AM
To: Framers List; framers@omsys.com
Subject: The military, MS, and XML (was RE: framers digest: April 28,

Chris Despopoulos, maker of fine FM plug-ins, wrote: 

> You ignore the reason XML exists...  SGML was made to satisfy 
> a need of 
> the US military - data that is not "trapped in proprietary file 
> formats."  In fact, this is a military requirement.  

I always thought the military requirement was for mind-numbing,
formulaic, rigid, and anal documentation -- which SGML certainly
facilitates. ;-)

<snip> I guess the military does blow smoke, but 
> considering they currently account for a full half of the US gvmt 
> budget, that's enough smoke to mean something.

Whoa, can't let that factoid pass. The FY03 federal budget is nearly
$2.2 trillion. The defense portion of that (_including_ intelligence and
Dept. of Energy nuclear-weapons-related spending) is $396 billion, or
_18%_ of the total. And that's up a percent, after a whopping $46
billion increase from FY02, the biggest one-year increase since Vietnam.
[These numbers are from Council for a Livable World
(http://www.clw.org/milspend/dodbud03.html), which generally opposes
military spending. The CBO's numbers are slightly lower.) 

As a Libertarian, I'd like to see lower defense spending (and a much
lower federal budget). But facts are facts, and the liberal canard that
defense is the biggest part of the federal budget has _never_ been true
-- at least, not since WW2. In fact, until FY03, it had been shrinking,
relative to total federal spending and to GDP, for many years. 
> In terms of data, proprietary means only accessible via an inherently 
> limited set of tools.  Maybe that's a problem, and maybe it's not.

Tool sets are always limited. Everything is, except ideas and opinions.
;-) As I said in earlier posts, future accessibility is affected more by
the ubiquity of the file format than by whether it's proprietary or not.
Whether you like it or not, MS is the current winner on that score. 
> I see it sort of like the argument between DC and AC current. 
<snip details of AC/DC analogy>  Imagine the cost of conversion to AC! 
> This is precisely what is referred to when people bring up the "MS 
> Office barrier of exit."  XML is like a transformer that can wire your

> DC appliances to AC, and your AC apliances to DC.  <more snip>

I'm not sure I buy the analogy at all. Isn't XML (like SGML) primarily
the storage medium (and secondarily transport)? If there's a
"transformer," wouldn't it be the schema, DTD, or whatever that converts
from storage to presentation? 

In any case, it seems to me that MS's "crime" once again consists of
enabling -- OK, encouraging -- their customers to use MS's proprietary
extensions of a standard. If most users find increased utility in those
extensions and adopt them, then the sheer numbers mean that virtually
every competing vendor for the next 20 years will offer tools for
dealing with the MS files. That's why I said earlier that there's safety
in numbers. :-) 


Richard G. Combs
Senior Technical Writer
Voyant Technologies, Inc.
richardDOTcombs AT voyanttechDOTcom
rgcombs AT freeDASHmarketDOTnet


** To unsubscribe, send a message to majordomo@omsys.com **
** with "unsubscribe framers" (no quotes) in the body.   **

** To unsubscribe, send a message to majordomo@omsys.com **
** with "unsubscribe framers" (no quotes) in the body.   **