[Wylug-discuss] Why Office Document Standardisation Matters to *Everyone*

[Dave Fisher]

A few posters have questioned why a few of us have a 'bee on our bonnet'
about this issue.

The appalling lack of coverage in large parts of the 'OSS-friendly'
media (Slashdot and The Register spring to mind) seems to confirm
widespread ignorance and confusion on the matter.

For example, if you search for 'open xml iso' on Google news:
  
  http://news.google.co.uk/news?num=100&q=%22open+xml%22+iso+&btnG=Search

You see a cluster of 70 articles in the first item returned.

If you click through the link called "all 70 news articles", you'll see
that the vast majority of articles are MS PR puff pieces celebrating the
arrival of a crippled OOXML-ODF translator, i.e. the standards issue has
been crowded out by something which has very little to do with it.

So why is the translator "nothing do with the real issue"? Or to put it
another way, what is the real issue?

The real issue is:

  *Networked* Data Integrity

At the most superficial level, the problem with Microsoft's proposed
standard is that it ensures that third parties can never exactly
reproduce the data and formatting in MS Office documents, while
pretending/promising the opposite!

In other words, most people will never be able to switch to alternative
products without accepting some degredation of their existing data ...
how many large organisations are ever going to accept that?

But there are even more important practical and philosophical issues at
stake here.

Like Will Newton, I hardly ever use office UN-productivity suites like
MS Office, Open Office or even KOffice.  

I write almost exclusively in plain text or plain text markup languages
(HTML/XML/LaTeX, etc.) ... precisely because I don't want to tie my data
to any particular application.

However, like everyone else in our networked world, I still find myself
having to read and write snippets of information/data that have been
composed or edited by someone else who did use office tools.

Moreover, my everyday life depends on software which automatically
(re)processes data which has been previously handled by other people's
office tools, e.g. my bank account, tax, websites, email, etc.

Tiny mis-translations in such procedures can have annoying, costly, and
very occasionally, catastrophic effects.

Messed-up currency symbols on websites is one example of a trivial
annoyance. At the other end of the spectrum, we could have events *like*
the Mars mission that crash landed because of a failure to properly
convert between SI units and Imperial units.

One way of avoiding such mistranslations is to have unambigous standards
for coding and processing the data and the metadata. So that
applications can automatically (and reliably) validate the data/metadata
*before* processing them.

If hidden and ambiguous data/metadata from MS Office is allowed to be
slung about willy-nilly from one networked app to another, without
third parties being able to check for its integrity, we'll all be a lot
worse off.  

Indeed, non-Microsoft and non-Office users will be far worse off than
the people who use the complete MS toolchain, since Microsoft users get
some 'benefit' from consistent brokeness.  It's similar to the hardware
drivers issue, only *much* more important.

It's probably only in the last 2-3 years that most people have begun to
use applications which occasionally exchange fragments of data with one
another over the net automatically. There will be a lot more of this in
future.

As such, it's only now that the deficiencies of proprietary office
formats have become a tangible issue, which governments have reacted to
with demands for open standards.

Unfortunately, it looks as though the vast majority of governments and
end-users (inluding OSS users) have failed to realise the shere depth
and breadth of the problem.

Dave