Re: [xml-dev] The year is 2027, and we need to examine archived XML docu

["Ken North" <kennorth@sbcglobal.net>]

Roger Costello wrote:
> Suppose the year is 2027. XML stopped being used years earlier, in
> favor of another format. The financial institution wants to examine its
> data archives (XML documents) from 2007. What capabilities must the
> financial institution have to read the archives?

> I think the only capability that the financial institution will require
> is the ability to interpret Unicode or ASCII, correct?

There are different scenarios that depend on the information of interest.

One scenario might be simply needing stock prices for a time series analysis. It
becomes more complicated if you need additional data for multivariate analysis,
for example the history of insider trades, financial reports, and so on.
Sarbanes-Oxley and other regulatory compliance will add another layer of data
and complexity to the requirement for a comprehensive archive.

For the simplest scenario (time series analysis), meaningful tags for the stock
symbol, company name, price, currency, and stock market will make the data
useful - even without a schema or parser.

But if you get into multivariate analysis and regulatory compliance, you have so
many different types of information you'll need metadata and documentation in
readable form to help decipher schemas, namespaces and so on. It would be a good
idea to archive design documents and source code so it's possible to pull up a
complete snapshot for any given point in time. Imagine a lawsuit or regulatory
action that drags out for 10-15 years, such as the IBM anti-trust case
(1967-80).

Organizations with foresight should invest in generations of archive servers
with massive storage capacity, and commit to converting the archives as new
storage technology replaces old.



======== Ken North ===========
www.KNComputing.com

www.WebServicesSummit.com
www.SQLSummit.com
www.GridSummit.com