Data Exchange Formats

[PAULO A. BUCKUP]

Jim Croft wrote:

>Julian is absolutely right here.  I can think on no institution with
>complex databases (or even simple databases with sundry lookup tables)
>based on different software that have meaningfully exchanged data.  When
>data exhange takes place at all, it is usually done painfully, as an ad
>hoc exercise, one table at a time, slowly, with much human coercion, and
>often at great expense.
...
>I would be interested in hearing if anyone has exchanged complex
>relational data between differennt databases, and if so how they did it
>and what formats were used.  Or if anyone knows of any workable standards
>or protocols for this task.

At the NEODAT Project we spend a great deal of time converting and
upgrading data from about 30 participating fish collections from South,
Central, and North America, and Europe.
Yes, it is true that data can usually be moved from one format to
another.  However, that is really a painful and highly costly process.
We spend 90% of our data conversion effort trying to figure out the
details of conversion between different software or even between
different implementations of the same Collection Management System.

For fish collections we have produced the NEODAT Reference Manual for Data
Exchange which lists international standards and protocols adopted by the
NEODAT Project.  The document is available in hardcopy from the NEODAT Project.

Paulo A. Buckup.