[texhax] OT: explaining the problems with long term storage of data in proprietary formats

[Christopher W. Ryan]

I apologize for the OT question, but I can't resist tapping the
knowledge of this group, who by definition tend to have a deeper
understanding of open-source software issues.  Further correspondence
could be off-list if preferred.

At the medical school where I work, we are about to take up the topic of
tracking students' experiences:  how many of what kind of patients they
see during their four years in school.  This is a new regulatory
requirement from our accrediting body.  I'm a member of a small working
group looking at strategies and options.  Most would involve some sort
of electronic database, of course.

There are many new companies that are offering web-based
products/services to do the tracking. Students enter info (age, sex,
diagnosis, etc.) and it is all transmitted to the company's servers.
The companies offer periodic reports back to students and faculty.
There are varying amounts of eye-candy and features to differentiate one
product from another, all trying to get market share.

I know little about these companies so far. Specifically, I don't know
in what format they store the data.  (I suppose SQL would be a good
bet.)  But one of the concepts to which I want to introduce the
subcommittee is the idea of storage of digital data, file formats,
long-term usability of data, and ease of migration to other solutions.
In essence, I want to make sure that if a vendor we choose goes belly
up, that we can still have our data.

These issues are completely foreign to the other members of the
committee.  Very bright people, but that does not mean they have a
sophisticated understanding of computer information systems, and even
less so of open-source philosophies.

Can anyone point me to on-line resources that explain, in *very* simple
terms, the problems of storing data in proprietary, non-open, file
formats?  It would at least get them started thinking about the issue.

Thanks.
-- Chris Ryan