After more than 25 years in the business as a programmer and system analyst
for large corporations, the answer is yes, most programming in business is
front ends for data base applications, and, in my experience, about 85 to 90%
of business data bases are string fields.

Not in my experience.  Regular expressions are most important when the
incoming string is irregular or what you're looking for is irregular.  Most of
the string work I've done is parsing flat files into fields, where the field
length, starting position and ending position are known.  A simple matter of
substring'ing the fields of data out of the incoming string.  I can remember 2
or 3 times, however, when I wish I had regex available to me, but was using a
language that didn't support it.

You're welcome.

Lee Weiner
lee AT leeweiner DOT org

Yes that is correct. The majority of educated computer scientists that go
work for a large cooperate business get involved in the
database environment. That might database-programming, optimizing,
maintenance and development on current systems etc.
3 years it was deemed extremely important that education, where i come from,
focused extra on databases because the industry would
run short of people who had the capabilities to deal with databases. As a
result of that there is a vast number of database- and database-related
courses being offered.

Not really what to make of that.

Regex is useful if multiple pieces of data have been munged into one
string. The obvious solution to that is not to munge multiple pieces of
data into one string. Or at least do it in a standardised way (CSV, XML,
www-form-urlencoded, etc).

Tom Hawtin

Doesn't fit my (fairly extensive) experience.  Some do work in that kind of
environment, some don't.

This is just plain wrong, even for someone who does work with lots of strings.
Regexps are a bad tool for most purposes (the do have valid uses, but those are
relatively rare).

   -- chris

Let me jump in here and say that while a lot of programming is indeed
about working with databases, one of the important things about
databases is that allow you to store information in semantically
meaningful structures.  By cramming everything into strings, you lose
the benefits of doing so.  Of course, some kinds of data (names and
such) are inherently text, in which case you're using strings... but
regular expressions are mostly used in contexts where information was
improperly crammed into text form.  Proper use of databases is a cure
for this, so it doesn't happen as often in database programming.

My experience is that regular expressions are a marginally useful tool,
which is just flexible enough to allow it to be used poorly to solve
other problems as well for which it's not actually the best choice.  
When regular expression matching is used all over the place, it's time
to get skeptical and ask why.