Then updates to the customer's configuration requires modifying code, compiling it, and shipping the result. It's easy to put an interface on configuration file modification and in many cases, updates can be applied without needing to restart a running system.
That's the problem with soft-coding, you trick yourself into believing that modifying a configuration file is not modifying code, but it is. It's code that isn't compiled, isn't regression tested.
As far as re-shipping, nearly every language nowadays has a method for automatic updates. Here is what Chrome uses: https://code.google.com/p/omaha/
But it's modifying code that your support team is allowed to modify in the field. By contrast, depending on the business policies in place any code that has to go through a compiler might also have to go through meetings, reviews, QA testing, a release cycle, IT department scheduling, acceptance testing, blah blah blah.
By this logic, every user is "programming" when they change their facebook account settings or configure the appearance of their desktop. Call it what you want, but configuration via "config files" of some sort (perhaps managed by a GUI) is a totally different beast from modifying 'main'.
