Introduction to Revision Control

Revision control is a means of recording incremental steps in the development of software or other text-based documents. By using revision control, it becomes possible to undo changes that have been made if the changes prove to be untenable. It can also provide a means to limit who can modify particular files as well as to identify who made a particular change.

This is potentially very useful in the development of instrument software where there are multiple developers who may be working in physically removed locations. Up until the present, we have operated in an environment where control of the source code was maintained entirely by word-of-mouth and not enforced by the operating system. The individuals working on an instrument would agree on who should be working on a particular file at a particular time. As it happens, that approach was largely successful, but there was little in the way of record keeping. If any source code was checked in, it was usually by Norton, and there was no record of who actually editted the software.

As part of an effort to provide somewhat more comprehensive documentation of the software for instruments, I would like to promote the use of revision control. The implications are as follows:

See Also:

Some other references are:


Return to Guides Index Page
(c)1998 Norton T. Allen