What subjects benefit from textbook editions?

Take undergrad economics. While standard micro- and macroeconomic theory likely doesn't change quickly enough to warrant a new textbook edition every few years, students might be... irritated... if recent economic events (the US housing crisis, the Great Stagnation, right now the ruble meltdown) were not reflected and discussed.

Suppose the last example of a major crisis in your econ textbook (printed in 2004) were the dotcom bubble bursting in 2001 - today's college students were barely walking back then. This would be ancient history for them.

Yes, of course a motivated instructor could work with an older textbook and provide the updates based on his own notes. This is a lot of effort, though, and apparently few instructors go to this trouble.

With regards to "the most up to date," it certainly depends on the frequency of versioning: for example, I cannot imagine a subject in which a new version every year would be justified from a scientific consensus perspective.

There are, however, fields where the consensus is advancing quickly enough that a new version every 5-10 years would certainly make sense. A number of biomedical sciences, for example, would have this property, as there has been a continuing rapid advance in our understanding of the mechanisms of control within individual cells and their relationship to organism-level behaviors.

Of course, a highly motivated instructor might collect notes and surveys themselves such that a textbook was not needed, but that's an independent axis from your question, I believe...

Communications engineering, from what I understand, is rapidly changing (what with the Internet at all). In fact, any textbook related to computer technology is bound to be severely outdated in several years (with some exceptions). There can be issues, for example, when students are led to believe (as is the case in my telecommunications textbook published in 2005 and used in 2014) that Token Rings are common alternatives to Ethernet--something that this article from 2007 quite firmly denies.

While introductory calculus hasn't changed much in the past couple hundred years, fields that are rapidly changing, e.g. anything involving computers, require up-to-date textbooks. I don't imagine anyone will benefit much from a textbook on internet communications published 5+ years ago as opposed to a current one.