<p>

[quote]

i mean software engineers are ultimately engineers, but computer scientists (the ones who work in R&D) don't seem to fit the mold

[/quote]

</p>

<p>One could even debate whether pure R&D computer scientists are even truly *scientists*, rather than just a subtype of mathematician. The demarcation of science and mathematics is that the theories of the former must, in principle, be able to be tested and falsified/verified against empirical evidence, whereas those on the latter can rely purely on self-consistent logic derived from a set of axioms. Just as nobody can ever falsify the statement that the derivative of sin(x) with respect to x is cos(x), similarly, nobody can ever empirically verify whether a particular set of numbers is a computable set with a Turing Machine, simply because we don't have, and never will have a Turing Machine. </p>

<p>However, in fairness, it should be said that certain branches of the sciences are arguably not scientific either. Other than perhaps those relating to AdS/CFT correspondence, large swathes of string theory may be unscientific in that they don't (yet) produce clear predictions that would distinguish them from other theories of physics. Chemical graph theory may also arguably be unscientific. Both of them may simply be exercises in mathematics rather than sciences per se.</p>