The RUP is iterative, incremental and in many ways looks like an agile methodology, but in many others it is different from any agile methodology. I will be listing down the points:
Number of artifacts
Martin Fowler’s "Using UML and Standards" book for understanding object-oriented design says that "The PU was not intended by its creators to be heavy or not agile, although its large OPTIONAL set of activities and artifacts have understandably led some to have this impression. On the contrary, it was intended to be adopted and applied in the spirit of lightness adaptability - an agile PU."
So, originally UP had no interest in inserting a large number of artifacts into the process. It only suggests the possible artifacts to be used. In some other parts of the book it is said that if you want to use other artifacts, running away a little from those suggested, there is no problem. That is, there is no rigidity as to the number and type of artifacts, only indications.
The big problem is implementations of UP, Like the RUP, they ended up forcing a large number of artifacts along their process and this differentiates it from agile methodologies that usually employ agile modeling with its principle of documenting only when it is necessary to communicate more specifically.
Iterations have all phases of the methodology
OBS FOR THE RUP: There are 4 phases, Design, Elaboration, Construction and Transition. Each phase consists of objectives and several iterations that will make this goal be achieved through applications of disciplines that correspond to Analysis, Planning, Modeling, Development and Testing.
The agile methodologies have in their iterations all the steps (Analysis, Planning, Modeling, Development and Testing), ie, the iterations are complete processes that will result in working requirements. In the RUP, each iteration will apply disciplines that correspond to these steps described above, but the level of application depends on the focus of the phase that the iteration is taking place. For example, in the Elaboration phase, disciplines related to modeling will appear with greater application in the iterations, but this same discipline does not have great application in the iterations of the Transition phase.
Scope
It is known that in agile methodologies the scope is not defined in advance or in full at any stage. In the OR there is a great focus on the definition of the scope in the first phase of the methodology, which is the Design phase. Although the RUP preaches the feedback of information throughout the process to refine the requirements throughout the process, the Design phase brings a greater focus the definition of the previous scope, which evades the characteristics of an agile methodology.
Finally, It is possible to approach it in a more agile way?. It is not difficult to approach the RUP as a more agile vision. Just return some concepts of PU in the OR, such as malleability in artifacts and make the Conception phase a little more open as the scope definition. What I believe is more difficult is to try to reduce the focus of each phase and iteration, addressing in each more of the whole process and making them more complete.
In addition, Martin Fowler says (R)UP may be agile under certain conditions: https://www.martinfowler.com/articles/newMethodology.html
– Piovezan
Great record, @Piovezan !
– Tássio Auad