The development of experimental software is rarely straightforward. If you start making something you don't understand yet, it is very unlikely you get it right at the first try. The iTask system has followed this predictably unpredictable path. In this system, where combinator functions are used to construct interactive workflow support systems, the core set of combinator functions has changed along with progressed understanding of the domain. Continuous work on this system led to the emergence of a new programming paradigm for interactive systems: Task-Oriented Programming TOP. In this paper we reconstruct the evolution of one of the core iTasks combinators to catch a glimpse of this emergence. The combinator is the parallel combinator that facilitates the composition of multiple concurrent tasks into a single one. We reconstruct its evolution from the written record in the form of published papers and discuss this reconstruction and what it tells about the progressed understanding of programming with tasks.