From a computer network scientist perspective, this is a very interesting paper, as it considers the use of time in business simulation. However, in this sense, its generic title can be a bit deceptive. In fact, Lainema nicely reviews time theory for economic applications and simulators. The presentation of such a review is interesting from any perspective, and it alone justifies reading this paper, which, by the way, seems just a bit too long by normal engineering standards.

Yet, something is missing. At the beginning of the paper, Lainema considers such games as Harry Potter and the Prisoner of Azkaban, The Sims, and Civilization III. But that’s it; the problems that are supposedly addressed in these simulators are never mentioned again.

Furthermore, although Lainema focuses mostly on financial and workflow simulation issues, his main concern is the debate between batch processing and real-time processing. His objective is to prove that “a continuous flow of time in a simulation game may allow the achievement of objectives not possible in batch-processed simulation games.”

Unfortunately, this paper does not clearly present the main point of building simulators and it is misleading about how simulators are built. Furthermore, without additional sources--for example, Perros’ free ebook [1]--readers will not be able to grasp the basic mechanics of algorithms that address the management of events (rather than the management of time).

Only some very specific simulators address the batch processing of tasks. In fact, simulators deal with time flow mostly in two manners: in an event-advance manner or in a time-advance manner. Each of these approaches means that, in the course of simulation, there is a starting event--the event from which time starts counting, such as the sale order or the player log in--but the ending event may not exist as such, since the simulation may be terminated due to clock overrun or because a goal is achieved.

Multi-gaming, online gaming, and, in fact, all of the recent enriched-environment games are not compatible with a batch versus real-time approach, mostly because the concept of batch is extremely fuzzy--for example, what would be the batch tasks to perform when a player in Harry Potter casts a spell on an animated and randomly behaving opponent?--and the concept of real-time does not really exists in computers, let alone on the Internet.

Moving on, Lainema puts forward some extremely interesting claims. For example, he states that “businesses today often operate in competitive environments that are increasingly turbulent and unpredictable,” and then cites authors from 1997 to 2002. While reading this, one cannot help but wonder what, 40 years ago, business owners would say about their business environment. It seems as though Lainema misses the point. Businesses that have been around for a very long time are now subject to a much wider range of interactions with a much wider business environment; this--not the change of the nature of businesses--explains the increase in complexity. Again, this suggests that the author’s batch versus real-time debate does not make much sense.

In conclusion, the paper’s description of the problem and its classification of time are worth reading; that being said, its initial reference to computer games is inappropriate. Readers should complement Lainema’s paper with a solid computer simulation document.