Ready.

Page 36. This page is in the page-by-page raw-dumps section.




New names without any [new] substance/abstractions

On page 36, copycat82 starts telling about itself. (Upto that point, it is literature overview.) The copycat82 lists its adopted modifications to Petri nets, in five parts, and right there, it should be noticed that the structure is not anything new, and the later, more detailed, pages will only make this clearer.


Modify_1: places/locations as control variables

The first "modification" is a renaming of the Petri net places as "control state variables." The name "control graph" is not new with copycat82. VD78, for example, names its Petri net graph, as the "control graph", as opposed to the data-graph, which has data/variables. The renaming does not bring any new abstractions, any new ways of managing the Petri net tokens, that E-nets and/or VD78 did not already have.


Modify_2: "component"s

The Peterson (1977) tutorial on Petri nets, tells of the hierarchical-modeling idea. The copycat82, as the second modifiedness, again suggests a renaming (p.36):

"(ii) Software components correspond to the nonprimitive transitions of a Petri net, where a nonprimitive transition has a Petri net subgraph as its inner structure and does not fire instantaneously [Petri 75]."

A "software component" is an arbitrary subnet, that is shrunk, and replaces a transition. Other than the possibility of partial-firing, which the copycat82 itself does not study, the assumptions that will be attached to a "software component" will be making it appear as if it were an E-nets T-transition, and/or VD78 subnet, optionally, with some i/o control logic subnets at the entrance and/or exit of a subnet. These gate-macros are drawn similar to SARA (UCLA graphs), and are a faulty/quirky set of cut-and-pastes from E-nets, VD78, and SARA, translated into Petri nets, with quite a few problematic behaviors, in the copycat82 implementation.

The Macro E-nets have two types of abstraction methods. The first, isolates a procedure within a transition, and with a given elapsed-time. The second method, defining macros as arbitrarily shaped subnets, is a further modeling tool. All macros are expanded, before the net is passed to an E-net verifier.

Danthine (1980), for example, when integrating the X-transition of E-nets, to time Petri nets, isolates/manages/encapsulates the internal states (the decisions based on context/environment/data), within the procedures of some X-transition(s).


Modify_3: control transfer

The renaming goes on. With the first "modification," the Peri net places were renamed "control state variables." The ones here fit into the control and data distinction, as presented by VD78, and as merged together by the copycat82 (but see next section, for a discussion of an existential problem, about this). It does not bring any new abstractions, any new ways of managing the Petri net tokens, that E-nets and/or VD78 did not already have. A list of other such trivial name-replacements may help the readers acquainted with E-nets, and/or VD78:

Needless to say, E-nets and VD78 are different researches, but this is a common subset of features between them, among a few others, and the copycat82 is workable, only somewhere within that subset.

And the names preferred by the later published copycat82 appear verbose, but less meaningful, in that context. The name "data-transfer specification" is especially irrelevant, when it means data-access and/or data-updating. (Especially, as the copycat82 considers shared-memory computation, within the "distributed systems" range. Its "example application," at the end of the copycat82, uses a shared memory area, for keeping the (shared) variables, although the description on page 65 appears to suggest a better isolated system).


Modify_4_5_and_6: Representing Data

With E-nets, data is associated with tokens (as token-attributes), which reside at locations (places), and travel among the places (as an array). The copycat82 (on p.36), tells the similar: "(v) Data objects can be associated with places and places can also be interconnected through data objects in modified Petri nets."





Further Reading

(page-by-page raw-dumps section) . . . . . previous . . . . . next




Forum: . . (Fair Menu . . . . . Fault Report? . . . . . Remedy for your case . . . . . Noticed Plagiarism?)

Referring#: 0
Last-Revised (text) on June 10, 2004 . . . that was http://www.geocities.com/ferzenr/decalun.pg036.htm
mirror to mid80.net, on June 18, 2009
Written by: Ahmed Ferzan/Ferzen R Midyat-Zila (or, Earth)
Copyright (c) 2004, 2009 Ferzan Midyat. All rights reserved.
mirror