replicator.org ... continuous garbage collection

Well, not expected that it really works out, I was very impressed: 1500 lines of code of the state machine engine made it possible to replace all (around 20) hand crafted switch/case based state machines in one of the asynchronous frameworks I am working at. The state machines are specified by a syntax directly reflecting the UML state chart terminology, making the state machines much more transparent and extensible.

Yet, the state machine module is not finalized: missing are history feature and a proper documentation. If you know UML state charts and would like to try it out, please drop me a note.

But that’s not actually what I wanted to write about today. As the title implies, this post is more about object orientation and object models.

It was about time: David Harel and the Rational/IBM/UML gang fully specified state machines syntax and semantics and its graphical representation. Combined with a network transparent asynchronous messaging system based on Robin Milner’s Pi Calculus, the justification to model state machines to be used in C++ component development is reborn.

As Alan Cox pointed out
“A computer is a state machine. Threads are for people who can’t program state machines.”
creating a statemachine is not always that easy. But with a little help, we – the small group of 3 million C++ hackers – might get it right for the first time modeling a state machine by breaking it down into states, events and transitions instead of creating chaotic switch case monsters.

So, for the last few days I was not continuing working at the Paramatrix project and focused on some things to be finished going back to as early as in November 2004.

First of all, I found a solution to create Docbook documentation WYSIWYG. After evaluating several software packages I found that XMLMind is the editor of choice.

Secondly, there was a new cyclic list container I have created and used in the last few months with very much success, so I finished the paper and made a contribution to the Libsmart Project. The actual announcement is at this page. You can download the header file for free, it has a BSD style license. There is even a forum set up to collect your experience and questions.

So, to implement the Focusmanager I found out, that there is another small technology needed. Basically, I need some kind of an association system to link arbitrary typed instances to the scenegraph nodes. While rendering the scenegraph, this could have multiple uses:

• OpenGL or other scenegraph renderer need some kind of optimisation attachments, based on the scenegraph node-level. I don’t like the idea of creating inheritance hierarchies for the scene nodes to specialise them. I want them to be allocated and attached separately, and I want to attach instances based on classes not derived from the same baseclass.

• Additionally, the Focusmanager may attach information on a per node granularity, for example to store the relevant information to interpolate quaternions.