dotgnu-general
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [DotGNU]What's most important for DotGNU developers to focus on


From: JonesJR
Subject: RE: [DotGNU]What's most important for DotGNU developers to focus on
Date: Wed, 26 Dec 2001 19:56:49 +0100

> * Avoid huge monolithic systems which are difficult to modify
>   because everything depends on everything.  Instead, build
>   modular systems where new functionality can be obtained by
>   adding components, or exchanging one or two components.

In my years of experience this has always been my first goal in
projects that I wrote and designed. I totally agree with this
statement. I always say: "First see the global idea and start
making the global framework, then start implementing the modules".
But be sure that anything can connect to the global framework.
I call this kind of development generic design. My favourite statement
about this kind of components is the flowing: "If you want this
component to eat your sandwich it will do so". I've written more than
one library on this concept and that are the libraries that stay up
to date and stay consistent for many years. The users of the libraries
can extend them any way they want.

> * Each component must verify all data which it receives from
>   other components to make sure that the assumptions which are
>   made about this data are correct.

On this point I have to disagree in some way, not every component
needs to do this, some of the components only used in bigger components
and totally controlled by this bigger component don't need to do this
kind of verifing, this is leading to overhead and slow interaction between
the components. Make sure the outer component is data-safe and the innner
components will receive only valid data. I think the principal of
encapsulation is valid not only on a single component (class) but also on
many components controlled by one or more components. The outer components
encapsulate the child components.


Next, I'm looking for a project in which I can find satisfaction for my
hunger
to new technologies.
I want the following features :
  1. Coding: I want to code, no matter what language (prefer C, C++, Delphi)
  2. Design: I want to design big generic framworks so others can benefit
from them.

Interested ?? I'm your man.

I was looking for some work on the pnet and pnetlib projects but found no
todo-list
anywhere, if there is one please let me know, so I can take a look at it.

The following are my specs

 - 2 years coding in C, C++ and ASM mainly on embedded systems, also DLL
writing, no GUI.
 - 3 years coding in Foxpro, mostly on libraries which MS "forgot" to
implement.
 - 5 years design of frameworks for libraries and components for
communication (RS232, TCP, UDP, GSM)
 - 3 years coding in Delphi, mainly communication layers and threading
solutions.
 - 3 years network configuration on NT, Novell and some Linux.

Do not make the mistake of thinking I'm working for 16 years, all this was
done the last 5 years.
This was all on the same time on different projects at work.

Best features:

 - Delphi, C, C++.
 - Writting libraries based on frameworks.
 - Designing generic frameworks.
 - Designing OO components.
 - Communication between applications, software- and hardwarelayers.

In my free time I did the following.

 - Maintained two units of Delphi Jedi Code Library and wrote several
components
   for win32 system programming for the same library.
 - Wrote some python code.
 - Designed some websites, with PHP, Perl and ASP.
 - Wrote an article for The Borland Community about threading.
 - Wrote a course on advanced Delphi programming.
 - And many other experiments.

So please let me know something if there is some work to be done. I'm always
willing
to learn something new.

Greetings,
  Wim De Cleen aka JonesJr
  Software Engineer/Technical Analyst



reply via email to

[Prev in Thread] Current Thread [Next in Thread]