Rapid Application Development for the masses.
censored distributed easy extendable fast informative intelligent lean maintainable open (source) portable pretty rfc'ed robust secure transparent version numbers virtual
So far, all there is to see is a clock and a simple looking chat client, so when
will all this fancy network stuff happen? To answer that question, we have to
cover a little bit of history first. Note, there is a lot more than a clock and
chat client under the covers.
The development of matrix-RAD is largely tied up with the personal history of
onefang since 1999. The main reason is that everybody else was too busy with other
things to do much for the project, although pontiac and runestone, along with
their business, provided much needed support when the going got rough. runestone
now has a wife and a new baby, so he is unlikely to have free time any time soon.
Prior to 1999 I had spent twenty years as a professional software developer coming
up with strange ideas about how to make development go faster, and itching to try
them out.
In 1999 I started writing a login client for the Telstra Big Pond Cable
network (BPC). Since the login client provided by Telstra was very crufty, for
16 bit Windows and Mac only, and their service was unreliable, I had two major design
goals - robustness and portability. Using some of my strange development ideas
produced something that was extremely portable, and bullet proof.
moriarty, pontiac, and runestone were other BPC users that were active in the
BPC community, and becoming friends of mine. BPC had certain well known problems
that quite a few of us tried to find solutions and work arounds for. pontiac and
runestone came up with the original matrix-DFS concept to solve most of those
problems, invited me in for a talk, and together we nutted out the basic outline
of what it should be. The result of that talk is the basis for the high fallutin'
networking ideas detailed on this web site. Since one of the good sides of BPC was
the great community, we decided to make matrix-DFS a community building tool as
well.
We decided that, as the most experienced coder, I would be the head coder (code
nazi they call me), and that we would use my login client as a beginning. At about
that time I was putting some of my other strange ideas into practice in an
unrelated project called Skang. Since we were going to need a GUI for all this
fancy code, Skang got roped into the project too.
While working for Schoolsnet, they asked me to write a web based chat client. After
much investigation and discussion, Schoolsnet decided to let me base it on Skang, and
let me own the chat client, so long as I kept it open source. Naturally, since
matrix-DFS was open source, and Skang was now a part of it, I added the chat client.
Our first community building tool was born, I named it Yabba.
At about the same time, I found out that my digital watch was on it's last legs. Parts
for it are no longer available, so it was on it's last battery. This watch is a
solar powered rechargable type watch, which I have had for twenty years. I have been
looking at watches for most of that time, and have never found anything better. I
decided to build a replacement. Since Skang was now useful as a rapid prototyping
tool, I used it to build a prototype of my watch. I had also built a decent NTP
network for BPC users, so it was a no brainer to add the watch to matrix-DFS.
Getting paid lots of money to write a part of an open source project that was
rapidly becoming my personal opus was great. I had also met a woman, who moved
in with me, and I suddenly had a sex life for a change.
Life was good.
You don't really need to know the details of what happened next, suffice it to say that
the woman turned out to be a crazy violent junkie, and she destroyed my life. Actually, I
shouldn't give her all the credit, others helped independantly. There was a month or two
when I couldn't do any work on anything, and then I escaped with my computer, and not
much else. runestone stored a few of my remaining things at his house, and I got a
cubicle in his office that I could use to continue developing matrix-DFS.
Life was bad.
Being sick of our whole twisted society (see rant.txt), I was planning on running away from it
and living in the Australian bush. I had to do something about matrix-DFS before I left
though. While I was desperatly trying to hold my life together, and getting screwed over yet
again, I went through several months of dumping my brain into the project. As quickly as I could
I would either describe ideas in the TODO document, or implement them roughly in the code.
The plan was to put as much as possible in before I saved enough money to move north, no matter
how rough and bug ridden it was. Amazingly, I didn't break much, except for portability.
This was when I added the throbber, web server, development environment, help system,
web site, and lots of documentation.
I had to sell my printer to runestone, and my computer to pontiac, so I could afford a
backpack and the move to Brisbane. The only computer I had left was a palm top, no
development environment, only suitable for taking notes.
After a few months living up north, I got lucky. HUMBUG gave me an old P100 with no HD.
I got the barest amount of work, and a loan from the department of social insecurity, just
enough to upgrade the computer to something I can develop with. I spent a lot of time
cleaning up the mess I had made of matrix-DFS. For the purposes of the work, which I
managed to organise using Skang for, I added SQL, the beginnings of the widget set, a
servlet module, and the beginnings of security. I started to write an active resume
as a Skang module, which I will use to apply for jobs. I have lost contact with
everybody else. The focus has changed to RAD, so it has now been renamed matrix-RAD.
This is how things are as I write this. The code looks and works a lot better than it did
before I moved, the TODO list is a lot shorter, and I am starting on the docs. That's
why I'm writing this document.
I have no network, one and a half computers, only one OS, I barely have email, and once a
fortnight I get to carry my computer in my backpack on a two and a half hour walk to the
University of Queensland so I can use their network for ten hours. Writing network code
will be easy, testing it will be tricky, and slow.
I'm trying to get some small contracts, so that I can have enough money to
quit this rat race. $10,000 is all I need to get me enough equipment to go bush
and continue programming this thing. I'll come back every now and then with updates.
I intend to finish the active resume, and apply much polish to the system so that I
can use it as a shining example of my work. This should help me stand out from the
competition in the currently highly competitive IT job market. I am currently working
on a major release before the end of the year.
I am now officially dedicating my life to making matrix-RAD THE killer network development app.
This file was last modified on Friday, 12-Nov-2004 10:06:57 EST