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

The Linux of Financial Trading Systems



PureBytes Links

Trading Reference Links

[posted to the list by request]

Wow! We are getting some great responses via email and on the newsgroups. 

What most of us want out of this project is a tool for trading that will do
almost everything we want, and -- since we will have access to the source --
can be made to do all those things we want to extend it to do in the future.
We want to implement our best trading concepts on our own terms, not wait
years until some company like Omega Research decides to give us more power
so we can trade on our best ideas. I believe Omega had the right idea way
back when (and their success to date indicates so), but as a TradeStation
user I know they aren't doing it for many of us anymore.

While TradeStation has been the market leader, it has also been a
frustrating tool to use. EasyLanguage really pisses me off because I would
rather work in a real language without so many limitations. TradeStation has
also been frustrating because we've been stuck with a 16 bit Win 3.1 program
for way too many years -- and now we've been waiting and waiting... and
waiting for TS 5.0 (which has been delayed again, with the release moved
into next year). Obviously, we'll get only what Omega says we can have and
when they say they are ready to give it to us -- as long as we only have
TradeStation to turn to.

As traders, we want to control our destiny. That's why we feel so much
frustration when Omega Research leaves us in the lurch. We trade for profit,
but it is more than profit. We like the freedom, the lifestyle, and the
independence. We really do thrive on controlling our own destiny. That is
why we, the community of computer literate technical traders, deserve to
have the source code to our own open trading platform that lets us implement
our most creative trading strategies without limitation!

This will really be a unique project among open software development
projects. The Linux community is admirably altruistic. But here's what's so
great about our project to build an open trading platform. We can do it
altruistically and put the source out there for the benefit of all. We can
do something great and feel good about what we've done for the welfare of
our entire community. I have no doubt that the world will be a better place
when we complete this project. However, we can take what we've created and
use it to maximize our own trading profits. Talk about having your cake and
eating it too! We could even create an elected board to trade various
systems on our platform with donated money and use the proceeds to support
causes we vote to support. Try doing that with an operating system!



Here's a preliminary statement of purpose:

We want to build a trading platform with capabilities similar to and
eventually extending beyond TradeStation where the source is open and freely
available (but organized and maintained). The project will be in the spirit
of Linux, but instead of building an operating system, we will be building a
trading platform.



Some more rantings:

Given Omega Research's arrogance in the arena of trading platforms it is
easy to draw the comparison to MS in operating systems. But while Linux has
a huge mountain to climb, and who knows how successful it will ultimately
be, we can taste victory one by one as we successfully trade on the open
platform we've created that lets us build, test, and implement trading
systems without limitation. We don't necessarily need the world to write
applications that run on our platform, although that can happen. As traders,
the best "applications" are the trading systems we build around our own
style and personality. A trading platform can become a standard much more
easily than an OS. And, of course, a trading platform can let us make our
maximum profits even without any widespread adoption. What it all adds up to
is this: we have a project that can't fail.

I know that this community can build a trading platform that will remove the
boundaries from our system building and technical trading. If you are like
me, you know that the right tools will increase the profitability of your
own trading and you are incredibly excited about getting those tools.
TradeStation has certainly made a lot of people a lot of money from trading.
But many of us have wanted more, and for years and years Omega has not
responded. No other vendor has given us the power we want either. We'll take
our destiny into our own hands with the power of a community building an
open system. The Linux model will serve us well. Open source benefits
everyone.

Furthermore, we want to avoid the frustration felt when quirks (or bugs) in
TS interfere with the trading strategy we want to pursue. I know many day
traders are irritated by the unknown time stamp accuracy of the real time
quote server. I know many technical traders who are frustrated by the
arbitrary 13000 bar limit per chart. I know other traders who can't
implement their best day trading systems because EL doesn't have the
mathematical accuracy to support their algorithms. Personally, I hate the
idea of building 16 bit DLLs to extend EasyLanguage when I can't even buy a
good 16 bit compiler anymore. As good as it is, there are thousands of
issues (problems) with TradeStation. Sure, there will be problems with the
trading platform we build, but at least the community of users will have the
power to make improvements. With TradeStation the community of users is
locked out -- only Omega Research controls the evolution of the product. I
can't even get Developer Support at Omega to answer my email reliably, they
don't take phone calls, and they don't call back in response to issues.
EasyLanguage support for users is just as bad. If I had the source, I could
support myself better. I'm sure many others feel the same way.

Here's my own example of want I cannot do with TS. I trade options. I like
mechanical trading systems. I would like to backtest option spread
strategies just like I can backtest equity or futures trading systems.
TradeStation won't do it. OptionStation is a joke. (I hear the new
OptionStation 5.0 will feature EasyLanguage programming and system testing,
which is exactly what I need... exactly, that is, if we could throw out the
EasyLanguage and replace it with a real language, and if we had access to
the source so I could overcome Omega's limitations and expand the features
over time.) Well, you get the idea. Even the new version will be more of the
same old stuff. I want to start making money from my new approach to
systematically tested option spread strategies as quickly as I can. In other
words, I want to control my own destiny, not be hamstrung by Omega.

There's no reason to limit our conceptualization of this trading platform to
the TradeStation mold either. I reference TS merely because it is something
familiar to most of us.


Here are some preliminary thoughts on approaching the project:

Let's organize the project like a business. It should probably be a
non-profit organization. I like the idea of an organization because we want
coordination and cooperation. We want efficiency of effort. We certainly
don't want bureaucracy, and we don't want to waste our time in meetings. But
we want efficient coordination of our efforts. We want a virtual
organization existing across the 'Net (like Linux).

Once we get a firm commitment from, say, 20 qualified people we will form a
legal entity that will control the source code and put it into the public
domain (copyleft, or whatever appropriate method).

We will elect a board to guide the project and help organize the efforts of
all participants so that time and energy isn't wasted. The board will make
policy decisions. However, since all contributors are independent, no
decisions exert creative control over the participants and contributors.
(There will be the usual restrictions against an individual taking the
source and making a commercial application.) The result of this structure is
that any single contributor can have a significant impact on the project by
contributing an insanely great idea. The main function of the board and the
organization will be to hold the project together over time. The main
function of the legal stuff will be to keep the source truly in the public
domain and keep new contributions open for all to use.

I volunteer to do whatever is required to get the project started.