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

We need a "Universal" Market Data Server!



PureBytes Links

Trading Reference Links

Re: Omega or TraderWare??? thread

In all this discussion about trading software platforms, one thing needs to
be made very clear.  We, as a trading community, need a product like
MarketStreams universal Market Data Server(M3) in order to foster
competition between the data providers and charting/back-testing software
developers. Take the data providers for example.  Today eSignal is getting
praise and Quote.com is getting trashed on this list.  Tomorrow, after a
buyout or management change it may be the reverse.  In order to foster
competition between the data providers and keep them on their toes, we need
the ability to quickly switch between them with a minimum of bother and
without having to look for historical data in another proprietary format or
to convert our existing data.  M3 maps all the symbols of all the different
feeds to it's own common symbol library.  Changing data providers is as
simple as loading a new driver. It's the ultimate guarantee that a data
provider using proprietary technology won't abuse or milk us while it raises
it's prices or degrades service by under investing in its infrastructure.
The same principal applies to charting/back testing software developers.  If
one charting software package no longer measures up to the competition,
delete it, install another one that interfaces with the M3 and your off.

Needless to say, we also need a data server that's reliable, efficient and
stable.  In order to achieve this, it needs to be developed and maintained
by a company whose sole focus and expertise is in this one product niche.
Firstly because it's so technically complex that we would be better served
by a company dedicated solely to it's development and improvement.
Secondly, so there won't be any conflict of interest between itself and
other data suppliers and/or charting software developers that typically
leads to the above-mentioned abuses.  Currently the only contender for the
job appears to be MarketStream.

OR is none of this -it's going in entirely the opposite direction- and now
it's even branching off into the brokerage business.  It wants to become
this vertically-integrated 300 lb. gorilla in trading services.  Problem is,
it can't even get it's own core product/services right.

In the future, it will be in all of our long-term interests to use only data
feeds supported by a common universal data server and to use only
charting/back testing software that interfaces with it.  This will give us
clout we need with data providers and software developers.

Personally, I very much like the TradeLab model for these reasons (with all
due respect to Lawrence Chan whose NeoTicker uses the same model, I
believe). It will use M3 thereby offloading data server development with all
it's incumbent difficulties and costs -both present and future- onto
MarketStream.  It will use VB thereby offloading the programming IDE
development onto Microsoft.  It's developer, Bob Brickey, only has to focus
on the tick database and charting components and interfacing them with M3
and VB using the latter's respective SDK's.  This is a viable model from
both the software engineering and business point of view; now if he could
only get it out the door sometime soon!