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

Re: To MarkBrown: Real-time datafeeds in Python



PureBytes Links

Trading Reference Links

At 11:59 AM 5/5/2004, MichaelSuesserott wrote:

>Also, IMHO it is not quite fair to insinuate that Joel might be afraid of
>making actual trades, and that that was his reason for wanting to develop
>trading software. 

I don't think I said that. Sorry if it sounded that way. So much for trying to be amusing...

I thought that, like everybody else at some time or the other, he wanted to develop the new software for his own use to get features he couldn't get in presently available software.

My point was that he would be far better off using any existing platform than trying to develop one himself for his own use. It always seems easier to "roll-your-own" but from hard experience, it never is. 

I didn't understand that he wanted to sell it as a part of a new business effort. I must have missed that point somewhere along the way so I decided to check his past messages. I see that he began selling QuantStudio last summer:

>>From: "Joel Reymont" <joelr@xxxxxxxx> 
>>To: <omega-list@xxxxxxxxxx> 
>>Subject: Quantitative trading / TS alternative 
>>Date: Sat, 26 Jul 2003 15:49:40 +0400 
>>
>>Folks,
>>
>>Those of you who are looking for a TS alternative or something that gives
>>you a lot of quantitative trading possibilities, please take a look at
>>QuantStudio. It enables you to prototype, test and implement real-time
>>quantitative strategies across the currency, stock, commodity and bond
>>markets. It comes with MyTrack and IB interfaces for real-time data and
>>execution. You can program it in .NET languages such as C#, JavaScript.NET
>>and VB.NET.
>>
>>Please see http://www.smartquant.com/forums/viewforum.php?f=4 for more
>>details.
>>
>>Disclaimer: Yes, I do sell QuantStudio. I did play with TS in the past and
>>have a copy. I also own MetaStock, NeuroShell Trader and Trading Recipies. I
>>know that people are always looking for alternatives, specially now that
>>Omega is looking to discontinue support for TS 2k.


Then as late as April 15 he was still promoting QuantStudio:


>>To: omega-list@xxxxxxxxxx 
>>From: Joel Reymont <joelr@xxxxxxxx> 
>>Subject: QuantStudio testimonial 
>>Date: Thu, 15 Apr 2004 23:44:32 +0400 
>>
>>Folks,
>>
>>Those of you who looked at QuantStudio a while ago might want to take a look at this testimonial:
>>
>>http://www.smartquant.com/forums/viewtopic.php?t=670
>>
>>This is from someone who has been very critical of us before.
>>
>>        Thanks, Joel


Then this new project started a couple of weeks later:


>>From: "Joel Reymont" <joelr@xxxxxxxx> 
>>To: <omega-list@xxxxxxxxxx> 
>>Subject: Uptick: Tick-based futures trading engine .NET or Java 
>>Date: Wed, 28 Apr 2004 18:36:36 +0400 
>>
>>Folks,
>>
>>I want to build from scratch a lean and mean tick-based futures trading
>>engine. The engine will be open-source and include the basics such as
>>data storage, backtesting, optimization. This is my personal project
>>outside of the SmartQuant umbrella.

I guess that is where I got the idea that this was for his own use.


>We have had quite a few
>developers on this list, of systems as well as of trading tools. Some trade,
>some don't, and they were respected anyway.

Well, it is the Omega List so I sort of though most people were still interested in using TradeStation...     :-)



>You are a developer, too, Bob, aren't you? (And I bet you would love Python
>:-))

I used to be in a past life. I even considered developing a trading platform back when Omega couldn't seem to get their act together. But after a brief analysis I decided against it. Below is a summary of what I thought it would take at the time. This was years ago and the barriers-to-entry have certainly gotten much higher since then.

But if he wants to "get rich by selling software" as he says, far be it for me to discourage him. Maybe he can do it and it would be nice if we had more viable choices of suppliers...

Bob Fulks


===================================

Part of an old private message:
------------------------------
...

My last "real job" was managing the engineering department of a major 
software company so I have some experience in this area.

Writing the code is probably 10% of the effort of developing 
software. There is also:

Documentation, 
Instruction books, 
Training courses, 
Promotional materials, 
A complete library of the usual functions, indicators, and systems
Interfaces to data vendors
Interfaces to brokers systems
Testing on:
   all common versions of the operating systems in use, 
   all common graphics cards, and 
   other hardware configurations.
Customer support:
   Helping customers become effective
   Answering questions
   Handling enhancement requests
   Fixing bugs
Modifications to respond to:
   changes in the data sources
   changes in broker interfaces
New versions from time to time to keep the revenue coming in.
Business aspects
   Negotiating contracts with vendors
   Legal paperwork
   Tax paperwork
   Keeping employees happy
   Employee benefits
   Marketing the product
   Technical presentations
   Getting endorsements
   Raising money
   Keeping investors happy
   Keeping banks happy
   Meeting the payroll
etc., etc., etc.
 
I estimate that it would require a full time staff of about 20 people to 
build a new commercial trading application competitive with TradeStation. 

And the market is very small by most software-business standards so the 
potential return-on-investment for any new entry is very low (or most likely 
negative)...