Hey
I posted a question on a usenet forum the other day about the availability of a unix c++ api and somebody emailed me this link https://sourceforge.net/projects/tws-c-api and suggested I ask on elitetrader.
So far so good, better than nothing. On the other hand this code is not supported by IB and seems to be maintained by some russian guy in his spare time. So I wonder if I can trust my ATS to an api like that and whether anyone has used it.
I'm not sure I can trust it, so what other recommendations can you make?
Please no java and no windows!!!
Why is it that some people can't get it that some of us programmers demand c++, unix and matlab?!
The other thing I don't like about the IB api is that they require a tws application to sit in the middle of all communications which means you can't connect to their servers directly. Hopefully it doesn't add too much to the latency. Anyone know what the latency is? I really need a fast link to the broker's servers.
I posted a question on a usenet forum the other day about the availability of a unix c++ api and somebody emailed me this link https://sourceforge.net/projects/tws-c-api and suggested I ask on elitetrader.
So far so good, better than nothing. On the other hand this code is not supported by IB and seems to be maintained by some russian guy in his spare time. So I wonder if I can trust my ATS to an api like that and whether anyone has used it.
I'm not sure I can trust it, so what other recommendations can you make?
Please no java and no windows!!!
Why is it that some people can't get it that some of us programmers demand c++, unix and matlab?!
The other thing I don't like about the IB api is that they require a tws application to sit in the middle of all communications which means you can't connect to their servers directly. Hopefully it doesn't add too much to the latency. Anyone know what the latency is? I really need a fast link to the broker's servers.
This is most basic in any minimum programming task - in Matlab it's always a kludge. Simply try to rewrite a piece of MatLab serious matrix code into Python - you'll quickly find out and will never want to touch ML again. Further take a look at things like SciPy, Numeric, Rpy, Gnuplot, matplotlab/pylab for starters. You'll be equipped like the smartest research facilities in the world (all for free).