I have the same problem. I'm slowly re-implementing most of my stuff to use SOAP web-services. Eventually my plan is to just host the data on my own server at home using my cable-modem connection, and then have my software access the data via the web services. That way whenever I travel my financial database will be accessible as long as I have internet access. Since, I'm the only one who'll be using the SOAP services, I don't think I'll need any more bandwith than my outgoing cable connection.
After 25 years software development and working with the big 3 software development firms I guess I have to say your experiences and opinions are not in line with the facts. The indian software development centers of these firms, properly staffed and managed, produce the same quality of code as their US based counterparts, for about 1/4 - 1/5 the cost. Of course there are just as many poorly run software firms in india or with overseas operations as there are US based. As for independent software people ... I dont know anyone that would work for 25 dollars an hour: you can make the same amount of money driving a truck with no college education. However, your assessment of rates is way off the mark. In real terms, rates for even esoteric skill sets and projects have fallen below their 1995 normalized and adjusted for inflation rates. This includes rates for the big 4 consulting firms. So, we have an industry where pricing pressures will continue, in real terms, downward. This is largly due to the ability to offshore work which, when done correctly, yields huge cost savings that more than makes up for the downward pressure on pricing. The other factor is that most software specializations are no longer secret: most of the techniques are known - even for quant types of applications. Anyone that claims otherwise is just blowing smoke. If I were you, I would get out a bit more: Unless you have very specialized skills - beyond the typical quant - your future is one of downward pressure on compensation and a shrinking US based labor pool.
Software development is quickly becoming blue collar work. I wouldn't want to be in this industry 20 years from now. Stuck in a cubicle competing with 1 billion Indians & 1 billion Chinese for a crappy, boring, average pay job. If I had to choose, I would be a plumber with my own company or some other skill that cannot be outsourced.
Well, in another 5 or 7 years the industry will look quite different. There will still be a demand for engineering but the work will most likely involve very little programming and much more testing and integration of components and systems: this has been the trend over the last 15 years or so and it has dramatically accelerated over the the last 7 years. Actually I see the title of computer scientist or software engineer going away or being replaced with a different title reflecting the changed environment. Computer science at the end of the day really was largly just mathematics anyway ....
You're quite correct. I would add that you do not need 'computer science' skills (not as many anyways) for testing & integration. In fact, you need functional domain skills in order to fully understand the problem that has to be solved by the software. For a job like that, you are better off hiring someone with field experience rather than a 'software engineer'
why not started with a job/intern posting at the local state university, describing briefly that you need a senior student with (insert your language), whereby you will pay $15-$20/ hour to design and integrate 1 simple stategy that you have and need to automate. And, post that you can provide sample code for your platform (which many provide). Post that the project will likely be 60-80 hrs of work, and will require outside meetings. In the process of getting intern credits, the student is paid and you will teach them some of the basics to the equities/futures market. And, if you have some brains, you might keep the entire package of the strategy from them, and put the final touches on it yourself. why pay $150-$250 an hour, to someone in the industry who will likely steal your automation especially if you know that you are profitable? yes, working with the Intern will take longer. and, you will have to put in time learning as well. obviously the programmers that are already established will not agree, and that is exactly why you should error on the side of caution with them. (but this is IF you have a proven profitable strategy that you simply want to automate.) just my 2 cents, patrick