Would you collaborate with a programmer if it meant sharing your system/research?

[ondafringe]

A workaround for this could be that I first hire a programmer to suggest a general framework/architecture without revealing details of the calculations.

Finally, I give the job to someone else according to the specifications of the prior programmer. But, since I don't need to explain to him how it works, I can possibly still keep things secret. For example, I can say that this is just a collection of statistics. If he doesn't understand the underlying framework/potential, I don't think he'll bother digging deeper.

You could break your "secret sauce" down into one or more discrete parts. Then hire one or more programmers to code the logic and build a DLL for each of those parts, and then deliver the source code and DLL for each of those parts to you. Then hire another programmer to code the main application, add the DLL(s) to that application, pass relevant data to each DLL, and process what the DLL(s) return.

Since a DLL is compiled code, the main app programmer can't access the logic contained within the DLL unless they reverse engineer the DLL, which is unlikely, especially if you go through Upwork and hire a programmer in India.

For an extremely simplified example:

You want to bring in one-minute price data on a particular symbol. You also want to create a 30-Bar SMA on that data. At the end of each minute, if the price is above the SMA, you want to be Long. If the price is below the SMA, you want to be Short.

Programmer B codes the logic for the DLL. All this programmer knows is she/he will receive two numbers and check to see if Number1 is greater than Number2. If it is, she/he codes the DLL to return a value of "True," otherwise, return a value of "False."

Programmer A codes the logic for the main app and adds a reference to the DLL. This programmer brings in the price data from the API and calculates the SMA. All this programmer knows is that at the end of each minute, she/he needs to pass the current price of the symbol and the current price of the SMA to the DLL, and if the DLL returns a value of "True," the code should put you Long, otherwise, the code should put you Short.

Third-party DLLs are used quite often to calculate and handle all kind of things. Maybe that approach might work for you. If so, it would help keep your "secret sauce" secret.

 

[Laissez Faire]

You could break your "secret sauce" down into one or more discrete parts. Then hire one or more programmers to code the logic and build a DLL for each of those parts, and then deliver the source code and DLL for each of those parts to you. Then hire another programmer to code the main application, add the DLL(s) to that application, pass relevant data to each DLL, and process what the DLL(s) return.

Since a DLL is compiled code, the main app programmer can't access the logic contained within the DLL unless they reverse engineer the DLL, which is unlikely, especially if you go through Upwork and hire a programmer in India.

Third-party DLLs are used quite often to calculate and handle all kind of things. Maybe that approach might work for you. If so, it would help keep your "secret sauce" secret.

That's a great suggestion, @ondafringe. I can see that as a way to go if I wanted to keep things as secret as possible.

However, due to the nature of this project it would definitely make things harder for me to put together. I'd definitely prefer if one programmer took on the whole job.

Due to how it's all organized, I think it's possible I could have a programmer do this and don't really see the value in it and just move on without really understanding it. They could simply discard it as a bunch of data and statistics without any predictive potential.

I think I will write up a project outline on UpWork in as much detail as possible without revealing any secrets and see if I can get some good ideas/offers.

I'm sure I spent > $25K in total on the current application, but a lot of that was trial and error, subsequent developments, etc. Since the general framework and calculations now are mostly complete it should only be a matter of re-programming it from scratch within a more efficient architecture/framework.

If someone thinks they can do it for < $10K I would definitely have to consider it.

A collaboration deal might ensure a better finalized solution as we can bounce ideas of each other and the programmer fully understands what I want to accomplish. But the drawback is that he'll get my work for "free" and share equal rights. How big of an issue that would be I need to ponder a bit more, but I'd certainly prefer to keep to myself...

 

[Laissez Faire]

So, I put up a basic job description of what's needed online and the bids I'm receiving range from $10-20K and I imagine it would require/cost even more once all details are revealed as I had to keep it general. I don't think $25K would be an exaggeration.

So, that's the monetary value of my end of the deal.

Main concern with a collaboration on my end is simply that I will now share the rights of my IP and final product with someone else, but it's possible that's a price that's worth paying.

I'm not sure if it's possible that I could retain the rights to my IP while sharing equal rights to the finalized solution.

The one thing we're certainly in agreement on is that the finalized solution should remain private between the two of us.

 

[Nine_Ender]

So, I put up a basic job description of what's needed online and the bids I'm receiving range from $10-20K and I imagine it would require/cost even more once all details are revealed as I had to keep it general. I don't think $25K would be an exaggeration.

So, that's the monetary value of my end of the deal.

Main concern with a collaboration on my end is simply that I will now share the rights of my IP and final product with someone else, but it's possible that's a price that's worth paying.

I'm not sure if it's possible that I could retain the rights to my IP while sharing equal rights to the finalized solution.

The one thing we're certainly in agreement on is that the finalized solution should remain private between the two of us.

In this market there is a shortage of talent so you have to decided how badly you want it done now or not. The IP may be a bargaining chip because these kind of projects can be a potential headache for IT consultants unless you are willing to pay open ended on an hourly rate. I don't know anything about your project but have done a lot of high end development work in the past. The programmer won't value the IP like you do but in this line of work you often get what you pay for; mediocre programmers often provide mediocre results and a strong developer/analyst will do great work if you do your part well. That means no huge scope creep, nickel and diming.

 

[Laissez Faire]

Well, this collaboration looked promising for a while, but ended up being a failure. Long story short, the programmer vastly underestimated the scope of the project not taking the sufficient time to set up and learn my existing system properly before he started the project of re-writing what I already have. Further, it's quite possible he overestimated his own skills and time available to work on the project in addition to having a full time job and family obligations. I expressed concerns about this early on, but he assured me it would not be a problem. Turns out I was right.

At 9 months past the agreed upon deadline I suggested that we ended the collaboration and he agreed, although he did say he would honor his obligation and complete it if I wanted it. I'd estimate we were at around 70 % of completion when ending it, but seeing how slow progress was and how unmotivated he seemed at this point, I just didn't have any interest in extending it. I don't want a collaboration with someone that's not willing to put in the same type of effort and sacrifices I'm willing to do.

I thought it would be the perfect match as he seemed skilled enough with a strong interest in finance and learning to trade. And I do think he was very motivated and inspired early on, but seemed to lose that along the way.

I think this is a common pattern for most people in life regardless of their pursuit. It's easy to be passionate and inspired early on dreaming about success, but most people seem to be idle dreamers who give up along the way. Some sooner than others.

If I ever decide to move forward with improving this application/system, I think I will have to pay it out of my own pocket just like I did with the original application that I already have in my hands.

 

[spy]

A little more creativity and you've just re-invented the Component Object Model. Congrats, it's 1993 again!

You could break your "secret sauce" down into one or more discrete parts. Then hire one or more programmers to code the logic and build a DLL for each of those parts, and then deliver the source code and DLL for each of those parts to you. Then hire another programmer to code the main application, add the DLL(s) to that application, pass relevant data to each DLL, and process what the DLL(s) return.

Since a DLL is compiled code, the main app programmer can't access the logic contained within the DLL unless they reverse engineer the DLL, which is unlikely, especially if you go through Upwork and hire a programmer in India.

For an extremely simplified example:

You want to bring in one-minute price data on a particular symbol. You also want to create a 30-Bar SMA on that data. At the end of each minute, if the price is above the SMA, you want to be Long. If the price is below the SMA, you want to be Short.

Programmer B codes the logic for the DLL. All this programmer knows is she/he will receive two numbers and check to see if Number1 is greater than Number2. If it is, she/he codes the DLL to return a value of "True," otherwise, return a value of "False."

Programmer A codes the logic for the main app and adds a reference to the DLL. This programmer brings in the price data from the API and calculates the SMA. All this programmer knows is that at the end of each minute, she/he needs to pass the current price of the symbol and the current price of the SMA to the DLL, and if the DLL returns a value of "True," the code should put you Long, otherwise, the code should put you Short.

Third-party DLLs are used quite often to calculate and handle all kind of things. Maybe that approach might work for you. If so, it would help keep your "secret sauce" secret.

 

[spy]

100% of the logic in the system I use was written by me. I'm just not a coder and able to do that part myself.

NO!
If you wrote the logic, then you wrote the code... they are identical. What you wrote was a bunch of near gibberish that a "professional" re-interpreted for you into some kind of Frankenprogram. Don't lie to yourself.

 

[Laissez Faire]

NO!
If you wrote the logic, then you wrote the code... they are identical. What you wrote was a bunch of near gibberish that a "professional" re-interpreted for you into some kind of Frankenprogram. Don't lie to yourself.

That doesn't make sense.

I can write complex IFS and formulas in Excel, but I'm not able to do the same in C# or Python since I'm not a programmer.

All the calculations/math/formulas/IF statements were supplied by me.

 

[spy]

I wrote the pseudo-code/math for all of these myself.

To me, this seems like a very complex task to handle for someone just learning to code.

I think you're contradicting yourself terribly here and sadly mistaken. If you can't go from the pseudo-code to functioning code with just a little elbow grease... you didn't actually pseudo-code anything to begin with. It sounds to me like your idea is based on a bunch of hunches and whims.

 

[spy]

That doesn't make sense.

I can write complex IFS and formulas in Excel, but I'm not able to do the same in C# or Python since I'm not a programmer.

All the calculations/math/formulas/IF statements were supplied by me.

It doesn't make sense to you because you're kidding yourself. The moment you enter a formula into Excel or place a task in a scheduler or set the time on the damn VCR you become a programmer. You're just not that good yet, give it some time.

You don't have to port your software... keep using Excel/VBA or whatever Turing complete language you want. Of course, if you knew what you were talking about you wouldn't need me to explain this.

You just don't want to do the actual work involved because being that precise is, in fact, very challenging.

I've heard the story over and over and over a gazillion times. Oh, I've got a "great programming" idea for you... then the person tries to describe it and all that comes out is hemming and hawing and vague garbage.

As if you can't tell I'm a bit PTSD'd from it. "Woe is me, if only I could find a good, trustworthy programmer"... look deeply inside yourself first.