I went through the same thought process just a little while ago with my new startup. I'm proficient in .Net but I kept an open mind and explored other web technologies.
I thought I was somehow missing the boat by not using RoR or something open. But I had discovered what others have and that is that .Net scales just fine. I have government agency as a customer and they use .net and they have some very heavily utilized sites and the hardware spec to run them is quite manageable. Also there are legions of talented .net developers the world over.
In terms of the development cost, the express versions of the tools are free and they are not missing features that you would need unless you have a big team.
The server costs can be mitigated by joining the bizspark program.
The other reason I chose to use the .Net platform is that there are 3rd party components that I require. I would rather use these natively rather than wrapping them in a web service or something like that.
Personally I think the price between “free” and using .Net is in the noise floor.
I use the Jabra C820s. They basically copied the Bose NC circuitry. I tested then side by side at a Bose Keosk at the airport and they sound the same and are just as comfortable.
I don’t like the way Hollywood portrays people in this thing of ours. The typical computer guy in the movies is introverted, socially inept, unpleasant and weak.
I would like to see a cooler, happier portrayal of us. I realize that our reality is actually countless hours of toil. But I wish movies would err on the side of cool rather than "nerdy"
Step 1 assess the level of fraud for the given type of product your selling. I ran and ecommerce site that sold over 100 Million dollars with of product in its lifetime and had only 1 case of credit card fraud.
If you still think its a problem check out verified by visa.
I know your using Paypal but you might get some pointers there.
Here is an example:
User uploads and image to be OCR'ed and the resulting text is to be displayed (in process). Say I have the OCR algorithim (the business) in a c library. In the .Net world I would create a dll from the library, reference it and call the ocr function and display the results. Does RoR have the ability to include external late binding libraries where the "business" can be implemented?
I've try a bunch of little micro business experiments like making iphone apps etc. I wouldn't bother with any legal framework until the $$ starts rolling in. Keep it real.
There are other people telling you what a unit test is so I won't bother. What I will tell you is that you will have a eureka moment when your unit test unexpectly fails. Then you will say "Boy I'm sure glad I have unit tests."
App store is the only game in town. The first bad reviews will make you want to jump off a building. Best thing to do is to realize that your in this for the long term. So suck it up and keep improving.
IMHO a PhD like any other education is just potential energy. It's meaningless unless you can apply it, or someone else can. I would save some money and do a startup however small. This way you know if your making something of true value rather than something you find just interesting. One of my cofounders has a PhD but he worked on various startups during his thesis.
1. Figure out the kind of stuff your interested in.
2 Determine the kind of tools others use to build said stuff.
3. Learn those tools.
4. Build something cool.
Because when your dead your dead.
I will know that I created something greater than myself or died trying.
Some might retort "what about happiness?" Driving my Porsche makes me pretty freaking happy!