If you would not control the market (i.e. obtain a patent) after inventing a drug, competition would drive down the price to marginal costs. /Marginal/ costs do not reflect development costs, so you would never earn back your development costs.
You're missing the fact that Mailchimp offers an entire email marketing app whereas Amazon offers an email deliverability web service. They each have completely different features and are in a different market.
They could, but I assume SendGrid has more checks in place to prevent their users from sending unwanted emails. That way SendGrid has little risk of getting blacklisted, whereas a presumably uncontrolled service like Amazon's would.
Launchy is a similar app for Windows and Linux. It's not super great for finding and opening files, but it's a great timesaver for opening websites and applications, especially if you use your keyboard more than your muse (as I presume most hackers do).
Build a comparison site or two: look for a market where affiliate programs are ubiquitous, structure the data and write a small application that helps customers decide. Get traffic using some basic SEO. Then pocket the commission from sales.
I did this for the personal savings market and "sim-only" mobile telephony subscriptions market in the Netherlands. It brings in a couple of hundred euros a month with only 1-2 hours of maintenance.
I wrote a script sometime ago, which sells about 4 copies a month for $25 each. The only way I marketed was by adding it to directories such as HotScripts. That will you give you some traffic if you're providing a solution in a field that's not highly competitive (a niche).
You could try that.
I always thought Google had a browser for the masses in mind with Chrome. Something that is easy to use, stable and fast: a good alternative for IE.
User scripts are something for power users, who will always prefer Firefox. Won't this create the risk that average users will accept installing all kinds of scripts without much regard for security implications? User scripts are a powerful yet low-threshold tool which could easily lend itself to purposes such as identity theft.
If this gains a little traction, you might find your inbox full of requests for your services. (Edit: I missed the dark gray bar at the top of the page and now see that you're aiming for that. Good luck!)
Additionally - I am not certain about the mentioned scientists' exact definition of loss - you might argue that since the damage was already done (Joe had already been charged), the 11 dollars was a gain rather than an adverted loss.
While "20% of the activity is responsible for 80% of the problem," I doubt if "dialing the wrong number," "opening the wrong document" and "making spelling mistakes" are the kind of problems you should be spending lots of attention on.
Introspection and self-improvement can both be powerful tools, but it is best used against much larger problems, many of which will not become apparent by keeping track of writing down every little stupid mistake.