"2. Use plain linear models for non-linear interaction"
It should be noted that Linear models are only linear in the model parameters, while the features can be transformed using non-linear functions. This trick makes linear models very powerful. Also if you have big data (in millions/billions) then you are better off with linear models, as SVM is very difficult to
scale.
In my experience (all in big data), I rarely seen people use SVM, usual choices are logistic regressions and tree based models. In some finance and insurance industries you are restricted to use only interpretable models, which linear models are.
Salary is just a part of the total cost of an employee. You have to take into account: recruiting, office space, food, medical insurance, payroll tax, stock options, etc. Usually these things will be equivalent to the salary.
Yes, you are right we can modify our JS and get your keys. This vulnerability is in most of the commercial tools out there too. Thats why it is open sourced so you can run it in your own servers. Running it on your own heroku account is close to zero dollars.
Pros:
1. Open source tool, you can run internally in your company. https://github.com/saravanacp/secureshareme
2. Very secure: it encrypts the data in the browser and the key is stored in the URL anchortag which is not sent to server at any point of time. Only the sender and the receiver has access to the keys.
3. You can also opt to send a secondary verification code to receiver's mobile for two layers of security.
4. Option to self distruct message based on time or if an attack is detected.