MIT License – Does MIT License Require Derivatives to Be Public?

This Ticket Regarding the Boost Open Source License v1.0 lists 4 different things that make this license different than the MIT license.

Two of them have to do with warranties. If you use a BSL licensed library, you'll need to write your own disclaimer stating that you aren't responsible if the software turns into a three-headed monster and does some bad stuff to your customers. The warranty disclaimer isn't extended to licensees. The advantage is that you can also apply your own warranty for a fee if you'd like, should you want to provide one.

One of them has to do with the copyright notice. You are right that it doesn't need to be included in binaries.

The last one makes it clear that licenses can be issued to either individuals or organizations.

If I were to release a project under such a license, would it be automatically shunned by every developer on the face of the earth, or, assuming it is actually a useful project, does it have a fair chance at getting contributions from Joe Programmer?

Your project won't necessarily be shunned by everyone. However:

• A lot of people will shun it on principle.
• The chances of unattached developers contributing for free are low. There are more "worthy" projects with better open source credentials. (This might change once your project / software has proven itself.)
• A lot of potential paying customers will think twice about whether your project has long-term viability.

(Actually, the last point applies to all customers, paying or not. The point is that if a customer application depends on your code, and your funding model fails, they can be stuck depending on a codebase that is effectively dead. Since the dual license approach tends to discourage community contributions, the chances of a self-sustaining community appearing after your company goes belly up are significantly reduced.)

The second part of this question can easily become subjective, but any well argued point of view will be highly appreciated. For example, do dual licensed projects made by commercial entities have success with the open source communities?

This is highly subjective. Success is relative. Success is subjective / debatable.

I think that the best that can be said is that dual-licensing does tend to put off contributions of effort, especially from unattached developers. Developers working for other companies may still contribute, if they / their company sees an economic benefit to doing so.

But these things are almost impossible to measure, for any number of reasons. And I'm not aware of anyone having attempted to do the measurements.