The one true way

You've all seen it: religious wars over technical solutions. emacs versus vi. C++ versus Java. VxWorks versus Embedded Linux. These are decisions you often can't make right because there is no right solution. And whichever you choose, you will alienate a percentage of your technical people. I've tried various ways to solve this dilemma with disappointing results.

What should you do when faced with these?

You will have to do due diligence to make sure that there isn't a clear winner because of some property you haven't yet thought of. Pick a senior person or a small team as technical advisors, and have them research the options, but give them a due date and set the expectation for how much resources they should spend on it. You want to understand whether you can get mindshare or not. If you can't you want to discover that quickly, recognize there probably isn't a "right answer", then simply decide or appoint someone to decide.

Here's the critical part though: make sure the team knows that the decision will be final, who will make the decision, when, and what any known decision criteria is. Let them know that you will support the decision and that you expect the entire team to support it as well. Moreover, that there will be negative consequences for anyone who doesn't go along with the decision. Make sure you are captain of your ship and that the crew isn't going to sabotage the decision. They are either on your ship or not. Their choice. All for one and one for all.

Then you can go on to working on things that add value.