An entertaining quote from the book I'm reading, "It is better for developers to be surfing than writing code that won't be needed."

They aren't referring to technical spikes or learning opportunities, they are talking about features to a real product.  In other words do option 1*, its cheaper.  Adding unneeded features to your code base will cost you more money over the lifetime of the product than paying the developer to do nothing at all.  Each feature you add complicates the code base making each additional change more costly.

* There are always 3 options to any decision, and option 1 is always "do nothing."