Originally Posted by
furballi
That's irrelevant as long as there are at least 3-4 good turns for engagement. Feel free to experiment using a tensile testing machine.
I know Furballi is an an engineer, so this isn't for him, but for others who might read this.
I don't want to belabor the point, but standard practice is for the spoke to engage the head by 2-3mm. This is not arrived at by accident but is based on a century of experience.
Yes, a brass nipple has a tensile strength of 80% of the spoke in the shank area (Furbali's assertion, accepted at face value), but that only tells a part of the story. Since the brass is notched in a number of places, especially at the rim line by the threads, this means that there'e be notches at the fulcrum of any bending moments. These will lead to premature fatigue failures. (the type of brass typically used for nipples isn't especially suited to repeated bending stresses)
Nipple breakage at the rim line is a well documented phenomenon in both brass and aluminum nipples, though far less so in steel nipples. We don't have to reinvent the wheel here, the CPSC conducted extensive research when they formulated the first bicycle standards. They focused on nipple failure because of a number of cascade failures of wheels leading to injury, and mandated a standard that short (not into nipple head) spokes with brass nipples generally failed, though those with steel nipples passed. Fully threaded spokes and nipples (2-3mm into the head) pass easily regardless of nipple material.
It's always tempting to shortcut standard practice or rules of thumb, but before ignoring them you need to consider why folks have done things a certain way for eons.
Now if the Op wants to ride this front wheel based on Furballi's advice vs. following well established practice he's free to do so.