I prefer learning all possible ways, then chose according to the situation.
Others might prefer always taking a single known good way.

No programmer worth his salt has only one way; however if it is at all possible to do it "the right way" I can see absolutely no reason to do something else.

As to "learning all possible ways", If you see the need go ahead and "learn" them, I know many ways I have not "learned" and should an absolute need arise, I will then "learn" them. Till then let them lie.

We are still fighting that an engineer once employed here designed something that was "a better choice" than the Power Trends PT series. Yes, he saved appeox $2 on about 10.000 units, but for the last 3 years we have made umpteen service calls to - initially replace with working same - now replace them with a redesign using again a (newer) PT switcher. Last someone dared to figure out the cost of the "savings" we had passed the $50.000 mark. and that does not even include the cost of the units we are throwing out.

I adhere to the old saying "better safe than sorry".

Do I ever go out on a limb? of course, when no existing technology exist there is no other way. I have a very proprietary color sign that is absolutely glorious and can change display faster than you can blink. Did that happen by using "tried and true"? of course not - it did not exist.
Would I redesign a RS232 interface using a MAX232(equivalent) chip to see if there was a better solution NO. Did I do that when there was a better solution? yes, I am using now the "industrial strength" 60V resistant transcievers.

Erik