That's terrible advice. If you're unsure of what you're supposed to be doing, you should close the computer and start exploring your options with pen and paper before doing more damage.
If you keep going based on assumption, you drive halfway across the country in the wrong direction like in Dumb and Dumber.
Sometimes it's better to do nothing at all than to inflict damage because you didn't know what you were supposed to be doing.
Basically, when I hit a roadblock that I can't get around by working on something else while I wait for clarification, my process is as follows:
1. List my options.
2. Pick the best one.
3. Send an email to my boss outlining the possible options, and note that I've picked what I think is the best possible option.
4. Continue coding and know that I might have to rewrite that feature if my assumption was incorrect.
If you're right, great - it wasn't a roadblock. If you're wrong, you modify or rewrite... which isn't too big of a deal, as you would have been picking your belly button lint while waiting for an answer anyway.