Try and determine a solution on your own, but do not spend more than a few hours on it. If you get can't make any progress on a problem seek help – there's no shame in doing so. Ask advice on your approach to solving the issue or your ideas on a proposed solution, but have a go at nutting it out yourself. This will earn you more respect than giving up completely.
I gave Jim the task of fixing an XML feed bug we had been seeing. Two days later I went to see him to find out why I had not seen or heard from him.
“Hey Jim, how's that XML feed processor coming along?”
“I'm stuck. I don't know where the feed comes from”.
“Ok. Who do you think might know where it comes from?”
“er...IT support?”
“Yes! Well done Jim!”
“So what do you think the next step for you might be?”
“er...contact IT support?”
“Yes!”
Jim contacted IT support and all was well.
The next day in the team meeting I asked him again.
“Hey Jim, did you get that XML feed processor all finished and checked in?”
“I'm stuck. I get a compiler error.”
“When was that?”
“Just after I spoke to IT support”.
“I can come and help you, but you need to raise these things the moment you get stuck ok?”
The next day in the team meeting I asked him again.
“Hey Jim, did you get that XML feed all finished and checked in?”
“Yes, pretty much done!”
“What does pretty much mean?”
“Well, there's this little bug...”
The next day in the team meeting I asked him again.
“Now it's 100% finished!”
“Great! I'll have a look straight after the meeting at the junit tests and give it a run through. Then I'll check your javadoc and the wiki documentation. I'll also assign you to something new right now”.
“er...that's probably not a good idea. I haven't done any of those final things yet.”
Notice a familiar pattern? Do you think this developer makes management of a project more or less difficult? Try and find solutions to problems for a few hours, have a go at nutting it out yourself, but don't spend five days stuck on something.
