planetscape has compiled a nice list of nodes to answer this kind of question here - hope that is of some use.

Also, feel more than free to post some of your solutions (to any problems you might find) here at the monastery where you're quite likely to receive advice on the three or four best ways to accomplish something, with the intent of learning how some of the great minds here might approach the problems.

s**lil*;  $*=join'',sort split q**;  s;.*;grr; &&s+(.(.)).+$2$1+; $; =
qq-$_-;s,.*,ahc,;$,.=chop for split q,,,reverse;print for($,,$;,$*,$/)
[download]

In my mind, the job of a manager or leader is not to work, but to motivate and coordinate the work of others. The main point being, if you supervise a team of five coders, your team should be able to create output exceeding the output of a group of six coders each coding independently (if not a little more, since I would imagine your salary is probably higher than a coder's).

I don't know how much work you do managing and how much extra time you have, but if I were a manager, I would be reading management books and websites. How do you attract talent? How do you help the employees which are slipping? How do you attain and then mantain a high level of work? How does your team fit into the grand scheme of things at your company? etc. I'd be way more worried about those questions than, say, what's the -T switch for?

The main point being, if you supervise a team of five coders, your team should be able to create output exceeding the output of a group of six coders each coding independently...

Good theory, but I think the numbers are a bit unrealistic. I would hope that a manager with a team of five coders would be able to create the output of about three coders working independently. Okay, maybe four, but that'll be tough.

The trick is that if you actually have three or four coders working within the same group without a manager or other organizational scheme, it'll be a complete mess. They will be doing redundant work, then spending time integrating their work and eliminating redundancy, then deciding how to interface their pieces together, and so forth. The point is that unless you have truly independent projects, there's no way for your coders to work completely independently. (And if you do have several truly independent projects, your company probably ain't gonna make it. It's spread too thin. Focus on what you do differently, and beg/buy/borrow/steal the rest.)

Ok, so maybe the numbers aren't quite right, but you get the idea. Think of it as parallel processing, which it is. The ideal would be linear scaling, but dependencies between processors (coders) always drop the speedup below that. One job of a manager is to eliminate as many dependencies as possible, and minimize the disruption of the rest. That may mean scheduling first the pieces that need to be resolved to avoid blocking others, or it may mean setting some things in stone to make sure the parts work together while leaving other things loose and flexible so they can adapt to their circumstances.

I like your analogy, and it's true to a large extent. There's more to it than that. A manager is more than a scheduler for processors. CPUs don't break up with their mates. CPUs don't request (?:p|m)aternity leave, don't have 'bad hair days,' don't overclock themselves when you pay them a sincere compliment about their work. A good manager not only schedules, coordinates, and plans, but motivates, congratulates, and inspires. I think I can state categorically that there are not enough good managers in the world.

---

I like computer programming because it's like Legos for the mind.

I agree with the other poster about needing to learn more about management. Doing it right is as hard as coding, might be harder. And if you want to keep your skill up-to-date, you can do a little side project or something. Contributing to open source or CPAN would be very rewarding experience as well.

If you believe that the market for labor is efficient, then managing is harder than coding. Otherwise there would be more managers than coders and/or the coders would be paid more than their managers. (Neither of which is generally true, though there are exceptions to the latter).

I love coding. If I had to give up coding in order to give managing a try, I'd take it.

---

I like computer programming because it's like Legos for the mind.

I'm not a manager myself, but I find the insights very useful.
(updated: minor correction to the URL - thanks kyle!)

newsgroup comp.lang.perl.misc
maillist        perl.beginners
web            perlmonks question nodes
[download]

It is sort of the same idea as the old teaching technique called "story problems" (IIRC), where instead of asking you to prove the Pythagorean Theorem, you are asked something like " find the height of a mountain, if you are 5 miles away and the angle to the top is 15 degrees". They are both essentially the same problem, but one is "more realistic" than the other.

You could, as long as the OP wasn't really serious when they said "free".

Because BrainBench isn't.



--chargrill

---

s**lil*;  $*=join'',sort split q**;  s;.*;grr; &&s+(.(.)).+$2$1+; $; =
qq-$_-;s,.*,ahc,;$,.=chop for split q,,,reverse;print for($,,$;,$*,$/)
[download]