On 5/1/07, Dave Smith <[EMAIL PROTECTED]> wrote:
[. . .] Can you give me any suggestions on how I can convince him to
join the rest of the civilized world and use revision control daily,
without being a jerk and going over his head?

All I can offer is my own story: I was doing an assembly code
optimization project for that 300-level computer architecture class at
BYU.  Being somewhat of a loner I chose to do the project alone and
was actually doing quite well.  Unfortunately I hadn't quite met the
target yet and the only thing left was loop-unrolling, which I
honestly didn't understand at the time.  I asked the professor if he
thought I should turn it in for partial credit or dive in and try
loop-unrolling.  He was very confident that I'd do just fine so I
plunged forward and gave it a try.  The whole thing blew up in my face
completely to the point that I actually started to try to rewrite the
dang thing from scratch.  I actually managed to do that and it was
ridiculously fast, but it lacked the errors the original algorithm had
so the output wasn't the same and I failed the project completely.  It
was such a blow to my ego at the time that I gave up and flunked out
of that class.  I also had really bad headaches for a while.  I was a
defining event in my life.  I have used revision control faithfully
ever since.  It would have saved me in so many ways.

Actually I can offer one other bit of advice:  separate production
builds / deployment from the engineer such that nothing actually gets
built / deployed without being checked in.  Many companies see this as
a requirement anyway and you can sometimes use Sarbanes Oxley (sp?) to
scare them into it.  Anyway, making it a procedural change should help
make it less personal.

/*
PLUG: http://plug.org, #utah on irc.freenode.net
Unsubscribe: http://plug.org/mailman/options/plug
Don't fear the penguin.
*/

Reply via email to