[ANN][IMPORTANT] New leadership formed!

Russell Penney russell.penney at tincanct.com
Fri Feb 18 00:53:14 UTC 2005


Goran said:
<SNIP>
> I liked your idea on mentoring and since you want me to listen I want to
> hear more. Can you write us some concrete proposal on how we all could
> get some mentoring going? How do newbies find mentors? How is the
> mentoring done? How do mentors make themselves available? How do we
> present such an idea to the community? etc.
> 
> I have taught OO etc to programmers and like to teach, so this is an
> aspect I like. And there are lots of people into teaching in this
> community. You have my ear.

Damn, now *I* have to do something rather than just sit and snipe from the
sidelines. You take the fun out of everything! :)

Ok, here goes.

I see a lot of people willing to help and usually the response is "Great,
start doing something". This is the wrong approach IMHO because there is NO
documentation that is up to date on what needs doing, NO help forthcoming
when they ask for more information and NO central body whose job it is to
support these people.
So ...
Have a look at the Debian process (seeing people keep talking about it as a
model). http://qa.debian.org/howto.html is one place to start.

We need a list somewhere of all the tasks that need doing which is kept up
to date by somebody or somebodies. Create statistics that show how many
tasks were completed, "Hey everybody 125 bugs were solved this month!!!!
That's 50 more than last month" or "Another 43 class comments were written"
or ... you get the drift.
If someone wants to help, assign one of the "mentors" to them. The mentor
can guide them, based on their skills, on the jobs that need doing and
supply motivation. Start them with some basic tasks to find the level they
are comfortable at. "Hey can you write the class comment for class X" or
"can you categorise this class's methods into core and enhancements" or etc.

Of course the biggest problem is that all these things are useless without
some way of getting them into Squeak. IMHO that is the biggest problem you
face and no amount of rushing to get to 3.9 will make any difference because
you haven't solved the underlying problems.

Russell






More information about the Squeak-dev mailing list