Oh Most Wise and Majestic Emacs, please be in -*- text -*- mode!
* Add link to http://wiki.java.net/bin/view/Javanet/CommunityReferences
* Add note about importance of recording patch credit in a parseable
way from the start.
* Add http://www.selenic.com/mercurial/ to the list of VC systems.
* Add a note about the bogus-authority problem:
So-and-so new volunteer posts a patch, based on suggestions from
some developer. When other people review it, they criticize the
very things that the experienced developer had suggested. So-and-so
cries foul, saying "But I *asked*, and I was *told* to do it this
way." So-and-so has made the Authority Mistake. It doesn't matter
who authorized it, that's not the way things work. The change is
being discussed as itself. It's okay to point out that that first
developer was in favor of it, as an indication that you're not
totally insane, but don't think that gets anyone off the hook.
* This could be worked into Chapter 5 or 8 somewhere, but where?
(Inclusion of internal issue numbers in public
commits. Though not strictly about money, it is about the
relationship between internal/profit-driven, and external/volunteer
development.)
* Poul-Henning Kamp's post really slams Brett Glass by name. Do we
want to either blank that out, or find out some more about Brett
Glass and put in a disclaimer that one's impression of him shouldn't
be formed solely from this one post of Kamp's?
* In chapter 5, there are some deep lessons to draw from the whole
(insert projname here, you know the one) debacle. There is no such
thing as an independent project. One cannot just assume that the
official public structures are the only ones that are relevant; all
interested parties have to get involved. This is politics, it
cannot be avoided.
This case history also has relevance to Andy's suggested topic of
how to handle a difficult developer who happens to work for you.
* The Apache glossary http://www.apache.org/foundation/glossary.html
might be nice to refer to as an external glossary. It's a good
browse.
* do a grep for "issue", replace with "bug", just in case
* for Chapter 2: don't make your home page be something like
http://www.cc.utah.edu/~nb3367/. People will assume it's not going
to be around after the end of the school year. (That particular
home page is about the worst project front page in history, btw.)
* "Software Development and Cosmic Rays" essay, ~/mail/sp-outwent
(where else? oh, duh, in www/writings)
* Ben's notes on People Behaving Badly (see ben.txt). This probably
isn't appropriate for the treeware, but I'd like to link to it from
the online version.
* examples.mbox
* the "lone genuis" exception: you can get a lot done if you're just
*that* good. But then, you probably don't need this book.
I have several friends who know [omitted] to some degree. One of them
said "he often walks the fine line between genius and lunatic." The
problem is, genius is such a commodity these days, that it's not
acceptable to be an eccentric any more.
-Greg Hudson
* The Debian New Maintainer process as a fantastic example of
conscious training / indoctrination.
* Jim's point about co-ops that weren't content just to solve a
limited problem, but had to solve *everything*. "Okay, I can see
how this ensures we have access to whole grains through the winter,
but how does it solve the problems of world hunger, and is it doing
anything to bring relief to oppressed peoples?"
* The emacs-devel bug-tracking discussion (whew).
* Greg's "no power plants" rule; the Mozilla "super review" system.
* What is openfoundry.org? They host the bugtracker for svk, apparently.
* Erik's comments on the IRC section in Chapter 3:
been reading from your link a bit. interesting reading.
you may want to state that longer texts in the info bot
are usually easier to understand from the recipients POV. Also,
those can be sent privately to take some 'burden' off the channel at
busy times.
a nice bot example is 'minion' in #lisp: a bot with an attitude :-)
thanks for the tips!
* Replace all subversion.tigris.org URLs with
subversion.oss.collab.net, if the migration happens before
publication.
* Don't make people always fight their own battles. Especially if you
can predict the outcome, then you'll get a _lot_ of appreciation at
a relatively cheap price, if you go to bat for them sometimes.