Re: 7.1.95 instant comments

[Massimo Campostrini <campo@mailbox.difi.unipi.it>]

Stan Shebs <shebs@cygnus.com> writes:

> I can see putting warnings in the blurb about things that are known to
> be incomplete, but I'd have a hard time assigning ratings to some
> games.  For instance, tokyo.g *was* playtested, but then I enhanced it
> to have burning buildings that explode and spread fire, and the
> resulting game hasn't been playtested as much.  Also, how much testing
> is needed to quality as "playtested"?  What about human vs human as
> well as human vs AI?  I note that commercial game companies have the
> same problem - they playtest what they think is a lot, then real users
> crash the game right away...

I was thinking of something very rough.  As a minimum let's say a
rating of "well tested" (players can complain if it's
broken/unbalanced), "complete but not well tested" (players can
probably enjoy it, please try it and provide feedback), "incomplete"
and "not an entry point" (both designer only).  A blurb for incomplete
games is a must, but a IMHO "rating" file in the lib dir would be
nice.  How much testing is needed?  I would think "it has been played
a few times human vs AI and human vs human, and people found it
interesting and did not quit early on because it was
broken/lame/unbalanced"; after all, a game is not a compiler. ;-)

> I suppose an "xconq-bugs" would make sense.  "xconq7" is open to generic
> discussion now, but I don't see people drowning in mail.  Or have people
> been inhibited because they think this list is for developers only?

I don't see "xconq7" mentioned anywhere in the docs, so I guess that
people have been inhibited because they don't know of the list; or
because they don't know of the game, but we will run a little
advertisement this time. :-)

	Massimo