Re: Testing ideas as 7DRLs

At Tue, 23 Sep 2008 23:23:59 -0700 (PDT),
Antoine wrote:

On Sep 24, 2:43 pm, Slash <java.ko...@xxxxxxxxx> wrote:
On 23 sep, 20:30, Antoine <antoine.from.r...@xxxxxxxxx> wrote:
3 - create a variant of an existing major open-source RL (ie Nethack,
Crawl or *band).

IMO #3 requires the player to know (possibly to know a lot) about the
base major open-source RL, knowledge which may be wasted time if the
dev wants to go his own way, me thinkz...

I see your point.

But (1) it is good programming experience to find out about how
another RL works, so maybe not 'wasted' time in the end, and
(2) maybe a dev who thinks they want to write their own game will
realise in the end that they'd rather be involved in developing an
existing game. In many ways this seems to be more fulfilling - look at
the Stone Soup guys.

I noticed that people seem to be, for some reason, afraid of reading
other people's code. Or maybe just disgusted with the idea. Or maybe
there is some other sort of barrier there; no matter what it is, I
find it very hard to convince or even force people to read exsting
source code. With *understanding*, of course.

This is a very unfortunate fact, because reading and analyzing existing
code is one of the beast ways to learn and improve one's coding skills
and style. It occured to me what I read one of interviews with Donald
Knuth, where he mentions how he was intrigued with one of the first
compilers and how he printed all of its sources out and analyzed during
his vacations.

Programmers who refuse to read existing code (the justification doesn't
really matter) are like people who never read any books but would want
to write their own novel. They might be able to write an original story
and maybe even some good fragments, but without all the experience from
reading books they will just lack the technique required to write a good
book. Note how even reading bad books helps here.

So here's the advice for all of you: download and read source code from
some roguelike game now. Think of some small change you could make in it,
and try to find the right place in the code to make the change. See if it
works. (This finding the right place is a good way of ensuring that you
actually understood the source.) Try adding it in several different ways.
Try to fix some annoying bug or add some convenience feature to the user
interface (you can send the patch to the author once you are at it).

It will really help you with your own game.

--
Radomir `The Sheep' Dopieralski <http://sheep.art.pl>
"Whenever you find yourself on the side of the majority,
it's time to pause and reflect." -- Mark Twain
.

Relevant Pages