Re: [Adonthell-devel] Sign of life

[Kai Sterker]

On Thu, May 31, 2012 at 10:41 PM, Chris Frey <address@hidden> wrote:
> On Thu, May 31, 2012 at 08:40:41PM +0200, Kai Sterker wrote:
>> But first, the state were we have Waste's Edge again needs to be
>> reached, at least in my opinion.
>
> I think like that too... "if I just get here, a release will be suitable"
> but I'm not sure that is correct anymore.  More like, "what's the minimum
> I can polish to make a release?"  Because each new feature or bugfix is,
> in theory, useful to somebody, but not very many people will know about
> it if it is just a git commit or a mailing list notice.  It's the release
> that makes that feature widespread and useful.

This may be the case for utilities and tools, but for a game, as user
I would expect at least some kind of gameplay to call it useful. I
guess you could demo small aspects of a game (conversation with NPC,
NPC pathfinding, game controls, items, inventory, ...) and this is
what worldtest (and all the other bits and pieces in the test/
directory) is all about.


> Ideally, it would be good if end users were driving the feature requests
> to some degree, because that's when you've transitioned from scratching
> your own itch to solving problems that other people have.  And when
> people's problems are solved, users increase, and when users increase,
> contributors increase.

Again, this approach is more difficult for a (complex) game than other
software. What you strive for is entertain the users, not perform some
specific task for them. So I guess user feedback is very valuable and
in case of games a good way to give additional polish. (For example,
if DB will ever see the light of the day, I would love to hear
suggestions about alternative solutions for quests or seemingly
missing choices in dialogues and implement those).

But there needs to be a foundation for that kind of feedback.
Something, however small it is, that's functional. Worldtest shows
features of the engine, but I wouldn't call it functional. WE, albeit
no true RPG, is functional IMO. It's known to entertain people for a
couple hours :-).


>> Something in style of the previous three 0.4 alpha releases
>> (worldtest) could be done anytime. And with an official engine release
>> it would also be possible to have an official tools release (as the
>> tools require the shared libraries from the engine.
>
> That sounds promising!

I assume that might be the path to go then. It also crossed my mind to
maybe split the test stuff from the engine itself. People shouldn't
have to recompile the whole moloch each time a small new thing is
added to the test data. OTOH, most  new features added to test depend
on new features added to the engine ...

But, being able to split engine and test would allow to give more
visibility to each individual test.


> I think that might be the wrong question.  What we want is to spark users,
> or in this case, experimenters.  The contributions come afterward, in
> my experience.
>
> Maybe a series of demo releases, instead of game releases, is better.
> Each release would have some demo (maybe with docs) of a small feature
> of the engine that is newly complete, showing how it works, and how to
> use it, from a game designer's perspective.  With each release, that
> feature gets widespread publicity, and as each small feature is demoed,
> people's creative juices get tickled as they play with the demo.  Plus,
> each release is an excuse to announce your project again on places like
> Freshmeat/Freecode, which can increase potential users, and therefore
> contributors.

I strongly dislike this kind of attention-grabbing behavior, but I
guess it can be quite effective :-). The more active a project
appears, the more additional interest it attracts.

But your general idea brought me to consider splitting engine and
test, as noted above. There's also still the idea of demonstrating
stuff in videos instead of releases, too. Easier on people to just sit
and watch for a few minutes than download and compile for an hour,
then test the new feature for half a minute. Maybe combine the two,
though. Have the "lazy" option for the broader masses, and the code
release for those with actual interest in all the details.


The question then is, should WE be put aside for a while and instead
other aspects of the engine get the focus? Item/inventory GUI and
world integration comes to mind. Crafting afterwards. Pathfinding in
the 3rd dimension would also be a candidate, which would also be
useful for WE. Then, when those features gained the required
attention, WE could be finished much more quickly and additional
features required by DB would already be complete.

Opinions?

Kai