Yeah, me and Toen were having a bit of a discussion lately, and I agree. Planning, besides initially, is definitely something I don't do enough.

I'd like to say this as well: Gruedorf is commendable for its encouragement to make developers do something every week.

But there's something I forgot as soon as I embarked on this: I lacked planning. Forcing weekly updates without having goals set about what to do was a good way to waste progress.

I'm taking a new approach now, to ensure my ideas are properly communicated and remembered. The following are rules I want to employ on my development projects from now on.

Anything that is planned must also be written down.

Stuff that isn't written gets forgotten, and it's also how you end up deviating from your original goals and get sidetracked. Anything design-wise that changes shall also be written down, and logged separately.

Designs should be worked on until there's enough clarity about the task at hand. As ambiguities arise, write down more detail on topics, so these things are clear. If something doesn't work go back.

When the design is fleshed out enough, start working.

As problems arise, go back to the design, and write down the changes you've made. This should be done as a separate update, without erasing the old copy of the design. This way if a change was a mistake, you can go back because you still have records of the old information.



With this degree of constantly recorded changes and current design plans, it should be especially effective with teams. Collaborations would not only employ discussion, but afterwards, note-taking of the events. Someone could keep minutes as you're going, on a wiki or something. The wiki updates would then be pointed to during the meeting and disagreements in the writeup would make many ambiguities clear right away.

I dunno, this is just a dump of ideas, really, but I think it has some merit. I agree with Toen's views, and I hope that discussion on this will at least make it clear to the struggling they need to try a different attempt or be stuck like so many of the experienced-yet-no-finished-game veterans in our community.

You can go crazy with too much design structure and meetings and approval processes as well. I won't be one to rail against planning but a lot of Diver Down was made up as I went along. I mean, I had the entire plot scripted out and there were scenes I knew I wanted, but i don't think there is something wrong with not knowing what the crap dungeon #4 is gonna be like when you are starting production on a verge title.

Still, had I known everything beforehand it would have turned out a better game. So planning is great. But I think focusing on moving forward and fighting feature creep are worthy pursuits as well.

In the indie game scene, we all tend to be dreamers. At some point you need to say, "maybe these zombies don't need to be limbed 20 different ways and instead I could start on the next town." Finishing the game is a tough thing to do, but you can break it down into manageable chunks. If I could sit down and bang out the art and programming for a monster encounter then I was on top of the world. That's one step closer to completion. So it really boils down to a drive to the end.

I totally agree with you all.
For my own experience with VERGE, especially V1 (good old DOS time), I was able to concentrate my effort on making the structure for my demos. The pre-sets functions parts of V1 was really useful, I've more the artist fiber than coder one.

When V2 cames out, I was really excited to port my works on it, since the engine added nice news features. I was really motived, with lot and lot of ideas I hoped to realise. But... yeah, when I understand V2 need more Code structure, I was disturbed, and all my ambition gone away.

Again with V3, nice features I would like to use for a remake of an old demo, higher resolution, nice effects, etc... But now, Im not able to code even very basics scripts to launch the game.

Organisation is very important, in every projects (commercials, independants, amateurs), I had the chance to have Wooly on my side, who procured me structure and advise, and nice contact with Seren (from Revelation). Good old time eh. Without them, I would not be able to do something.

Im actually looking to V3, but the code side just make me crazy !

(Im sorry for my bad english ><)

I'll finish this post in french sorry :/

/french part/
J'ai eu la chance d'intégrer peu de temps après mes aventures sur Verge, une formation 3D qui bossait avec des programmeurs, sur des projets communs le temps d'une année. La structure et l'organisation sont vitales pour aboutir à quelque chose de satisfaisant.

Je me rappel avoir pesté à maintes reprises lors de mes essais pour Menolly, ou encore Life, qui utilisaient tout deux beaucoup de codes que j'avais pu trouvé dans les exemples accompagnant 'Sully Chronicles', sans cela, je n'aurais jamais rien fait.
/end french part/

I had very good moments with the whole community, I remember the old Repo, all the efforts makes by Vergers, and the great quality (and quantity ^^) of demo/projects/games.

For now I'm a little 3D Animator for a french DS developers, and when I see some artists making tiles and assets for 2d games, the only thing I want is to launch Verge at home !

Im sorry, I don't really know why Im boring you with all of this. I comes often here to see what old friends/contacts became.

(And I hope to see Wooly again grrr :p)

Thanks for you time, thanks to Vecna, McGrue, Overkill, Seren, Wooly, and a lots more for the good moments I spendt with you all !

There is my MSN for contact : david.kuhmel@wanadoo.fr (yes it is :p)