A lot of people asked me to blog as this volunteer project progressed, share some insight into how things were going. I’ve not had enough time until just now, and it’s a mix: Some good news, some bad news.
- It’s a friendly group of people, and we seem to get along well
- Alex Crouzen picked up Objective C and OpenGL, and Shi Daniels threw together some wall and floor textures. That gave us a very basic “here’s some walls and floors, running in OpenGL, on an iPhone”, but no walking, no movable camera, no interaction, etc. It actually worked! :)
- Brian Green and Wendy Despain (both games industry pros) joined up and did loads of great game-design and plot work. As a group, lead by them, we generated lots of pages of good design – although (my fault) the remit probably wasn’t as well constrained as it should have been
- Lots of good ideas and commentary from the other people in the group
- We setup a private wiki, a 24-hour skype group-chat, and a private google group, and had some pretty good communications going between the active contributors
- I picked up Alex’s code and rewrote the underlying OpenGL stuff (I’m very rusty; haven’t done any OpenGL for > 4 years), but got it integrated with iPhone GUI stuff, so you could use drag gestures to move around the map in 3D, and added some programmatic map generation
- A lot of people (I think more than half, in the end!) had to step back – or even drop out – due to time constraints / external commitments (work, life, etc). Ultimately, through no fault of any individual … our team was somewhat gutted before we’d got beyond the first milestone :(
- …as a result of that, and some mistakes I personally made on the project leadership side, we stagnated, and the project has ground to a halt with too few active coders
It’s not clear where we’ll go from here, but I’m now looking at whether I can take some of the lessons learnt from this and start it all over again, do it better, more simply, and more clearly directed, as a commercial project.
One of the biggest mistakes I made personally is that I tried to do this as a volunteer project – but simultaneously tried to do it unmanaged, Scrum-style. You could fairly argue that this was foolish, given that I’ve run a lot of volunteer projects before, and they have *always* needed huge amounts of vicious, aggressive project management. Not because people are lazy or dumb, but merely because volunteer projects are always the lowest item on people’s list of priorities (way below “day job” and “family”, for instance – which is fair enough).
But I’ve got so used to working this way now (team-based, distributed responsibility, etc) that I ignored my previous experience and actively avoided aggressive PM.
I’m putting together a detailed project plan now – as if I were doing this commercially. i.e. working it out based on me personally paying other people to work on it – it may not be viable / possible, but if nothing else it’s a good practice to plan to that level of diligence. It’s just a theoretical exercise right now, although I’m sharing it with some industry colleagues and some professional coders and artists, seeing if it’s convincing enough for any of them to want to try it.
One of the interesting things I’ve noticed already while writing it up that way is that – if I have to pay people – I’d start with a much smaller team, almost entirely coders, and only bring on other people once we have (*if* we have…) a substantial working prototype. Which is the opposite of how I was taught to make games originally. But it’s exactly how I was taught to do commercial startups. This could be interesting…
PS: here’s a screenshot of one of the first renders:
4 replies on “Volunteer project: a simple RPG for iPhone – UPDATE”
The picture’s a broken link for me!
Of course, as soon as I post that comment, the picture appears. I think it’s a Google chrome / self-signed cert thing.
The link should be http not https is all. (And firefox 3.5 refuses to show it as https)
It also says this site’s certificate has expired
Hm, I was under the impression from the UI screenshots that you were going landscape, not portrait..