MMO Development Lesson #20
It’s better to start with a great shell than a jumbled mass of crap. Okay, that wasn’t articulated well, but the point is this: If you are outlining a system or some content or whatever it is, don’t mash a bunch of random ideas together and hope they will work out. Instead, create a notes document in which you jot down all of your random ideas for whatever you are creating, then write a structured and cohesive document as the document. You will thank yourself later, because trying to revise a jumbled mess is even harder than writing something completely new (I tend to just wipe things clean and start over if I don’t have the notes doc).
In other words: project management?
A great reference here would either be the PMBoK (Project Management Body of Knowledge) as outlined by the PMI (Project Management Institute – http://www.pmi.org). That is, if you’re utilizing a more traditional waterfall style SDLC (software development lifecycle) methodology.
Of course, you could also try a more interative style project, using Agile style management methodology, like SCRUM. (I hear that SOE is starting to experiment heavily with SCRUM).
Regardless of which methodology you choose (you could pick any number of PM flavors here dotted with additional disciplines – everything from a version of Lean, SixSigma, Critical Pathing, etc), but in the end, the basic point is the same: plan before you design; design before you construct.
This is one of those areas where I think many projects fail – and by failure, I don’t just mean bad releases or buggy products, though those are certainly more obvious failures. I mean failure in a more generalized sense, where a “failure” can include scope creep, budget overruns, or missed milestone dates. Though not immediately viewable by the ultimate user, these failures can be just as damaging to the underlying bottom line.
One can launch a great game, but if the budget was signicantly boosted during the process, or if designers had to work inhuman hours to make up for bad planning (not just design planning, mind you), then these issues can take a very tangible toll on the company driving the creative process.
From a purely design perspective (which is more of what I think you were driving at), then it only makes sense to plan before you design. You have to move beyond the “Wouldn’t it be cool if…” stage of brainstorming, and at some point post a solidified and approved set of requirements.
That requirements document should then be referenced as a contract between the various areas of the development team, from content designers to artists to engineers and coders. Development a framework first, then expand upon it as needed – but have that framework approved and agreed upon early on in the project, to set and reaffirm expectations.
There are a variety of templates and and processes which already exist to help design requirements of this type – and in fact, an entire profession virtually exists for the purposes of designing such documentation. Your friendly neighborhood project managers, project coordinators, and business analysts can help you to learn more.
Personally I like the mind-mapping technique for the first part of this. Instead of writing in a linear form (like in Notepad or Word), you just start with a blank sheet of paper and brainstorm in every direction. Use colors, highlighters, and follow every train of thought to its end. Then you can look at the huge jumbled mass and start grouping and organizing, refining the idea until it’s well thought out.
I use a similar idea when writing. Before I start a new novel, I sit down and map out all the plot threads, all the scenes I’m thinking about, the characters, the backstory, organizations, etc. When the story “looks right” from this bird’s eye view, then I get into the trench and start writing the text. The map is loose enough to allow for spontaneous creativity, but enough of a guide so the story doesn’t derail (where I waste a lot of time writing a branch that I’ll later have to scrap for pacing).
Planning is very important for any kind of project. It would be nice to have the writeup done first and get everything on the paper done before looking at additional things to add rather than saying “ooo, let’s do this!” and then have to cut it at the last minute because there isn’t time.
Feature creep is definitely a CR15 encounter.
Thermoses, there are some pretty good brainstorming/mind mapping applications out there. I’ve personally used Curio on a couple of smaller projects. It did what it was supposed to do well enough:
http://www.zengobi.com/products/curio/
Mind if I use this as a bumper sticker?
Every proejct I work on I try and pound this in to the folks I’m working with. Sadly, some just don’t “get it”