When F. Scott Fitzgerald wrote, “The rich are not like you and me,” he could have been just as easily referring to the gap between programmers swaddled in their nice little cubicles and those of us who for better or worse are out here in the cold and dark peddling our microISV wares and startup services.
Gone are the easy days of knowing exactly what is the absolute! best! programming methodology: it’s the one your manager tells you to use. Whether it’s a million bucks of RUP or a big whiteboard in the conference room, that’s the right one, forever and amen. Suck it up and salute – besides they all work more or less as well in a group, right?
But what if the lead developer and the clueless new guy and the anal-compulsive test engineer and the artsy-fartsy UI artiste are all staring back at you in the mirror each and every morning? Does counting my cats in my programming team and pair programming with my bonsai plant count?
No, they don’t.
One of the less pleasant aspects of the startup/microISV life is not only are we as naked as new-born chicks of all those nice corporate programming policies, procedures and methodologies yet we know deep in our programming souls that we should be following some sort of methodology because a good methodology is the only thing that stands between you and complete code dementia.
Since I started my first microISV app in 2003, I’ve searched for a way to avoid code dementia (some would say too late, but I digress). I didn’t find it while writing MasterList Professional, sure as hell didn’t find it as I’ve fitfully enhanced it after its initial release. And really, really wanted to find something when I started scoping Project X nearly a year ago.
The problem with even the lightest of methodologies in my opinion is they just don’t scale down well to one or two developers. And they sure as hell don’t scale down timewise when the only time you have to work is on the side of your day job, whatever that may be.
At least for me, the light at started flickering at the end of the microISV programming methodology tunnel when in February I got my hands on “Head First Software Development (Brain-Friendly Guides)” By Dan Pilone and Russ Miles.
You need to read this book if you’re doing a microISV. Not because it’s cute, funny, realistic. But because it is development methodology for real small teams – like 3 people – without the corporate cruft. The chapters on approaching user stories and tasks, project planning and good design are definite startup must reads. And the burn down process – including calculating your actual velocity – will get your head out of the clouds of cool features to play with and back down to customers want software that just works, dammit, right quick.
So what else have you got?
Actually, something really nice. While HFSD gave me some good ideas on how to implement a ultralight development methodology, doing paper sticky cards for user stories and moving them around on my side desk just wasn’t cutting it. I needed software, and I needed it now!
So taking a day I went digging for agile development methodology software that a) didn’t suck b) scaled down below the corporate market c) would either be Mac based or run decently on the web. I found Acunote, and six weeks later I’m still in love.
Acunote is a Ruby on Rails Scrum-centric software project management SaaS. Between its keyboard support for data entry, inline editing and other ajax-goodness and burndown charts it does exactly what I need without one wasted click or keystroke.
And you can’t beat the price: “You’ve been featured on TechCrunch. You have revenues and are ready to share a little. (with us)” – $99/month. Or, “You’ve launched, raised the seed round and your data would benefit from the security of SSL. – $49/month. Or better still, “You are a small startup building the first version of your product. Save the money for ramen.” – Free. Free is good, understanding who I am is better and finding the right a company that understands my world is absolutely best.
Bottom Line: building a startup or microISV is tough enough without the deadweight of hauling around some massive development methodology left over from your corporate days. Ditch it for just enough structure that keeps you focused on coding code that is going to make a difference to your customers. For me, it was Head First Software Development plus Acunote. Give them a try.