Ryan Sloan

January 11th, I will begin my next adventure at Georgia Tech. I’m not really talking about Spring Semester 2010, which should be another chapter of Georgia Tech life. I’m referring to the fact that I will be embarking on my first adventure in project management. As the Information Technology chair for Georgia Tech’s Student Government Association I have a variety of responsibilities. Until this time, most of those responsibilities involved adding/modifying content on SGA’s website (and sitting in a lot of meetings, of course.) Next semester my committee is going to undertake a comprehensive rewrite of Course Critique. Course Critique provides grade distributions and survey information for every course offered at Georgia Tech. Unfortunately, the current incarnation is prone to outages (particularly during registration periods) and several portions of it are not functional at all (namely the mechanism that uploads and processes grade distribution and survey info — a critical component.) After assessing the code base, I decided it was best to begin a complete rewrite. But this time we are going to do it right.

Even though this project doesn’t officially begin until January 11th, I am using what little spare time I have had over the winter holiday to begin what is arguably the most boring phase: requirements gathering and specification. While this may not be the most interesting phase, I really believe that it is one of the most critical (if not the most critical.) Proper design and documentation will (hopefully) prevent this project from going the way of its predecessor.

I titled this post “An Unlikely Project Manager” because until recently (like, really really recently) I saw myself as a software developer. I figured I’d be a developer forever, which I was sort of ok with. It was in the past year when people started suggesting that I could combine my technical skills with my skills as a communicator and bring balance to the force (Hopefully it goes better for me than it did for the Jedi.) This is my first opportunity to get my feet wet and see how good I really am at this whole “management” thing.

In an effort to avoid becoming a pointy-haired boss, I am going to be doing as little managing of people as possible. I’m going into this experience with three key ideas/quotes:

1. Management is taking out the trash.
2. “Don’t tell people how to do things, tell them what to do and let them surprise you with their results.” — George S. Patton
3. “In preparing for battle, I have always found that plans are useless but planning is indispensable.” — Dwight D. Eisenhower

Management is taking out the trash

I wish I could remember where I heard this quote, but I searched for a long time and couldn’t find it anywhere. (Maybe I invented it? Unlikely.) Regardless of its origin, this statement captures what I believe (and remember, I’m a college student who hasn’t done any professional managing) is a key idea in project management. I will be doing little-to-no coding on this project. Instead I’m relying on the ingenuity of my team. My primary purpose will be to remove objectives from their paths. I’ll be doing most of the preliminary planning, communicating with third parties, initially setting up their environment (issue tracking, source control, etc.) and all the other boring stuff. I guess what I’m getting at is that you should let them do what they’re good at, and don’t bother them with the other stuff. Sometimes that means taking the can out to the curb.

“Don’t tell people how to do things, tell them what to do and let them surprise you with their results.”

Patton was dead on here. As I mentioned before, I will be doing almost no programming. I’m ok with that, but it’s important that I avoid the micromanagement trap. This project will be fairly large, and there is no way for me to have my hands in everything. I picked this team because I felt they were capable, and I think they can put together some great software without being pestered. This isn’t a promise that I won’t offer my opinion, of course, I just want them to argue with me when they disagree. I’ll probably cave.

“In preparing for battle, I have always found that plans are useless but planning is indispensable.”

Well, people liked Ike. I thought if I included a quote from him people would like me, too.

In all seriousness, I think this is one of my favorite quotes. Not just as it relates to software engineering or project management, but in general I think it’s a pretty solid guideline. I am going to make sure that my team is given a complete and clear requirements document as well as functional specification. We’ll be laying out a schedule of milestones with target dates. And in spite of all this, approximately one week into the project we’ll realize we’re behind/the requirements have changed/there is no spoon. But that’s ok. It happens. Things will almost certainly not go according to plan, but thorough planning will give us a deeper understanding of the problem, making it easier for us to adapt to change. We can’t realistically count on meeting all our deadlines, or the server setup working just the way we would like it to. People will get sick, or become overwhelmed with school. There’s really no way to know what’s going to happen when we embark on our first sprint. Adaptation is the key to a successful operation. So we’ll adapt.

Here we go…

So I’ll be keeping those three things in mind as I start this adventure. I’m planning to use this blog to chronicle the progress of the project (partially as a personal record, and partially because it seems like a good way to generate some content and discussion.) Wish me luck!

or The One in which I Rebuke all things Shiny and Cutting Edge

Living your life on the bleeding edge can be exciting (I mean heck, Aerosmith wrote an entire song about it.) The Edge is an exciting place — gadgets are shinier, smaller, and more expensive there. It’s hard to resist the allure of a new gadget, or the features of a new technology. We’re all guilty of it from time to time. All it takes is one press release, one vague and over-hyped magazine article, or (for some of us) a late night TV commercial with a 1-800 number plastered across the bottom of the screen. Next thing you know you’re busting out the plastic. Trust me, I own a Snuggie or two. I think there’s some law of nature that says “He who Dies with the Shiniest Stuff Wins.”

It’s bad enough to spend roughly the money supply of Turkmenistan on new gadgets, but it’s when this mentality carries over into other fields that it becomes really dangerous. When I say “other fields” what I really mean is “software development.” I have had two experiences with this, and both of them had one thing in common (other than the fact that the projects were overwhelmed by the shininess.) In both cases, it was decided (once by an executive decision and once by a vote) that we would be using Ruby on Rails for a new project. There is nothing wrong with Ruby on Rails, but on both teams, there was only one person who had written so much as a line of Ruby in his life. I’m going to talk about the most recent project in this post, because it was equally stressful and frustrating but it has a happier ending. Blogging is a lot like writing a summer blockbuster in that respect.

DISCLAIMER: This post is about to descend into what many will call “RoR bashing” but I assure you it’s not. Rails is a badass framework, and I hope it’s going to be around for a long time. This isn’t about the technology itself, it’s about the “let’s use what’s hot right now” mentality.

One of the Computer Science courses that most every Computer Science student at Georgia Tech has to take is CS3300 - Introduction to Software Engineering. It’s the second course where students are assigned a large team project. (The first being CS 2340, which I wrote about in my previous post “Adventures in Teamwork“) In this course, each team is instructed to come up with an idea for a software project, and go through the Software Development process to bring it to life. We decided to build a web app that would detect the user’s location (using HTML 5 geolocation services) and find the nearest restrooms with crowdsourced cleanliness ratings and comments. (We decided to call it “Poopt,” for anyone who was wondering. I can never resist a pun. Apologies to Sam Altman.) It would run in any web browser that supports HTML 5, but the target would be mobile phone users. I got the idea from an episode of Seinfeld where George told Jerry to give him any street corner in New York, and he could tell him the nearest clean public restroom.’

Before we even started design, one of my group members turned to us, and with a twinkle in his eye, suggested we develop our project in Ruby on Rails. The other three of us got pained looks on our faces — we had never written a line of Ruby in our lives. He assured us that both Ruby and the Rails framework were easy to learn, and that RoR could essentially do the hard parts of our project for us. Nothing gets a bunch of overworked Georgia Tech students on board like the promise of less work. On top of that, this could be a cool opportunity to learn something new! So, with some free RoR resources on hand, we decided to dive in.

Now, group work is a funny thing. It can bring out the best and worst in people. In any team, you can find at least one or two slackers (if you can’t find one, it’s you.) In my case, every member of the Poopt (it might be juvenile but I bet you grinned a little) team had agreed to learn Ruby and do their part. Things started off well — our RRoRG (Resident Ruby on Rails Guru, pronounced “Rawrg”) gathered the three of us together and started explaining things. Unfortunately, after day 1, interest started to dwindle in learning Ruby. There is supposed to be this little voice in your head named “Common Sense” that says things like “Now Ryan, maybe deciding to build a (tightly scheduled) project in a language no one knows is a bad idea.” Unfortunately, my little voice was too busy drooling and saying things like “so…shiny…”

Needless to say, by the second week of our project, the team had realistically dwindled to two: RRoRG, and poor Ryan, just trying to keep up.
In the end, we finished our project, and managed to pull an A out for every milestone, but it was largely because Mr. RRoRG worked really hard. I did what I could, but we both recognized that because of our timeline, there was no time for me to try to learn, the best I could do was tweak displays and templates.. RRoRG worked long nights, and I sat at the table and try to take care of all the bureaucratic and administrative legwork. I was totally willing to learn, but with the time constraints and my other coursework, there just wasn’t enough time.
Now, RRoRG was right about one thing — Ruby on Rails DID have a lot of features that made our project much easier. It would have been perfectly suited for our project if the rest of us knew what we were doing. Sometimes, you just have to settle for the old fallback. Maybe it’s less shiny and less exciting, but when it comes down to it, you’ve got to work within the constraints of your system.
I think I can sum up this long diatribe with one sentence: Next time you find yourself rushing out to buy a Snuggie, ask yourself: will a blanket do?

Epilogue
I’d like to say thank you, RRoRG, for working long hours, trying to teach me when you could, and not lumping me in with any other dead weight that we may or may not have had. You know, some day (when I have the free time) I’d really like to learn Ruby on Rails…