Thursday, February 14, 2008

What Interview Questions Do You Ask ?

Here are some of the things I might ask in an interview. I admit, I'm not the best interviewer, I'm a developer. But, I am part of the hiring process for new developers (and should be). I know many other developers are placed in the same situation. I hope to share my experience and learn from others.

They aren't listed in any particular order and this is neither comprehensive or complete. I'm not worried about publishing it because if a fraction of the people I interview actually took the step of reading the first few pages of Google results for whatever they are interviewing for they'd be ahead anyway.

Most of these are pretty Java specific. We use other languages too but, the majority of dev is with Java right now.

My goal is not to torture a candidate with trivia questions. My goal is find find out if they can write code and how well they can fit in with the team.

I'm curious to see what kinds of things others are asking too.

I actually ask a lot more code questions but, the nature of work makes those proprietary :-(

Start With Something Simple

(A good article from J.D. Meier's Blog)

What's the best way to build momentum and get results?  Start with something simple.  Seriously.  I get to see folks who get results and those who don't.  The difference nine times out of ten isn't smarts.  It's simply action.  The smart folks who don't get results, either get stuck in analysis paralysis or add too many dependencies up front.  The folks who get results start taking action and adjust along the way.

Why This Works
Starting with something simple works.  It's not that thinking up front doesn't help.  It certainly does.  The problem is, three things can happen along the way:

  • At some point, your thoughts are based on way too many assumptions and you don't know what you don't know.  You then find out too little too late.
  • Somebody faster came along.  While you're thinking, they're doing.
  • In the absence of results, your idea slowly dies inside.

The best way to fuel your fire is to incrementally get results.  Start with something simple.  Results feed on themselves.  If you start with something small, you'll learn faster and you'll start to adapt.  You'll inform your thinking.

How To Start
Start with the smallest thing you can personally do.  If you don't know where to start, here's key questions to help:

  • What's the simplest thing you could do?
  • What could you do today?

Personally, I find asking what I can do today to be the most effective.  Time is a great forcing function.  It's very easy to cut scope using time.  If you don't respect time, then it's very easy to add way too many things that will never happen.

Fail Fast
While starting with something simple helps build momentum, you'll also want to quickly spike on your risks.  You can do this separately, after you have some success under your belt. 

To fail fast, cut your idea into thin end-to-end slices and test your results.  For example, take one story or usage scenario and try to instantiate it.  Even before you build the solution, simply doing a dry run will reveal a lot of questions you can use to shape your approach.

The purpose of failing fast isn't to fail.  It's to uncover your risks and pick better paths.

(Read more at... http://blogs.msdn.com/jmeier/archive/2008/02/10/start-with-something-simple.aspx)

Friday, February 08, 2008

10 myths about innovation

Myth 1 Creativity is innovation:
Creativity is glamourous. "But getting a new idea is just 1% of the job. The rest is execution,"says Govindarajan."Companies don't pay much attention to execution because they think that innovation equals creativity,"says he.There is another problem related to execution: "What they know how to execute is business-as-usual or the core business. Organizations don't lack great ideas, but they struggle in innovation because of over-focus."
That's why Wal-Mart outmanoeuvred Sears in the discount retailing format. Or in the more recent Tata Nano case, as soon as the car was unveiled,Ford made a half-a-billion dollar investment in their Indian operations to come up with a small car priced at $7,000 by 2014."They are still not getting it!" says Govindarajan. "Someone has already fired the first shot with a $2,500 car. They ought to be thinking of something better."

Myth 2 You need a great leader:
it's a commonly held notion that we need a hero who champions the idea and works against all odds. "Innovation doesn't require a great leader, it obviously requires an above average leader," says Govindarajan. As for the notion that s/he fights bureaucracy, "It breeds a very wrong view because to succeed, you don't have to fight bureaucracy—you have to learn to coexist with it, to tap into the resource base."

Myth 3 It equals s k u n k wo r k s :
Skunkworks is a notion that the innovation team works in the "basement". It follows none of the rules of the corporate office. "But innovation in big corporations is centred around ways they can leverage their enormous capabilities to work on complex problems which a start-up would never do," says he. While innovation projects do need some space of their own, they also need to be connected to the main organization so as to leverage its resources. This also forces a degree of accountability on them.

Myth 4 You need major organizational changes:
Remember IBM under Lou Gerstner? "As soon as he came, he changed everything," says Govindarajan. IBM, under him, became a metaphor for how to catalyse innovation. "But that doesn't mean innovation requires major changes. Innovation requires targeted changes," says Govindarajan. "Changes have to be made in that part of the organization where that innovation is being cultivated. By wholesale change you can destroy the business-as-usual machinery." Because business-as-usual is what keeps an organization going but innovation is what keeps its future profits secure.

Myth 5 Innovation equals chaos:
"To say innovation equals chaos is dangerous because that's only about the 1%. The 99%— or the commercialization part—requires extreme discipline," says Govindarajan.
But isn't Google like that? From outside, the organization has nothing but chaos. "In Google what you see as chaos is empowering people at lower levels to conduct experiments. But they always ask them what they learnt from these experiments and how one can incorporate the lessons," says he.

Myth 6 It's the senior executives' job:
"The job of the people at the top is to come up with the strategy while people at the bottom execute it. People in the middle make sure things happen. But the problem with this approach is that innovation becomes nothing more than responding to changes." The people closest to the action are the so-called doers. So Ford, for instance, has missed a huge opportunity because it didn't empower the people in India to drive the innovation for the people's car. "Their approach is that people in Detroit know what works best," says Govindarajan. "The CEO's job is not to get actively involved in innovation, but to manage the organizational context."

Myth 7 It implies customer-orientation:
In the 1970s Xerox was making a lot of money with their $2,000 copiers when a bunch of engineers ("the doers") thought of $1,000 personal copiers.
Xerox asked its customers if it wanted these. The problem with that is: who is the customer of Xerox at that point? It is a big company's central copying department. So the answer is obviously no. "That's the problem of the big three automobile companies in India. The customer who is driving a BMW will never tell you he wants a one lakh car."

Myth 8 Planning is irrelevant:
"There is an element of truth to this because if you are in a rapidly changing industry, when do you plan?" says Govindarajan. But planning is important. "One thing we know is that the future is going to be full of surprises. The worst thing you can do in an organization is to be surprised by a surprise. You need to create a capacity to respond to surprises. Build mechanisms that even if the future doesn't happen like I plan, I have some capacity to respond to it. It is a learning tool."

Myth 9 It requires new products and technologies:
Many of the innovations have nothing to do with new products. The Mumbai dabbawallahs, for instance, distributed the Reliance Energy IPO marketing material with the dabbas to 200,000 households and delivered the filled-up forms to the Reliance office. Says Govindarajan, "Even Apple's iPod, one of the most popular innovations in recent times, is nothing but a handheld hard drive."

Myth 10 There are generalisable tools that companies can use to drive innovation:
Innovation comes in different forms and requires a nuanced approached. "If you apply general principles, you can actually destroy it," says Govindarajan. So the tools to measure various indicators such as performance of, say, GE's Six Sigma which is a continuous performance improvement process vis-à-vis a Tata Nano will wary.

- Vijay Govindarajan (in Times)
(Govindarajan is the Earl C Daum 1924 Professor of International Business at Dartmouth College's Tuck School of Business and director of Tuck's Center for Global Leadership. He is one of the four Indians to appear on the Thinkers 50, a ranking of the Top 50 business and management thought leaders in the world)