On Breadth First and Depth First thinking

Great talk by Don Norman, a founder of The Cognitive Science Society, and widely considered to be the first to apply advanced human factors to design via cognitive design on The three ways that good design makes you happy:

The three ways he talks about are:

  • Visceral: How pleasant things seem to work better.
  • Behavioral: How good design lets you feel in control.
  • Reflective: How people prefer design that reflects their personality.

One interesting point he makes is how a happy state of mind is puts the brain in “breadth first search” mode and is more conducive for out of the box thinking, but tension and some pressure puts you in a “depth first search” mode (as dopamine gets added into the system). Moment of self realization here for me personally: as a programmer I have often preferred tighter deadlines (note: not to be confused with unrealistic) to open ended project timelines. I need the pressure to focus. A bunch of my programmer friends say they feel the same way. Thats why the burn down charts for closed work tickets is almost never the ideal one:


(note the change of slope as projects approach end dates = more pressure)

There is often the temptation (by me anyway) to carry the model of deadline driven development to design, but Don Norman’s talk indicates that this would be a bad idea, especially if you are conceptualizing new experiences. How much pressure in terms of deadlines and deliverables can anyone apply on design teams before you kill lateral thinking thats absolutely required for creative teams. However its just as easy to fall into the time sink of finding the perfect experience. Voltaire’s quote about Good being the enemy of Great comes to mind again.

So concept in breadth first mode, execute in depth first.

The concept also seems applicable to application/web design around browse and consume. Browse pages/screens, where there isn’t an end goal really and a user is out looking for “something interesting”, should put people’s brains in breadth first mode, with more emphasis on visuals, etc, but when the user is in content consumption mode, like say on video consumption pages or web search, pages, a muted design may be a better idea. It certainly explains why Google Search pages, which most of us use to get to something we need and not really browse, work so well, but on all their other applications where users are more about entertainment than work, their designs feel awful.

Notes from Jonathan Rosenberg’s talk on Rules to Success

Fantastic talk by Jonathan Rosenberg, Senior Vice President, Product Management at Google on Rules to Success. The talk was so heavy with aphorisms that I ended up watching it twice and noted down his various points. They may be slightly off missing a couple of points here and there, but still pretty educational (for me at least 🙂 )

Starting note: I loved this statement at the beginning of the talk:

If you want to build a ship
don't herd people together to collect wood
and don't assign them tasks and work,
but rather teach them to long for the 
endless immensity of the sea.

Antoine-Marie-Roger de Saint-Exupery

Rules on Communication

  • Overcommunicate always all the time. You cant communcate enough.
  • Openly share everything with your collegues. Trust your people and give them this info. Trust breeds loyalty.
  • Repetition does not spoil the prayer.
  • Each word matters. Be crisp and direct and choose each word wisely. Its is not rambling: Leave out the parts that people skip. “I would have written a shorter letter if I had the time” : Mark Twain
  • Great leaders are great teachers and great teachers are great storytellers. Narrative is what that matters.
  • On Talking:
    • As leaders you learn more by listening than talking. Listening makes you humble and smart. When you listen you learn how things work, when you talk, you echo how you think things work.
    • If you must talk, ask questions. People learn more about you from your questions
    • If you actually know the answer in a business situation, talk, but back up with data.
  • Try to respond instantly. If you dont respond instantly, everything stalls.

Rules on Culture:

  • Avoid Hippos: (Highest paid person’s opinion)
  • You should not be able to read an org chart by looking at the product (for example, you cant look at the apple org chart when you look at the ipod).
  • Healthy orgs crush bureaucracy in all forms.
  • Ask for winning strategy and look for good tactics.
  • People are more productive when they are crowded. Social groups moderate bad behavior.
  • Empower the smallest of teams. Small teams accomplish more. Read the mythical man month. Create teams about the size of a family.
  • Working from home is cancer. Ban it.
  • Engineers and product managers add complexity, marketing adds management layers, sales adds coordinators.
  • Knights are knights and knaves are knaves. Tom Peters: There is no momentary lapse of integrety.
  • Focus on value rather than costs.
  • Never suggest copying a competitor. You can do better.
  • Hope is not a plan.
  • Success breeds the green eyed monster. Take away with surprise and humility.
  • Do all re-orgs in a day

Rules on Hiring:

  • Know how to interview well.
  • Gimmicks like free food, games, etc aren’t that important. People come to google to work with great people.
  • Managers don’t hire people. Committees hire people.
  • Promotions should be a peer review process.
  • Instead of laying off the bottom 10% dont hire anymore.
  • Dont hire specialists, esp in high tech. Change is the only thing permanent. “I have no special talent just passionate curiosity.” : Einstein.
  • You cannot teach passion. Nothing great was ever achieved without enthusiasm. If the enthusiasm isn’t palpable in a room, its not there.
  • Urgency of a role isnt important enough for a quick hire.
  • Identify and purge bad eggs.
  • Diversity is your best defense against myopia.
  • You cannot punt the management training program.
  • Life is not fair. Disproportionately reward risk takers and performers. Don’t tell people they did a good job if they didn’t. Real life is a meritocracy. Celebrate and reward what you want to see more of.
  • Build around the people who have the most impact.

Rules on Decision Making:

  • Decision making is about concesus and not unanimity. Dont spend hours towards unanimity. Good enough is better. Voltaire: “Perfect is enemy of the good”.
  • There is no consensus w/o dissent. Patton: If everyone is thinking alike then someone isnt thinking.
  • If there is doubt about what to do think abt customers perspective.
  • Choose your goals wisely. If the goals create conflict change the goals.
  • No-one of us is smarter than all of us.
  • Where there is harmony there is no innovation. Discuss and arguement leads new ideas and new meaning. Innovation comes from disagreement.

Rules on Fostering Innovation:

  • You cannot manage creativity to manage risks. Innovation comes from creativity.
  • Create a culture of yes based on optimism and big thinking.
  • Never stop someone from a good idea for a better one. Darwinian rule works. Best ideas win and others fail.
  • A leaders job is not to prevent risk but to recover from failures fast. Good failure happens quickly.
  • A good crisis is a terrible thing to waste. Nothing teaches like a crisis.

Rules on Humility

  • Learn something new to remember how hard its to learn. Teach something so you can learn.
  • Never stop learning. School is never out.
  • Humility is correlated with age. Arrogance is inversely correlated with age.
  • You get personal leverage from delegation and inspection. Smart people suorround themselves with smart people.
  • Judgement comes from experience and experience comes from errors. You learn more from your msitakes. At google screwups are written and archived to learn from for the future.
  • Smart ppl can smell hypocracy. Make sure you spend ur time on things you say is important. Culture is set from the top.
  • Don’t burn bridges.
  • Would you work for yourself?
  • Write a self review and be critical about yourself.
  • Communicate and confess when you make a mistake.

Final note:

Develop your own style.Styles that worked for others may not work for you (ends with the story of Cortez’s burning the boats, which may have worked for him, but Johnathan prefers the way of the Isreli tank commanders who lead by shouting “Follow me”)

Value vs. Revenue

I have been working for Comcast Interactive Media for over 5 years, thats a lot by some measures and barely anything by others. This is my first full time job and as those go, not particularly shabby. Off late I have been involved in more R&D/labs/prototypes kinds of projects ever since my move to the UX Technologist position. Sometimes I still struggle with the amorphous nature of this position. Things were a lot clearer when you executed on someone else’s plan. I am closer to product development now which is great and gives me a new look at how some of the things we build get concepted. And you have these interesting conversations around things like ROI on User Experience, the importance of time invested in design vs the need for getting features out of the door quickly, the love/hate feelings towards 300×250 ads of dancing polar bears on our sites. Its fascinating to see the camps erected on either sides of these debates.

Trying to pitch a concept to a friend of mine recently, our conversation moved towards revenue models. Usually these are answered by either “well, there is slot for an ad here…” or “We’ll increase user engagement by x “. However, I realized for that at the core, I didnt really care for the rev model immediately. I was really trying to work on something valuable. Something I can look back in my later years and be proud of, not because it would help my career but because I would have made a difference (in as significant a way as a web programmer can I guess) and people would use it.

The thought of course had been crystallized by the book I have been reading: Daniel H Pink’s “Drive”, I haven’t finished it yet, but its a great read so far on what motivates people to do what they do.

Looking back, I realize I have been making these distinctions for a long time, just not realizing it: why I was glad I didn’t write an app like “iFart” even if it made a lot of money and why working on projects that I perceive no value irritates me like crazy. “Drive” says this is a fundamental human characteristic. Nothing kills morale than knowing you are working towards revenue oriented goals and not value oriented ones. That morale is hard to substitute even by money. And you can get a lot of people working on something if you convince them of the value of that. Value and revenue don’t have to be in opposition. Something valuable can and will generate revenue, but the goal of your project cannot be to make something revenue-able, it has to be to make something valuable. If your goal is revenue targeted only, your success if any would be pretty small.

Apple lost someone today too, Director Jerome B. York, and they showed it by changing their entire homepage to pay respect to him:


Bet they lost some revenue on that one, but I bet it makes them a lot more valuable a company

Collected links from the whole Apple/HTC lawsuit thing

So, Apple sues HTC with over 20 patents, most of them idiotically trivial, including for example, Time-based, non-constant translation of user interface objects between states, which to the lay-developer should read “transitions”. Ah so Apple patented transitions in 2001, somebody better tell all the developers who are using things like jQuery effects, Tweenlite etc.

Steve Jobs says “We think competition is healthy, but competitors should create their own original technology, not steal ours.” , which is hilariously ironic when you see an earlier video with him saying how Apple has been shameless about stealing great ideas

This time even the die hard mac-heads call the craziness: Wil Shipley, Mac developer and author of the very successful Delicious Library application, writes an Open Letter to Steve Jobs on the whole thing.

Wait, what, you dont know who Wil Shipley is? Well he has a Wikipedia entry AND has starred in his own Penny Arcade strip. Yeah, so he’s kind of popular like that:

But I digress. Die hard mac fanboy John Gruber posts a good writeup on this whole mess, and more folk have started associating Apple with Evil, as evidenced by Paul Grahm’s comment on Hacker News and Tim Bray’s recent Tweet.

Meanwhile, Android Central posts a pretty awesome infographic on patent litigation in the mobile space

while someone tries to patent the whole practice of patent trolling. How meta…!

On Innovation in big companies

These notes are from BarcampPhilly ’08 that were lost in the depths of my computer’s hard disk and I only just found again. At the event, a few of us from CIM held a round table discussion on innovation, especially developer led innovation in big companies. Here are the points that people mentioned:

  • Innovation only comes from free cycles, even the best ideas need time to be thought through
  • Leverage something thats existing rather than a brand new idea
  • Agile development can actually hurt innovation with too much transparency. Its hard to justify something that you may feel has a long term win
  • Technical managers should whet the ideas of their team members
  • Innovation comes from bottom almost never from top
  • Innovation = Risk
  • True story: At AOL, the AOL IM service was declared a risk
  • Constraints actually push innovation
  • Innovation wont happen without proper incentives/rewards.
  • How about cross team innovation. How do designers/developers collaborate to innovate?

So this has only taken almost a year to put up, with BarcampPhilly ’09 coming up soon. Since then there has been a few opportunities within CIM to push bottom-up innovation, including a labweek for all of engineering.