Tech/Art

"I believe that love is a better teacher than a sense of duty"

Sun, 15 Jan 2012 19:43:33 +0100

“I believe that love is a better teacher than a sense of duty”

- Albert Einstein about learning the violin after listening to Mozart’s sonatas

Another very cute tiny story

Wed, 04 Jan 2012 05:53:50 +0100

Another very cute tiny story

"The older I get, the more I see how much motivations matter. The Zune was crappy because the people..."

Sun, 01 Jan 2012 21:39:33 +0100

“The older I get, the more I see how much motivations matter. The Zune was crappy because the people at Microsoft don’t really live musuc or art the way we do.”

- Steve Jobs

Great paragraph by Jobs about unknown similarities between tech...

Sun, 01 Jan 2012 20:38:38 +0100

Great paragraph by Jobs about unknown similarities between tech and art worlds

laughingsquid: The Most Important Events of the Last 100 Years...

Sun, 01 Jan 2012 03:23:01 +0100

laughingsquid:

The Most Important Events of the Last 100 Years in 10 Minutes

Pretty intense video. A bit depressing because it focuses mostly on negative events but feels like a rollercoaster!

"My intuition told me that joining Apple would be a once-in-a-lifetime opportunity to work for a..."

Sat, 31 Dec 2011 05:21:36 +0100

“My intuition told me that joining Apple would be a once-in-a-lifetime
opportunity to work for a creative genius. Engineers are taught to make a decision analytically, but there are times when relying on gut or intuition is most indispensable”

- Tim Cook, Apple’s COO about its initial itw with Jobs

"Jobs replied that if he weren’t working with computers, he could see himself as a poet in..."

Mon, 26 Dec 2011 02:49:10 +0100

“Jobs replied that if he weren’t working with computers, he could see himself as a poet in Paris.”

- Walter Isaacson in Steve Jobs, discussion between Jobs and Sculley

One day before breakfast, an orange rolled off the counter and...

Sun, 25 Dec 2011 17:21:23 +0100

One day before breakfast, an orange rolled off the counter and escaped its fate, bounding happily through the kitchen door.

Filled with hope,
the egg followed.

"It’s better to be a pirate than to join the navy"

Sun, 25 Dec 2011 02:56:54 +0100

“It’s better to be a pirate than to join the navy”

- Steve Jobs to the Macintosh team

"The journey is the reward"

Sun, 25 Dec 2011 02:55:53 +0100

“The journey is the reward”

- Steve Jobs to the Macintosh team

"where you acquiesce to the world around you and you conform, or you sort of defiantly break whatever..."

Sun, 18 Dec 2011 17:28:36 +0100

“where you acquiesce to the world around you and you conform, or you sort of defiantly break whatever remaining bonds connect you to that world and create for yourself a different set of values.”

- Sean Parker

My first cover published on YouTube! Great song from the great...

Wed, 30 Nov 2011 17:31:21 +0100

My first cover published on YouTube! Great song from the great movie Crazy Heart. Hope I’m not killing it :p

The US struggle to get more science and engineering graduates

Sun, 06 Nov 2011 09:39:10 +0100

The US struggle to get more science and engineering graduates:

Good article explaining the current shortage of students majoring in science and engineering in the US (it might actually be the same in Europe). The reporter points out that some students do start college with an engineering degree in mind but change major because it’s too hard and theoretical.

I do agree with the need to design science and engineering curriculum better. Take maths for example. Loved it in the beginning and then got tired because I couldn’t really see where Calculus and Probability Theory were going to take me. Now that I’m starting to study Machine Learning, one of the most exciting subset of Computer Science (it’s what spam filtering or self-driving cars are based on), I realize that when applied, Math is one of the coolest and most powerful tool and I’m dying to go back to my old Statistics classes to understand better what’s happening. My point is: break up science classes between theory and application from the very beginning and you will deter more students from changing major.

You could argue more or less the same thing for engineering. I’ll take an example from CS because that’s what I know but I’m sure you could find similar cases in Mechanical or Electrical Engineering. The introductory CS class in every college should be pure FUUUUN. At Stanford’s CS106A class, most assignments are about making cool games and guess what: the class has more than 600 hundred students enrolled this quarter, biggest class ever taught at Stanford! I know some schools in France where still recently, the first programming class was in C, which is one of the hardest language. No wonder students aren’t into programming after that!

"I always thought of myself as a humanities person as a kid, but I liked electronics,” he said. “Then..."

Fri, 04 Nov 2011 09:48:40 +0100

““I always thought of myself as a humanities person as a kid, but I liked electronics,” he said. “Then I read something that one of my heroes, Edwin Land of Polaroid, said about the importance of people who could stand at the intersection of humanities and sciences, and I decided that’s what I wanted to do.””

- Steve Jobs

Good set of values to apply to your company! Thanks @randhindi

Fri, 04 Nov 2011 06:01:37 +0100

Good set of values to apply to your company!

Thanks @randhindi

"I tell my kids, what is the difference between a hero and a coward? What is the difference between..."

Thu, 27 Oct 2011 22:27:43 +0200

“I tell my kids, what is the difference between a hero and a coward? What is the difference between being yellow and being brave? No difference. Only what you do. They both feel the same. They both fear dying and getting hurt. The man who is yellow refuses to face up to what he’s got to face. The hero is more disciplined and he fights those feelings off and he does what he has to do. But they both feel the same, the hero and the coward. People who watch you judge you on what you do, not how you feel.”

- Cus D’amato, legendary boxing trainer

Eric Ries' talk at Stanford

Thu, 27 Oct 2011 20:17:49 +0200

Eric Ries' talk at Stanford:

I had seen the name of Eric Ries in lots of places before. Basically everywhere the Lean Startup principles are mentioned. But I had never taken the time to listen to him personally. This talk is a good one-hour wrap-up of the key concepts he’s trying to communicate.

Thanks @feross

This video is a few years old and already very famous (a now...

Thu, 07 Jul 2011 20:26:04 +0200

This video is a few years old and already very famous (a now passed-away Carnegie Mellon CS professor gave a 1-hour talk on how to lead your life). But the counter says 14M views which means there are still LOTS of people who have never seen it. This needs to be fixed ! Believe me, it will change your life !!!!

My Spotify playlists

Thu, 23 Jun 2011 10:37:32 +0200

Hi everyone !

Hadn’t blogged in a while…I’ve been playing a playlist game with some friends lately : every participant had to publish a 10-song PL every week. At the end of each week, a vote would determine the three best !

I’m gonna be honest, my playlists weren’t so successful in the game but I’m still posting them here as they might be appreciated by some. It’s mostly easy-listening, electro, rock. You can add me on Spotify, I’m johanisma and you’ll also get access to my friends’ playlist, some of them are real gems !

WUPC #09 (soundtracks only)

Enjoy and send feedback !

Pair programming is not the best way to learn from a pet project

Sun, 06 Feb 2011 23:24:58 +0100

Now close to finalizing my first iPhone app, it’s a good time to look back on my first experience developing a pet project and share the mistakes I made in a series of posts. This first one will be about pair programming.

Hypothesis : pair programming is the best development technique to maximize technical learning from a pet project.

As I started programming only recently - which will be the subject of another post - I wanted to take advantage of that first pet project to learn as much as possible about the iOS SDK, Rails that we are using for the back-end, etc etc. So along with my co-developer, we decided to pair program, ie code everything on one screen so both of us would be involved in every step of the coding.

Facts : IT’S NOT !

After a while, I realized it was a bad idea. I believe dividing up tasks is a more efficient way to learn. Here is why :

1)Higher motivation

Two factors lead to higher motivation when adopting solo programming : faster execution and sane competition between my co-developer and I.

When the team is small, pair programming inevitably leads to sluggish execution which in turn hurts morale. The faster you execute, the more you’re motivated as you see your product with new features and getting closer to release !

Moreover, there is a virtuous between you and your partner : throughout the development, you see him nicely coding up new features and you feel like you have to at least do as good of a job as his on your tasks. This sane competition is a great source of motivation to efficiently write proper code.

Motivation is the most important thing you need if you want to improve your programming skills through a pet project. Without it, you easily get discouraged when facing problems which inevitably come up all the time when you’re a beginner ! If you give up, obviously no more learning !

Dividing up tasks sets up an appropriate development environment reinforcing your motivation at every step. Releasing your first product can be a long process where discourage is the most dangerous enemy. Fighting it will be a major step towards release…and learning !

2)Better self-reliance

Pair programming can be great in certain contexts in order to produce proper code. However, when you’re almost new to programming, it’s important to learn to solve problems by yourself without expecting the solution to come from someone else. In other words, if you have a good co-developer, pair programming can allow you to be sometimes passive which is definitely not a good way to learn.

When you begin, I think you should focus on “vertical learning” versus “horizontal learning”. Pair programming corresponds to the latter one : you’re involved in all the problems but you always share the responsibility of solving them. In “vertical learning”, you solve the problems you’re facing by yourself which requires a complete understanding of the situation.

So granted, I didn’t learn much about the tasks that were accomplished by my co-developer, for example the subtleties of using UITableViews, but that’s clearly counter-balanced by the obvious advantages of splitting up tasks.

Conclusion

Of course, pair programming can be really the appropriate technique in certain situations. I’m just saying that when your goal is to maximize technical learning from a pet project, it is not the best strategy.

The technique we ended up adopting was kind of a mix : solo programming most of the time and pair programming when it came to the most important and difficult features of our app.

What do you think is the most appropriate development technique for a pet project when you are not very experienced ?