Vendredi, c'était API Hours à la Cantine, avec notamment d'excellentes présentations de  Mehdi Medjaoui (Webshell) et d'Aurélien Fache (faberNovel) sur les API, hier, aujourd'hui, demain et après-demain.

Des API partout, dans votre lit, vos lunettes, votre coeur, voilà ce que nous promettait avec enthousiasme Aurélien Fache, en rappellant à quel point dans la création d'API aujourd'hui, c'est l'hybridité qui est de mise : mash-ups, croisement de données venant de sources hétéroclites… Un art combinatoire qui s'incarne dans des objets du quotidien, aujourd'hui Siri (un mélange de 40 API !), demain Google et son projet "Glass". Autant dire que le numérique d'aujourd'hui et de demain, via les API, est bien un "sacre de l'hybride" comme le suggère Milad Doueihi dans Pour un humanisme numérique.

Comment alors explorer toutes les possibilités qui s'offrent à nous ? Comment stimuler la création et proposer de nouveaux dispositifs, de nouveaux modèles d'usages ? Il faut peut-être pour cela repartir d'une donnée toute simple, le fait que créer une API, c'est d'abord écrire. Et écrire, c'est composer avec des ressources (l'alphabet latin pour le français, les données des services connectés pour des API) selon une règle commune (la grammaire française, les langages informatiques pour les API, dont le Json semble être aujourd'hui le principal). Or, de telles questions ont déjà été posées dans un passé pas si lointain…

En 1960, François le Lionnais, mathématicien et écrivain (c'est important…), fonde avec Raymond Queneau l'OuLiPo, l'Ouvroir de Littérature Potentielle, qui a pour vocation de stimuler la création littéraire par la contrainte arbitraire et bien souvent inspirée de règles mathématiques : combinatoire vertigineuse des Mille milliards de poèmes de Queneau, interdiction du "e" dans La Disparition de Perec, méthode "S + 7", poèmes booléens… tout y passe. Tout n'est pas devenu mémorable bien sûr, mais l'essentiel est là : un véritable foisonnement de textes inédits.

Pourquoi ce détour par l'OuLiPo ? Car je crois que le problème est fondamentalement similaire pour les API d’aujourd'hui (et de demain). Nous avons la chance de disposer d'un langage quasi-standard, d'avoir de plus en plus de données disponibles. Comme l'a montré Aurélien Fache, la logique combinatoire est une tendance de fond qui trouve dans les méthodes de l'OuLiPo un écho troublant. Pourquoi alors ne pas imaginer un OuApPo, un Ouvroir d'API Potentielles, qui s'inspire des démarches oulipiennes pour stimuler par la contrainte la création d'API ?

Bien sûr, il existe aujourd'hui de nombreuses manifestations qui ont un objectif similaire : hackaton, barcamp etc. L'idée est ici de proposer une nouvelle formule, qui prenne en compte la dimension combinatoire et systématique de l'écriture. A chacun de créer des règles et des contraintes, selon sa fantaisie, en essayant d'éprouver tout ce que les API peuvent permettre de créer.

Et pour finir, un petit rappel sur la population oulipienne. Les Roubaud, Le Lionnais, Perec ont été des transfuges, des hybrides pris entre littérature et mathématiques, entre logique et langage. Qui de mieux que des développeurs pour comprendre et approfondir cette démarche de création ?

80% of everything your organization does is a waste of time.

Don’t believe it? Most don’t, even when confronted with the following statistics:

• 1.3% of movies earn 80% of revenues
• the wealthiest 20% people globally own 82.7% of the world’s wealth
• about 20% of U.S. patients consume 80% of hospital resources
• 91.1% of plays on Last.fm are made up of 8.9% of bands
• 30.4% of websites receive 69.9% of hyperlinks
• 26.9% of email users send 73.1% of all emails
• 25% of Facebook users cover 75% of all friendship links
• 22.9% of users make up 77.1% of all “@” mentions on Twitter
  
  

It follows, and research has repeatedly shown, that in most complex systems about 20% of the inputs are responsible for about 80% of the outputs. In other words, 20% of the work done in your organization is responsible for 80% of the results, however they are defined.


In 1951, Joseph Moses Juran used the rule to start the Quality Revolution that swept the Japanese manufacturing machine into power, observing that over 80% of defects are caused by much fewer than 20% of causes.

In 1963, IBM discovered that 80% of a computer’s time is spent executing 20% of the code, using this insight to completely redesign its software and paving the way for its dominance of the industry.

The numbers may indicate something more like a 70/30 or 90/10 split, but the idea is the same: the most effective portion of the resources at our disposal is approximately 16 times more productive than the least effective, exerting a powerful influence that few people and companies take advantage of.

The difficult part, of course, is figuring out which assets are which. If you had a simple table ranking the productivity of your assets from least to greatest, it would be a simple task to focus your energy on the productive few and spend less time on the unproductive many.

Luckily, there are methods for extracting this information dynamically, without wasting time on compiling numerous data sets and correlating variables (effort which would, paradoxically, fall in the least productive 80%).

Let’s look at four specific ways to use these principles in the area of project management.

1. Simplify mercilessly

Simplicity is not just a design or user-experience imperative. It is fundamentally about productivity, since projects obey the Law of Organizational Complexity: as the number of project aims increases, the effort to accomplish them increases not in proportion, but geometrically.

There is no hard-and-fast rule for how many objectives to include, but for each one ask yourself: does adding this new objective add value in proportion to the time spent on it? If not, you’re probably already in the Land of Diminishing Returns. Stop and reconsider.

2. Impose constraints

As Eric Ries affirms in his book The Lean Startup, “having too much money is just as big of a problem as having too little.” This is not hyperbole. In a world of technology that seeks to remove as many constraints as possible, these limitations enable us to focus on the most super-productive tasks.

Whether you use tight deadlines, short product lifecycles, or stretch targets, the important thing to remember is that constraints shouldn’t be used to squeeze one more drop of work out of your employees, or impose unsustainable work schedules. Think of it as the “gamification of project management”: you are working together to dynamically discover the highest value gems in the vast landscape of priorities.

3. Plan for optimization

Although the current trend is to spend less time planning and more time iterating, planning plays a big role in harnessing the 80/20 Rule. And if you think about it in the terms we have discussed, it makes sense. As long as you are confident that you are optimizing your work at the 20% inputs level, you can spend up to 80% of the total project time in planning while maintaining most of the result. Of course, in practice you won’t need nearly 80% of the time for planning, and any amount less than this is time saved or quality improved.

4. Design, design, design

It has become almost a cliché, but Design Thinking is one of the clearest examples of 80/20 logic. It is well-recognized that mistakes in design result in the largest number of defects, contribute the most to cost overruns, and perhaps most seriously, are most detrimental to customer satisfaction.

It helps in the design process to think of each small improvement not as an inconsequential detail, but as a super-productive, factor-of-16 optimization. Ask yourself: how many such super-optimizations are worth looking for in the design phase? The answer is probably “quite a few.”

It’s important to understand, however, and perhaps this will give you new impetus to reboot the search for productivity in your organization, that these recommendations are not based on fuzzy notions of employee empowerment or me-too stories of past successes.

The most advanced research on chaos theory and complex systems affirms the essential truth of the 80/20 Rule, and challenges us to discover the hidden dynamics behind the projects we put so much work into.

Before our annual analysis of the strategy of a tech giant, we wanted to share with you the reason why we're so enthusiastic about the Russian technology market.

Looking at the main sources of global news, you won't find many positive article on Russia: just check out Wall Street Journal or New York Times and you will get an idea.

But that didn't prevent us from looking behind the curtain... and what we found was surprising. Not only surprising in a positive manner, but also in a very promising way: there is a very fast expanding digital market in Russia, that not only is catching up on western level, but is also extremely innovative.

When we found out, we decided that we had to be part of it and we opened an office in Moscow. Our Russian office is now helping companies bridge the gap between them and the Russian market.

Today, we are releasing an extensive market study that will show why we are committed to Russia, and why you should have a different look on the Russian market.

Read the Study




Download this free presentation

You can get this 103-slide PDF presentation just for sharing!

Don’t be afraid to click, you can change what the tweet says. If you don’t have a Twitter account, or if you don’t want to share it, you can still download it on slideshare.

Key ideas

Why should you read this presentation?

- You still think that the Russian market is underdeveloped and that Russian startups are only copycats
- You are hesitating on expanding your business to Russia
- You have heard interesting stories about the Russian digital market, and you want to learn more
- You are Russian and you think that no Westerner understands what is going on there

At the end of the presentation, we give you 3 reasons why you should expand to Russia.

Did you know?

- Russia accounts for the largest pool of internet users in Europe, before Germany.
- Russians use more mobile payment solutions than UK citizens.
- Online advertising has already overtaken print advertising in Russia, a year before the US.

Debunking prejudices

You will dive into the dynamism of the Russian digital market and will discover that

- This dynamism is not superficial: it is based on a deeply-rooted culture of technology and science.
- Even if the overall market is still emerging, it has strong pillars to stand on: role-models such as Yandex or Mail.ru, and a diversified network of investors.
- The ecosystem is boiling, with many successful adaptations of western best-practices and a lot of local innovations.

And finally we will share with you our vision of the Russian technology specialties.

The tech world is abuzz with Facebook’s recent acquisition of Instagram for $1B. The debate on why Facebook paid such a hefty price rages online -- everywhere from the NYT to Wired.

Let’s summarize the product Facebook bought:

• Instagram is a dead simple, lightweight photo sharing app that launched in Oct 2010
• They have raised 57.5M in funding from big names like Sequoia to Jack Dorsey (cofounder of Twitter & Square)
• 35M users (they racked up over a million of Android users within 12hrs of their Android app launch!)
• “Pre-revenue” i.e. Instagram doesn’t make any money

Speculated reasons for the price-tag range from competitive play to renewed talks of a bubble.

Cost per User

Source : Wired

Facebook paid ~$28 for each user, which isn’t outrageous compared to many other acquisitions. The real question is how much these users are worth. If you consider that many of these users are already on Facebook and that they don’t generate any revenue, it’s hard to say quantify their actual value.

                                                              Cost per Employee

But maybe we can take Facebook CEO, Mark Zuckerburg at his word: “We don’t plan on doing many more of these, if any at all. But providing the best photo sharing experience is one reason why so many people love Facebook and we knew it would be worth bringing these two companies together.” This reasoning supports the arguments made in The Facebook Effect, which suggests that photo-sharing and tagging were crucial in Facebook’s viral growth.

If controlling the photo sharing world was their motive, the move is simultaneously offensive and defensive. Offensive in that it allows Facebook to improve its photo capabilities (don’t be surprised to see Instagram’s iconic filters made available for our FB photos). It’s also defensive because it prevents Google+ or Twitter from competing with Facebook’s position as the dominant social network for photo sharing.

At the end of the day, we can’t know why Facebook bought Instagram; we can merely speculate. Nonetheless, watching how this will play out will definitely be interesting.