Wavyx's shared items in Google Reader

The Rands Test

2011-10-11T17:17:24Z

It's hard to pick a single best work by Joel Spolsky, but if I was forced to, I'd pick The Joel Test. It's his own, highly irresponsible, sloppy test to rate the quality of software, and when anyone asks me what is wrong with their team I usually start by pointing the questioner at the test. Start here.

It's a test with 12 points and as Joel says, "A score of 12 is perfect, 11 is tolerable, but a 10 or lower and you've got serious problems". More important than the points, his test clearly documents what I consider to be healthy aspects of an engineering team, but there are other points to be made. So it is completely an homage to Joel that I offer The Rands Test.

I was employee #20 at the first start-up and the first engineering lead. Over the course of two years, the team and the company exploded to close to 200 employees. This is when I discovered that growing rapidly teaches you one thing well: how communication continually finds new and interesting ways to break down. The core issue being the folks who've been around longer who also tend to have more responsibility. As far as they're concerned, the ways they organically communicated before will remain as efficient and simple each time the group doubles in size.

They don't. A growing group needs to continually invest in new ways to figure out what it is collectively thinking so anyone anywhere can answer the question: "What the hell is going on?" This is the first question The Rands Test answers. As I'll explain shortly, the second question The Rands Test helps you answer is selfish. The second asks: "Where am I?"

12 Points

Let's start with bare bones versions of the questions and then I'll explain each one.

Do you have a 1:1?
Do you have a team meeting?
Do you have status reports?
Can you say No to your boss?
Can you explain the strategy of the company to a stranger?
Can you explain the current state of business?
Does the guy/gal in charge regularly stand up in front of everyone and tell you what he/she is thinking? Are you buying it?
Do you know what you want to do next? Does your boss?
Do you have time to be strategic?
Are you actively killing the Grapevine?

(Note: While I'll explain each point from the perspective of a leader or manager, these questions and their explanations apply equally to individuals.)

Do you have a consistent 1:1 where you talk about topics other than status? (+1)

I think you'd be hard pressed to find anyone who would suggest 1:1s are a bad idea, but the 1:1 is usually the first meeting that gets rescheduled when it hits the fan. I'm of the opinion that when it hits the fan, the last thing you want to do is reschedule 1:1 time with the folks who are likely either responsible for it hitting the fan and/or are the most qualified to figure out how to prevent future fan hittage.

Furthermore, as I wrote about in The Update, The Vent, and The Disaster, conveyance of status is not the point of a 1:1; the point is to have a conversation about something of substance. Status can be an introduction, status can frame the conversation, but status is not the point. A healthy 1:1 needs to be strategic, not a rehashing of tactics and status that can easily be found elsewhere.

A 1:1 is a weekly investment in the individuals that make up your team. If you're irregularly doing 1:1s or not making them valuable conversations, all you're doing is reinforcing the myth that managers are out of touch.

Do you have a consistent team meeting? (+1)

The team meeting has all the requirements of the 1:1 -- consistency and a focus on topics of substance -- but don't give yourself a point just yet.

Status does have a bigger role in a team meeting. As we'll talk about shortly, the Grapevine is a powerful beast and a team meeting is a chance to kill it. I have a standing agenda item for all team meetings that reads "gossip, rumors, and lies" and when we hit that agenda item, it's a chance for everyone on the team to figure out what is the truth and what is a lie.

After that's done, my next measure of a team meeting is: did we make tangible progress on something? I don't know what you build, so I don't know what's broken on your team, but I do know that something is broken and a team meeting is a great place to not only identify the brokenness, but also to start to discuss how to fix it.

If you're killing lies and fixing what's broken in a team meeting, give yourself a point.

Are handwritten status reports delivered weekly via email? (-1)

If so, you lose a point. This checklist is partly about evaluating how information moves around the company and this item is the second one that can actually remove points from your score. Why do I hate status so much? I don't hate status; I hate status reports.

My belief is that email-based status reports are one of the clearest and best signs of managerial incompetence and laziness. There are always compelling reasons why you need to generate these weekly emails. We're big enough that we need to cross-pollinate. It's just 15 minutes of your time.

Bullshit. The presence of rigid, email-based status reports comes down to control, a lack of imagination, and a lack of trust in the organization.

I want you to count the number of collaboration tools you use on a daily basis to do your job -- not including email. If you're a software engineer, I'm guessing it's a combination of version control, bug tracking, wikis, CRM, and/or project management software. All of these tools already automatically generate a significant amount of status regarding what has tactically gone down each week.

When someone asks for a status report, my first thought is: "I'm already generating piles of status on these various tools, why not just ask those?"

Well, there's a lot of noise in those tools. Well, write a report that takes out the noise -- collaboration tools are built around reporting. The status information is out there. In what managerial textbook does it say it's a good idea to "Distribute the task of figuring out what is going on to the people who are performing the work?" That's, like, your job.

Well, what I really want is your high level assessment of the week. Three things that are working, three things that aren't, and what we're going to do about it. Ok, now we're talking. I can do a strategic assessment of the week, but why don't we just put that at the beginning of the 1:1? That way when you have questions (and you will), we can have a big fat debate.

But I'd like to have a record I can review later. Super, feel free to write down anything we talk about.

Yes, status reports are a hot button for me. I've written hundreds of them and each time I've begun one, I start by thinking, "Why in the world do I feel like I'm performing an unnecessary act?" Status reports usually show up because a distant executive feels out of touch with part of his or her organization and they believe getting everyone to efficiently document their week is going to help. It doesn't. Emailed status reports say one thing to 90% of the people who wrote them: "You don't value my time". This leads us to our next point...

Are you comfortable saying NO to your boss? (+1)

Perhaps a better way to phrase this point is: do you feel your 1:1 with your boss is somehow different than every other meeting you have during the week? Part of healthy communication structure is when information moves easily around the team, organization, and company, and if you walk into a meeting with your boss always on your best behavior and unwilling to speak your mind I say something is broken.

Yes, he or she is your boss and that means they write your annual review and can affect the trajectory of your career, but when they open their mouth and say something truly and legitimately stupid, your contractual obligation as a shareholder of the company is to raise your hand and say, "That's stupid. Here's why..."

Easier said than done, Rands.

Ok, don't say it's stupid.

Here's the deal. I believe that leaders who think they're infallible slowly go insane with power created by the lie that being wrong is a sign of weakness. I screw up -- likely regularly -- and I've been doing various forms of this gig for twenty years. While it still stings when I stumble upon or others point out my screw-ups, I'd sooner I admit I fucked up, because then I can figure out what I really did wrong faster, and that starts with someone saying "No".

Can you explain the strategy of your company to a stranger? (+1)

Moving away from communications, this point is about strategy and context. If I was to walk up to you in a bar and ask what your company did, could you easily and clearly explain the strategy?

This is the first point that demonstrates whether you have a clear map of the company in your head, and you might be underestimating the value of this map. If you're a leadership type, chances are you can draw this map easily. If you're an individual, you might think this map is someone else's responsibility and you'd be partially correct: it is someone else's job to define the map, but it's entirely your responsibility to understand it so you can measure it.

As we'll see with the following questions, The Rands Test isn't just about understanding communications, it's about understanding context and strategy. How do you think the employees of HP and Netflix feel given the strategy flip-flops over the past few months? Safe or suspicious? Let's keep going...

Can you tell me with some accuracy the state of the business? (Or could you go to someone / somewhere and figure it out right now?) (+1)

It's a brutal exaggeration, but I think you should independently judge your company the same way that Wall Street does: your company is either growing or dying. Have you ever watched the stock price of a publicly traded company the day after they announce that they are going to miss their earnings numbers? More often than not, no matter what spin the executives have, the stock is hammered. It's irrational, but what I infer when I see this happen is that Wall Street believes the company has begun a death cycle. If the executives can't successfully predict the state of their business, something is wrong.

I realize this isn't fair and there are myriad factors that contribute to the health of the business every single day, and I encourage you to research and understand as many of those as possible. But when you're done, I'd also like you to have a defensible opinion regarding the state of the business, or at least a set of others whose opinion you trust.

This is a picture that you are constantly building, and this is an easier task if you've given yourself a point on the prior question regarding company strategy. If you have a map of what the company intends to do, it's easier to understand whether or not it's doing it. This leads us to...

Is there a regular meeting where the guy/gal in charge gets up in front of everyone and tells you what he/she is thinking? (+1) And are you buying it? (+1)

Our last point regarding context involves the person in charge. In rapidly growing teams and companies there's a lot going on -- every single day. When the team was small, the distribution of information was easy and low cost because everyone was within shouting distance. At size, this communication becomes more costly at the edges. Directors, leads, and managers -- these folks tend to stay close to current events because it's increasingly their job, but it's also their job to take steps to keep the information flowing, and it starts with the CEO.

On a regular basis, does your CEO stand up and give you his impressions of what the hell is going on? Whether it's 10 or 10,000 of you, this is an essential meeting that:

Gives everyone access to the CEO.
Allows him/her to explain their vision for the company.
Hopefully allows anyone to stand up and ask a question.

If the value of this meeting isn't immediately obvious to you, I'd suggest that you are one of those lucky people who already has a good map of the company as well as a sense of the state of the business. That's awesome -- here's a bonus point for you: does the CEO's version of the truth match yours or is he/she in a high Earth orbit with little clue what is actually going on? Give yourself a point if it's the former and if it's the latter, what does that say about the state of the business? Growing or dying?

Can you explain your career trajectory? (+1) [Bonus: Can your boss? (+1)]

Next, switching gears a bit, give yourself a point if you -- right this very moment -- can tell me your next move. You're already doing something, so explain what you're going to do next. It's a simple statement, not a grand plan. One day, I'd like to lead a team.

Part of a healthy organization isn't just that information is freely moving around; it's what the folks receiving and retransmitting it are doing with it. You're going to mentally file and ignore a majority of this information, but every so often a piece of information will come up in a 1:1, a meeting, or a random hallway conversation, and it will be strategically immediately useful for you to know what you want to do next.

Angela got a promotion and her team is great and I've always wanted to be a manager.
Jan just opened a requisition and his group is working on technology I need to learn.
They fired Frank. That creates a very interesting power vacuum...

You can argue that even without a plan you'd make the same opportunistic leap, but I've found that having a map is usually a better way of getting to a destination.

There's a bonus point here as well. Does your boss know what you want to do next? He or she likely has even more access to the information moving around the company, and whether they like it or not, have equal responsibility to figure out how to get you from here to there.

Do you have well-defined and protected time to be strategic? (+1)

If you gave yourself two points on the prior question, congratulations, I think you're in better shape than most, but there's one more point. Are you making progress towards this goal? Can you point to time on your calendar or even just in your head where you are growing towards your goal?

I like being busy. Like really busy. Like getting in, grabbing a cup of coffee, and suddenly finding the coffee is cold, it's 6pm, and I forgot to eat busy. Busy feels great, but busy is usually tactical and not strategic. This is why I'm constantly maintaining my Trickle List -- it's my daily reminder of doing work that is larger than right now.

If you have time where you're investing in yourself while you're at work and your boss is cool with it -- give yourself a point.

Are you actively killing the Grapevine? (+1)

When Grace walks in your office, you know she knows something by the look on her face. She moves to the corner of the office and starts with, "Did you hear...?" and the story continues. It's a doozy, full of corporate and political intrigue, resulting in your inevitable response: "No. Way."

Being part of a secret feels powerful. In a moment the organization reveals a previously hidden part of itself, and in that moment you feel you can see more of the game board. So, that's why they fired him. I was wondering. Grace finishes with the familiar, "Don't tell anyone," which is ironic since that's precisely what was asked of her 15 minutes ago.

There is absolutely no way you're going to prevent folks from randomly talking to each other about every bright and shiny thing that's going on in your company. In fact, you want to encourage it. 1:1s and meetings are only going to get you so far. The thing you can change is the quality of the information that's wandering the company.

In the absence of information, people make shit up. Worse, if they at all feel threatened they make shit up that amplifies their worst fears. This is where those absolutely crazy rumors come from. See, Kristof is worried about losing his job so he's making up crazy conspiracy theories that explain why THE MAN IS OUT TO GET HIM.

Without active prevention, the Grapevine can be stronger than any individual. While you can't kill the Grapevine, you can dubiously stare at it when it shows up on your doorstep and simply ask the person delivering it, "Do you actually believe this nonsense? Do you believe the person who fed you this trash?" Rumors hate to justify themselves, so give yourself a point if you make it a point to kill gossip.

Magnitude and Direction

There is a higher order goal at the intersection of the two questions The Rands Test intends to answer: Where am I? and What the hell is going on? While understanding the answers to these questions will give you a good idea about the communication health of your company, the higher order goal is selfish. I'll explain.

I think of the two lines of questions as a vector. A simple vector can be drawn as two points connected by an arrow, but a vector is far more interesting. It's a geometric object that describes both direction and magnitude. Understanding how information moves, how you communicate with your boss, and being able to describe both your career strategy and that of your company sketches a vector in your head. The first point is you at this very moment and the other point is where you want to be. The distance and direction between the two start to explain how you're going to get there. I love vectors because they draw a picture about a complex problem and I hope as you were answering the questions above this mental picture began to appear in your head.

Like the Joel Test, the point of the Rands Test is not the absolute score, but the score is good directional information. If you got a 12, I'd say you're in a rare group of people who have a clear picture of their company and where they fit in. Between 8 and 10, you are likely troublingly deficient in either communications, strategy, or your development - it depends where the points are missing. Less than 8 and I think you've got a couple of problems.

There are a lot of different scenarios I expect folks to find themselves in as they explore these questions, which is why it's tricky to proscribe specific action. Your company may be doing well, but you may be unhappy and have no clue what you want to do next. You might love your job, but have no idea whether the company is actually growing. Your course is dependent on what you care about and the Rands Test points out good places to start.

Applications mobiles multi-plateformes: les approches PhoneGap et Titanium Mobile

2011-10-17T05:10:28Z

Introduction

Le développement d’applications pour terminaux mobiles (iPhone, iPad, Android, Blackberry, Windows Phone, Nokia Symbian, Samsung Bada…) se heurte à la fragmentation des technologies de développements: environnement iOS/Objective-C pour l’iPhone et l’iPad, SDK Java spécifique pour Android, J2ME pour Symbian, etc.

Deux approches possibles lorsque l’on débute un projet d’application ciblant plusieurs de ces plateformes sont de développer une application pour chacune d’elle, ou de développer un site Web compatible.

Dans le premier cas, l’inconvénient concerne bien évidemment le coût des développements. Dans le deuxième, on sera limité en richesse de l’application par les possibilités du Web.

Ce fut l’objet d’un précédent article sur notre blog: http://blog.octo.com/debat-web-apps-vs-natif/.

Entre ces deux approches se situe une offre assez fournie de solutions de développement multi-plateforme, proposées par des éditeurs proposant leurs propres plateformes d’exécution et leurs outils de développement.

Parmi celles-ci, nous nous sommes concentrés dans cet article sur PhoneGap et Titanium Mobile, qui sont aujourd’hui parmi les plus abouties et sont représentatives des deux principales approches de développement multi-plateforme: l’utilisation des moteurs de rendus Web pour PhoneGap, et la translation de code source vers la plateforme cible pour Titanium.

Présentation des solutions

PhoneGap

Nitobi, l’éditeur de PhoneGap, a annoncé son rachat par Adobe le 3 octobre 2011. Cela confirme l’intérêt que l’on peut avoir envers cette solution, même si ce rachat surprise apporte une certaine incertitude sur l’avenir de PhoneGap.
PhoneGap est un outil open-source, et devrait d’ailleurs rejoindre la fondation Apache.
Le principe de PhoneGap est de fournir des API JavaScript aux navigateurs Web standards, permettant d’appeler des fonctionnalités natives non disponibles autrement: accéder à l’appareil photo, à l’accéléromètre, au système de fichiers…
Cela nécessite d’embarquer le code source HTML/CSS/JS dans une application native, grâce au composant permettant d’inclure une vue Web dans une application, disponible dans chaque SDK.

Titanium

Développé par Appcelerator qui vend du support et des formations sur Titanium, c’est également une solution libre.

Le principe de Titanium est de fournir une machine virtuelle JavaScript permettant d’accéder au système natif, et ainsi de développer des applications natives mais en JavaScript. C’est la promesse d’applications plus réactives et à l’expérience utilisateur plus proche du natif que celles basées sur des pages Web, comme PhoneGap.

Points communs

L’intérêt principal de l’utilisation d’une plateforme de développement multi-plateforme comme PhoneGap et Titanium Mobile réside en 2 points:

pouvoir déployer l’application sur les magasins (AppStore, Android Market…) afin de bénéficier de ce canal de distribution et de communication. Cela est réellement assuré par ces deux outils;
réduire les coûts de développements en mutualisant du code. Cela est assuré par ces deux outils à condition de bien les utiliser et de limiter la part des développements spécifiques à chaque plateforme et donc de se limiter aux fonctionnalités supportées « out of the box ». On rogne donc sur l’expérience utilisateur et la richesse fonctionnelle.

Comparaison

Plateformes supportées

PhoneGap	Titanium
iOS, Android, BlackBerry, Bada, WindowsPhone, WebOS, Symbian	iOS, Android et depuis très récemment BlackBerry

Notre avis:

En pratique, les développements PhoneGap nécessitent des adaptations pour chaque plateforme, dont les capacités et l’expérience utilisateurs diffèrent fortement. Etant donné l’état du marché (avec l’écrasante domination de l’iPhone, d’Android et de Blackberry), on ciblera généralement ces 3 plateformes uniquement. C’est ce que l’on a pu constater dans la plupart des cas où une entreprise ciblait initialement toutes les plateformes: l’augmentation du coût pour les supporter au fur et à mesure rendait cela impossible et finalement, au maximum, seules les 3 ou 4 principales étaient vraiment adressées.
Concernant Titanium, le support de Blackberry est encore récent et disponible uniquement sous Windows, alors que la plupart des développeurs mobiles sont sous Mac (pour l’iPhone). Cela est dommage mais on espère que ce sera bientôt résolu.
Pour cibler tous les terminaux mobiles à moindre coût, le meilleur moyen reste de réaliser un site Web simple, qui permet de diffuser de l’information mais ne permet ni de fournir des fonctionnalités avancées (appareil photo, etc.), ni de proposer une expérience utilisateur riche, ni d’être visible sur l’AppStore.

Richesse de la plateforme

PhoneGap	Titanium
Possibilités des navigateurs Web + API PhoneGap (principalement appareil photo, système de fichier, accéléromètre, liste des contacts, géolocalisation)	Accès aux composants d’IHM natifs + API Titanium (base de données, géolocalisation, gestion des contacts, intégration Facebook, appareil photo, lecture et enregistrement audio/vidéo…)

Notre avis:

Titanium est indéniablement plus riche fonctionnellement et fournira une apparence plus proche du natif, ce qui est en général l’objectif des concepteurs d’applications.
PhoneGap est dans l’absolu plus limité en terme de fonctionnalités, et on doit concevoir les écrans comme des pages Web et non des écrans natifs. C’est d’ailleurs une erreur que l’on retrouve souvent sur les projets mobiles utilisant des technologies Web: il ne faut surtout pas tenter de reproduire des comportements natifs, mais plutôt viser des comportements Web performants.
Les deux plateformes sont extensibles, Titanium plus facilement que PhoneGap, mais nous déconseillons globalement d’aller dans cette direction car cela devient très coûteux à maintenir (on estime en général à 20% la part maximale de code spécifique tolérable), et nécessite en outre encore plus de compétences que de faire des applications natives spécifiques à chaque plateforme.

Plateforme de développement

PhoneGap	Titanium
HTML/CSS/JavaScript standard. On développe comme si c’était une application Web normale. Les API PhoneGap sont bien documentées, par-contre on devra chercher sur le Web la documentation pour les technologies HTML/CSS/JS et les éventuelles librairies JavaScript utilisées.	JavaScript en utilisant des API spécifiques à Titanium, que ce soit pour créer les IHM ou pour écrire des appels au système. Le site d’Appcelerator propose la documentation des API ainsi qu’une application démo (KitchenSink) et des vidéos.

Notre avis:

On remarque l’utilisation de plus en plus large de JavaScript, un langage décrié par ses côtés peu structuré et trop tolérant. Il est nécessaire, pour développer ce type d’applications, de se former profondément à JavaScript et de connaître les design patterns de structuration/modularisation du code dans ce langage, qui existent mais sont souvent mal connus des développeurs.
C’est d’ailleurs le principal objet de la formation à Titanium donnée par leur expert Kevin Whinnery, que nous avons eu la chance de suivre à l’UpdateConf 2011.
De manière globale, l’environnement de développement de Titanium est mieux intégré et plus documenté.

Productivité des développement

PhoneGap	Titanium
PhoneGap nécessite de développer pour chaque plateforme ciblée dans l’IDE correspondant, et ne fournit pas d’outils supplémentaires.	Titanium fourni un IDE dédié, basé sur Aptana, un des meilleurs IDE Web reposant lui-même sur Eclipse (acheté récemment par Appcelerator)

Notre avis:

Titanium fournit un environnement de développement « clé en main » et performant, là où PhoneGap nécessite d’installer et configurer soi-même plusieurs environnements.
La productivité est clairement côté Titanium aujourd’hui.

Gestion des déploiements

PhoneGap	Titanium
PhoneGap propose depuis peu un outil de build sur le cloud: PhoneGap Build, qui compile un projet PhoneGap vers les différentes plateformes cibles, que vous pourrez ensuite publier sur l’AppStore, l’Android Market, etc.	Titanium ne fournit pas d’outil de packaging multi-plateforme. Vous devrez donc créer manuellement les packages applicatifs.

Notre avis:

L’automatisation du packaging et de la livraison des applications sur les différentes plateformes (magasins publics type AppStore ou comptes privés « in house ») est un véritable enjeu concernant la productivité des développements et le coût global d’un projet. C’est d’ailleurs pour cela qu’OCTO propose Appaloosa, qui facilite la distribution en entreprise d’applications mobiles dans des stores privés.
L’outil PhoneGap Build (compatible uniquement avec les applications PhoneGap), qui vient d’être ouvert en version finale, était une véritable nécessité.

Synthèse

Fonctionnalité	Synthèse
Plateformes supportées	Avantage PhoneGap
Richesse (fonctionnalités + IHM)	Avantage Titanium
Langage de développement	Egalité
Productivité des développements	Avantage Titanium
Gestion des déploiements	Avantage PhoneGap

Les points forts de PhoneGap sont donc aujourd’hui son support de plus de 6 plateformes, et le fait qu’il utilise au maximum les technologies standards du Web.
Ses points faibles concernent ses limitations fonctionnelles et sa moindre richesse d’IHM du fait qu’il n’utilise pas les composants natifs, ainsi que son manque d’outillage de développement.

Les points forts de Titanium sont sa richesse fonctionnelle et graphique, ainsi que la productivité de son environnement de développement.
Son point faible concerne le peu de plateformes supportées.

Conclusion

En conclusion, on a pu voir que Titanium est globalement plus productif et fournit plus de fonctionnalités que PhoneGap dans le but de développer des applications mobiles multi-plateformes. On le préfèrera en général, si l’on ne doit pas supporter plus que iOS + Android + BlackBerry.
Au-delà du choix de l’outil, notre conseil majeur restera d’utiliser les outils pour ce qu’ils savent faire, et éviter de « tordre » leur usage pour leur faire faire autre chose ou de manière différente. Cela s’approche des problématiques bien connues d’utilisation trop personnalisée de progiciels, qui rend rapidement très coûteuse la moindre évolution.

Et d’ors et déjà, le choix de la technologie de développement multi-plateformes doit aussi se faire par-rapport aux applications Web HTML5 qui apportent de plus en plus de fonctionnalités et viennent sérieusement concurrencer les approches natives et hybrides, et les solutions Web existantes comme les moteurs de rendus (BkRender, Wokup) et les nombreux frameworks Web pour mobiles (par exemple Sencha Touch.

La problématique du développement mobile multi-plateforme est loin d’être encore résolue, et nul doute que nous reviendrons encore dessus rapidement car c’est un enjeu fort pour les entreprises aujourd’hui.

Suggestion d'articles :

Performance is a Feature

2011-06-20T23:38:41Z

We've always put a heavy emphasis on performance at Stack Overflow and Stack Exchange. Not just because we're performance wonks (guilty!), but because we think speed is a competitive advantage. There's plenty of experimental data proving that the slower your website loads and displays, the less people will use it.

[Google found that] the page with 10 results took 0.4 seconds to generate. The page with 30 results took 0.9 seconds. Half a second delay caused a 20% drop in traffic. Half a second delay killed user satisfaction.

In A/B tests, [Amazon] tried delaying the page in increments of 100 milliseconds and found that even very small delays would result in substantial and costly drops in revenue.

I believe the converse of this is also true. That is, the faster your website is, the more people will use it. This follows logically if you think like an information omnivore: the faster you can load the page, the faster you can tell whether that page contains what you want. Therefore, you should always favor fast websites. The opportunity cost for switching on the public internet is effectively nil, and whatever it is that you're looking for, there are multiple websites that offer a similar experience. So how do you distinguish yourself? You start by being, above all else, fast.

Do you, too, feel the need – the need for speed? If so, I have three pieces of advice that I'd like to share with you.

1. Follow the Yahoo Guidelines. Religiously.

The golden reference standard for building a fast website remains Yahoo's 13 Simple Rules for Speeding Up Your Web Site from 2007. There is one caveat, however:

There's some good advice here, but there's also a lot of advice that only makes sense if you run a website that gets millions of unique users per day. Do you run a website like that? If so, what are you doing reading this instead of flying your private jet to a Bermuda vacation with your trophy wife?

So … a funny thing happened to me since I wrote that four years ago. I now run a network of public, community driven Q&A web sites that do get millions of daily unique users. (I'm still waiting on the jet and trophy wife.) It does depend a little on the size of your site, but if you run a public website, you really should pore over Yahoo's checklist and take every line of it to heart. Or use the tools that do this for you:

We've long since implemented most of the 13 items on Yahoo's list, except for one. But it's a big one: Using a Content Delivery Network.

The user's proximity to your web server has an impact on response times. Deploying your content across multiple, geographically dispersed servers will make your pages load faster from the user's perspective. But where should you start?
As a first step to implementing geographically dispersed content, don't attempt to redesign your web application to work in a distributed architecture. Depending on the application, changing the architecture could include daunting tasks such as synchronizing session state and replicating database transactions across server locations. Attempts to reduce the distance between users and your content could be delayed by, or never pass, this application architecture step.
Remember that 80-90% of the end-user response time is spent downloading all the components in the page: images, stylesheets, scripts, Flash, etc. This is the Performance Golden Rule. Rather than starting with the difficult task of redesigning your application architecture, it's better to first disperse your static content. This not only achieves a bigger reduction in response times, but it's easier thanks to content delivery networks.

As a final optimization step, we just rolled out a CDN for all our static content. The results are promising; the baseline here is our datacenter in NYC, so the below should be read as "how much faster did our website get for users in this area of the world?"

In the interests of technical accuracy, static content isn't the complete performance picture; you still have to talk to our servers in NYC to get the dynamic content which is the meat of the page. But 90% of our visitors are anonymous, only 36% of our traffic is from the USA, and Yahoo's research shows that 40 to 60 percent of daily vistors come in with an empty browser cache. Optimizing this cold cache performance worldwide is a huge win.

Now, I would not recommend going directly for a CDN. I'd leave that until later, as there are a bunch of performance tweaks on Yahoo's list which are free and trivial to implement. But using a CDN has gotten a heck of a lot less expensive and much simpler since 2007, with lots more competition in the space from companies like Amazon's, NetDNA, and CacheFly. So when the time comes, and you've worked through the Yahoo list as religiously as I recommend, you'll be ready.

2. Love (and Optimize for) Your Anonymous and Registered Users

Our Q&A sites are all about making the internet better. That's why all the contributed content is licensed back to the community under Creative Commons and always visible regardless of whether you are logged in or not. I despise walled gardens. In fact, you don't actually have to log in at all to participate in Q&A with us. Not even a little!

The primary source of our traffic is anonymous users arriving from search engines and elsewhere. It's classic "write once, read – and hopefully edit – millions of times." But we are also making the site richer and more dynamic for our avid community members, who definitely are logged in. We add features all the time, which means we're serving up more JavaScript and HTML. There's an unavoidable tension here between the download footprint for users who are on the site every day, and users who may visit once a month or once a year.

Both classes are important, but have fundamentally different needs. Anonymous users are voracious consumers optimizing for rapid browsing, while our avid community members are the source of all the great content that drives the network. These guys (and gals) need each other, and they both deserve special treatment. We design and optimize for two classes of users: anonymous, and logged in. Consider the following Google Chrome network panel trace on a random Super User question I picked:

	requests	data transferred	DOMContentLoaded	onload
Logged in (as me)	29	233.31 KB	1.17 s	1.31 s
Anonymous	22	111.40 KB	768 ms	1.28 s

We minimize the footprint of HTML, CSS and Javascript for anonymous users so they get their pages even faster. We load a stub of very basic functionality and dynamically "rez in" things like editing when the user focuses the answer input area. For logged in users, the footprint is necessarily larger, but we can also add features for our most avid community members at will without fear of harming the experience of the vast, silent majority of anonymous users.

3. Make Performance a Point of (Public) Pride

Now that we've exhausted the Yahoo performance guidance, and made sure we're serving the absolute minimum necessary to our anonymous users – where else can we go for performance? Back to our code, of course.

When it comes to website performance, there is no getting around one fundamental law of the universe: you can never serve a webpage faster than it you can render it on the server. I know, duh. But I'm telling you, it's very easy to fall into the trap of not noticing a few hundred milliseconds here and there over the course of a year or so of development, and then one day you turn around and your pages are taking almost a full freaking second to render on the server. It's a heck of a liability to start 1 full second in the hole before you've even transmitted your first byte over the wire!

That's why, as a developer, you need to put performance right in front of your face on every single page, all the time. That's exactly what we did with our MVC Mini Profiler, which we are contributing back to the world as open source. The simple act of putting a render time in the upper right hand corner of every page we serve forced us to fix all our performance regressions and omissions.

(Note that you can click on the SQL linked above to see what's actually being run and how long it took in each step. And you can use the share link to share the profiler data for this run with your fellow developers to ~~shame them~~ diagnose a particular problem. And it works for multiple AJAX requests. Have I mentioned that our open source MVC Mini Profiler is totally freaking awesome? If you're on a .NET stack, you should really check it out. )

In fact, with the render time appearing on every page for everyone on the dev team, performance became a point of pride. We had so many places where we had just gotten a little sloppy or missed some tiny thing that slowed a page down inordinately. Most of the performance fixes were trivial, and even the ones that were not turned into fantastic opportunities to rearchitect and make things simpler and faster for all of our users.

Did it work? You bet your sweet ILAsm it worked:

That's the Google crawler page download time; the experimental Google Site Performance page, which ostensibly reflects complete full-page browser load time, confirms the improvements:

While server page render time is only part of the performance story, it is the baseline from which you start. I cannot emphasize enough how much the simple act of putting the page render time on the page helped us, as a development team, build a dramatically faster site. Our site was always relatively fast, but even for a historically "fast" site like ours, we realized huge gains in performance from this one simple change.

I won't lie to you. Performance isn't easy. It's been a long, hard road getting to where we are now – and we've thrown a lot of unicorn dollars toward really nice hardware to run everything on, though I wouldn't call any of our hardware choices particularly extravagant. And I did follow my own advice, for the record.

I distinctly remember switching from AltaVista to Google back in 2000 in no small part because it was blazing fast. To me, performance is a feature, and I simply like using fast websites more than slow websites, so naturally I'm going to build a site that I would want to use. But I think there's also a lesson to be learned here about the competitive landscape of the public internet, where there are two kinds of websites: the quick and the dead.

Which one will you be?

Strip Games

2010-02-01T05:00:00Z

Comment concevoir vos applications iPhone ? Avec le post-it OCTO !

2010-02-01T10:42:39Z

La conception est une étape clé de la réalisation d’une application iPhone. C’est à cette étape que l’on valide la disposition des écrans, les objets affichés, les enchainements, etc.

C’est également à cette étape que l’on va éventuellement revenir sur des fonctionnalités ou en ajouter de nouvelles. Pourquoi ? Parce que l’écran est trop petit, que tel bouton n’est pas assez gros, que telle fonctionnalité n’est accessible qu’après 4 écrans en bref pour rattraper les erreurs et valider le reste …

Le story board permet de régler certains de ces problèmes et chez OCTO, nous maitrisons cet exercice. Nous avons donc appliqué nos bonnes méthodes avec un story board sur nos tableaux blancs. Le problème rencontré est que la taille du tableau autorise toutes les folies et pousse à mettre des widgets qui, réduits 15 fois, s’avèreront être trop petits.

C’est à la suite de ce constat que nous avons décidé de créer des papiers pré-imprimés à la taille de l’iPhone avec les composants standards (écran, tab bar, navigation bar, status bar, clavier). Voici le résultat :

Voici un exemple de l’usage que nous avons fait de ce post-it sur l’application OCTO Talks:

Ce post-it sera distribué aux participants du petit déjeuner « iPhone as a business platform » organisé le 9 Février à l’Atelier BNP Paribas. Les consultants OCTO en sont bien entendu équipés sur les projets iPhone de nos clients.

2011 Federal Budget: The Details

2010-02-01T15:51:30Z

President Obama proposed his 2011 budget today. At $3.8 trillion, it won’t trim deficits in the near future, but will focus on longer-term deficit reduction. We compiled the details below.

$3.8 trillion total budget for fiscal year 2011 (it starts October 1)

Defense
$159 billion for wars in Pakistan and Afghanistan
$43.6 billion for Homeland Security
20% more for Department of Veterans Affairs

Economy
$100 billion to combat unemployment and create jobs
A small business tax credit is in the works to encourage increased wages and hiring

Taxes
$2.6 trillion revenue from taxes and other sources (18% more than 2010)
Tax cuts will remain for those making less than $250,000/year
Other tax cuts will expire December 31
The Making America Work tax cut will remain
Capital gains tax for new small business investments will end
Multinational corporations will pay $122 billion more on earnings they make overseas

Education
$28 billion for the Elementary and Secondary Education Act (that’s 12% more than 2010)
$17 billion more for Pell grants.

More components
A big bank fee
No more subsidies for gas, oil, and goal producers
A 3-year spending freeze
Slashing or paring down 120 government programs (including the Army Corps of Engineers, 27 education programs, NASA, and more)

According to the Wall Street Journal,
Mr. Obama plans to rely on a new debt commission, created by executive order, to come up with recommendations on how to meet his promise to bring the figure down to the equivalent of 3% of GDP by 2015, according to budget analysts briefed on the proposal.

(Other information comes from the Philadelphia Inquirer.)

5 Business Lessons from “Up in the Air”

2010-02-01T17:19:36Z

Jason Reitman’s Golden Globe-winning Up in the Air, a recent film starring George Clooney, is a lesson in what happens when you become your job. If you’re like Ryan Bingham (Clooney’s character), you end up a business vagabond, absorbed in the artificial status of frequent flier miles and hotel points, always skimming the surface of things.

Through Ryan’s story, Up in the Air shares wisdom for businesses and career-minded people–mostly by teaching you what to avoid. Here are five of the best business takeaways.

1. Listen to your customers. Early in the movie, Ryan and ambitious new employee Natalie Keener go to one downsized company together. Natalie, still in training, fires an employee who expresses suicidal ideations. When Natalie shares her reservations about the woman, Ryan tells her not to worry about it. Weeks later, Natalie finds out the woman killed herself. She is so upset she quits her job and leaves town.

Why wasn’t Ryan alarmed by the woman’s talk of harming herself? Because he was so used to firing employees that he stopped truly listening. If he had been more aware, less habituated, he might have been able to help the woman, or at least direct her to someone who could. Instead, he didn’t take her seriously. As a result, he was unable to help prevent her death.

The lesson here? Even if you deal with thousands of customers, you’ll best serve them if you make listening part of your personal or company DNA.

2. Have a Plan B. Up in the Air is full of characters who lack backup plans. Ryan Bingham, for example, bases his entire value system into his job. To Ryan, flying and travel represent freedom. Frequent flyer miles and hotel points make him feel valued. His status at work makes him feel competent. He has his basic human needs met through his unattached working lifestyle.

But what happens when his boss stops him from traveling, thanks to new technology that lets him fire people remotely? He faces an empty apartment and an empty life.

What happens when Ryan goes out on a limb and tries, for once, to show serious interest in a woman he met on the road? He fails. Then he defaults back to his old shallow system.

If Ryan was to be fired, his entire world would collapse. He never considered life outside of the travel circuit. He avoided cultivating anything but travel points. If he loses his job, he loses his outlook, his belief system, his life. The lesson? Don’t let your business or career be like this. Have a fall-back plan or two.

3. You can’t digitize everything. In today’s information age, we try to make everything digital. That includes social interactions, customer service, interviews, and other functions that used to occur face-to-face. But certain impactful communications, like firing people, can’t effectively be done from a distance.

That’s what Craig Gregory, Ryan Bingham’s boss, finds out the hard way. Gregory hires Natalie Keener, a young business school grad, to revamp the company’s system of hiring and firing people. Instead of doing it face-to-face, Natalie implements a remote system. The person getting fired is informed of her fate by a face on a screen.

After Natalie quits, Gregory (Jason Bateman) throws in the towel on her method of firing people. Gregory wanted to change his traditional business model to cut company costs. He realized afterward that his company actually worked better the old way. Going digital isn’t the answer to everything, after all.

4. Don’t expect security. When Gregory implements remote firing, Ryan, one of the company’s best employees, is grounded in Omaha. Ryan, a star performer, thought his position was secure. If this movie and this economy prove anything, however, it’s that security is a thing of the past. Not even stellar performance can guarantee it for you.

5. Make your values proactive. Ryan’s mission and values involved avoiding stuff and commitment. He dreaded being weighed down, so he avoided anything that made him feel that way.

Values based on avoidance won’t get you very far. Such values force you to depend on external conditions for your success. For example, if Ryan were to lose his job, he would have to find another situation that enabled him to avoid a sense of being weighted down.

If, on the other hand, Ryan’s strategy were based on proactive values, he could stay true to his beliefs while being flexible about external situations. If, for example, his values were based around fostering an internal sense of freedom, he could live or move anywhere. It wouldn’t make a difference. He could be grounded in Omaha, forced to stay in one place, but find freedom in pockets of the city.

It’s a mental outlook. Ryan’s was limiting him. Don’t let your business operate on the same premise of avoidance. Be proactive.

Why Is Governance Pain So Common With SharePoint?

2009-02-05T21:49:13Z

Blogger: Craig Roth

Governance problems have plagued all sorts of websites, but in my experience they seem to come up disproportionately in SharePoint installations. In researching and writing my new document "Website Governance: Guidance for Portals, SharePoint, and Intranets" (slated for publication in March) I wanted to figure out why that is. Here is what I found out about why SharePoint has proven to be particularly vulnerable to chaos when ungoverned:

Ease of deployment: SharePoint is easier to license and install than other portal products. That's great, except more parts of the organization will be tempted to set up servers. Decentralized installation and setup of the servers often leads to siloed installations that do not conform to the organization’s best practices or technology standards.
Grass roots nature: SharePoint’s ease of use has proven to be a double-edged sword. While it opens up self-help collaboration and content capabilities for a broader swath of information workers, it also places creation in the hands of a large number of non-IT users who are only minimally monitored. This can lead to poor findability and an inconsistent user experience.
Lack of multi-farm management: SharePoint lacks enterprise-wide management features that other portal products have had for years. The highest level of management in SharePoint is the server farm, but enterprises wanting unified policies and governance across multiple server farms have few tools to accomplish this. Microsoft has made a step to remedy this situation with a Cross-Site Configurator that was specifically developed “in the context of IT management challenges that have arisen with the rapid growth of SharePoint deployments.” However this product is unsupported for now and is not a part of the official WSS build.
Frequent overlaps with other installed capabilities: SharePoint provides an integrated set of capabilities that often exist in separate products that an organization may already have installed. A team that has been managing a content management, search, collaboration, or portal system for years may wake up one day to find users starting to leverage SharePoint for the same capabilities. The result is information segregation and a quick call to the CIO to make a decision on coexistence or shutting down one of the overlapping alternatives.

This doesn't mean that SharePoint cannot be governed. But they do point to the importance of creating a statement of governance early in the planning cycle for SharePoint. While some large SharePoint deployments rise above all these problems, it is rare and difficult for them to do so without a governance structure in place.

Note: This is a cross-posting from the KnowledgeForward blog.

30 Simple Family Pleasures

2009-02-05T10:00:08Z

Editor’s note: this is a guest post by mommy blogger Vered DeLeeuw.

Do you remember to have fun every day?

I’m sure you remember to brush your teeth, to eat, and to go to work. I’m confident you have a handwritten or a technology-based to-do list that you carefully go through each day. But if you read this blog, it means you want more from your life. You want to de-stress, to be present, to enjoy the journey instead of rushing through it like a madman.

Having fun with your family, especially the simple kind of fun - the one that doesn’t drain you emotionally and financially, is a wonderful way to relax and enjoy life. And in times of economic uncertainty, it is especially important to enjoy family time that does not involve spending lots of money. Here are 30 ideas for high-quality, low-cost, stress-free family fun.

1. Play together in the snow or in the rain, then go inside for a cup of hot chocolate made with real chocolate and marshmallows.

2. Make a large bowl of popcorn and snuggle in front of the DVD to watch a family movie.

3. Have a picnic. In wintertime, it’s just as fun to have a picnic indoors.

4. Snuggle together in bed on a cold Sunday morning.

5. Go out for a leisurely Sunday brunch. Many restaurants offer great Sunday buffets, and prices these days are more reasonable than ever. Kids often eat free of charge. Although buffets can be challenging if you’re trying to lose weight, or maintain your goal weight, once in a while is probably OK. Since kids usually have very limited choices as far as food goes – they eat what we give them - they really appreciate the selection and choice a buffet offers.

6. Just talk. The older my kids are, the more interesting our conversations become.

7. Try tongue twisters. Last week when we waited forever for the check to arrive at a restaurant, my husband started a tongue twisters game. We laughed so hard, I nearly choked. My favorite tongue twister: “She sells seashells by the seashore.” Repeat three times, FAST!

8. During dinner, ask each family member to tell one good thing and one bad thing that happened to them that day. The good things are obviously fun to share. The “bad things” are a great opportunity for your kids to talk about things that bother them and for you to be more involved in their lives.

9. Create a story together. Each family member gets to add one sentence at a time. These stories can be seriously funny, especially if one of the kids happens to be in a silly mood.

10. Look at a picture album together and tell your kids the stories behind the pictures.

11. Make cookies or cupcakes together and decorate them. Try this fabulous recipe for chewy chocolate chip cookies.

12. Watch old family movies.

13. Visit the library. Browse the children’s books together. Read them a story or two. Listen to Story Time if the library offers it. Browse the grownups’ selection too: it’s good for your kids to see you’re interested in books.

14. Have a family make-your-own-pizza night. Prepare several topping options. Let each child decide what kind of pizza they want. More tips here.

15. Play a board game. I don’t know about you, but my husband and I play to win. We don’t believe in letting the kids win. After all, their friends won’t let them win. The good news (or bad news, depending on your point of view): they often win anyway!

16. Go for a hike. We love to go to a local park that has beautiful redwood trees in addition to a small playground and a shaded picnic area.

17. Go to the museum. If your kids are younger than 5, take them to the children’s museum. If they are older, they can probably handle an hour or two in a “real” museum. A natural history museum is great, but many kids enjoy modern art museums as well. Many museums are free of charge for kids under 12, and many offer free admission for the entire family on monthly “family days.”

18. Go ice-skating.

19. Go to the movies. While going to the movies is not as cheap as it used to be, matinee tickets are usually reasonably priced. It’s entirely possible to find PG-rated movies that are appropriate for elementary school kids yet are fun (or at least not mind-numbingly boring) for the grownups. During winter break we took the kids to see “Marley and Me” and “Bedtime Stories.”

20. Make breakfast foods for dinner, together. It never fails to make the kids happy. We like to make pancakes together. When they’re ready, we decorate them with berries to make fun “faces.” If your kids have milk and fruit with those homemade pancakes, it’s a pretty balanced meal. And if you’re really brave, try making whole-wheat pancakes.

21. Go to the beach. The beaches here in Northern California are cold and windy, even in summertime, and very rocky. But even though they don’t provide the classic beach experience, we love to go tidepooling. We often manage to spot interesting sea creatures and shells.

22. On a warm summer night, it’s fun to go camping in your own backyard and sleep in sleeping bags, under the stars.

23. Jump rope. We do it outside in the yard in summer and inside during the winter. My kids are already experts. I’m still learning, or rather relearning.

24. Make the most of power outages. During a recent power outage, we all sat together in the kitchen. Since we were bored (no technology!) and had to use candlelight to light the kitchen, we used the candles to make s’mores. Seriously. It was one of our best nights ever. More tips for making indoor s’mores here.

25. Exercise together. I exercise at home using a yoga mat. My kids love to watch me exercise and they love even more trying to imitate what I’m doing on our extra mat.

26. Dance together. Play some loud music and go crazy with the moves. Don’t tell anyone, but I love to dance to ABBA music.

27. Bake homemade bread. You can use a bread machine if you have one, or you can knead the dough by hand. This takes time and is great for a winter day when you’re inside anyway. There’s nothing like the smell of bread as it’s baking, and there’s nothing like eating it fresh with homemade butter on top. Try this bread recipe. To make homemade butter, simply whip heavy cream with a pinch of salt until it turns into butter. Separate it from the remaining liquid (buttermilk) and spread on the freshly baked bread.

28. Write letters to the grandparents. Even if they live nearby, they will love getting letters from their grandchildren. If you join your kids and write a few words too, you will not only set a great example for your kids but will also make your parents very happy.

29. Read a book together. You can read to your child, or she can read to you, or you can simply snuggle together, each of you reading her own book.

30. Watch TV together. TV is not just a convenient babysitter: it can be an opportunity to share an experience with your kids. Once in a while I join my kids when they watch one of their favorite TV shows. They snuggle close to me, and whenever something funny happens on the show, they watch me closely to make sure I get it. When we’re done watching, we talk about what we saw. I often use it as an opportunity to talk about advertising, since there’s so much of it these days aimed at children.

To read more from Vered, see her blog: mommy blogger.

Photo courtesy of Jeremy Barwick

Qu'est ce que le Lean ?

2009-02-04T19:04:19Z

Qu’est ce que le « Lean » ?

Cet article est une présentation générale du Lean Management. Vous pouvez le lire en suivant le fil de la conversation qu’il retrace, ou aller directement là où une notion est abordée. Les notions abordées :

Qu’est ce que le « Lean » ?

Ce terme « Lean » désigne un ensemble de principes et de pratiques de management et d’organisation. On emploie aussi le terme Lean Management. Cet ensemble est issu de TOYOTA et assure son succès depuis 50 ans.

Quels sont ces principes ?

Le premier, le plus fondamental, est le respect des gens. Il consiste à créer les conditions pour que les collaborateurs s’épanouissent dans ce qu’ils font. Le travail doit être inspirant et motivant. Il doivent être formés, responsabilisés et encouragés à résoudre eux-mêmes les problèmes qu’ils rencontrent dans leur travail quotidien. C’est fondamental car si les principes Lean sont du bon sens, la plus grande difficulté consiste à créer un environnement dans lequel suivre ce bon sens n’est pas dangereux mais standard.

Ah.

Par exemple, un développeur livre son module à la date prévue alors qu’il a de fortes raisons de penser qu’il ne s’intègrera pas au reste de l’application, ce qui nécessitera un effort important de diagnostic et re-livraison par la suite. Le contexte qui le fait agir contre le bon sens est que son responsable hiérarchique est objectivé sur le respect de cette date de livraison.

Ok. Un autre principe ?

Eliminer le gaspillage sur le flux de valeur. Il faut se concentrer sur ce qui crée la valeur pour laquelle le client de l’entreprise est prêt à payer. Toute l’entreprise est au service de cette création de valeur. Elle est créée le long d’un flux de transformation. Dans l’idéal, ce flux est tiré et sans stock pour réagir aux fluctuations de la demande et minimiser l’investissement. Cela signifie que l’on ne commence à produire quelque chose que lorsqu’il y a une demande. On ne produit alors que ce qu’il faut pour cette demande.

J’ai entendu parler de cette histoire de gaspillages. Il y a 7 types. C’est ça ?

Oui. Surproduction, délai, stock, transport, mouvement inutile, processus inutile et défauts. Par exemple attendre 3 jours pour réunir les acteurs en vue de décider d’un choix d’architecture pour un projet est un gaspillage de type « délai » pour le flux de valeur de ce projet.

Je te suggère le petit exercice suivant : quand tu rentres chez toi à la fin de la journée tu cherches un gaspillage auquel tu a été confronté dans la journée pour chacun des 7 types.

Ok, je vais essayer.

Autre principe du Lean : l’amélioration continue pas à pas des performances. Il y a 3 façons de changer :

Innover : faire quelque chose de nouveau
Imiter : reprendre quelque chose qui a déjà marché ailleurs
Améliorer en continu pas à pas : partir de la situation actuelle et la changer petit à petit en s’assurant que chaque changement est une amélioration.

C’est quoi les avantages et inconvénients de chacune de ces façons ?

L’innovation est risquée mais peut faire faire des grands bonds. Elle peut être aussi dommageable pour une organisation car elle est souvent initiée par quelques élus au sacrifice potentiel de l’engagement et du leadership des autres. Par exemple, changer de technologie en informatique, ou changer de processus de développement.

L’imitation n’est pas risquée. Elle est connotée péjorativement en occident. Par exemple à l’école « tu ne copieras pas ». Pourtant c’est un type de changement très efficace qui permet de profiter des innovations des autres sans courir de risque. Aujourd’hui adopter un processus de développement comme SCRUM rentre dans cette catégorie pour beaucoup d’entreprises.

L’amélioration continue pas à pas n’est pas risquée. Elle nécessite un engagement de tout le monde, et une discipline qui assure que chaque pas posé est bien une amélioration. L’amélioration continue est difficile à imiter. TOYOTA la pratique depuis 50 ans. C’est pour cela qu’ils enseignent le Lean à leurs concurrents : ils n’ont pas peur d’être rattrapés.

Le Kaizen c’est quoi ?

Justement. Kaizen est le terme japonais pour « amélioration continue pas à pas ». Dans le Lean la méthode pour appliquer le Kaizen est PDCA.

C’est quoi PDCA ?

PDCA est une méthode d’amélioration continue inventée par E. DEMING dans les années 1950.

Ca veut dire quoi PDCA ?

C’est la première lettre des 4 étapes de la méthode. Elles forment un cycle qui se répète à chaque amélioration : PLAN – DO – CHECK – ACT.

Que fait-on dans chaque étape ?

L’étape PLAN contient 4 parties. Définir ce que l’on veut, comprendre l’écart avec ce que l’on a, choisir une contre mesure et faire une prédiction.

D’abord tu dois identifier ce que tu veux faire : quel est le problème ? Un problème est un écart entre ce que tu observes et ce que tu souhaites. C’est à ce moment que certains problèmes se résolvent juste en explicitant qu’est ce que l’on souhaite (et notamment : qui le souhaite), et quel est le standard actuel.

J’ouvre une parenthèse ici : la notion de standard est quelque chose d’important dans le Lean. C’est très différent de l’image que l’on en a dans les entreprises pré-Lean. Dans ces dernières, les normes et les procédures sont définies par des experts qui les formulent pour qu’elles soient appliquées par d’autres. Dans l’industrie c’est le moyen d’imposer des cadences de travail.

Je demandais à un client « Ce document fait partie du processus standard de gestion de projet ? » « Oh non, le processus standard n’est jamais appliqué. Il marcherait dans l’idéal mais n’est pas applicable ici. » « Depuis combien de temps n’a-t-il pas évolué ? ». « Heu, il n’a pas évolué depuis sa création il y a 3 ans ».

Dans le Lean un standard est l’ensemble des meilleures pratiques que l’on connaît à un moment donné pour effectuer une tâche qui se répète. Ces bonnes pratiques émergent du terrain, de la pratique des opérationnels eux-mêmes. Il n’y a qu’un standard à un moment donné, et il évolue constamment en fonction des leçons que l’on apprend de l’expérience.

Le modèle du Triangle D’or du Management Visuel permet de faire fonctionner un standard. Le management visuel consiste à rendre visible les écarts par rapport au standard. Comme cela il n’est pas besoin d’être courageux pour parler de ces écarts, ils sont traités tout de suite soit en renforçant la formation aux bonnes pratiques, soit en ajustant le standard.

Un triangle ?

Oui. Il faut 3 composantes pour que le standard fonctionne.

D’abord le standard est écrit, de façon accessible, simple et opérationnelle. Il doit être un support de formation. A ce stade il y a un consensus entre tous ceux qui vont l’appliquer qu’il s’agit de la meilleure façon de faire la tâche.
Ensuite il faut un mécanisme pour rendre visible les écarts par rapport au standard pour ceux qui l’appliquent. C’est un auto contrôle de qualité. Par exemple un gyrophare qui s’allume quand l’intégration continue échoue au sein d’une équipe de développement.
Troisième sommet du triangle : la gestion des écarts. Elle doit être provisionnée et organisée afin de traiter les écarts efficacement, soit en renforçant la formation soit en faisant évoluer le standard.

Dans de nombreux cas il manque un élément au triangle et le standard ne fonctionne pas. Par exemple : les développeurs ne sont pas d’accord avec les conventions de code, ou le gyrophare s’allume mais personne ne pense avoir l’autorité d’arrêter ce qu’il fait pour investiguer le problème, etc.

Ok pour le standard. Nous en étions où ?

Oui, refermons la parenthèse du standard. Nous en étions à la phase PLAN du PDCA. La première étape de cette phase consiste à exprimer le problème comme un écart entre ce que l’on observe et le standard.

Ah oui. Ensuite ?

Ensuite, tu analyses cet écart, pour identifier les causes profondes dans l’organisation et dans les processus qui sont responsable du problème. L’important ici est de dépasser le traitement du symptôme. Ce que l’on cherche c’est à ajuster petit à petit l’organisation du travail pour se prévenir définitivement des problèmes. Pour remonter la chaîne de causalité du symptôme aux causes profondes tu appliques la méthode des 5 pourquoi.

Ah oui, tu m’en a déjà parlé.

Il faut faire attention à ce moment de ne pas identifier l’absence d’une solution auquel l’on pense comme étant la cause du problème. Par exemple, quand je dis « le problème c’est qu’il manque des tests unitaires », j’ai une petite voix qui me dis « là, tu es en train de fermer des options ». Différer le jugement pour se donner l’opportunité de connaître le problème avant de passer à la solution permet d’ouvrir le champ des actions possibles.

Nice.

Je continue ?

S’il te plait.

3eme partie de cette étape PLAN : choisir une solution au problème. On appelle cette solution une contre mesure car nous ne sommes pas dans la résolution ponctuelle d’incidents imprévisibles mais dans l’ajustement de processus organisationnels imparfaits qui génèrent des gaspillages.

Pour le choix de la contre mesure nous envisagerons à quel niveau de profondeur il est le plus efficace et pratique d’agir. C’est important que l’analyse du problème se base sur des faits concrets, sans interprétation ni généralisation.

Ensuite nous cherchons la contre mesure en essayant d’abord ce qui permettrait de supprimer par conception la source du problème. C’est le principe de POKA YOKE.

POKA YOKE ?

Oui. Cela signifie « error proofing », ou « anti erreur ». L’exemple le plus parlant est celui des premiers connecteurs série sur les ordinateurs. La position des broches électroniques était symétrique : il était possible de brancher la prise à l’envers, ce qui grillait les composants. Rapidement une solution POKA YOKE fut adoptée en mettant plus de broches en haut qu’en bas, ce qui a rendu impossible un branchement inversé.

Ah oui, je comprends.

Donc le premier type de solution que l’on recherche est de type POKA YOKE car elles sont les plus efficaces. Ensuite nous essayons de trouver comment détecter automatiquement les premiers indices d’apparition de ce problème. Enfin, si nous n’avons toujours pas de solution alors nous organiserons une revue manuelle de détection.

Dans l’informatique, utiliser 2 méthodes de classe plutôt qu’une seule méthode avec un argument optionnel est POKA YOKE. Les tests unitaires automatiques se rangent dans la seconde catégorie de solution. Un installeur qui s’arrêterait automatiquement dès qu’il rencontre un problème entre aussi dans cette seconde catégorie. Les revues de code se rangent dans la troisième catégorie.

Ok.

Dernière partie de l’étape PLAN : après avoir défini le problème, l’avoir analysé et choisi une contre mesure il faut faire une prédiction.

Une prédiction ?

Oui. Jusqu’à maintenant nous avons fait des hypothèses sur le problème, ses causes et l’efficacité de la solution. Un prédiction consiste à décrire ce qui devrait être observé une fois la contre mesure mise en place, en supposant que les hypothèses sont valables. Elle permettra d’infirmer ou de confirmer l’efficacité de la solution. Ce n’est qu’après avoir fait cette observation que nous pourrons dire que ce changement est une amélioration.

Voici une petite histoire qui circule dans le milieu de l’automobile : un manager dans une usine Française souhaite améliorer le temps de traitement des pièces sur un poste critique. Après analyse avec l’ingénieur, ils estiment que l’on peut gagner 12 secondes par pièce en chauffant la pièce à une température particulière avant de la traiter. Ils expérimentent et observent qu’ils gagnent 18 secondes. Les gains financiers sont importants. Ils sabrent le champagne. Au Japon, cette expérimentation serait un échec. Le manager et l’ingénieur serait reparti faire leurs calculs pour comprendre l’écart de 6 secondes qui invalide leurs hypothèses.

Je vois.

Ensuite la phase qui suit PLAN est DO. C’est le moment de l’expérimentation ou l’on va appliquer la contre mesure.

Expérimenter ? J’entends déjà mon boss dire « c’est pas avec mon projet que vous allez jouer à l’apprenti sorcier ! ».

Oui. Et pourtant sans la rigueur de confronter les hypothèses qui fondent chaque changement à l’expérience du terrain nous sommes dans des actes de foi, bien plus magiques et « sorciers », qui participent à faire végéter les organisations dans leurs ornières intellectuelles.

Right.

A la fin du DO nous passons à la phase CHECK où l’on confirme ou infirme que ce changement est bien une amélioration en confrontant ce que nous observons avec la prédiction initiale de la phase PLAN.

La 4eme phase ACT consiste à standardiser le résultat de ce changement. Si c’est un succès alors la contre mesure rejoint le standard et nous pouvons passer à une autre opportunité d’amélioration. Sinon, retour à la compréhension du problème et nouvelle boucle PDCA.

Tu te souviens ce que l’on a dit du standard ?

Oui : il n’y en a qu’un et il évolue constamment.

Exact. PDCA est une méthode simple et son application doit rester simple. Elle doit être applicable par n’importe qui dans l’organisation, à propos de problèmes triviaux comme de restructurations organisationnelles. Le plus complexe c’est de garder la discipline de l’appliquer. Et pour cela, rien de tel que de le faire à deux ou en équipe. Et surtout, garder en tête que c’est sur le GEMBA que l’on apprend.

Le GEMBA ?

Oui, c’est le terme japonais qui désigne le terrain où se produit la valeur ajoutée. Dans une usine, c’est la chaîne de production. Dans l’informatique c’est le plateau projet ou s’écrivent les specs, le design, le code, ou s’exécutent les tests et les mises en production.

Ok. Il y a un autre terme que j’entends aussi : KANBAN

KANBAN est une méthode pour organiser un flux tiré. Après avoir traité un élément, chaque étape demande à l’étape précédente de lui fournir un autre élément à traiter, quelque soit la taille du stock de pièces qui lui reste à traiter. Cela crée un flux de demande qui remonte le flux de création de valeur, depuis la demande client jusqu’à la première étape du processus.

Par exemple dans l’informatique un exemple de flux tiré serait : un client appelle son interlocuteur informatique pour lui dire qu’il veut automatiser une nouvelle tâche dans son application. Cet interlocuteur appelle le responsable de la production pour lui dire qu’il veut que soit installée cette nouvelle fonctionnalité. Celui-ci demande à l’équipe de développement de développer cette fonctionnalité. En fonction du coût intrinsèque que nécessite chacune de ces opérations il peut être nécessaire de créer du stock à chaque étape. Si l’on ne peut faire qu’une seule mise en production par mois, alors il faut empiler les fonctionnalités à livrer. L’approche Lean ne va pas à l’encontre de ce stock, elle oriente l’amélioration vers la diminution de ce coût intrinsèque qui régit la taille du stock.

Merci pour toutes ces explications.

Pas de quoi.

Et la TOC, c’est quoi ?

Heu… une prochaine fois peut-être.

Bert Busch, "Health Care"

2009-02-03T06:39:54Z

Bert Busch's mother Jeanette was the second of my grandfather's five sisters. His brothers Malcolm and Ron also contributed jokes. When we were growing up in the seventies, Bert had a groovy moustache that made him look a little like Gomez from the Addams family.

Tags: video, jews, jokes, comedy, funny, catskills

Ten Ways to Use LinkedIn to Find a Job

2009-02-02T16:57:34Z

Searching for a job can suck if you constrain yourself to the typical tools such as online jobs boards, trade publications, CraigsList, and networking with only your close friends. In these kinds of times, you need to use all the weapons that you can, and one that many people don’t—or at least don’t use to the fullest extent, is LinkedIn.

LinkedIn has over thirty-five million members in over 140 industries. Most of them are adults, employed, and not looking to post something on your Wall or date you. Executives from all the Fortune 500 companies are on LinkedIn. Most have disclosed what they do, where they work now, and where they’ve worked in the past. Talk about a target-rich environment, and the service is free.

Here are ten tips to help use LinkedIn to find a job. If you know someone who’s looking for a job, forward them these tips along with an invitation to connect on LinkedIn. Before trying these tips, make sure you’ve filled out your profile and added at least twenty connections

Get the word out. Tell your network that you’re looking for a new position because a job search these days requires the “law of big numbers” There is no stigma that you’re looking right now, so the more people who know you’re looking, the more likely you’ll find a job. Recently, LinkedIn added “status updates” which you can use to let your network know about your newly emancipated status.
Get LinkedIn recommendations from your colleagues. A strong recommendation from your manager highlights your strengths and shows that you were a valued employee. This is especially helpful if you were recently laid off, and there is no better time to ask for this than when your manager is feeling bad because she laid you off. If you were a manager yourself, recommendations from your employees can also highlight leadership qualities.
Find out where people with your backgrounds are working. Find companies that employ people like you by doing an advanced search for people in your area who have your skills. For example, if you’re a web developer in Seattle, search profiles in your zip code using keywords with your skills (for example, JavaScript, XHTML, Ruby on Rails) to see which companies employ people like you.
Find out where people at a company came from. LinkedIn “Company Profiles” show the career path of people before they began work there. This is very useful data to figure out what a company is looking for in new hires. For example, Microsoft employees worked at Hewlett-Packard and Oracle.
Find out where people from a company go next. LinkedIn’s “Company Profiles” also tell you where people go after leaving the company. You can use this to track where people go after leaving your company as well as employees of other companies in your sector. (You could make the case that this feature also enables to figure out which companies to avoid, but I digress.)
Check if a company is still hiring. Company pages on LinkedIn include a section called “New Hires” that lists people who have recently joined the company. If you have real chutzpah, you can ask these new hires how they got their new job. At the very least you can examine their backgrounds to surmise what made them attractive to the new employer.
Get to the hiring manager. LinkedIn’s job search engine allows you to search for any kind of job you want. However, when you view the results, pay close attention to the ones that you’re no more than two degrees away from. This means that you know someone who knows the person that posted the job—it can’t get much better than that. (Power tip: two degrees is about the limit for getting to hiring managers. I never help friends of friends of friends.) Another way to find companies that you have ties to is by looking at the “Companies in Your Network” section on LinkedIn’s Job Search page.
Get to the right HR person. The best case is getting to the hiring manager via someone who knows him, but if that isn’t possible you can still use LinkedIn to find someone inside the company to walk your resume to the hiring manager or HR department. When someone receives a resume from a coworker even if she doesn’t know the coworker, she almost always pays attention to it.
Find out the secret job requirements. Job listings rarely spell out entirely or exactly what a hiring manager is seeking. Find a connection at the company who can get the inside scoop on what really matters for the job. You can do this by searching for the company name; the results will show you who in your network connects you to the company. If you don’t have an inside connection, look at profiles of the people who work at the company to get an idea of their backgrounds and important skills.
Find startups to join. Maybe this recession is God telling you it’s time to try a startup. But great startups are hard to find. Play around with LinkedIn’s advanced search engine using “startup” or “stealth” in the keyword or company field. You can also narrow by industry (for example, startups in the Web 2.0, wireless, or biotech sectors). If large companies can’t offer “job security,” open up your search to include startups.
Build your network before you need it. As a last tip, no matter how the economy or your career is doing, having a strong network is a good form of job security. Don’t wait until times are tough to nurture your network. The key to networking (or “schmozing”), however, is filled with counter-intuitiveness. First, it’s not who you know—it’s who knows of you. Second, Great schmoozers are not thinking “What can this person do for me?” To the contrary, they are thinking, “What can I do for this person?” For more on schmoozing, read “The Art of Schmoozing.”

Here are two more ways I can help you in your job search. First, for an aggregation of hundreds of newly posted jobs, check out Jobs.alltop. Second, to really stay on top of what’s the latest news about LinkedIn, go to Linkedin.alltop; this will turn you in a true LinkedIn power user. Just remember me when you’re rich and famous!

Addendum: "Using LinkedIn to Find a Job" by Kaye Monty

8 Tools to Track Your Footprints on the Web

2009-02-02T00:57:40Z

Last week we looked at how easy it is to leave footprints on the Web; today we'll show you how easy it is to track them.

Although search engines provide a great starting point when you're searching for someone online, with all of the new social sites that have popped up over the past few years, they're often just not enough.

Sponsor

In our recent State of Blog Search 2009 post we discussed the various reasons you may choose to use any or all of the following blog search tools: Technorati, Google Blog Search, Ice Rocket, Ask.com Blogsearch, and FriendFeed. While these blog search engines are great to fill specific needs, they're also another great place to look for your footprints on the Web.

However, you can drill down even more.

1. BlogPulse: Trends in the Blogosphere

Part of Nielsen-Online, BlogPulse highlights the top trends in the blogosphere and is mostly used to determine the hottest topics on the Web and how they got to be that way. But, its value as a personal monitoring tool can not be disregarded. Search for your name then grab the RSS feed to see who is talking about you and what they're saying.

2. Pipl: Searching the Invisible Web

Pipl claims to search the deep or invisible Web to find documents, blog entries, photos, publicly available information that other search engines don't serve up. It's a great, fast search engine that we like; the only disadvantage is it offers no RSS feed.

3. Spy: Watching what Happens on the Web

According to the site, Spy can "listen in on the social media conversations you're interested in." This clean visualization search tool watches Twitter, FriendFeed, blog posts, Google reader shares and Flickr for any term you want. An RSS feed is available.

4. Serph: The Social Web Right Now

A brilliant tool for searching the social Web, Serph shows you what is being said about you "right now." Serph gathers results from blog search engines, social media sites, social news sites and social bookmarking sites and offers an RSS feed for the results.

5. Social Mention: Mentions of your Name on the Social Web

Another great tool for searching the social Web, Social Mention offers a quick glance at mentions of your name on the Web. Just enter your name and switch between blogs, microblogs, bookmarks, comments, events, images, news or all of them at once. Slower than Serph, but occasionally offers different results. An RSS feed is available.

6. Monitter: Tracking Twitter

Monitter is one of the coolest looking monitoring tools for Twitter and one of the most useful. We've written about it before and although most people are using Twitter's own search tool for search and alerts on Twitter, Monitter offers a little bit more. Giving you the option to search for three different keywords at once, Monitter is great if you want to keep your eye out for mentions of your name, your username and your company all at the same time. It also offers an RSS feed.

7. BoardTracker 2.0: The Ultimate Search Tool for Forums

BoardTracker is a forum search engine, message tracking and instant alert system that offers relevant results quickly. One of our favorite search tools for forums and message boards, BoardTracker currently tracks in excess of 1.2 billion posts.

8. Google Alerts: The big G

We couldn't end this post without mentioning Google Alerts, although likely most of you are familiar with it. Although Microsoft and Yahoo have alert tools, Google's offering beats them hands down. It offers e-mail and RSS alerts for any set of keywords including your name.

While we're still waiting for that perfect product that will associate our names with our brands with our usernames, and send us the results instantly, we don't expect to see it anytime soon (although we've got our fingers crossed), but we do hope that this list provides you with some alternatives to track your footprints across the Web.

If you've got a great tool you want to share, please let us know in the comments.

Discuss

Chad Hurley, Craig Mundie, and Mark Zuckerberg Talk Mobile At Davos (Updated With Video)

2009-01-30T17:00:38Z

Today at Davos, Mike just finished moderating a panel on the Next Digital Experience with Chad Hurley (YouTube), Craig Mundie (Microsoft), Mark Zuckerberg (Facebook), Shananu Narayen (Adobe), Hamid Akhvan (T-Mobile) and Eric Clemons (Wharton). I will put up the video as soon as it is available.

The wide-ranging discussion focused on how the mobile Web is becoming increasingly important, and in many countries, the primary way people connect to the Internet. A good chunk of the panel was spent talking about the implications of sharing where you are and what you are doing all the time because mobile phones make it easier to do so. Chad Hurley noted that the rate at which YouTube is serving videos to mobile devices is growing at a faster rate than the site as a whole. Mike asked Chad Hurley how long would it be before people started using their mobile phones to upload videos to YouTube in a serious way. Hurley’s response:

It is available on some devices, but mostly you still have to connect to your computer to upload. But it will become a larger part of what we show. People on the street, sharing their thoughts and experiences.

For Facebook CEO Mark Zuckerberg agrees, mobile access to the Internet simply makes it easier for people to share information. Speaking like a true technologist, he thinks this is good because it will make society more efficient:

People are becoming comfortable sharing more and more because it is valuable. By sharing where you are at a certain time you might make it so you can meet up with people who are around you or get advice on what to do. This happened to me in New York before I came to Davos. I didn’t realize it was one of my friend’s birthdays. It is really useful, it creates a lot of efficiency within society. We want to push people to share more and more information, because we think people will evolve to use more and more of these [services], which is why we want to go to mobile.

There are, of course, huge privacy concerns here which every company making a push into mobile needs to be careful about, but the bigger issues the panelists agreed are simply technical. There are too many mobile devices to support. Facebook CEO Mark Zuckerberg argued that the primitive state of the technology is holding companies back more than the privacy issues, which he thinks are manageable if you let users control who sees what. Says Zuckerberg:

The platforms aren’t there yet. With all the mobile platforms—iPhone, Blackbery, Android, the mobile web—it is difficult to develop for all of them. When the number of platforms consolidate it will become a powerful thing. And when GPS becomes more available in phones. It is not the privacy issue.

Microsoft’s chief research and strategy officer agrees “what people need is notice and choice.” He also thinks that the bigger barrier to making the mobile Web more like the one on our OCs is technical right now

You are too early in the cycle. No uniform way to get at those capabilities even within the devices. As people who write the apps will exploit it more. When there were few cameras in phones there was little photo integration. Now you can click the photo and move it to an app in a few button pushes. It is just a matter of evolution.

Crunch Network: CrunchBoard because it’s time for you to find a new Job2.0

Confirmed: Apple and Adobe Collaborating on iPhone Flash

2009-02-01T07:36:36Z

The ongoing debate over Flash on the iPhone appears to be over after Adobe CEO Shantanu Narayen announced last week at the World Economic Forum that Adobe and Apple are working together in an effort to bring Flash to the iPhone.

"It's a hard technical challenge and that's part of the reason Apple and Adobe are cooperating to try and get it done as soon as possible," Narayen said in an interview with Bloomberg.

Sponsor

While Narayen didn't answer the question of whether Apple had strengthened its commitment to putting Flash on the iPhone he did point out that the ball was now in Adobe's court. "We have the developer kit and the onus is now on us," he said.

Flash now may be good enough for Apple

The debate has been long and tough. In March last year Narayen said that with or without Apple's blessing, Adobe planned to develop a Flash player for the iPhone/iPod touch platform. Unfortunately Apple CEO Steve Jobs had already made it clear the current Adobe offerings were not acceptable. Flash Lite, the version designed specifically for mobile devices, just wasn't good enough and Adobe's desktop product was just too slow on the iPhone. According to Jobs, what was needed was a "missing product in the middle."

Has Adobe been working on that product in the middle? In October 2008 Adobe Senior Director of Engineering Paul Betlem confirmed that a Flash player was in development for the iPhone, but pointed out that Adobe would still need approval from Apple to get it onto their products. Clearly, approval has now been granted.

During the Bloomberg interview, Narayen said that Adobe plans to have over a million devices shipped by the end of the year that have Flash and he's not just talking iPhone; the company is working on the Android platform, the Windows mobile and the Symbian platform.

As to when we'll see Flash on the iPhone, your guess is as good as ours. No time frame has been offered but given Narayen's comments last week, we hope to see it this year.

Watch the five minute interview below.

Discuss

Layoffs Send People and Knowledge Packing

2009-01-31T17:00:00Z

The scale of layoffs over the past few weeks is unprecedented. The impact on these people who have been shown the door and on the companies that have let them go will linger for years to come. Besides the emotional damage that occurs when people are forced out, there is a tangible cost to companies when knowledge and experience walk out the door. Once that knowledge and experience are gone, no amount of TARP money will bring them back. It may be too late for some companies to prevent this now, but putting measures in place will lessen the blow in future.

Sponsor

One of the few ways to address the problem is to adopt collaborative tools and processes that capture the information companies need to be able to thrive.

Every company in the world relies on the bits of information that live in its employees' heads. The accounts payable clerk may know a small nugget on how to negotiate the best price with an important vendor. A salesperson may repeat the same pitch when selling to a certain kind of customer that works every time. An engineer may be the only person who knows why software was originally built with ColdFusion. The branch manager may know the precise spot to kick the box compactor when it acts up.

All of these bits of knowledge add up to a vast amount of information that a healthy business requires. But the vast majority of companies do a poor job of gathering this information in a systematic way. When people leave, either by choice or not, what inevitably happens is that the company and its future employees are forced to relearn it all through trial and error. The cycle repeats over and over again. The scale of this problem is huge in the current environment.

So, what does a company do to address the problem?

1. Use tools and process. Companies must have infrastructure in place to encourage, or force, employees to share information. It doesn't matter what it is, as long as it is at least marginally effective. Anything is better than nothing. How many managers around the world are at this moment digging through abandoned email folders trying to figure what their employees were working on and what they knew.

2. Measure collaboration. Again, it doesn't matter how, just do something that you think has a good chance of success. You could measure the number of contributions to a knowledge base, the frequency of mentoring sessions, the number of white papers written, whatever. What you measure is less important than doing it consistently over time and measuring improvement. Of course, the metrics are not irrelevant, but don't wait to choose the perfect ones to track. Start small and simply, and go from there.

3. Reward employees for sharing. If you don't measure and reward productive behavior, it isn't going to happen. Collaboration is a bit fuzzy and can't be measured like the number of phone calls answered per hour, but there are ways.

4. Focus on informal information. This is often where the best information resides. For example, many employees send emails back and forth answering questions and trading best practices. You need a way to harvest these nuggets of information.

None of this stuff is rocket science, but few companies nail this process. We recently spoke with David Coleman of Collaborative Strategies, who said only 10% of his clients focus on "back-end" collaboration. The majority have invested in front-end collaboration technology, like web conferencing, to save money and reduce travel. This is also important but doesn't help companies retain information when people leave. The current economic conditions put even more pressure on companies to wring as much information as possible from remaining and future employees. The important thing now, if you are a business owner or manager, is to do something before it is too late again.

Discuss

Snipt.org: Easily Share Code on Twitter

2009-01-30T18:59:57Z

Sending a piece of source code for troubleshooting to one of your friends or colleagues can be a hassle. Snipt.org provides a new solution for this. Just copy and paste your code into Snipt, tell it what programming language it is in, and Snipt will give you a short URL for your code snippet to hand out on Twitter. The developers want you to think of it as "twitpic, but for code and long text," though it is really a lot more flexible than that.

Sponsor

Besides giving you a short URL, the most important feature of Snipt.org is that it preserves the formatting of your code and also colors it according to your specified programming language for easier reading. Snipt.org understands over 125 different programming and scripting languages, ranging from C++ and Perl, to LaTeX and Python.

Overall, this looks like a very versatile tools for programmers and everybody else who wants to easily share short snippets of code.

Snipt.net

Of course, there is also snipt.net, which is not related to snipt.org at all (at least as far as we can tell), and which focuses on creating a repository of code snippets. Snipt.net does not feature any Twitter integration, though, according to the snipt.net team's latest Twitter updates, they are thinking about this.

Discuss

Google's Mythical GDrive Surfaces Once Again: Will It Bring Cloud Storage to the Mainstream?

2009-01-30T18:00:42Z

At this point, most signs point toward Google releasing its rumored GDrive in the near future. In many ways, this mythical GDrive is simply the missing puzzle piece in Google's online strategy. While Google offers a number of online services with a storage component, it still doesn't offer a unified storage solution that brings Gmail, Picasa Web Albums, and Google Docs together.

Sponsor

Why it Matters: Bringing Cloud Storage to the Mainstream

You can already store your photos on Picasa Web Albums (though the amount of free storage is very limited). However, once you pay Google, your Gmail storage and Picasa storage limits become one - so Google clearly has at least some unified infrastructure for doing some of the back-end work in place already. According to a document (PDF) unearthed by Google is Watching You, that is exactly what Google is planning to do with the GDrive.

Google has enough clout to take online storage mainstream. While this directly benefits Google, it will also benefit the cloud storage industry in general, as the big name behind the product will drive up the general comfort level with online storage.

There are, of course, numerous small and medium sized-companies that offer online storage in some form or another. The smartest ones integrate directly with your desktop, so that you can seamlessly move data between the cloud and your own machine. A large number of other services also offer backup services, though without directly integrating this with your desktop. While all of these offerings are interesting, none have really made it into the mainstream yet.

What the GDrive is Up Against

Google's biggest competitor in this business is most likely going to be Microsoft, which has just started its push for cloud computing and storage. With its Live Drive, Microsoft offers 25GB of storage to all of its millions of Windows Live users. But Microsoft wouldn't be Microsoft if it didn't also offer ten different online storage solutions that can't speak to each other. You can also use LiveSync to transfer data between your own computers, Live Mesh for syncing and online storage, and Office Live Workspace for managing and storing office documents. And these are just Microsoft's consumer products in this space.

If Google gets the GDrive right, it will be able to offer one single online storage solution that does all of what Microsoft's plethora of tools does, but through one unified user interface and service. If the descriptions of the GDrive that have surfaced over the last week turn out to be true, then Google wants to offer a solution for all your files, including documents, photos, and (interestingly) music.

If Google can also offer solutions to access these files on mobile phones (besides Android) and if it offers a good integration with the desktop, then it could surely become the company that takes cloud storage into the mainstream.

Now we just have to wait for the actual release of the GDrive...

Discuss

The Sad Tragedy of Micro-Optimization Theater

2009-01-30T07:59:59Z

I'll just come right out and say it: I love strings. As far as I'm concerned, there isn't a problem that I can't solve with a string and perhaps a regular expression or two. But maybe that's just my lack of math skills talking.

In all seriousness, though, the type of programming we do on Stack Overflow is intimately tied to strings. We're constantly building them, merging them, processing them, or dumping them out to a HTTP stream. Sometimes I even give them relaxing massages. Now, if you've worked with strings at all, you know that this is code you desperately want to avoid writing:

static string Shlemiel()
{
    string result = "";
    for (int i = 0; i < 314159; i++)
    {
        result += getStringData(i);
    }
    return result;
}

In most garbage collected languages, strings are immutable: when you add two strings, the contents of both are copied. As you keep adding to result in this loop, more and more memory is allocated each time. This leads directly to awful quadradic n² performance, or as Joel likes to call it, Shlemiel the painter performance.

Who is Shlemiel? He's the guy in this joke:
Shlemiel gets a job as a street painter, painting the dotted lines down the middle of the road. On the first day he takes a can of paint out to the road and finishes 300 yards of the road. "That's pretty good!" says his boss, "you're a fast worker!" and pays him a kopeck.
The next day Shlemiel only gets 150 yards done. "Well, that's not nearly as good as yesterday, but you're still a fast worker. 150 yards is respectable," and pays him a kopeck.
The next day Shlemiel paints 30 yards of the road. "Only 30!" shouts his boss. "That's unacceptable! On the first day you did ten times that much work! What's going on?"
"I can't help it," says Shlemiel. "Every day I get farther and farther away from the paint can!"

This is a softball question. You all knew that. Every decent programmer knows that string concatenation, while fine in small doses, is deadly poison in loops.

But what if you're doing nothing but small bits of string concatenation, dozens to hundreds of times -- as in most web apps? Then you might develop a nagging doubt, as I did, that lots of little Shlemiels could possibly be as bad as one giant Shlemiel.

Let's say we wanted to build this HTML fragment:

<div class="user-action-time">stuff</div>
<div class="user-gravatar32">stuff</div>
<div class="user-details">stuff<br/>stuff</div>

Which might appear on a given Stack Overflow page anywhere from one to sixty times. And we're serving up hundreds of thousands of these pages per day.

Not so clear-cut, now, is it?

So, which of these methods of forming the above string do you think is fastest over a hundred thousand iterations?

1: Simple Concatenation

string s = 
@"<div class=""user-action-time"">" + st() + st() + @"</div>
<div class=""user-gravatar32"">" + st() + @"</div>
<div class=""user-details"">" + st() + "<br/>" + st() + "</div>";
return s;

2: String.Format

string s = 
@"<div class=""user-action-time"">{0}{1}</div>
<div class=""user-gravatar32"">{2}</div>
<div class=""user-details"">{3}<br/>{4}</div>";
return String.Format(s, st(), st(), st(), st(), st());

3: string.Concat

string s = 
string.Concat(@"<div class=""user-action-time"">", st(), st(),
    @"</div><div class=""user-gravatar32"">", st(), 
    @"</div><div class=""user-details"">", st(), "<br/>",
    st(), "</div>");
return s;

4: String.Replace

string s =
@"<div class=""user-action-time"">{s1}{s2}</div>
<div class=""user-gravatar32"">{s3}</div>
<div class=""user-details"">{s4}<br/>{s5}</div>";
s = s.Replace("{s1}", st()).Replace("{s2}", st()).
    Replace("{s3}", st()).Replace("{s4}", st()).
    Replace("{s5}", st());
return s;

5: StringBuilder

var sb = new StringBuilder(256);
sb.Append(@"<div class=""user-action-time"">");
sb.Append(st());
sb.Append(st());
sb.Append(@"</div><div class=""user-gravatar32"">");
sb.Append(st());
sb.Append(@"</div><div class=""user-details"">");
sb.Append(st());
sb.Append("<br/>");
sb.Append(st());
sb.Append("</div>");
return sb.ToString();

Take your itchy little trigger finger off that compile key and think about this for a minute. Which one of these methods will be faster?

Got an answer? Great!

And.. drumroll please.. the correct answer:

It. Just. Doesn't. Matter!

We already know none of these operations will be performed in a loop, so we can rule out brutally poor performance characteristics of naive string concatenation. All that's left is micro-optimization, and the minute you begin worrying about tiny little optimizations, you've already gone down the wrong path.

Oh, you don't believe me? Sadly, I didn't believe it myself, which is why I got drawn into this in the first place. Here are my results -- for 100,000 iterations, on a dual core 3.5 GHz Core 2 Duo.

1: Simple Concatenation	606 ms
2: String.Format	665 ms
3: string.Concat	587 ms
4: String.Replace	979 ms
5: StringBuilder	588 ms

Even if we went from the worst performing technique to the best one, we would have saved a lousy 391 milliseconds over a hundred thousand iterations. Not the sort of thing that I'd throw a victory party over. I guess I figured out that using .Replace is best avoided, but even that has some readability benefits that might outweigh the miniscule cost.

Now, you might very well ask which of these techniques has the lowest memory usage, as Rico Mariani did. I didn't get a chance to run these against CLRProfiler to see if there was a clear winner in that regard. It's a valid point, but I doubt the results would change much. In my experience, techniques that abuse memory also tend to take a lot of clock time. Memory allocations are fast on modern PCs, but they're far from free.

Opinions vary on just how many strings you have to concatenate before you should start worrying about performance. The general consensus is around 10. But you'll also read crazy stuff, like this:

Don't use += concatenating ever. Too many changes are taking place behind the scene, which aren’t obvious from my code in the first place. I advise you to use String.Concat() explicitly with any overload (2 strings, 3 strings, string array). This will clearly show what your code does without any surprises, while allowing yourself to keep a check on the efficiency.

Never? Ever? Never ever ever? Not even once? Not even if it doesn't matter? Any time you see "don't ever do X", alarm bells should be going off. Like they hopefully are right now.

Yes, you should avoid the obvious beginner mistakes of string concatenation, the stuff every programmer learns their first year on the job. But after that, you should be more worried about the maintainability and readability of your code than its performance. And that is perhaps the most tragic thing about letting yourself get sucked into micro-optimization theater -- it distracts you from your real goal: writing better code.

[advertisement] Did your buddy just get his ear chewed off for another server crash? Help him out by recommending PA Server Monitor. He just might buy you lunch. Download the Free Trial!

Wavyx's shared items in Google Reader

The Rands Test

Applications mobiles multi-plateformes: les approches PhoneGap et Titanium Mobile

Introduction

Présentation des solutions

PhoneGap

Titanium

Points communs

Comparaison

Plateformes supportées

Notre avis:

Richesse de la plateforme

Notre avis:

Plateforme de développement

Notre avis:

Productivité des développement

Notre avis:

Gestion des déploiements

Notre avis:

Synthèse

Conclusion

Performance is a Feature

1. Follow the Yahoo Guidelines. Religiously.

2. Love (and Optimize for) Your Anonymous and Registered Users

3. Make Performance a Point of (Public) Pride

Strip Games

Comment concevoir vos applications iPhone ? Avec le post-it OCTO !

2011 Federal Budget: The Details

5 Business Lessons from “Up in the Air”

Why Is Governance Pain So Common With SharePoint?

30 Simple Family Pleasures

Editor’s note: this is a guest post by mommy blogger Vered DeLeeuw.

Qu'est ce que le Lean ?

Bert Busch, "Health Care"

Ten Ways to Use LinkedIn to Find a Job

8 Tools to Track Your Footprints on the Web

1. BlogPulse: Trends in the Blogosphere

2. Pipl: Searching the Invisible Web

3. Spy: Watching what Happens on the Web

4. Serph: The Social Web Right Now

5. Social Mention: Mentions of your Name on the Social Web

6. Monitter: Tracking Twitter

7. BoardTracker 2.0: The Ultimate Search Tool for Forums

8. Google Alerts: The big G

Chad Hurley, Craig Mundie, and Mark Zuckerberg Talk Mobile At Davos (Updated With Video)

Confirmed: Apple and Adobe Collaborating on iPhone Flash

Flash now may be good enough for Apple

Layoffs Send People and Knowledge Packing

Snipt.org: Easily Share Code on Twitter

Snipt.net

Google's Mythical GDrive Surfaces Once Again: Will It Bring Cloud Storage to the Mainstream?

Why it Matters: Bringing Cloud Storage to the Mainstream

What the GDrive is Up Against

Who's Online and What Are They Doing There?

Who Uses Email?

Social Networking Dominated by the Young

Older Generations Research, Shop, and Bank

The Sad Tragedy of Micro-Optimization Theater

It. Just. Doesn't. Matter!