Do I know you?

Dear Mike,

I suppose you have read my blog post about cold-calls I regularly receive from recruiters. Or perhaps you haven’t, as you’ve only seen my CV few minutes before sending your e-mail, as you admitted. Anyway, even if you did, I’ll try to help you understand what I meant.

To tell you the truth, I found the job posting utterly and completely boring. Why don’t you give some material to your client on how to write attractive job descriptions? There are plenty of close-to-identical offers. Why would I care about that one?

For example, have them look at my public repositories and ask them to tell me what they like and don’t like about it. Get me involved. Show me some code that you think is freaking awesome. Let me know what companies inspire you. Tell me about your culture. Tell me about you.

In the meantime, I’m going to work on these little personal projects that help me stay up-to-date in the software industry and at the same time shows people like your client how I code. Remember: you are actively searching – I’m not.



Do you know me?

Dear Ben,

I’m flattered that you wanted to talk to me about a job offer, and I’m sorry I wasn’t able to talk longer. Here’s a written response that I hope you will appreciate.

Five years ago, I applied for a Java offer by writing to Miss J. from your recruiting firm. The process was not successful, and I accepted an other offer.

Few months later, Miss J. asked if I can refer someone for an developer.

Two years later, Miss J. talks with me on the phone. I explain that I am interested in Java development, not .NET. I am not open to new job opportunities at the time.

Nearly a year later, I got a generic job offer from Mr. M. My response asked to be removed from your firm’s mailiing-list.

Last year, I got an invite to join Miss. D. on LinkedIn with a generic message, which I ignored.

Today, a person named Ben calls me during the day, as if I didn’t have a full-time job.

Who are you? Who’s my contact at your firm? Who was Mr. M? Why did Miss. D.? Where’s Miss J.? Why would I want to talk to you now?

Do you know me at all? Did I pop in your search results because I had « Java » in my file? Has someone referred you to me? Have you researched me on the internet, before telling me about that great offer I should not ignore? I have a blog, a Google+ profile, a Facebook profile, Twitter & Diaspora accounts, a LinkedIn profile, a Stack Overflow profile with questions I asked and answered, a GitHub profile with public contributions on stuff that I find exciting, and even account and you can also learn about me by browsing my photos on my Flickr account. There are 38,300 results if you Google me, and there are very few people with the same name. You already know all this, right?
By now, you know what you should do. Master your craft. Once you have enough reasons to believe I may want to talk to you, go ahead introduce yourself, and tell me something exciting. Until then, I’m having a good time with my life, and I hope you do as well.
Nicolas (it spells with no « h »)

Evaluating Structure101 with Storm

I won a licence for Structure101 at the last Code Retreat, watched introduction videos, and decided to give it a try. I looked for a decent Java project to run it on, and I picked one of the most starred projects on GitHub: storm.

First, Structure101 wants to know where the bytecode is, so I need to compile storm. I look for compile instructions and I find it contains Clojure sources. Since that should compile to bytecode, I take a chance and keep going.

Compiling Storm

I see storm has a project.clj in the top-level directory mentioning lein, so I install leiningen. The latest is version 2, but I need to use version 1, as per the project.clj and the error I get otherwise:

~/storm (master)? $ lein help
ERROR: requires Leiningen 1.x but you are using 2.0.0-preview10

So I type lein compile, but it ends with a failure:

Compiling backtype.storm.messaging.zmq
1 [main] ERROR org.apache.zookeeper.server.NIOServerCnxn - Thread Thread[main,5,main] died
java.lang.UnsatisfiedLinkError: no jzmq in java.library.path, compiling:(zmq.clj:1)

I find bin/ and execute it, and get:

checking for ZeroMQ… no
checking zmq.h usability… no
checking zmq.h presence… no
checking for zmq.h… no
configure: error: cannot find zmq.h

It turns out the script does not check if commands failed and goes on with the rest. Before that error, zeromq complained about missing uuid-dev, and I `sudo apt-get install uuid-dev`.

This helps, but then jzmq wants to make org/zeromq/ZMQ.class but can’t find the classdist_noinst.stamp « target ». It turns out this is a known issue with a simple fix to do in src/ Finally `lein compile` completes.

Creating Structure101 project

I go back to Structure101 and create a new local project, picking Static classpath because I’ll point it to the generated `classes` directory. I choose the Logical breakout because I don’t really care how storm organizes its classes for the moment.

I choose the detailed granularity because the code base is not very big, and I guess I can reduce the amount of detail later if the view looks too cluttered.

For now I’ll hide external dependencies to concentrate on the storm codebase.

No excludes, no additional source files, no transformations.

Project Summary

It takes only few seconds to generate the summary.

Storm Project Size

Storm Project Flat Tangles

It also says « The architecture for this project is not defined ». From the text following that note, I understand this is me that needs to define the desired architecture rules that must be obeyed, and which will be used to generate a violations report, possibly viewable in an IDE such as Eclipse or IntelliJ.

Full summary here.

Structure: Exploring Tangles

In the summary, one big tangle that adds Excessive Structural Complexity (XS) is the `backtype.storm` package. Navigating to that package under the Structure tab, from the Package Explorer, I get this:

Dependency graph: backtype.storm

This is the Composition view. The dependency graph shows, at the top, that command, planner, and drpc packages depend on the tangle. At the bottom, the tangle depends on cluster, event, messaging, stats, nimbus, util, and ui. The `generated` package is even lower because it depends on nothing else.

Within the tangle, the first observation I make is that we see both the `backtype.storm` package as well as its subpackages. We can see that classes in `backtype.storm` refer to classes within sub-packages, and vice-versa. This discourages letting a package depend on its parent, where each level in the package hierarchy builts on its leafs.

The reverse of that – leafs depending on its parent – would mean that leafs couldn’t be refered to, other than by a package in a different root package. I see this can lead to interesting discussions, and how enforcing rules can help large teams build maintainable systems.

Back to the diagram, the `tuple` package is clearly a dependency of all other packages. The `utils` package could almost be a leaf, but it has a single reference to the `task` package (going up), although there are 13 references going the other way. This clearly is a smell, and I want to know what is that reference. Highlighting the arrow going up from utils to task, the Dependency breakout pane shows the culprit:

Dependency breakout

It shows that utils.ShellProcess takes a TopologyContext as a parameter. Navigating (with the context menu) to consumers of TopologyContext, I find that it is referenced 160 times from all over the place. This is often what happens with « context » classes: they tend to become a very important holder of state, and it is eventually passed to whoever needs access to any state. As a first step to untangle this, I could move ShellProcess out of utils into its own package, but I can’t find a way to do this in Structure101. This is most likely only possible in Restructure101.

Structure101 also has a standard Class Hierarchy viewer, to which I can navigate from the context menu of any class or interface represented in the dependency graph. For example, here’s `backtype.storm.spout.ISpout`:

Class hierarchy graph: backtype.storm.spout.ISpout

The Good

I feel like I’m navigating classes in Eclipse, but with a bird’s-eye-view. I can see what call what, but instead of jumping from one method to the other, I get lists that interact with my selection in the graph. It feels natural, and lets me explore the code base.

I have been looking at a fairly big project, after all. It is all foreign code to me, so that does affect how fast I can learn to use Structure101 effectively. I’d be interested to use the tool for a new project, seeing the dependencies evolve and provide architectural feedback during development. This introspection would certainly help me understand how to write without tangles from the start.

I have not tried to define an architecture diagram, but I can see how defining a target architecture can be useful in identifying tiny steps to get toward the goal. Moving classes around is relatively easy with a good IDE, but if I was to reduce the dependencies on the TopologyContext class, for example, I would need to keep track of that over a long period.

The Bad

Clojure generates many classes. This makes the experience of looking at class dependencies a bit difficult for that part of the code. However, the navigation and display shows packages first, hiding them unless they are part of a tangle.

Getting the most of the tool requires a bit of reading and trial and error. With experience, key observations would be faster to extract.

The Walkthrough guide refers to an older version of Structure101, and is difficult to follow. The user interface has changed, and there are dead links.

The right-click menu is broken, at least on the latest Ubuntu 12.10. The menu pops, but the mouse does not hightlight the items. Using the keyboard to navigate the contextual menu works, though.

I wish there were more keyboard bindings. Using arrows, enter, space, and context menu would help.


I won’t stop here. I’ll try Restructure101 to resolve problem I see in dependencies, in a « what if » mode of exploration. And the next step would be to use it on code I own and know well.

This is not something I would tell everybody to install, as the US$600 price tag is prohibitive for individuals – Restructure101 is even more expensive. The 30-day trial version can be useful for spot architectural work, but I’m glad I won a licence, as this will give me the opportunity to try it with various projects over a longer period of time.

Commentaires dans SmellyFish

J’avais parlé il y a longtemps de permettre aux visiteurs d’évaluer les recettes dans SmellyFish. Aujourd’hui, Andréanne m’a demandé s’il était possible de connaître le nom de celui qui a ajouté une recette. Je mémorisais déjà l’information, mais elle n’était pas affichée, alors c’était très facile à faire.

Et tant qu’à avoir les mains dedans, j’ai ajouté un commentaire textuel aux évaluations numériques. Prenez par exemple la recette de Filet de porc aux noix de pin .

To a man with a wiki, everything looks like a page

Certains ont eu la bonne idée de créer un wiki pour stocker des informations généalogiques, et j’ai fait la même chose. Il y a WikiTree et Rodovid, qui utilisent tous deux MediaWiki avec chacun une extension qui interprète les informations qui doivent être entrées dans un format bien particulier, dans une boîte texte. Je ne suis pas allé aussi loin avec mon installation de MoinMoin, car je ne crois pas que l’interface d’un wiki soit appropriée pour du travail intensif. Les programmes comme GRAMPS font un bon travail en présentant une interface rapide, facilitant l’entrée de données, et ayant une navigation soignée.

Ce que WikiTree et Rodovid reprochent à ces programmes, c’est qu’ils travaillent en « silos », sans permettre de bâtir une base de donnée mondiale. Ils peuvent certes exporter en GEDCOM, mais l’importation de multiples fichiers est un casse-tête que personne n’a vraiment résolu.

Ce qu’on voit, c’est des sites web en HTML, générés à partir du fichier GEDCOM, très peu ergonomiques, souvent laids et incompréhensibles, desquels on ne peut pas facilement extraire les données qui nous intéresse pour les importer dans notre base personnelle. Il y a aussi des sites qui permettent au public d’y déposer leur fichiers et aux visiteurs de rechercher les fichiers des autres. Ces sites ne bâtissent pas une base de donnée mondiale, mais offrent simplement un engin de recherche. Tout ça vient souvent avec des abonnements où l’on doit fournir un numéro de carte de crédit…

Une des solutions, celle adoptée par WikiTree et Rodovic, est d’utiliser un Wiki, et d’y ajouter des fonctionalités au-delà du langage de formattage. Dans une boîte texte, on peut donc écrire « Spouse = Jeanne Marchand ». Ça fonctionne plutôt bien, mais c’est très loin d’offrir une interface efficace pour de l’entrée de données.

Ce qu’il faut, c’est une application Web 2.0 pour la généalogie, utilisant donc HTML, CSS, JavaScript, DOM et Ajax, et qui soit collaborative, dans la ligne de pensée du Wiki, sans être un Wiki.

À partir de ce soir, plus que jamais, je chercherai les bases d’une telle application sur internet, si quelqu’un l’a déjà commencée. Comme bases, il existe le logiciel GRAMPS mentionné plus haut, écrit en Python et dont la structure de la base de données peut probablement être ré-utilisée. Il y a aussi The GENTECH Genealogical Data Model, une mine d’or de 100 pages. Pour le reste, je maintiens une liste de signets sur

Développer avec une équipe d’assurance-qualité

Durant le développement d’un projet, mon travail est testé par une équipe d’assurance-qualité, et ça se traduit par des interactions avec une personne en particulier. C’est la première fois que quelqu’un passe véritablement en arrière de moi pour vérifier que l’application fonctionne comme il le faut. Or, au fil du temps, j’ai noté l’effet que ça a sur mon propre développement.

J’avais commencé à écrire des tests unitaires, mais l’assurance-qualité du point de vue de l’utilisateur va encore plus loin. Ça met le focus sur les résultats, d’autant plus que je suis incité à mettre à jour l’environement de test plus souvent, dès que j’ai ajouté une nouvelle fonctionalité (quelques heures de travail, ou peut-être quelques jours).

En d’autres mots, je « release early, release often ».

Enfin, l’interaction fréquente avec l’équipe de test assouvit ce besoin presque vicéral que nous avons de montrer ce qu’on a fait à quelqu’un. Il ne faudrait pas que ça remplace les présentations régulières au client-utilisateur, mais il n’en reste pas moins que cette relation est très motivante, particulièrement pour améliorer les défectuosités de moindre importance, mais qu’on néglige trop souvent.