Lawgier.net

L’utilisation du data-mining dans la lutte contre la fraude fiscale : l'article 57 du PLF 2020

Sun, 08 Dec 2019 10:22:00 +0000

L’article 57 du projet de loi de finances (ci-après « PLF ») pour 2020, présenté par le Gouvernement le 27 septembre dernier prévoit la possibilité pour l’administration fiscale et douanière d’utiliser un procédé de datamining. Le datamining est un processus qui permet d’extraire des informations pertinentes à partir d’une certaine masse de données.

Le procédé n’est pas nouveau. La DGFiP utilise déjà depuis 2013 un « ciblage de la fraude et valorisation des requêtes » (CFVR) pour lutter contre la fraude à la TVA, qui a ensuite été élargi en 2016 pour viser les dirigeants d’entreprise.

La nouveauté tient ici au caractère général du ciblage : le procédé de lutte contre la fraude fiscale proposé se ferait à partir de données publiées par tous les internautes et librement accessibles sur les réseaux sociaux et les plateformes de mise en relation (leboncoin, Airbnb).

L’objectif présenté par le Gouvernement est « d’autoriser la collecte en masse » des données¹. Cela n’est pas sans poser d’interrogations en matière de respect de la vie privée, notamment à la lumière du premier anniversaire du RGPD. Comme l’indique la CNIL dans son avis du 12 septembre dernier, on assiste à un « changement d’échelle significatif »² des prérogatives confiées à Bercy dans le cadre du contrôle fiscal. La collecte de données, autrefois ciblée sur les fraudeurs suspectés, deviendrait généralisée.

Cette expérimentation serait mise en place pour une durée de trois ans et viserait les infractions relatives à la domiciliation fiscale et aux activités occultes.

Lors de son adoption en première lecture à l’Assemblée Nationale le 13 novembre 2019, l’article a été légèrement amendé.

Tout d’abord, le recours à la sous-traitance pour collecter et analyser les données a été proscrit. Si le souhait de Philippe LATOMBE, rapporteur spécial de la Commission des lois, était d’interdire tout recours à la sous-traitance, Gérald DARMANIN, Ministre de l’action et des comptes publics, a estimé que l’Administration n’était pas équipée pour concevoir seule le mécanisme de détection. Autrement dit, si l’Administration pourra sous-traiter la conception de l’algorithme, la collecte et l’analyse devra être directement réalisée par les services de Bercy.

Le sort des données sensibles, telles que les opinions politiques ou les convictions religieuses, pose également problème. L’article 9 du RGPD proscrit leur collecte, sauf à justifier, entre autres, d’un « motif d’intérêt public important ». Or, ces données ne sont pas nécessaires à la constatation d’une fraude fiscale, puisque le ciblage est généralisé. Dans ces conditions, comment justifier d’un « motif d’intérêt public important » ? La version initiale du texte permettait une conservation de ces données pendant 30 jours. L’Assemblée a réduit ce délai à 5 jours, ce qui ne résout pas l’incompatibilité du texte avec le RGPD.

Il faut aussi préciser que le rôle de la CNIL serait renforcé dans l’ensemble des étapes de contrôle de l’expérimentation.

Le Groupe LaREM a souhaité insister sur le caractère d’outil du système, au service des agents fiscaux. L’objectif est d’exclure la transformation de l’expérimentation en une machine de redressement fiscal automatisée. Toutefois, force est de constater que si l’outil n’induit pas nécessairement un redressement fiscal (il n’y a « que » signalement de « risque de fraude »), celui-ci repose sur une collecte généralisée des données de tous les citoyens, et non plus des seuls fraudeurs.

Par ailleurs, le Conseil d’État s’est interrogé sur la place de cet article dans le PLF. Dans un avis, publié par Nextinpact, la haute juridiction relève que le texte, ne créant pas de prérogatives spécifiques, ni de procédure nouvelle de contrôle ou de recouvrement, ne relève pas du domaine de la loi de Finance. Nous serions face à un cavalier législatif. En cas d’examen par le Conseil Constitutionnel, l’article 57 pourrait donc être écarté.

Le texte amendé a été adopté et transmis au Sénat, et sera discuté d’ici le 10 décembre.

Les sénateurs auront donc à étudier la pertinence de ce système et de l’opportunité de laisser Bercy libre d’exploiter les données des individus. Lors des débats à l’Assemblée, Gérald DARMANIN a insisté sur le fait que la lutte contre la fraude fiscale était un objectif à valeur constitutionnelle. Aux sénateurs de déterminer si cet objectif doit prévaloir sur le droit au respect de la vie privée des individus.

Article rédigé dans le cadre de mon stage dans le cabinet VIGO. Publié chez Dalloz Actualités

Projet de Loi de finances pour 2020, n°2272, 27 sept. 2019, p. 220-221. ↩︎
CNIL, Délibération n°2019-114 du 12 septembre 2019 portant avis sur le projet d’article 9 du projet de loi de finances pour 2020, p. 2. ↩︎

Intelligence artificielle : justice prédictive et réforme de la justice

Sat, 18 May 2019 12:30:00 +0000

Cet article se place dans une série de projets rédigés dans le cadre de mon Master 2 en Droit des activités numériques à Paris Descartes. Une version pdf de celui-ci est disponible en téléchargement.

La sécurité juridique est un principe directeur de l’État de droit. Comme le faisait remarquer le Professeur Boulouis, « la formule sonne en effet comme une sorte de redondance, tant il paraît évident qu'un droit qui n'assurerait pas la sécurité des relations qu'il régit cesserait d'en être un. »¹ Outre l’inflation législative, ce principe de sécurité juridique se heure parfois à la réalité des tribunaux et à la difficile évaluation de ses chances de succès en cas de litige. C’est ce à quoi la justice prédictive tente d’apporter une réponse.

Le Professeur Dondero définit la justice prédictive comme « des instruments d’analyse de la jurisprudence et des écritures des parties, […] qui permettraient de prédire les décisions à venir dans les litiges similaires à ceux analysés »². Elle se compose de deux éléments essentiels : une intelligence artificielle (IA), qui effectue des diagnostics de manière automatisée, à partir d’une base de données, qui compile un très grand nombre de décisions de justice.

Cette idée d’une analyse statistique des décisions de justice afin d’en tirer des probabilités n’est pas nouvelle. En effet, elle fait suite aux travaux de Simon-Denis Poisson, de Condorcet et de Laplace³.

Aujourd’hui, il existe plusieurs sociétés qui utilisent ces outils de justice prédictive. Par exemple, aux États-Unis, la société Legalist propose de financer des actions en justice qu’elle sélectionne grâce à un algorithme qui évalue les chances de succès et la durée probable de la procédure, en échange de la moitié des sommes attribuées par le juge⁴.

Si cette question est redevenue actuelle, c’est en grande partie grâce à la conjonction de deux éléments : à la fois, l’ouverture des données jurisprudentielles, en principe accessible librement depuis la loi Lemaire de 2016⁵, qui s’intègre au phénomène de big data ; et le développement de l’IA, en particulier celui de la technique de machine learning⁶. Cette conjonction a permis l’apparition de nouveaux acteurs : les legaltechs⁷, qui enrichissent à leur tour les données jurisprudentielles et le développement de ces IA.

L’intérêt de la justice prédictive serait de permettre aux justiciables de prévoir efficacement les conséquences d’un potentiel litige afin de se prémunir contre une potentielle action infructueuse ou pouvoir réaliser de meilleurs arbitrages économiques.

1. Les potentiels bénéfices de la justice prédictive

A. Permettre une meilleure accessibilité du droit

La justice prédictive répond tout d’abord⁸ à un problème de lisibilité et d’accessibilité du droit qui, souvent victime d’une inflation législative proposant de nouvelles normes peu lisibles, renforce de facto le pouvoir et la nécessité du juge d’interpréter la règle. Or, ces règles de second rang sont beaucoup moins accessibles que les règles dites de premier rang⁹.

En analysant automatiquement de grandes quantités de données, les outils de justice prédictive ont pour but de participer à remédier à ce problème de lisibilité. Leur analyse permettrait à la fois de formaliser les règles d’application de la norme, mais aussi de prévoir le traitement réservé par les juges à des cas particuliers¹⁰. Pour A. Garapon, contrairement à la statistique, qui approchait le problème « de manière trop générale », et aux juristes, qui ne pouvaient attendre ce résultat que par « un savoir intuitif », le big data permettrait de conférer une « consistance mathématique » à la prévision d’une décision de justice¹¹.

De même, le recours à des algorithmes, cantonnés au traitement des dossiers répétitifs, pourrait encourager le règlement de nombreux litiges en développant des modes alternatifs de règlement, afin de décharger les juges des tâches les plus chronophages¹². Cela pourrait permettre d’améliorer les temps et les conditions d’accès à la Justice, cruellement en manque d’efficacité en France.

B. Permettre une prise de décision efficace

L’un des principaux attraits de la justice prédictive, c’est d’optimiser le travail des juristes, clients, avocats et juges en anticipant certains résultats, afin « d’obtenir la meilleure rentabilité avec le minimum de moyens »¹³.

Grâce à ces outils, le justiciable pourrait évaluer de manière efficace les risques auxquels telle décision l’expose, mais aussi les chances de succès d’une procédure juridictionnelle et les moyens les plus pertinents à soulever lors de l’action¹⁴.

L’utilisation et le développement des outils de justice prédictive n’est cependant pas sans conséquence.

2. Les conséquences potentiellement néfastes de la justice prédictive

La soustraction progressive de l’humain dans la prise de décision, commune aux secteurs concernés par les avancées de l’IA, soulève de nombreuses questions et n’est pas sans conséquence. Néanmoins, il faut nuancer le développement potentiel de l’IA, qui est parsemé d’embuches.

A. Les problématiques techniques liées au développement de la justice prédictive

Si Antoine Garapon écrit que « le numérique permet de chiffrer la réalité, c’est-à-dire de la coder de telle manière que des réalités hétérogènes peuvent être lues et décryptées ensemble »¹⁵, concernant la justice prédictive, cette analyse est à la fois prospective et quelque peu problématique.

En effet, afin d’interpréter la loi (absconse) et les chances de réussite, les legaltechs se proposent de se reposer sur les décisions de justice qu’elles auront collectées, ce qui, contrairement au droit anglais, donne un rôle normatif à la jurisprudence, qui n’est normalement « qu’une application à des cas individuels de la règle de droit »¹⁶ et non pas la règle de droit¹⁷.

De plus, la modélisation a priori du raisonnement juridique, nécessaire pour pouvoir aboutir à une prévision la plus juste possible par la machine, risque de se heurter au manque de rationalité de certaines normes juridiques, à l’absence d’imbrication de certaines sources et à la très forte évolutivité des règles, rendant la tâche extrêmement ardue¹⁸.

Cette modélisation s’avère encore plus complexe en ce qui concerne le raisonnement des juges¹⁹. Comme l’indique X. Linant de Bellefonds, « la complexité du droit tient à son caractère téléologique et contentieux : deux raisonnements cohérents peuvent conduire à des arbitrages différents en fonction de deux ordres d’impératifs »²⁰. Cette absence de cohérence d’ensemble des décisions de justice²¹ révèle donc un problème capital : comment déduire de ce jeu de données un résultat probable avec une marge d’erreur acceptable ?

Enfin se pose la question de l’analyse de la conformité en droit de telle ou telle solution. En effet, l’IA, en sa conception actuelle, ne procède « qu’à des calculs de probabilité et ne [peut] pas discriminer entre les raisonnements juridiques valides et ceux qui ne le sont pas »²².

B. Les risques d’une prise de décision fondée sur une prédiction

Le risque principal de la justice prédictive est que la prévision donnée par l’outil influe sur la décision du juge²³. Cette influence, outre les problèmes qu’elle peut causer si la prévision faite ou les chiffres donnés sont faux ou biaisés, peut mener à une uniformisation de la pensée judiciaire.

Le résultat de l’analyse de la machine viendrait alors corrompre le jeu de données utilisé par celle-ci lorsque la décision prise sous son influence est ajoutée. Le biais initial de l’algorithme s’en trouverait alors renforcé, et renforcerait une sorte de cycle s’autoalimentant.

L’autre danger viendrait de l’incertitude de son fonctionnement²⁴. L’utilisation d’un tel outil, s’il tend à se développer, ne serait pas sans conséquences juridiques, puisque, à la suite d’une interrogation, une entreprise par exemple pourrait lancer une action en justice promise au succès par l’algorithme, mais couronnée d’échec et avec des conséquences financières graves.

Il est donc important de pouvoir savoir comment et pourquoi l’outil arrive à de telles conclusions afin de pouvoir véritablement apprécier le risque.

Conclusion

« There is no way that the law can avoid the scrutiny of science. If the lawyers and judges do not participate in this work, it will all be done by others »²⁵. Les propos de Lawlor, tenus en 1963 à propos des avancées scientifiques, correspondent parfaitement à la Révolution numérique et à ses conséquences sur l’ensemble de la société.

L’ensemble des mutations induites, notamment par la démocratisation de l’Internet, ont eu du mal à être appréhendées par beaucoup de juristes ; ce qui a laissé le champ libre à des hommes d’affaires, des mathématiciens et de férus d’informatique ; dotés d’une logique souvent très éloignée de celle du droit²⁶.

En se dessaisissant de ses problématiques, les professionnels du droit ont laissé apparaître des externalités négatives contre lesquelles ils ont peu d’outils pour les combattre. Le risque est de se retrouver confronté à une situation où la place du juge, face à l’efficacité de surface de la machine, voit son rôle remis en question.

En l’état actuel des choses, il ne semble pas souhaitable de prévoir une réforme de la justice en profondeur. Néanmoins, le bilan n’est cependant pas entièrement négatif et la justice prédictive dispose de certains atouts qu’il sera intéressant de développer, sous certaines conditions.

J. Boulouis, Quelques observations à propos de la sécurité juridique, Du droit international au droit de l'intégration : Liber amicorum, Pierre Pescatore, Nomos Verlag, 1987, p. 53. In J.G Huglo, La Cour de cassation et le principe de la sécurité juridique, [en ligne], Cahiers du Conseil constitutionnel n°11, déc. 2001. Disponible sur : https://www.conseil-constitutionnel.fr/nouveaux-cahiers-du-conseil-constitutionnel/la-cour-de-cassation-et-le-principe-de-la-securite-juridique . ↩︎
B. Dondero, Justice prédictive : la fin de l’aléa judiciaire ?, D. 2017, p. 532. ↩︎
Voir E. Barbin & Y. Marec, Les recherches sur la probabilité des jugements de Simon-Denis Poisson, in Histoire et Mesure, 1987, vol. 2, n°2, p. 39. In B. Dondero, art. prec., D. 2017, p. 533. ↩︎
J. Marin, Legalist, la start-up qui finance les plaints des entreprises, [en ligne], Le Monde, 29 août 2016. Disponible sur : https://www.lemonde.fr/economie/article/2016/08/29/legalist-la-start-up-qui-finance-les-plaintes-des-petites-entreprises_4989263_3234.html . ↩︎
Loi n°2016-1321 du 7 octobre 2016 pour une République numérique (dite également « Loi Lemaire »). ↩︎
Le machine learning, ou « apprentissage automatique » permet à un ordinateur d’apprendre de lui-même, en mettant à sa disposition un algorithme et une grande base de données étiquetées, lui permettant de reconnaître des données et de prendre des décisions en conséquence. ↩︎
Le terme de « Legaltech » vise les jeunes entreprises de type « Start-up » qui exercent dans le domaine du droit. ↩︎
Elle s’y heurte également, V. infra Section 2. ↩︎
A. Garapon, art. préc., §8. ↩︎
Eod Loc, §9. ↩︎
Ibid. ↩︎
J. M. Sauvé & S. Houiller, La justice prédictive, [en ligne], Colloque organisé à l’occasion du bicentenaire de l’Ordre des avocats au Conseil d’État et à la Cour de cassation, Cour de cassation, 12 février 2018, p. 2. Disponible sur : http://www.conseil-etat.fr/Actualites/Discours-Interventions/La-justice-predictive . ↩︎
A. Garapon, art. préc., §7. ↩︎
J. M. Sauvé & S. Houiller, art. préc., p. 2. ↩︎
A. Garapon, art. préc., §14. ↩︎
B. Dondero, art. préc., p. 535-536. ↩︎
Eod Loc., p. 536. ↩︎
Y. Meneceur, Quel avenir pour la « justice prédictive » ? Enjeux et limites des algorithmes d’anticipation des décisions de justice, JCP G 2018, 190, §15. ↩︎
Eod Loc., §16. ↩︎
X. Linant de Bellefonds, L’utilisation des systèmes experts en droit comparé, Revue internationale de droit comparé, 1994, vol. 46, n°2, p. 703-718. In Y. Meneceur, art. préc., §16. ↩︎
Y. Meneceur, art. préc., §16. ↩︎
Eod Loc., §17. ↩︎
B. Dondero, art. préc., p. 537-538. ↩︎
Eod Loc., p. 537. ↩︎
R. C. Lawlor, What computers can do: analysis and prediction of judicial decisions, American Bar Association Journal, 1963, 49, p. 339. In B. Dondero, art. préc., p. 533. ↩︎
A. Garapon, art. préc., §1-2. ↩︎

Intelligence artificielle et responsabilité

Sat, 18 May 2019 12:00:00 +0000

Depuis l’adoption de l’agriculture, la densité et la taille de la population s’est accrue. Cet accroissement a entrainé deux choses : l’augmentation du nombre d’individus, qui conduit à une augmentation du nombre d’idées, et la concentration croissante, qui permet à la fois à ces idées de se transmettre plus facilement et aux personnes de se spécialiser¹. Ces mutations ont conduit à une accélération du développement humain, entrainant les différentes révolutions industrielles et aujourd’hui notre entrée dans l’ère numérique.

Le développement des capacités de calcul permises par la « loi de Moore »² a entrainé l’apparition de nouvelles possibilités de traitement automatisé de tâches et de nouvelles méthodes. L’intelligence artificielle, que le chercheur Yann LeCun³ définit simplement comme « faire faire aux machines des activités qu’on attribue généralement aux animaux ou aux humains »⁴, s’extrait progressivement de la science-fiction pour devenir réalité.

Cette évolution de la machine est permise en particulier grâce au développement de méthodes telles que le machine learning et le deep learning, crée dans les années 1990 et revenue sur le devant de la scène en 2012⁵. Le machine learning, ou « apprentissage automatique » permet à un ordinateur d’apprendre de lui-même, en mettant à sa disposition un algorithme et une grande base de données étiquetées, lui permettant de reconnaître des données et de prendre des décisions en conséquence.

Cette prise de décisions automatisée n’est cependant pas sans conséquences. Au départ plutôt bégnines, l’augmentation des capacités des machines ou robots⁶ fait que les décisions prises deviennent de plus en plus importantes et de plus en plus incontrôlées. Un exemples est celui du high frequency trading, où les ordinateurs prennent des décisions de placement financiers dans un laps de temps de quelques microsecondes⁷. Autre exemple, les voitures autonomes, qui approchent leur phase de commercialisation dans certaines villes des États-Unis⁸.

Cette prise de décisions automatisée emporte donc certaines interrogations juridiques. Si le robot agit seul, qui est responsable de ses actes ? Les régimes de responsabilité actuels conviennent-t-ils à ce cas de figure ? Quel régime de responsabilité pour l’intelligence artificielle ?

Cette question se pose, d’autant plus que le droit français a des difficultés à appréhender cette question⁹. Dans un arrêt de 2013, la Cour de cassation avait exclu la responsabilité de Google car « la fonctionnalité aboutissant au rapprochement critiqué est le fruit d’un processus purement automatique dans son fonctionnement et aléatoire dans ses résultats, de sorte que l’affichage qui el résulte est exclusif de toute volonté de l’exploitant du moteur de recherche d’emmètre les propos en cause »¹⁰.

Pour répondre à cette question, une analyse en deux temps sera réalisée. D’abord, il faudra voir si le droit commun de la responsabilité est apte à envisager les décisions prises par une machine, puis, s’il est nécessaire de créer une personnalité juridique pour les robots.

1. Intelligence artificielle et droit commun de la responsabilité

A. Intelligence artificielle et responsabilité pour faute

Le cadre classique de la responsabilité civile est celui de la responsabilité pour faute, prévu à l’article 1240 du Code civil. Il suppose 3 choses : une faute, un préjudice et un lien de responsabilité.

Il semble néanmoins très difficile d’appliquer ce cadre à l’intelligence artificielle. En effet, il faudra démontrer une faute, qu’elle soit intentionnelle ou non, dans la programmation du robot ou dans son utilisation¹¹. Si le robot commet une faute en prenant une décision sans qu’il y ait un comportement litigieux de la part du propriétaire/utilisateur ou du fabriquant, l’article 1240 ne trouvera pas à s’appliquer.

B. Intelligence artificielle et responsabilité du fait des choses et d’autrui

La responsabilité prévue par l’article 1240 étant exclue, il faut envisager celle de l’article 1242 alinéa 1^er^ du Code civil, qui prévoit la responsabilité du fait des choses et d’autrui. Cette responsabilité permet de « faire remonter la responsabilité vers la personne qui dispose d’une autorité sur le responsable de la faute ou du fait dommageable »¹². Elle emporte la responsabilité du fait des préposés, du fait des enfants ou du fait des animaux¹³. Cette responsabilité du fait d’autrui se justifie néanmoins par « le pouvoir de contrôle du gardien », qui intervient pour « juguler les écarts de comportement » de ce qu’il a sous son contrôle¹⁴.

Cette responsabilité ne semble également pas convenir à l’intelligence artificielle, à la fois parce qu’elle a été conçue pour des objets corporels et s’adapte mal à l’immatérialité de l’intelligence artificielle¹⁵, mais aussi car elle semble aller à l’encontre de la fonction consubstantielle à toute chose autonome : « servir l’homme en prétendant le soulager d’un contrôle qui […] lui reviendrait sinon »¹⁶, la relation entre homme et robot allant de plus en plus vers une situation où l’homme se repose sur le contrôle du robot plutôt que l’inverse¹⁷, et où le robot a de plus en plus tendance à s’émanciper de son gardien¹⁸.

C. Intelligence artificielle et responsabilité du fait des produits défectueux

Le régime de responsabilité du fait des produits défectueux, issu d’une directive européenne¹⁹, est un régime « particulièrement favorable à la victime, qui n’a pas besoin de rapporter la preuve d’une faute du producteur »²⁰. Pour engager la responsabilité du fabriquant, la victime devra apporter la preuve d’un dommage, de la défectuosité d’un produit, et un lien de causalité entre les deux.

S’il semble que ce régime ne pose pas véritablement de difficultés à s’appliquer à l’intelligence artificielle²¹, se pose la question de l’exonération du producteur. Si le fabriquant prouve « que l’état des connaissances scientifiques et techniques, au moment où il a mis le produit en circulation, n’a pas permis de déceler l’existence du défaut »²², il pourra s’exonérer de sa responsabilité, ce qui empêche l’indemnisation de la victime du comportement du robot. Même si les hypothèses d’exonération sont faibles²³, elles restent applicables au cas de l’intelligence artificielle, dont l’amélioration constante et automatique pourra conduire à des cas où le robot échappera à son fabricant.

Ces régimes de responsabilité ne sont donc pas suffisants si l’on considère que les robots, apprenant sans l’intervention de l’homme, prennent des décisions qui ne pouvaient être anticipées par ses concepteurs, les faisant entrer dans un vide juridique si la décision prise entraine un dommage à un tiers. Il faut donc envisager d’autres solutions.

2. La création d’une personnalité juridique pour les robots

Outre la création d’un nouveau régime de responsabilité pour l’intelligence artificielle, une autre possibilité, proposée par des auteurs²⁴ et reprise par le Parlement européen, serait de créer, à l’image des sociétés et non pas à celle de l’homme²⁵, une personnalité juridique propre aux robots.

Dans un Résolution du 16 février 2017 concernant les règles de droit civil sur la robotique²⁶, le Parlement européen a demandé à la Commission européenne d’examiner la possibilité de créer « à terme, une personnalité juridique spécifique aux robots, pour qu’au moins les robots autonomes les plus sophistiqués puisse être considérés comme des personnes électroniques responsables, tenues de réparer tout dommage causé à un tiers »²⁷. De ce fait, le robot deviendrait responsable, ce qui lui permettrait de passer des contrats, notamment des contrats d’assurance²⁸.

A. Une proposition contestée

Cette proposition est cependant risquée, certains auteurs dénonçant une à la fois un manque d’intérêt et une possible déresponsabilisation des utilisateurs et des fabricants des robots²⁹. Manque d’intérêt car la réparation ne serait réellement garantie que par un contrat d’assurance qui peut être souscrit « directement par l’utilisateur/propriétaire du robot sans qu’il soit utile de faire du robot une personne juridique qui devrait s’assurer contre son propre risque »³⁰.

Déresponsabilisation, car les utilisateurs et fabriquant ne seraient plus incités à concevoir et utiliser des robots non-dangereux si « leur responsabilité personnelle pouvait être écartée au profit de celle des robots »³¹ et priverait « d’effet correctifs préventifs découlant du droit de la responsabilité civile »³² dès lors que « le fabriquant n’assumera plus le risque de responsabilité, celui-ci ayant été transféré au robot »³³.

B. Une proposition intéressante

La création d’une personnalité juridique pour les robots n’est cependant pas totalement dénuée de sens. En effet, elle n’exclue pas totalement la responsabilité des autres acteurs. En effet, « en cas de faute de conception, de mise à jour ou d’utilisation, le fabricant, le concepteur le propriétaire ou encore l’utilisateur du robot pourront toujours voir leur responsabilité engagée pour faute »³⁴.

De plus, le Parlement européen suggère, pour la chaine de responsabilité, qu’elle soit proportionnée au niveau d’instructions données, permettant de distinguer la période de formation par l’homme de celle où le robot se serait formé lui-même³⁵.

Conclusion

Même si l’augmentation des capacités de l’intelligence artificielle est rapide et le nombre de taches qu’il est possible de réaliser avec s’accroit, il semble cependant trop tôt, en l’état actuel des choses, de se positionner définitivement³⁶. Dans sa résolution, le parlement européen considère d’ailleurs que de nouveaux instruments ne pourront être mis en place qu’à un horizon de dix à quinze ans³⁷. La pratique, notamment assurantielle, permettra de se positionner de manière plus ferme sur « le point d’ancrage de la couverture du risque »³⁸.

Il est peut-être donc intéressant, afin d’anticiper les risques, de privilégier aujourd’hui une autorégulation³⁹ des robots pour anticiper certains risques, tout en réfléchissant à un cadre juridique pérenne. À moins que l’intelligence artificielle soit une illusion, ou, à l’inverse, surpasse celle de l’homme. Dans ce cas, la question de la supériorité plutôt que de la responsabilité devrait se poser.⁴⁰

N. Bostrom, Superintelligence – Paths, danger, strategies, Oxford University Press, 2016, p. 2. ↩︎
G. E. Moore, Cramming more components onto integrated circuits, [en ligne], Electronics, vol. 38, n° 8, 1965. ↩︎
Yann LeCun est un chercheur en intelligence artificielle, responsable du laboratoire d’apprentissage profond (FAIR) chez Facebook. ↩︎
R. Maggiori, Préface, p. 13-14. In : Collectif, Intelligence artificielle – Enquête sur ces technologies qui changent nos vies, Champs actuels, 2018. ↩︎
Idem, p. 15. ↩︎
S. Dormont, Quel régime de responsabilité pour l’intelligence artificielle ?, Comm. com. élect. 2018, n° 11, étude 19, p. 1. ↩︎
A ce propos, voir M. Lewis, Flash boys – A Wall Street revolt, Norton, 2014, USA ↩︎
K. Korosec, Waymo is testing what it should charge for its robotaxi service, [en ligne], Techcrunch, 26 octobre 2018. ↩︎
L. Archambault, L. Zimmermann, La réparation des dommages causés par l’intelligence artificielle : le droit français doit évoluer, Gaz. Pal. 6 mars 2018, n°315c3, p. 17. ↩︎
Cass, 1^ère^ civ, 19 juin 2013, n° 12-17.591. ↩︎
L. Archambault, L. Zimmermann, v. supra. ↩︎
G. Courtois, Robots intelligents et responsabilité : quels régimes, quelles perspectives ?, D. IP/IT 2016, p. 289 ↩︎
C. Civil, art. 1243. ↩︎
P. Brun, La responsabilité du fait des objets connectés. In : Lamy Droit de la responsabilité, 2018, n°350-60. ↩︎
L. Archambault, L. Zimmermann, v. supra. ↩︎
Lamy Droit de la responsabilité, 2018, n°350-60, v. supra. ↩︎
Idem. ↩︎
S. Dormont, v. supra. ↩︎
Directive 85/374/CEE du Conseil du 25 juillet 1985 relative au rapprochement des dispositions législatives, réglementaires et administratives des États membres en matière de responsabilité du fait des produits défectueux. ↩︎
S. Dormont, v. supra. ↩︎
S. Dormont, v. supra. ↩︎
C. Civ, art. 1245-10, 4°. ↩︎
CJCE, 29 mai 1997, aff. 300/95. Comme l’indique S. Dormont (v. supra) : « *Pour la Cour de justice, l’état des connaissances à prendre en compte n’est pas “l’état des connaissances dont le producteur en cause est ou peut être concrètement ou subjectivement informé, mais l’état objectif des connaissances scientifiques ou techniques dont le producteur est présumé informé” (arrêt, pt 27). La juridiction renvoie donc à un état des connaissances mondiales et “au niveau le plus avancé tel qu’il existait au moment de la mise en circulation du produit en cause”(arrêt, pt 26) *» ↩︎
A. Bensoussan, Plaidoyer pour un droit des robots : de la « personne morale » à la « personne robot », LJA 2013, n° 1134. ↩︎
A. Bensamoun, G. Loiseau, L’intégration de l’intelligence artificielle dans l’ordre juridique en droit commun : questions de temps, D. IP/IT 2017, p. 239. ↩︎
Résolution 2015/2103(INL) du Parlement européen du 16 février 2017 contenant des recommandations à la Commission concernant des règles de droit civil sur la robotique. ↩︎
Idem, § 59, f. ↩︎
A. S. Choné-Grimaldi, P. Glaser, Responsabilité civile du fait du robot doué d’intelligence artificielle : faut-il créer une personnalité robotique ?, Comm. com. élect. 2018, n° 1, focus 1, p. 2. et G. Loiseau, M. Bourgeois, Du robot en droit à un droit des robots, JCP G 2014, 1231, p. 2164. ↩︎
Lamy Droit de la responsabilité, n°350-60. ↩︎
G. Loiseau, M. Bourgeois, idem. ↩︎
Id. ↩︎
Comité économique et social européen (CESE), Avis d’initiative INT/806 sur l’intelligence artificielle – Les retombées de l’intelligence artificielle pour le marché unique (numérique), la production, la consommation, l’emploi et la société, 31 mai 2017, p. 11. ↩︎
Idem. ↩︎
A. S. Choné-Grimaldi, v. note supra. ↩︎
Résolution 2015/2103(INL) du Parlement européen, v. note supra, pt. 56*.* ↩︎
A. Bensamoun, G. Loiseau, v. supra. ↩︎
Résolution 2015/2103(INL) du Parlement européen, v. note supra, pt. 51. ↩︎
A. Bensamoun, G. Loiseau, v. supra ↩︎
Idem. ↩︎
N. Bostrom, v. supra. ↩︎

La Blockchain dans les services financiers

Sat, 18 May 2019 11:30:00 +0000

La dématérialisation des transactions a entrainé comme conséquence principale l’augmentation de la dépendance des utilisateurs aux établissements financiers. Cependant, plusieurs crises financières et un certain sentiment de méfiance envers le secteur ont conduit à une volonté de s’affranchir de ce contrôle. C’est dans ce contexte que la blockchain a réussi à se développer, principalement grâce à la création du bitcoin en janvier 2009¹.

Comme beaucoup de concepts issus de l’informatique et de l’Internet, c’est la locution anglaise qui est préférée pour désigner ce système, au grand dam de la Commission d’enrichissement de la langue française, qui lui préfère le terme de « chaîne de blocs »².

La blockchain peut être définie comme « une base de données décentralisée et sans intermédiaires qui permet d’automatiser une transaction, de l’authentifier et de l’horodater, tout en garantissant son immuabilité et son inviolabilité. Elle peut aussi assurer la confidentialité des données grâce au cryptage (sic) »³.

Le principal avantage de la blockchain, à la différence d’une base de données « classique », est qu’elle permet de s’assurer que les informations qui sont inscrites n’ont pas été modifiées par quelqu’un a posteriori, grâce à l’utilisation de procédés de chiffrements et à la présence d’une multitude de détenteurs de « blocs » d’informations.

Dans le cas d’une base de données classique, centralisée, il est possible de modifier ou de supprimer les informations enregistrées a posteriori, et donc de falsifier ces données. Il est donc nécessaire de faire confiance à la personne qui détient et gère la base de données, qui sera par essence souvent un tiers à la relation. La blockchain permet de se passer de la nécessité de faire confiance à un tiers.

La blockchain, ou la distributed ledger technology (DLT – technologie des registres distribués) a comme particularité de pouvoir être utilisé pour plusieurs types d’opérations : monnaie électronique, première utilisation à « grande échelle » avec le Bitcoin⁴ ; service de paiement ; mais aussi comme système de levée de fond : les ICO, qui font concurrence aux levées de fond menées auprès d’investisseurs institutionnels et traditionnels, et aux introductions en bourses.

Cette technologie fait donc concurrence aux principales activités de la Banque, qui est la colonne vertébrale des économies de marché et un secteur très réglementé. Si, parce que la technologie est relativement récente, elle paraît peu encadrée, le législateur commence néanmoins à se saisir du sujet, notamment avec le projet de loi dit « PACTE »⁵.

On peut alors se poser plusieurs questions : La DLT fait-elle peser une véritable menace sur le système bancaire ? Le cadre juridique des services financiers est-il adapté à cette mutation ? En somme, une réponse légale doit-elle être apportée à la blockchain ?

Nous verrons dans un premier temps quels sont les apports technologiques de la Blockchain dans le secteur financier pour pouvoir, dans un second temps, analyser l’encadrement juridique actuel et futur.

1. Les mutations technologiques provoquées par la blockchain

A. Les transactions financières

a. L’absence de tiers de confiance

La blockchain, en raison de sa nature distribuée, permet de s’affranchir de certains intermédiaires autrefois indispensables pour assurer certains échanges. En effet, il n’est plus nécessaire de passer par un établissement financier pour effectuer des transactions financières de manière dématérialisée⁶. Certaines versions de la blockchain permettent aussi de se passer de tiers de confiance tels qu’eBay ou PayPal pour réaliser certaines transactions : c’est le cas des contrats intelligents qui s’auto exécutent lorsque les conditions établies par les cocontractants se réalisent⁷.

b. La traçabilité des transactions

La blockchain peut également présenter certains intérêts pour les banques centrales dans le cadre de leurs activités⁸ ⁹. Elle permet de mieux retracer les transactions en pouvant suivre précisément chaque euro … en circulation, notamment grâce à un registre unique de transactions ; simplifier le processus de règlement et de livraisons de titres financiers, tels que les bons du Trésor ; une réduction des coûts de transmission et d’exploitation ; et améliorer la lutte contre le blanchiment d’argent.

B. Les cryptomonnaies

a. Le bitcoin, premier bien numérique « unique »

Si la blockchain permet de décentraliser le processus de confiance, elle permet également d’apporter aux biens numériques une caractéristique qu’ils n’avaient pas auparavant : le caractère « unique ». En effet, les biens numériques ont pour caractéristique de pouvoir être dupliqués à l’infini sans perte d’information, contrairement aux supports « traditionnels »¹⁰. C’est cette caractéristique qui a permis l’exposition de la contrefaçon d’œuvres, puisque celles-ci pouvaient être librement partagées sans coût.

Le bitcoin et la blockchain permettent de changer cela et de rendre un bien numérique « unique » et non reproductible, ce qui a un véritable intérêt dans le cadre du secteur de la finance : un titre doit être unique, tout comme la monnaie, afin de pouvoir s’assurer que le phénomène de double dépense ne se produise pas¹¹.

b. Cryptomonnaies et système monétaire traditionnel

Le Bitcoin, comme les autres cryptomonnaies basées sur une blockchain, représente un tournant dans l’approche « classique » de la monnaie, émise traditionnellement par l’État, qui dispose d’un pouvoir régalien sur la « frappe » de la monnaie.

Même si la monnaie-fiat comme l’Euro n’a pas de valeur intrinsèque¹² et peut donc être émise dans des quantités arbitraires, cet aspect régalien et unique de la monnaie (en France, seul l’Euro a cours légal¹³) est remis en cause par les cryptomonnaies, émises de manière décentralisée et sous de nombreuses variantes.

Ces cryptomonnaies ne sont d’ailleurs pas envisagées par le Code monétaire et financier (CMF), car elles ne rentrent pas dans la définition de monnaie électronique donnée à l’article L. 315-1, puisqu’elles ne sont pas « émises contre la remise de fonds aux fins d’opérations de paiement », mais contre la validation des transactions des utilisateurs de la blockchain de la monnaie en question.

C. Les Initial Coin Offerings (ICO)

L’Initial Coin Offering (ICO) peut être défini comme « une opération de levée de fonds par offre au public, donnant lieu à une émission de jetons numériques (ou ‘tokens’) destinée à financer le développement d’un projet par l’émetteur »¹⁴.

L’ICO tire son nom d’« IPO », l’Initial Public Offering ou introduction en bourse en anglais, dont elle s’inspire et concurrence de plus en plus aujourd’hui.

Maintenant que les apports techniques de la Blockchain ont pu être développés, il faut se pencher sur la régulation, actuelle et future, de la technologie de chaîne de blocs.

2. La régulation de la blockchain

A. Cadre juridique actuel

Même si la blockchain a entrainé la création de nouvelles pratiques, il n’en reste pas moins que ses produits restent soumis à la règlementation sur les instruments financiers en vigueur¹⁵, les régulateurs des marchés financiers faisant une approche au cas par cas¹⁶.

Concernant les ICO, il est difficile de les intégrer dans le cadre légal actuel¹⁷. C’est pour cela que l’AMF a, après consultation¹⁸, décidé d’opter pour un régime d’autorisation optionnel des ICO¹⁹.

B. Cadre juridique à venir

a. La loi PACTE de 2018

Ces vides juridiques tendent cependant à être résorbés par le législateur. C’est à l’occasion du projet de loi PACTE²⁰ qu’a été intégrée une disposition sur les ICO, qui reprend la position de l’AMF en délivrant des visas optionnels²¹.

Ce projet reste cependant insuffisant sur de nombreux points. En effet, il n’est pour l’instant pas prévu d’instaurer un régime fiscal et comptable pour ces opérations ; et la délivrance d’un visa ralentirait le lancement de l’ICO, dont la rapidité de mise en œuvre est l’un de ses principaux avantages²².

b. Discussions en cours

L’Ordonnance du 8 décembre 2017²³ prévoit la possibilité de transmettre des titres financiers par le biais de la DLT²⁴. Sont concernés les titres de créance négociables, les parts ou actions d’organismes de placement collectif, et les actions et obligations non cotées²⁵. Cette Ordonnance n’a cependant, à ce jour, pas été ratifiée par le Parlement²⁶.

Conclusion

Si la technologie blockchain peut être considérée comme une véritable valeur ajoutée face aux anciens modes électroniques de conservation et de transmission de titres financiers, il faut néanmoins relativiser certains de ses apports.

Même si les ICO rencontrent un succès fulgurant²⁷, leur intérêt, dans un cadre juridique protecteur des investisseurs, reste relativement concis. Techniquement parlant, la blockchain du Bitcoin, qui est la plus utilisée aujourd’hui, reste, pour gérer des transactions de la vie courante, inefficiente face aux infrastructures d’ores et déjà en place (en particulier celles de Visa et de MasterCard). Enfin, l’utilisation de cryptomonnaies en tant que véritable remplaçant des devises légales telles que l’euro n’est, aux vues de la relative fluctuation de leur valeur, toujours pas viable.

Cependant, même si la prudence reste de mise, les insuffisances techniques risquent d’être rapidement comblées. Il est donc important que le secteur financier et le législateur restent attentifs afin de se préparer à une mutation qui, comme celle lancée par l’Internet, semble de plus en plus inévitable.

B. Wallace, The rise and fall of bitcoin, [en ligne], Wired.com, 23 novembre 2011. ↩︎
JORF n°0121 du 23 mai 2017, texte n° 20. ↩︎
B. Barraud, Les blockchains et le droit, RLDI 2018/147, n°5206. ↩︎
S. Nakamoto, Bitcoin: A Peer-to-Peer Electronic Cash System, [en ligne], 2008. ↩︎
TA AN, 2017-2018, Projet de loi relatif à la croissance et la transformation des entreprises, n° 1088. ↩︎
« Une version purement électronique de l’argent électronique permettrait aux paiements en ligne d’être envoyés directement d’une partie à l’autre sans passer par une institution financière » : S. Nakamoto, id., p. 1. ↩︎
V. Buterin, Ethereum White Paper, [en ligne], The Ethereum Foundation, 2014. ↩︎
C. R. W. De Meijer, Blockchain and Central banks : a Tour de Table, Part 1, [en ligne], finextra.com, 3 janvier 2017. ↩︎
H. de Vauplane, Blockchain, cryptomonnaies, finance et droit : état des lieux, RLDA 2018/140, n°6526. ↩︎
B. Thompson, The cost of Bitcoin, [en ligne], stratechery.com, 3 mars 2014. ↩︎
S. Nakamoto, id., p. 2. ↩︎
H. de Vauplane, *id., *p. 7. ↩︎
Code monétaire et financier, article L. 111-1. ↩︎
J. Brosset & A. Barbet-Massin, La souscription de cryptoactifs et de jetons d’ICO : les recours des investisseurs, RLDA 2018/140, n°6531. ↩︎
Iris M. Barsan, Initial Coin Offerings : Comment réguler l’émission de cryptojetons ?, Revue Banque n°821bis, p. 61. ↩︎
AMF, Synthèse des réponses à la consultation publique portant sur les Initial Coin Offerings (ICO) et point d’étape sur le programme “UNICORN”, [en ligne], 22 février 2018, p. 5. Disponible sur : https://www.amf-france.org/Publications/Consultations-publiques/Archives?docId=workspace%3A%2F%2FSpacesStore%2Fa9e0ae85-f015-4beb-92d2-ece78819d4da (consulté le 4.11.2018). ↩︎
Iris M. Barsan, id. ↩︎
AMF, Discussion Paper on Initial Coin Offerings (ICOs), [en ligne], 26 oct. 2017. Disponible sur : https://www.amf-france.org/Publications/Consultations-publiques/Archives?docId=workspace%3A%2F%2FSpacesStore%2Fa2b267b3-2d94-4c24-acad-7fe3351dfc8a (consulté le 3.11.2018). ↩︎
AMF, Synthèse des réponses à la consultation publique portant sur les Initial Coin Offerings (ICO) et point d’étape sur le programme “UNICORN”, [en ligne], 22 février 2018, p. 5. ↩︎
TA AN, 2017-2018, id.[ ↩︎
R. A. Ozcan Les tokens pourraient devenir les titres financiers de demain, RLDA 2018/140, n°6527, p. 17. ↩︎
R. A. Ozcan, id., p. 17. ↩︎
Ordonnance n° 2017-1674 du 8 décembre 2017 relative à l’utilisation d’un dispositif d’enregistrement électronique partagé pour la représentation et la transmission de titres financiers. ↩︎
M. A. de Montlivault-Jacquot, Ordonnance Blockchain sur la transmission de titres financiers, [en ligne], alain-bensoussan.com, 27 décembre 2017. ↩︎
Compte rendu du Conseil des ministres du 30 mai 2018. ↩︎
Projet de loi ratifiant l’ordonnance n° 2017-1674 du 8 décembre 2017 relative à l’utilisation d’un dispositif d’enregistrement électronique partagé pour la représentation et la transmission de titres financiers, n° 997, déposé le 30 mai 2018. ↩︎
Selon le site Coinschedule.com, environ 21,5 milliards de dollars US auraient été investis sur les 10 premiers mois de 2018, contre 6,5 milliards de dollars US en 2017 : Coinschedule, Cryptocurrency ICO Stats, [en ligne], coinschedule.com. ↩︎

Is your brain really like a computer?

Sun, 22 May 2016 19:02:59 +0000

As I’m reading Nick Bostrom’s Superintelligence, I found this article from Aeon¹ about the brain and the bad analogy it suffers from computers.

In it, Robert Epstein writes that brains are not at all like computers, but their own organic things. For example, they don’t store information like computers: the author asked people to draw from memory 1$ bills and then to draw them with a reference. The results vastly differ, the first draw being very sparse and the second much more elaborate², showing that our brain has trouble to store that kind of information, unlike computers who don’t have any problem to do it.

Then, the author moves to a different thing: the idea of downloading your brain to a computer. And this is where the article rang a bell: this is a subject that Bostrom talks about in his book, but focusing mainly on the technical difficulties of doing it: how to scan a brain down to the protein scale, how to get the sufficient processing power to run it, …

But here, the author makes a totally different point: you just can’t do it.

This is because your brain and a computer aren’t the same things at all, not dealing with the same kind of things : you can’t store memories like a computer can, as you don’t deal with situations like a computer does: when you walk, you don’t calculate the size between your steps, the inclination of your legs, your balance, et cetera. You just do it; as a computer has to all of those things³.

The author then points out to Stephen Rose book⁴ : The future of the brain, who shows that to understand the current state of the brain, you have to get its entire history and the social context of its owner, because you can’t just replicate it like a digitized file : by itself the brain is just cells (neurons) interconnected (synapses), and cells can’t just store things.

There is a piece missing in our understanding of our brain — something that we will probably not get until a century⁵ — that make things work. But thinking that they work like computers could refrain it.

Thanks to Farnam Street “Brain Food” mailing list. You should subscribe to it. ↩︎
You should check the illustration on the article ↩︎
https://www.youtube.com/watch?v=rVlhMGQgDkY ↩︎
That I didn’t read and don’t know the worth. But it has good ratings on Amazon! ↩︎
I have some doubts about this, but first, I have to finish Bostrom book. ↩︎

Rich Cohen - The Fish That Ate The Whale

Tue, 10 May 2016 19:21:14 +0000

You may have casually heard the expression “Banana Republic” in a conversation, understanding vaguely its meaning but not really grasping where does it comes from, or why it does have this connotation.

If you take a dictionary, you would find this definition :

A small nation, especially in Central America, dependent on one crop or the influx of foreign capital.

Well, okay, but why is it used in a pejorative way? And why “Banana”, not something like a “Coffee Republic” or a “Sugar Cane Republic”?

This is because you didn’t get the whole story.

Meet Sam “The Banana Man” Zemmuray.

In his life, Samuel “The Banana Man” Zemmuray witnessed the wake of the first multinationals, the rise and the instability of Central America, the first and second world war, the cold war and the wake of the CIA. But he did not only watch those things happen, he was actively involved in them.

And this is much more impressive when you consider that he began as a peasant, and went single-handedly to the top of the world, being the only person competent in managing the biggest corporation of his time and even a kingmaker in some countries.

Now is the moment you’re wondering why you never heard of Zemmuray before. Don’t worry, you’re not the only one.

I read about him first in Mastery I believe, being presented “The fish that ate the whale”, the guy who started selling a bunch ripe bananas thrown away by United Fruit and ended up owning and controlling this same company, the behemoth who started the trade.

But Zemmuray life isn’t limited to this feat. He made the banana the most consumed fruit in the world, produced and sold at dirt cheap prices, even if it was a luxury item at the start of the trade. He also deeply transformed South America, being at the origin of the gringo’s reputation, and going so far as indirectly starting the Cuban revolution. When you see what the CIA did in the cold war in Latin America, you can see his shadow behind, trying to protect his interests, going so far as handling the smuggling of the weapons to Honduras and the Bay of Pigs. If you go further east, you’ll still find his fingerprint in Israël, a country he heavily helped to create.

What you will find in this book is the life of a man who wanted everything and got more, a man who stayed true to his principles, ho was willing to get his hands dirty to achieve the things that he wanted, finding solutions that no one could have thought, circumventing the problems when they couldn’t be directly torn down.

If you didn’t know Sam Zemmuray, now is the time.

Buy this book. This article is part of a series called The Book Corner, where I post short reviews of books that I found on the internet and thought they were worth sharing. If you’re interested, you can find more on this page.

As a young foreign student, I’m looking to perfect my writing. If you have some time to give me feedback about this article or anything else, please reach out on Twitter or via email. This is greatly appreciated. Thanks!

John Vaillant - The Golden Spruce

Sun, 08 May 2016 12:27:13 +0000

The first work from John Vaillant that I heard of was The Tiger, a story about the last great tiger in Siberia and of the mens who killed it. If the story looks interesting by itself, what was the most striking in this book was the amount of background and the depth of information in it; not only you learn about the story of how this beast was killed and why, but also the mindset of the people in this remote part of the world, the psychology of the animal, and much much more.The first work from John Vaillant that I heard of was The Tiger, a story about the last great tiger in Siberia and of the men who killed it. If the story looks interesting by itself, what was the most striking in this book was the amount of background and the depth of information in it; not only you learn about the story of how this beast was killed and why, but also the mindset of the people in this remote part of the world, the psychology of the animal, and much much more.

So that’s why, when I stumbled upon The Golden Spruce, I knew I’ll spend a great time.

The Great Spruce is Vaillant first book. The story is about why this tree, an oddity grown in the Queen Charlotte Island in British Columbia, Canada, was felled by Grant Hadwin, a logger-turned environmentalist, who wanted to make a point about the practices used by big logging companies like clearcutting and raise awareness on the subject.

But, like other Vaillant’s books, The Golden Spruce is way more than that. It’s a thorough account of the logging industry, from the first English settlers to nowadays; a description of the state of Native Americans in the US and Canada; but also a depiction of the North American rainforest, or what it’s left of, and it’s population: the biggest and tallest trees on earth, the ferocity of the elements, it’s diversity, …

This is why I love Vaillant’s books. He goes way deeper than the initial story to give a proper explanation of the event: the context, the mindset of its actors, … And because it’s linked to the initial story, it’s compelling : you want to find more with him: why this tree was so special for the island inhabitants, why the logging industry is so aggressive, why Hadwin thought it was a good idea to do it, who he was and how he did it, …

After I’ve finished it, I was happy to find that I could get the same kind of feeling that I got last summer when I finished The Tiger : knowing about something fascinating I had no idea happened, and having a better account of how people think and act like they do. Buy it on Amazon

This article is part of a series called The Book Corner, where I post short reviews of books that I found on the internet and thought they were worth sharing. If you’re interested, you can find more on this page.

Pierre Hadot - The Inner Citadel

Tue, 26 Apr 2016 14:31:36 +0000

Background

If there is someone who deeply influenced me in the last 6 months, it’s Ryan Holiday. Last summer, I stumbled on his monthly reading newsletter (you should subscribe here, it’s well worth it), and I found some great books to read. But if you have to start somewhere, go to his Reading List, and you will find some jewels.

The first book from his list is Marcus Aurelius “Meditations”, who he describes as :

To me, this is not only one of greatest books ever written but perhaps the only book of its kind. Just imagine: the private thoughts of the most powerful man in the world, admonishing himself on how to be better, more just, more immune to temptation, wiser. It is the definitive text on self-discipline, personal ethics, humility, self-actualization, and strength.

After this presentation, the only reasonable thing to do was to read it.

So I bought an abridged version of it in French last summer¹. I quite liked it, but I didn’t fully understand what Marcus Aurelius meant in some paragraphs, so this wasn’t the earth-shattering book that I was promised.

Still intrigued by it, I decided in March to get a copy of the next book on the subject, who purposely could explain to me what Marcus Aurelius really thought when he wrote it, why he uses this style of writing, and, by extent, what really is stoicism.

This book was Pierre Hadot’s The Inner Citadel², and I really recommend it to you if you’re remotely interested in stoicism³.

The book

If you first open Meditations, you will probably find yourself a bit lost. Why does the author write like this? What does he truly means when he says those things?

This is because this book wasn’t meant to be read by other people that the author himself. What you will read are excerpts from his personal journal that have been saved from oblivion. That’s why you will find him repeating the same things over and over in the book: he uses it as a way for himself to remember those principles and to apply them every day.

So, for a stranger, this could be a bit confusing, more so if you consider that the book was written almost 2000 years ago, and there are still parts that are forever lost. Take for example this excerpt:

“Discard your thirst for books, so that you won’t die in bitterness, but in cheerfulness and truth, grateful to the gods from the bottom of your heart.” (Meditations, Book 2, III)

What does he mean? Why on earth someone like Ryan Holiday, who read an unfathomable amount of books every month, would abide by this author and stoicism? This doesn’t make sense!

This is because Marcus Aurelius doesn’t write this for you, he writes it for himself. When he wrote this sentence, he was reaching the end of his life. So, he wanted to use his time left to act. Now, the sentence doesn’t mean the same thing anymore and doesn’t conflict with the book thirst that you should have in the early days of your life (who should extend in its entirety).

This is one of the insights that will give you Pierre Hadot’s book. It does a great job at explaining every difficulty that you could stumble on when reading it.

He goes to great lengths to describe difficult concepts and making them accessible. He also compares Marcus Aurelius writing to Epictetus’, from which he deeply inspired himself from, and use it to give a very thorough account of the stoic philosophy.

This is why if you’re interested in Marcus Aurelius book, you should read this one first. It removes the possibility of a misunderstanding, and give some rich background to the whole piece.


Buy it on Amazon	Buy the French version

It was an e-book, so I wasn’t really aware at the time that it was the abridged version. Shame on me. ↩︎
This book has two titles in the french version. The first one vas La citadelle intérieure, used for the english translation; and the second one, used for the re-edition, was Introduction aux Pensées de Marc Aurèle, who carries way better the true purpose of this book. ↩︎
As a disclaimer : I only read the original version in French, so I can’t vouch on the quality of the translation. If you are a francophone, you can find this version here. ↩︎

The Book Corner

Tue, 26 Apr 2016 14:30:24 +0000

When I was a child, I was reading quite a lot compared to my peers. At 10, I devoured the entire Harry potter collection (at that time, only 5 of them were published), was a fan of Both Emile Zola and Frederic Beigbeder¹ in my teenage years, but then stopped reading much when I got into college. I didn’t saw it as something to worry on, because I was reading a lot of news publications, mostly about tech, and considered that it could palliate this loss.

Also, I didn’t have time and energy to commit to those long pieces of text. What do you mean by reading the same piece for 10 hours? Who does this?. Seriously, I’m way better reading this New Yorker article on my Instapaper list, than doing this thing.

But after reading a lot of things on the internet², I convinced myself that magazine articles and news wasn’t enough to learn new things, and by being in a field who doesn’t really correspond to my interests, I would have to learn a lot by myself if I wanted to get this dream job of mine.

So, after stumbling on Ryan Holiday’s reading newsletter, I started following his advice on books, mostly those on his reading list. I’m at the beginning of this journey, and I’d like people to follow me on this one.

That’s why I created this Book Corner³. I want to use this blog where I put so much effort on creating to a good use, and to compel myself to write short reviews on the books I read, so I could have some kind of repository where I could see what I already read and what I thought about it.

Alongside this corner, I will apply the recommendation of Ryan Holiday to keep a commonplace book with quotes and excerpt that I really liked. The plan is to use index cards for this, but I don’t exclude to make a electronic version of them.

So if you’d like to follow me on this journey, you can read the review that I posted below. And if you’d like to contribute, feel free to send me some recommendations on Twitter or at paulhenri@lawgier.net, or check out my Amazon Wishlist and send me one that you liked or think I’d like. Thanks!

2016

May

The Fish That Ate The Whale, by Rich Cohen

The Golden Spruce, by John Vaillant

April

The Inner Citadel, by Pierre Hadot

They are totally not the same style. Just miles away from each other. ↩︎
As you could expect. ↩︎
If you are a Hello Internet listener, you know where this name come from. ↩︎

How to play audiobooks in Overcast

Sun, 20 Mar 2016 10:23:58 +0000

Like many people, I want to read more books that I do.

Many times while browsing the internet, I find myself exposed to books highly recommended by people I deeply admire, or people I just know being successful, but I always end up reading the first chapter and then forgetting to check them out, or catching myself daydreaming every two minutes while reading them. At the end of the day, the only kind of books where I can deeply immerse myself in are fiction books, like the new translation of The Lord of the Rings (that I highly recommend to French people, it’s just pure awesomeness), not books like Robert Caro’s biographies or Robert Greene’s Mastery.

But if there is a medium that I love, it’s podcasts. I listen to many podcasts during my longs commutes to college, and I don’t really have trouble to keep myself focused while listening to them. So, that’s why, enjoying a promotion for Audible’s subscription, I bought Ashlee Vance’s latest audiobook, Elon Musk.

What I liked the most in this medium is that you have to follow the pace imposed by the narrator, preventing you to drift into your thoughts and only read 10 pages in an hour. After finishing the audiobook, however, I realised something: the Audible iOS app is not as good as it could be. And this is mainly because I was spoiled with one feature: Overcast’s Smart Speed.

Overcast’s Smart Speed works by speeding up the playback during silent breaks, enabling you to hear more, while not having to distort the audio like by simply speeding up the playback. And while it’s a great feature to have in a podcast, it’s a killer one when it comes to audiobooks.

So, I spent some time looking on the internet how get Overcast to play my Audiobooks (until Audible calls Marco Arment, and just license his damn implementation!), and I found two different solutions, each having their own advantages and drawbacks. Here we go!

The easy one: Overcast “Upload“ feature

Arment’s announcement of this feature was for me both an exciting one, and a watershed. 2 days before, I was finishing to implement my own solution (more below) after spending quite some time to tweak it and making it work. Having him solving my initial problem right after I managed to tackle it was bad luck.

Anyway, the Upload feature has some advantages, but it’s not the perfect solution for audiobooks. It’s very easy to use and directly implemented in Overcast, but you are limited to 2 GB of storage, 250 MB files and you have to be a paid customer ($9,99 for a year, it’s well worth it) to be able to use it.

First, you’ll have to enable the feature in the iOS app:

*{ padding:0; margin:0; overflow:hidden } html,body{ height:100% } img,span{ position:absolute; width:100%; top:0; bottom:0; margin:auto } span{ height:1.5em; text-align:center; font:48px/1.5 sans-serif; color:white; text-shadow:0 0 0.5em black } ▶ " frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen >

Then, go to https://overcast.fm/uploads and log in. There, you’ll find a list of all your previous uploads and an upload button, where you can… upload your audiobook.

You can then find your file directly on the app, under the Upload section.

The most customisable: The self-hosted solution

If you find the native solution not customisable enough, or too limited, you can use Podcast Generator to host and publish your audiobooks directly to Overcast, or any podcast player that you want.

This solution works great, but suffers from few drawbacks:

You’ll need a remote server or a NAS to use Podcast Generator;
The feed is publicly available, which can be problematic due to copyright rules;
It’s not the simplest solution.

If you don’t already have a server, I recommend using the cheapest service that you’ll find. Podcast Generator is not a resource heavy program, so the hardware won’t matter much. The most important thing here will be the storage space.

I can recommend using Scaleway or Digital Ocean¹, but any VPS or EC2-like provider should work.

If you think that the advantages are worth the trouble, here’s how you can implement it.

Prerequisites

Before installing Podcast Generator, you’ll need to install some prerequisites on your server.

Podcast Generator needs :

A web server : The developers recommend to use nginx
PHP 7+

To install them, you can follow Digital Ocean’s great tutorial for Ubuntu 20.04. On their page, you will also find other versions for different OS.

Install Podcast Generator

Download and extract directly the Podcast Generator archive on your server.

First, you need to find the link to the latest version. Go to the project’s GitHub page and find the link to the .tar.gz archive.

For example, the link to download the v3.1 is :

https://github.com/PodcastGenerator/PodcastGenerator/archive/v3.1.tar.gz

Then, download the archive on your server with wget.

$ wget https://github.com/PodcastGenerator/PodcastGenerator/archive/v3.1.tar.gz

Extract the archive :

$ tar zxvf PodcastGenerator-3.1.tar.gz

The files are now located in the PodcastGenerator-3.1 folder.

Move the PodcastGenerator-3.1 folder to the /var/www directory:

$ sudo mv PodcastGenerator-3.1/PodcastGenerator /var/www/PodcastGenerator

As a mesure a precaution, set the right permissions on the /var/www/ folder:

sudo chmod -R 755 /var/www/PodcastGenerator

Make Podcast Generator recognise .m4B files

To get Podcast Generator to recognise .m4b files, you’ll first have to modify the supported_media.xml, located in the components/supported_media folder. Add this line between the <PodcastGenerator> tag.

<mediaFile>
	<extension>m4b</extension>
	<mimetype>audio/mp4a-latm</mimetype>  
</mediaFile>

Launch the web-based installation wizard

Open your web browser and type this address to launch the web-based install wizard (replace YOUR_IP_ADDRESS with your server’s IP).

http://YOUR_IP_ADDRESS/PodcastGenerator

Merge audio files

If your audiobook comes in multiple pieces, you can merge them using Audiobook Builder, which is also available on the Mac App Store.

It will give you a .m4b file with chapters in it.

Upload your audiobook files

To upload files from your computer to your server, you can use FTP or SFTP. Since you already have your SSH credentials, the easiest way should be SFTP, which use SSH to transfer files.

Upload the files to the /var/www/PodcastGenerator/media directory.

If you use macOS or Windows, I recommend Cyberduck, which is free.

Enable Podcast generator to fetch your files

To see the files in Podcast Generator, you’ll have to enable a feature. Go to the admin panel and choose the “FTP Feature (Auto-Indexing)” option to add the file to the directory.

After the files are properly indexed, you can modify the metadata directly on the panel.

Add the feed to Overcast

When everything is done, you can add your feed (http://YOUR_IP_ADDRESS/podcastgen/feed.xml) to Overcast, who will fetch the files.

Now, you can use Overcast’s great features to listen to your audiobooks!

The link for Digital Ocean uses their referral program, giving you a $100, 60-day credit as soon as you add a valid payment method to your account. Then, if you spend $25, I will receive a $25 credit. the referral free link to Digital Ocean is here. ↩︎

Serial Season 2

Sat, 02 Jan 2016 08:50:00 +0000

First season of Serial wasn’t very compelling for me because of my lack of interest in the subject. This year is about the Bowe Bergdal case : the only US private who was taken hostage in the Afgan war. The in depth coverage is astounding. I’m in the third episode now, and I’m completely into it. Give it a try. Serial Website - Listen in Overcast

To Scale : The Solar System

Mon, 21 Sep 2015 00:15:00 +0000

I couldn’t pass on sharing this, as I’m reading Stephen Hawking’s “A Briefer History of Time”.

Giving some perspective of the real scale of our solar system really helps understanding why things like getting on Mars or sending a probe to Pluto (which isn’t even represented in the video) are no small feat.

EU court rules “Skype” is too similar to “Sky,” blocks trademark application

Wed, 06 May 2015 23:34:00 +0000

Mark Walton, for Ars Technica :

Would you confuse “Skype” with “Sky”? It’s tough, I know, but in attempting to trademark the Skype name and bubble logo, the General Court of the European Union has ruled that the name of Microsoft’s Skype is so similar to British broadcaster Sky the public is likely to confuse the two.

As a French law student, I was taught to have a high opinion of the CJEU, because of it’s role as the European Union’s Supreme Court, who punishes countries’ misdemeanor and EU’s law transgressions.

Obviously, this is not the case today.

I doubt that anyone previously confused Skype and Sky as the same entity before. Okay, both use the “Sky” word, and have some representation of a cloud in their trademark, but I doubt they are the only ones to do so. More so, Skype was founded in 2003 and was a pure European product : it was made by Swedes, Estonians, and Danes before it was bought by eBay in 2005.

The best conclusion possible to this decision was made by Walton himself :

For now, just be aware that talking into your Sky+ remote isn’t the way to make VOIP calls, and clicking the video icon in Skype won’t bring up the new season of Game of Thrones.

Was anyone this confused?

Passion Pit new album : Kindred

Fri, 01 May 2015 00:24:00 +0000

Last week, Passion Pit released their new album : Kindred. As a big fan of their work, you could expect my excitement when I saw it on Spotify’s recommendations. And boy was I disappointed! After the first listening, I just wondered what I just heard. This wasn’t the Passion Pit that I knew, who could make amazing songs like Sleepyhead and The Reeling. This was just another bland pop record for me.

But I couldn’t give up on it. And I was right to do so.

*{ padding:0; margin:0; overflow:hidden } html,body{ height:100% } img,span{ position:absolute; width:100%; top:0; bottom:0; margin:auto } " frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen >

For me, Kindred might be the best Passion Pit album yet. Michael Angelakos do what he does the best : make catchy pop songs. And it shows in Lifted Up and My Brother Taught Me How To Swim (for me the 2 best titles in the album). Both are very catchy songs, with very ingenious electronic rhythms. The rest of the album is mostly good, with songs like Five Foot Ten using arcade game sounds (who give an interesting mix), Dancing On The Grave offering a melancholic tone, and Until We Can’t (Let’s Go) who makes you want to smash everything around you (great when you go for a run). And Angelakos voice is incredible, as usual. You keep wondering how he does it.

I’m not qualified to do a full review of the album, so all I can say is that if you know Passion Pit already, you should not be disappointed (and if you are, give the record another chance). If you aren’t, but you like bands like Foster The People, you should go listen it. And if you’re looking for some catchy songs, it’s for you.

Kindred is available on Spotify, Apple Music and iTunes.

Photos for OS X is here

Mon, 13 Apr 2015 10:30:00 +0000

I previously talked about the Photos app for MacOS, during my search for a better photo management system. As I said, I previously used (and still use for backup) a combination of Dropbox and OneDrive to store my pictures, but some of them were on both Google Drive and Google Photos because I previously used an Android device, and I also used Picturelife and was getting ready to upload all of them to Amazon Cloud Photos (because Prime membership).

As you can imagine, this was not an ideal setup. Because I used too many services, some of the files were not replicated correctly on my “Master save” (OneDrive), leading into situations like almost loosing half of the pictures I took from my trip to New York City with my 2 best friends (which, ironically, are the photos that I care of the most). Also, it was just a pain in the ass managing all of this.

So, when Apple finally released the stable version of Photos, I was kinda exited. And I wasn’t really disappointed.

I’m not going to do a review of the app, I’ll let people like Alex Guyot from Macstories doing this way better than I could. But here is a TL;DR version of it : If you have an iPhone and a Mac, and looking for a solution to keep all of your picture safe and avalaible easily on all your devices, this is the way to go.

Now, if you excuse me, I have to upload 5000 pictures over my very slow ADSL connexion (my peak upload speed is 70 kB/s). See you in 2 weeks!

The CIA campaign to steal Apple's secrets

Wed, 11 Mar 2015 12:51:00 +0000

Jeremy Scahill and Josh Begley, from The Intercept:

Researchers working with the Central Intelligence Agency have conducted a multi-year, sustained effort to break the security of Apple’s iPhones and iPads, according to top-secret documents obtained by The Intercept. […] By targeting essential security keys used to encrypt data stored on Apple’s devices, the researchers have sought to thwart the company’s attempts to provide mobile security to hundreds of millions of Apple customers across the globe. Studying both “physical” and “non-invasive” techniques, U.S. government-sponsored research has been aimed at discovering ways to decrypt and ultimately penetrate Apple’s encrypted firmware. This could enable spies to plant malicious code on Apple devices and seek out potential vulnerabilities in other parts of the iPhone and iPad currently masked by encryption.

This isn’t an unexpected revelation, with programs like XKeyScore and Prism, you could expect that the NSA or the CIA would target a company like Apple. What is more surprising is how they did it, and mostly one of the tool they used : XCode

The security researchers also claimed they had created a modified version of Apple’s proprietary software development tool, Xcode, which could sneak surveillance backdoors into any apps or programs created using the tool. Xcode, which is distributed by Apple to hundreds of thousands of developers, is used to create apps that are sold through Apple’s App Store. The modified version of Xcode, the researchers claimed, could enable spies to steal passwords and grab messages on infected devices. Researchers also claimed the modified Xcode could “force all iOS applications to send embedded data to a listening post.” It remains unclear how intelligence agencies would get developers to use the poisoned version of Xcode.

You can expect the reaction of iOS and MacOS developers. This is from Marco Arment:

Unpatriotic? Absolutely. Terrorism? Maybe. But those don’t quite capture what this really is: war. The United States intelligence agencies are at war against all U.S. citizens. President Obama, “the Constitutional Law president,” not only lets it happen, but supports it. Edward Snowden continues to be much more of national hero and a true American patriot than the President. And I see no future Presidential candidates in either party who are likely to be any better. I’ve said it before: history will not be kind to Obama on this.

But this kind of policy doesn’t lead to the CIA goal of making all of the information be easily readable by them:

As corporations increasingly integrate default encryption methods and companies like Apple incorporate their own indigenous encryption technologies into easy-to-use text, voice and video communication platforms, the U.S. and British governments are panicking. “Encryption threatens to lead all of us to a very dark place,” declared FBI Director James Comey in an October 2014 lecture at the Brookings Institution. Citing the recent moves by Apple to strengthen default encryption on its operating systems, and commitments by Google to incorporate such tools, Comey said, “This means the companies themselves won’t be able to unlock phones, laptops, and tablets to reveal photos, documents, e-mail, and recordings stored within.”

What led to this situation is the previous attacks made by the U.S. Government to illegally access the private data of millions (Apple said on Monday that it has sold more than 700 millions iPhones) of people, making no distinction between U.S. Citizens & its allies, and “terrorists”. The information that could previously be read when needed (a police investigation, a imminent threat to the National Security) by a simple warrant delivered by a judge now cannot be read at all, no matter what the urgency and the will of Apple.

“Encryption isn’t just a technical feature; it’s a marketing pitch,” Comey added. “But it will have very serious consequences for law enforcement and national security agencies at all levels. Sophisticated criminals will come to count on these means of evading detection. It’s the equivalent of a closet that can’t be opened. A safe that can’t be cracked.”

You can ask Tim Cook if that’s what he wanted when he revealed the new security features in iOS 8. I’m sure you know his answer.

“Obama’s comments were dripping with hypocrisy,” says Trevor Timm, executive director of the Freedom of the Press Foundation. “Don’t get me wrong, his actual criticism of China for attempting to force tech companies to install backdoors was spot on — now if only he would apply what he said to his own government. Since he now knows backdooring encryption is a terrible policy that will damage cybersecurity, privacy, and the economy, why won’t he order the FBI and NSA to stop pushing for it as well?”

This is what seems to be a good advice.

New Field Notes : Two-Rivers Edition

Wed, 11 Mar 2015 12:48:00 +0000

Field Notes just announced their new “Colors” quarterly edition for the spring : the Two-Rivers Edition. The biggest change from the usual is the number of variations : they are thousands of them, making your notebook quite unique.

We hand-set several designs using Hamilton’s collection of vintage type and ornaments. Hamilton then printed our designs in two random colors on a 1961 Heidelberg GT 13″×18″ windmill press. Randomizing the designs, papers, and colors resulted in thousands of variations. Further variations were introduced thanks to the nature of wood type, letterpress printing, and the music playing in the print shop during the 200+ hours on press.

The 3-packs are 9.95$ and limited to 3 per household.

If you haven’t already brought them, be sure to do it quickly : As usual, there is only 25,000 packs available and they should go quickly.

Getting better at writing

Fri, 06 Mar 2015 12:56:00 +0000

When you are publishing a blog, the writing may not be the easiest thing to do. This is more true when you are not writing in your mother tongue. You may have an idea, and some things to say on it, but you maybe don’t have the right words to express them in the best way.

This is a huge deal breaker, because mastering the language that you’re speaking in is what makes your text stand-up compared to the others. So you’re discouraged. You stop elaborating your ideas, and you end up not thinking about them at all. This is bad, because, then, you have nothing to publish on the website that you spend so many hours to make and polish, and you stop thinking about the things that you deal with everyday. You’re not so smart anymore.

You can’t just make insightful comments and analysis about complex topics on day one, and do this every day, 5 days a week. To be able to do these things, and to be relevant, you have to practice. To reach the level of someone like Ben Thompson, who succeeds to make a 10$ monthly subscription to his daily analysis on tech topics a huge bargain, you need to have some background, some practice. You need to know how all of those things works now to be able to predict on how they will work in the future.

All of this make me rethink how I work and how I write on the Internet. Because I’m too ambitious, and want to tackle complex topics, I find myself stuck when writing about them. For instance, after hearing Myke Hurley’s new series “Behind the App” on Relay.fm, I tried to write about ads in podcasts; because I thought that they didn’t work very well on this particular new show. But because I had to clearly explain my point of view to be taken seriously (and to not hurt the Relay.fm team, because I love what they are doing), and didn’t managed to get something satisfying enough, I didn’t published anything at all, even the arguments that I had and that I thought were good. And it wasn’t the first time that it happened.

This has consequences on other things. After loving to write on Day One or my Field Notes about my day, or what happened during the weekend, I lost this joy, and I found myself not writing anything at all during several weeks, even when I had more time of doing so. And after saying that you needed practice to write good stories, you see how preoccupying this is.

So I need to be less ambitious. Write less about more topics. Going to the point more quickly. Improving myself. Maybe not being famous at all can be helpful, and can make my blog more of a playground, an experimental lab, where I can try new things, make some mistakes, learn new things and be more forgetful at me.

This is why you’ve seen some pictures decorating this post. I am a newbie when it comes to photography (as I am with writing), but since I have at my disposal my dad’s DSLR (a Nikon D5200, if you’re asking, not my choice), I can use it to learn and try new things, and make your experience of reading on the website more interesting, like on The Newsprint, Tools & Toys, …

Coinbase is shutting its Tip Button

Wed, 11 Feb 2015 13:23:00 +0000

Coinbase Blog :

Today we’re announcing that we’ve decided to shut down our bitcoin Tip Button. The project has seen some good usage since we launched it, with about 10,000 users accepting bitcoin tips across a variety of websites.

Well, it didn’t lasted long (I’m removing all of them now).

Lawgier.net 2.0

Wed, 11 Feb 2015 13:13:00 +0000

Welcome to the new version of Lawgier.net.

After updating the website’s logo, I decided to change its design. Inspired by other Squarespace blogs like Above Avalon (discovered via The Newsprint’s Sunday Edition), I decided to go to a less centered view, and to put more things on the sides.

You’ll find the navigation links on the left with the new logo, instead of being at the top, and a new side bar on the right with a list of the most recent posts, my Twitter feed and a search bar.

I think that this design improves the ergonomics of the website and brings the content at the front. There’s less wasted space, and the content sits on the top, instead of being eclipsed by the massive logo and navigation bar. Now, on the splash screen, you’ll be able to see what this website is about on the first glance, and start reading articles directly, instead of having to scroll on the page.

I still have a few quirks to iron out, like the quote’s font (that I can’t change because of a Squarespace bug), or making an About page, but I’m quite happy of the result, and I hope that you’ll like it.

A preview of the upcoming Photos app for OS X

Sat, 07 Feb 2015 13:37:00 +0000

Dan Seifert from The Verge shows us how the upcoming Photos App for OS X Yosemite will work:

I was looking for a good Photo management app in the last few months, but I was quite disappointed by the solutions available. I wanted something easy to use and with a good UI, but I didn’t wanted to pay an hefty price on cloud storage if it was only used for pictures (like Picturelife). The sweet spot would have been an app using my Dropbox or OneDrive storage, but I didn’t found what I wanted. So I ended up using my OneDrive account to save my collection of digital pictures, because of the unlimited storage that I’ve got with my purchase of Office 365. But the UI of the iOS app is quite… bad.

That’s why I’m totally sold on the upcoming Photos app. I don’t mind paying for iCloud storage, and the UI/UX looks beautiful and powerful. The editing features looks amazing for people like me who fells completely lost with software like Lightroom. And the integration wit iOS is an huge benefit.

With my dad recent purchase of a DSLR, this should be a must-have app for me, even more when I realized that CD backups that my mother made 10 years ago was lost. This should never happen again, and I hope that iCloud and Photos.app will help me with it.

Now, all I need is to buy a Mac (😠 Intel).

Is unencrypting all your conversations would make the terrorists go away ? Of course it wouldn't.

Wed, 14 Jan 2015 13:25:00 +0000

Cory Doctorow, for BoingBoing.net, about United Kingdom Prime Minister David Cameron’s proposal of banning all “‘means of communication’ which ‘we cannot read’” :

This, then, is what David Cameron is proposing:

All Britons’ communications must be easy for criminals, voyeurs and foreign spies to intercept

Any firms within reach of the UK government must be banned from producing secure software

All major code repositories, such as Github and Sourceforge, must be blocked

Search engines must not answer queries about web-pages that carry secure software

Virtually all academic security work in the UK must cease – security research must only take place in proprietary research environments where there is no onus to publish one’s findings, such as industry R&D and the security services

All packets in and out of the country, and within the country, must be subject to Chinese-style deep-packet inspection and any packets that appear to originate from secure software must be dropped

Existing walled gardens (like iOS and games consoles) must be ordered to ban their users from installing secure software

Anyone visiting the country from abroad must have their smartphones held at the border until they leave

Proprietary operating system vendors (Microsoft and Apple) must be ordered to redesign their operating systems as walled gardens that only allow users to run software from an app store, which will not sell or give secure software to Britons

Free/open source operating systems – that power the energy, banking, ecommerce, and infrastructure sectors – must be banned outright

David Cameron will say that he doesn’t want to do any of this. He’ll say that he can implement weaker versions of it – say, only blocking some “notorious” sites that carry secure software. But anything less than the programme above will have no material effect on the ability of criminals to carry on perfectly secret conversations that “we cannot read”.

This is a great article about David Cameron’s demagogic proposal of making all the conversations -and by extension, every software- easy to read and to intercept. But his reflection, in the wake of Charlie Hebdo’s terrorist attack, applies to all countries who want to take similar steps.

Making encrypted software unlawful jeopardize freedom of speech, the essence of a democracy. Also, and even if Cameron will defend himself that he is pursuing it, encryption today is deeply needed to keep the web safe from hackers. And if you’re more worried for your money than you are for your liberties, try to think about purchasing something with your credit card on an unencrypted connection. You’ll make quite a lot of people happy (but not you).

Thanks to Next Draft.

Goodbye Charlie, see you next week.

Sat, 10 Jan 2015 14:30:00 +0000

***Note:***I didn’t planed to publish this article. I wrote it only for myself on Day One. But I couldn’t repress the need to publish it.

Shocked.

Yesterday morning, Charlie Hebdo was attacked by two heavily armed people. They killed 12 persons, including 2 policemen and a lot of cartoonists, including Charb,…. I loved these guys and their work. Charb’s cartoons, mostly those on Sarkozy, were hilarious. We’re gonna fucking miss them.

I learned the event by my brother, who told it to me when he got back from high school. A couple of hours ago, I received a notification from the Twitter app about @Maitre_Eolas and some people tweeting about #CharlieHebdo. After seeing a picture of a Police Car riddled with bullets, I thought that someone tried to attack the newspaper, since it wasn’t the first time it happened, but with little consequences. The picture was pretty explicit, though. The bullet impacts were extremely well targeted, and would have killed the passenger if there was one. But I had a lot to catch up on my RSS feed, so I moved on quickly.

The reaction on my Twitter feed was unanimous. There was a lot of sadness, because we were losing talented people; and incomprehension (Why were they killed? They were only people who drew things on paper, not assassins; and it’s not like they hated Islam in particular, they were satirical about anything, included every other religions.) But, after the sadness comes the anger, and all those people made me remember that this was the work of a handful of people, not the work of Muslims. And it made me realize that this act, which was supposed to avenge all the Muslims who were offended by the Mahomet’s caricature, put them in danger of a (unfounded and stupid) retaliation for something that they didn’t do and don’t approve.

What I can’t understand is the motive behind those killings. If those morons can’t grasp or handle that a society can be made with people with opposite ideas who can freely and openly discuss about them; and that disturbs them so much that they have to kill those people, well they have a long way to go, and I don’t think that all the ammunition in the world would be enough to eradicate the free world (who greatly outnumbers them). The most effective way of killing this liberty is to prove to us that this liberty is useless, or that it can greatly harm us. If you want to win us over, you have to make your ideas attractive. If you are offended by the publication of Mahomet’s caricatures instead of crying foul about them, just make the publisher irrelevant to the society by proving that you seriously don’t give a shit about it. Magazines like Charlie Hebdo lives from the public debate that they create. If there’s none, there is no interest in publishing them. If you make their work pointless, what they could do doesn’t matter at all, because no one cares.

Seriously, if someone who is writing all of this on a little Field Notes during his job as a cashier, while waiting for his next customer on a very calm evening; you would think that an entire organization could have thought about it. Because, to quote the West Wing¹: “Terrorism is 100% failure rate. Not only terrorist always fail at what they’re after, they pretty much always succeed in strengthening whatever it is they’re against”. Today, France and the French just come out more united than ever, just because of this.

Also, this proves that making ultra-repressive laws² in Parliament is useless. Because if today, with all the legal (and illegal -see the NSA) means that the Secret Services possess, you can’t track people who are carrying automatic weapons to kill people who are under constant protection from the Police, passing harsher laws doesn’t gonna improve the situation. They’re just gonna put civil rights, the one who are put in danger by those terrorist you are fighting against, under more and more pressure.

The West Wing Season 3 Episode 1 - “Isaac and Ishmael”. Watch it on Netflix ↩︎
“Données de connexion et Renseignement, Valls attendu pour explication” about 2014’s French military planning law (in French). ↩︎

Fixing iOS third party keyboard implementation

Tue, 06 Jan 2015 14:49:00 +0000

David Chartier, for Finer Things in Tech:

As interesting as iOS keyboards can be, their initial implementation at the OS level is severely flawed. They’re cumbersome to setup, switching between them is needlessly tedious, and limitations make it difficult to teach users about keyboard features. As far as I can tell, all these problems require solutions and improvements from Apple at the OS level.

You should read the full article : David Chartier sum up quite well the quirks of using multiple keyboards on iOS 8. They are painful to use, and don’t let me start talking about switching between them. Every time that I want to use David Smith’s Emoji++, I drop a tear.

But what’s interesting here too, is the propositions that he makes to improve it. Implementing them in iOS 9 could make what many people saw as the best example of Apple opening up, a feature that works even better than on Android.

Apple has lost the functional high ground

Mon, 05 Jan 2015 14:58:00 +0000

Marco Arment, on the state of Apple software :

I fear that Apple’s leadership doesn’t realize quite how badly and deeply their software flaws have damaged their reputation, because if they realized it, they’d make serious changes that don’t appear to be happening. Instead, the opposite appears to be happening: the pace of rapid updates on multiple product lines seems to be expanding and accelerating.

New features in a highly competitive industry like computer or smartphone marketing, are fundamental to keep your company relevant from your competitors. If Apple was selling the iPhone 6 with iOS 6 installed, I doubt that it would led to “the biggest iPhone launch ever”.

But the biggest priority that you should have when you’re making something, is that it should work as expected. What’s the point of having a Swiss-Army knife if the blade is so bad that you couldn’t cut anything?

Edit : Fixed the awful mistakes. Sorry about them.

Some thoughts on the App Store review system

Tue, 09 Dec 2014 14:59:00 +0000

Following yesterday piece, Cromulent Labs published a blog article explaining why the had to remove their app from the App Store recently.

Here’s some bits that I’ve selected:

During one of my conversations with someone at App Review last month, I asked if they could tell me if some of these new apps being accepted slipped through or if their use of widgets was deemed acceptable. I heard what had come to be a popular refrain from them. They couldn’t discuss other apps with me, they would look into those apps, and if I submitted a new app with that specific functionality they would be happy to review it and let me know if it was acceptable or not by either rejecting or accepting my new app. They steadfastly refused to tell me if a certain use of a widget was acceptable or not ahead of time.

Apple : “- Yeah, you have to spend dozens of hours to implement this idea to see if it doesn’t violate our -so ambiguous policies that even us understand them after the fourth revision of your implementation-”.

The Apple representative responded by saying that they prefer that the rules remain vague because that allows developers to come up with innovative ideas […]

Yeah, because uncertainty and vague policies is the best way to make your business work!

[…] and also allows Apple to be flexible in case they change their minds later.

This is the real reason why Apple is ambiguous.

When pressed on the issue of their policies leading to wasted developer time, I was told, “If you are afraid something you are working on will be rejected, then don’t work on it.”

So basically, they are telling you to not try to make your app better. Best way to “come up with innovative ideas”!

But this is the best piece :

During this same conversion, I also asked specifically why Launcher was removed from the App Store after 9 days when other similar apps are still available weeks later. The answer to this question was the most interesting and informative response I had ever heard from them. They basically said that Launcher was a trailblazer in uncharted territories and that they felt that they needed to make an example of it in order to get the word out to developers that its functionality is not acceptable without them having to publish new specific guidelines. And they said that the fact that they aren’t seeing hundreds of similar apps submitted every day is proof to them that taking down Launcher was successful in this regard. This was a pretty big revelation to me. After Launcher was rejected and the press picked up on it and started writing articles which painted Apple in a bad light, I was afraid that Apple might be mad at me. But it turns out that was actually the outcome they were looking for all along. They acted swiftly and made me the sacrificial lamb. And after that, removing other apps with similar functionality became a low priority for them.

No comment.

To finish, Joe Cieplinski summarize well the situation (check his article, it’s very short). I think it’s time for someone to vest his (or her) stock options. It’s not like Tim Cook hasn’t done this before.

Thanks Marco Arment.

Apple ask Panic to remove Transmit's iCloud Drive export feature

Mon, 08 Dec 2014 15:03:00 +0000

Panic’s update notes for Transmit :

Also, at Apple’s request, we had to remove the ability to “Send” files to other services, including iCloud Drive. In short, we’re told that while Transmit iOS can download content from iCloud Drive, we cannot upload content to iCloud Drive unless the content was created in the app itself. Apple says this use would violate 2.23 — “Apps must follow the iOS Data Storage Guidelines or they will be rejected” — but oddly that page says nothing about iCloud Drive or appropriate uses for iCloud Drive.

Since the release of iOS 8 and the Notification Center controversy, App Store’s App reviews keeps surprising people with non-sense decisions, like asking PCalc developer to delete it’s implementation of his Calculator in the Notification Center (which is clever), or for inserting buttons in your widget like in Drafts or Evernote.

But the worst thing, taken apart the absurdity of the rejections (see John Gruber’s thoughts on this case), are the a posteriori intervention of the reviewers : all the apps and the controversial implementations were endorsed by the review team, who got sometimes a full week between the submission and the release, and were updated many times before being rejected. This is a irresponsible behavior, and something that could be very costly for the developers who put time and effort in those features. And in this case, this is one of the critical feature of the app, who was sold as being able to transfer files to iCloud Drive or Dropbox/Box and now can’t.

By being this schizophrenic, Apple put in danger business (see the opinion of famous developers like Marco Arment) on which it relies heavily on to sell its products (an iPhone without the App Store is a very expensive piece of junk), and its relationship with its customers.

Maybe it’s time that Apple grab its shit together and do the right call : being friendly with third party developers (WWDC 14 was a good start) and being consistent.

New Wii U Zelda trailer shows Link horsing around

Sat, 06 Dec 2014 16:35:00 +0000

Kyle Orland, for Ars Technica:

After a very brief announcement and teaser at E3 this year, Nintendo used the inaugural Game Awards webstream and live presentation in Las Vegas tonight to show off a more substantial trailer highlighting new gameplay features in the next Legend of Zelda game, planned to hit the Wii U some time in 2015. In the trailer, Zelda creator Shigeru Miyamoto and project producer Eiji Aounuma discuss the game’s “massive world” designed with “a number of high places with great views” to view far-off goals (“It would take a very long time for Link to walk” across the map, Aounuma says). Distant landmarks can be highlighted through a first-person view, controlled by tilting the GamePad, then show up as markers on an interactive, zoomable map on the touchscreen.

Nintendo finally understood that their biggest strength wasn’t their Hardware (at least, for this generation of consoles), but their portfolio of games. I never considered buying a Wii U (and I have the original Wii) until today.

Now, the question is : Will I get a Wii U before the PS4? The release date of No Man’s Sky might be the answer.

What is the best delivery tracker for iOS / MacOS ?

Fri, 21 Nov 2014 16:38:00 +0000

The Sweet Setup stroke again today with a full rundown of the best ways to track deliveries on iOS and MacOS.

“Package tracking apps aren’t the kind of apps that we use 10 times a day. Our goal is to get the information we need and get back to work. All of these apps are doing something that we can do for free on many retail and shipping carrier websites, but a dedicated app makes tracking all of your packages a “one stop shop.” In this case, you are paying for the convenience. As I mentioned, I buy a lot online, so spending a few dollars on an app that can track all of those deliveries is well worth it for me.”

And Bradley Chambers found the best app : Deliveries Status (who was updated today for iOS 8).

“With widget support and ease of entry, Delivery Status does its job quickly and efficiently. Available on the Mac, iOS devices, and even the web, Delivery Status is our favorite app for tracking packages.”

I agree with him on this : I bought Deliveries Status a few months ago, and I like the way of displaying where items are on the map, and the notifications.

So, my advice is to read the whole piece to understand Bradley’s choice (and mine); and buy the app that suits you the best.

More insight about the Apple-GT Advanced deal.

Thu, 20 Nov 2014 16:43:00 +0000

Today, Daisuke Wakabayashi from The Wall Street Journal (the article is behind a paywall. To read it, just Google the title, and click on the link provided by Google) published more informations about GT Advanced deal with Apple, and some insights about why GT filed for bankruptcy one month ago (if you don’t know about this GT/Apple deal, Wakabayashi recaps the entire story well in this article).

First, it shows why Apple chose GT Advanced, a sapphire furnaces maker who had no experience whatsoever in making sapphire (and making furnaces is not the same as producing sapphire - I don’t think Whirlpool is the best company to call when you’re making food for 100 people) :

Apple also was having trouble finding a sapphire manufacturer. An executive at another company Apple approached last year said it couldn’t make a profit producing sapphire at the price Apple wanted.

Second, it shows what seems to be an huge mismanagement inside of GT :

GT quickly set out to hire 700 staffers. Hiring moved so quickly that at one point in late spring, more than 100 recent hires didn’t know who they reported to, a former manager said. Two other former workers said there was no attendance policy, which led to an unusual number of sick days. GT managers in the spring authorized unlimited overtime to fill furnaces with materials to grow sapphire. But GT hadn’t built enough furnaces yet, so many workers had nothing to do, two former employees said. “We just kept sweeping the floors over and over,” one of the former employees said. “I just saw money flying out the door.”

And third, GT didn’t even delivered what was asked : sapphire glass who can be used for the new iPhones :

Producing sapphire proved to be the biggest problem. It took roughly 30 days and cost about $20,000 to make a single boule. The people familiar with Apple’s operations said more than half the boules were unusable.

So, it seems that Apple is a ruthless partner who pushes for the latest innovations at many costs (see how they managed to be one of the first -if not the first- to get 5K displays from LG Electronics, who didn’t even put them on their TV). But, with those management problems and failure to do what they were asked to do, proper blocks of sapphire, GT seems to be the one to blame here.

You should read the entire article; it gives some great material to understand the whole piece.

Coinbase Introduces The Tip Button

Tue, 18 Nov 2014 16:48:00 +0000

The Coinbase blog :

Today, we are excited to release a new bitcoin tip button for accepting micropayments anywhere on the web. The tip button can be easily added to any web page and enables website visitors to tip using a Coinbase account or other bitcoin wallet.

Coinbase introduced today a very interesting way to make tips to people.This button doesn’t revolutionize anything, it’s very lookalike of the Flattr button, but adds the benefit of using Bitcoins for payments, which decrease the Credit Card fees, which can be a problem for transferring little amounts of money.

The default amount is 300 bits (Ƀ0.00000300), which is approximatively $0.10 as today, and can be changed to the amount that you want. You can use your Coinbase account to make donations, or any Bitcoin wallet.

I think that Coinbase’s idea is great; I’ve already implemented the button on all of my blog posts, if you’ve found some piece that you’ve liked, feel free to tip me.

But the biggest strength of this too is also it’s biggest weakness : Bitcoin. Because mot everyone has a Bitcoin wallet with Bitcoins on it, you can’t reach as many people as you could with a Credit Card based operation. This might change in a few years, but Coinbase idea is great to push people to use it.

Beautiful pictures of Philae landing on a Comet

Thu, 13 Nov 2014 16:51:00 +0000

After a 10 year voyage, ESA’s Rosetta spacecraft finally landed it’s Philæ module on the Churyumov-Gerasimenko comet. As you’ve might saw it on the news, this was a big achievement and a marvelous job made by the scientists at the European Space Agency. As you might think, I’m incredibly proud of them, and by the fact that this was made possible because countries, instead of fighting against each other, chose to be on the same side and make great things all together. This is why Science is great (and for many other things). I’m proud to be an European citizen.

If you want, you can read more on this beautiful article from The New York Times.

But today, I want to share some beautiful photos who were taken by both Rosetta and Philæ, who are breathtaking.

This is a montage made by The New York Times of 4 assembled pictures who represent the duck-shaped comet.

This is my favorite. the solar panels from the Rosetta spacecraft are rendered beautifully in this picture.

Philæ landed on flatter upper front of the comet. Photo edited by The New York Times

The photos are from the ESA, and some of them are edited by The New York Times.

You can find more of them on the ESA’s flickr account and on The New York Times feature.

Reeder update for iOS 8

Thu, 06 Nov 2014 17:03:00 +0000

If you’ve ever tried to use RSS on iOS, Reeder is an app that you’ve must have stumbled upon. Described as the best RSS app for iOS for a long time by most of the tech scene (and replaced by Jared Sinclair’s/Supertop Unread by some Apple podcasters), Reeder is still my go-to app when I have to tackle some of my untidy RSS feed. updated for iOS 7 with Reeder 2, Silvio Rizzi finally releases an update for iOS 8. It’s what can be described as a minor update, but it’s a much needed one.

The update includes:

New : - iPhone 6 and iPhone 6 Plus support - Support for landscape mode (iOS 8 only) - Standard share sheet → “More…” in Reeder’s share panel (iOS 8 only) Fixed : - Various bug fixes for iOS 8 - Failed authentication error handling

The standard iOS share sheet is hidden in Reeder’s custom share sheet, at the right bottom of the screen. It still privileges Reeder own share extensions, which are very good on their own (Instapaper implementation on Reeder is better than with the classic share sheet : you’re not stuck with a loading screen when you’re saving something, it does everything in the background.)

It also integrates the support of the landscape mode on iOS 8 devices, which fixes the inconvenience of having to launch a embed video on the app or the WebView on Safari if you want too see it in landscape -something you should definitively do when you’re viewing a 16:9 video-. This is very much appreciated.

So, this is a very solid update, but I think it’s missing some features, who would be very much welcomed :

Font choices : Helvetica Neue is boring, and doesn’t make text really easy and enjoyable to read. Adding some choices like [Ideal Sans][0] or some Serif fonts would be great.
Text justification : Even if some people doesn’t really like justification, I find it helps to make the text more easier to read, and looks better. A properly hyphenation would be great.
1Password full integration : This is not a top of the shelf needed update, but it would be great to have a full integration of 1Password (which is compatible with other password managers like Dashlane for third party sharing services (like Pinboard or Instapaper).

Anyways, this update still make Reeder as one of the best -if not the best- RSS reader on iOS. I don’t have a Mac, but if you have, you should definitively check out its Mac OS counterpart.

Nuzzel Workflow - Week One

Tue, 04 Nov 2014 17:35:00 +0000

So, last Thursday, I posted a blog post about the new Nuzzel iPad app, saying that this update (to the already existing iPhone app) and various feedback from famous writers made me rethink a bit my work-flow. So here I am, more than a week after.

The problem of not being a professional blogger.

So, first, I have to make a huge disclaimer before beginning to talk about my experience : things didn’t go as I would have expected.

To begin with, I’m beginning the last month of my semester, which implies an increasing amount of work that I have to tackle every day. This is something that impedes me to write and read as much as I want to. But everything should be okay after the exams, mid-December.

Then, I had some problems with my credit card, who was refusing all my online payments and, thus, my Ars Premier subscription, which, therefore, truncated all my RSS feed, making it a pain to read (even if Ars implementation of RSS is better than most of news’ websites), and discouraged me a bit to take time to read it all. Everything should be fine tomorrow (and I learned that my MasterCard had a threshold on the amount of payments that I make online. This is nuts.)

And I kept the best part for the end : Tuesday, I thought that it would be a good idea to make an Mac OS X virtual machine on my PC, a kinda half-backed Hackintosh. I did succeed to make a pretty good Mavericks install on VirtualBox, but I came to a conclusion that my laptop wasn’t the best computer to do a Hackintosh : I have a 1336x768 resolution screen, and my graphics card -a GeForce 610M- isn’t recognized by Mac OS, making the interface a little bit too much sluggish to be used properly. But the damage was done, and I lost two days in this project (it was kinda fun); and I didn’t even succeeded to update the thing to Yosemite (I have trouble to make VirtualBox boot from an USB, if anyone have some tip, it would be great!).

Keep it simple and useful.

But all of this doesn’t mean that I didn’t used Nuzzel or my RSS reader at all during this week. Nuzzel even proved itself very helpful in one case : it sent me an alert for Tim Cook’s coming out, because 5 of my followers twitted about it. I was delighted when I saw the notification, but the killer feature was this :

Instead of putting this article in my Instapaper queue, giving the link some context gave me the ability to see if it was relevant or not (in this case, it was). Amazing feature.

Clearly, I would have missed this article if I didn’t used Nuzzel at this time. It would have lost itself in my RSS queue until I took some actions to read it, even if it was relayed by most of the Apple bloggers that I have subscribed to. So you can imagine if it was a more low-key publication put on Twitter by a handful of people when I wasn’t connected? I would have missed it. So the app finds itself quite useful.

So now, let’s move to week 2!

Switching to Squarespace

Tue, 04 Nov 2014 17:28:00 +0000

The switch to Squarespace may have broke all the previous links made on Wordpress. To fix this, just put ‘/blog/’ between the domain name (http://lawgier.net) and the “slug” (/2014/10/XX/name-of-the-article) in your address bar. Thanks.

So, here we are, dealing with my need to experiment things. Testing everything. Never be satisfied. Avoir la bougeotte (have ants in one’s pants would be a good translation).

Before launching the website, I moved to many platforms : Ghost, Scriptogr.am, Postach.io, … I had settled for Wordpress for about a month, mainly because I didn’t found a better solution. I had tried Squarespace a bit, but didn’t found what I was looking for. Until today.

Squarespace 7 is a wonderful platform. The themes are awesome. I had trouble at first because I couldn’t do what I wanted. But those things were mainly a waste of time. Installing 5 plugins to make your blog a little bit reliable in case of dealing with a sudden huge amount of traffic is meh. Having to create a Rackspace Cloud Files instance only to make the website load a bit faster, and wont being able to upload files to it from your iOS device (I’m looking at you, Panic) isn’t really worth it. Won’t be able to know how much the website will cost you at the end of the month because Azure cost calculator is weird and gives you wrong numbers (my cloud instance runs for 4 hours a day -dunno why-, but Azure calculates an 24 hour usage) is bad.

So, Squarespace : 10 bucks a month (use a promo code like ATP to get 10% off your first purchase, but if you’ve already listened to any podcast, you already know this), 500 GB Bandwidth, no back-end to deal with, no CDN, blazing fast, great theming, great mobile app (would be even be better if it was supported by Byword).

So, I’m happy. But, until when?

How copyright law gives Amazon too much power over the book industry

Sun, 26 Oct 2014 17:43:00 +0000

Timothy B. Lee, for Vox :

I mostly agree with my colleague Matt Yglesias’s argument that Amazon is doing the world a favor by crushing book publishers. But there’s at least one way US law gives Amazon excessive power, to the detriment of publishers, authors, and the reading public: ill-conceived copyright regulations lock consumers into Kindle’s book platform, making it hard for new e-book platforms to gain traction.

Both Yglesias’s (linked in the quote) and Lee’s arguments are very interesting for me. But Lee is making a point there. DRM on Amazon eBooks and Kindle devices are awful consumer wise. They are punishing you for buying a book by putting so many restrictions, like not being able to use your book on an another device than an approved Kindle, that you don’t get when you pirate it. And those DRMs are ineffective : hackers always find a way to break them.

Tell me about how useful your DRMs are. Apple understood that putting DRMs on the music sold on iTunes was much more cumbersome that the benefits that it should have brought, and didn’t succeeded to. So that’s why they make pressure on the Music industry to remove them. But it’s still there for movies, TV shows, and ebooks too. So, if I buy a movie on the iTunes store, the only way for me to watch it on my TV is to buy an Apple TV. I can’t use my Roku 3 or my Playstation 3 or my Chromecast to see them. So, the only way for me is to buy a Blu-Ray, full of DRM too, and incompatible with most computers and ALL Apple devices, or download them on a shady website for free. Using the Playstation Store or the Google Play Store will lead to the same situation : being locked down to only one place. Now, tell me about how useful your DRMs are.

I think that the entertainment industry should be more focused on enhancing the user experience and satisfaction, instead of seeing bad things everywhere, and threat people like suspected thieves. Removing DRMs and letting people do what whey want with what they bought seems to be a good start.

Nuzzel on iPad

Thu, 23 Oct 2014 16:48:00 +0000

When you see a product that you know, but don’t expect being reviewed by both Jason Snell and Federico Vittci, it might give you a totality different approach on it. In this case, the product is Nuzzel.

What is Nuzzel?

Jason Snell introduces it way better that I could :

Nuzzel is a free web site and iOS app that mines your Twitter and Facebook networks and shows you the newest or most popular links. What Nuzzel does well is sort and filter and group the links in interesting ways, and present them as nicely formatted news-story items—not as tweets. The filtering makes a big difference. I can, for instance, see all links from the past four hours that have been recommended by two or more friends. This has a tendency to float the most interesting stuff to the very top.

What has changed?

You see, I’m a big RSS user. I’m a Feedbin member for a long time, and I pay subscriptions for websites just to have a full RSS feed (like for Ars Technica or Next Inpact, a French website). But there is a problem with RSS. I’m not interested in all Ars articles or The Verge ones, but they all get into my unread list. And, as a student, I have limited time. So, since I subscribe to a lot of feed (that I like), I often get to the point that I have more than 1000 new articles to read because I didn’t manage to read them all during the day; and that leads me to take at least 3-4 hours every weekends to clear the list (or give up and mark all as read). This is untenable, and it makes reading the news much more a pain that I genuinely get when my inbox is less cluttered (yep, I’m an Inbox 0 guy).

So let’s get back to Nuzzel.

When I subscribed to Nuzzel about 6 months ago, I saw clearly the benefits of the service, which was, at the time -and for me, an Android user-, a daily newsletter with a summary of the 5 articles the most shared by my Facebook friends and Twitter followers. Far for being a Twitter completist, I missed some great pieces well worth reading, and this saddened me a lot.

But my utilization of Nuzzel didn’t get past the morning newsletter, even when I got the iPhone and downloaded the app. So the usefulness of the service was very much limited. I’m sure that I wouldn’t notice if I didn’t got the daily digest for a week in my mailbox.

But the iPad app looks great on my iPad mini, and this might change something.

Anyways, I’m gonna try the iOS apps of Nuzzel for the coming week, and put my Feedbin account on diet. See you in the next episode!

And if you have any advices on RSS and stuff, just send be a tweet or an email. Thanks!

GT and Apple agrees on shutting down their plant and secrecy agreement

Wed, 22 Oct 2014 16:56:00 +0000

Re/Code :

Apple and GT Advanced Technologies struck an agreement on Tuesday that will let GT begin the shutdown of key Arizona operations and will shed some more light on why the former stock market darling abruptly filed for bankruptcy this month.

So it seems that we’re about to know what really happened in this case. This should give us some insight about Apple’s manufacturing deals with it’s parteners, if GT did or did not meet the expectations (and what were they), and under how much pressure was GT on.

Or, you know, there’s could be nothing.

Billy Madison, Happy Gilmore, and... Google Fiber?

Wed, 22 Oct 2014 16:54:00 +0000

Glenn Fleishman, writing for Six Colors :

The same principles that motivated Netflix to sign the Sandler deal should motivate Apple, Facebook, Google, and Microsoft to do the same thing—only for fiber-optic broadband Internet service in the United States, and maybe elsewhere. And for nearly the same reason. “Dear sir,” I hear you cry over the interwebs, “You’ve lost your mind.” No, no, hear me out, folks. As these tech giants find themselves increasingly dependent on Internet service providers like Comcast, AT&T, Cablevision, and Verizon, and with those companies wanting “fast-lane” exemptions to network neutrality rules (which are murky at the moment following court decisions) the easy way to keep a connection to consumers without being at the mercy of the broadband providers is to build it themselves.

Great article about net neutrality, and Google foray into the world of ISPs. Glenn is such a talented writer, and manage here to explain something complicated with a dubious analogy including Adam Sandler.

If you’re not familiar with the terms “Net Neutrality”, or asking yourself why did Google build a fiber service in Kansas City, and expanding it in Provo or Austin, this should really help you. If you already know why, read it too. And you should kickstart the Year 2 anthology of The Magazine, the publication created by Marco Arment and now edited by Glenn Fleishman.

New features on iOS 8.1

Mon, 20 Oct 2014 16:58:00 +0000

Federico Viticci :

iOS 8.1 brings bug fixes, speed improvements, and interface changes, but it also enables Continuity features such as Text Message Forwarding and Instant Hotspot, allowing iOS devices to better integrate with each other and Macs running OS X Yosemite. With iOS 8.1, Apple is opening access to its iCloud Photo Library beta – an iCloud service that stores all your photos from all your devices, in a single library that relays changes to every device. And last, iOS 8.1 marks the debut of Apple Pay, the company’s new payment service that rolls out in the US today.

If you kinda loose track of what new features is bringing iOS 8.1, this MacStories article from Federico Vittci is, as usual, great.

For me, I’m pretty exited for the “Text Message Forwarding feature”, which will bring the ability to respond to my SMS friends (the majority) directly from my iPad, and, someday, my hypothetical Mac.

I just activated the iPhoto Cloud Library to see the improvements (and it’s uploading a lot of files), but this is not a feature that I’m particularly exited for, since I already back up my photos with Carousel (from Dropbox) and OneDrive (thanks to my 1 To storage space from my Office 365 subscription), so not a big deal from now. But we’ll see the Apple implementation.

Also, the big feature that I’m waiting for is Apple Pay, but unfortunately, it’s not available in Europe, and, anyways, since I have an iPhone 5, I’m not eligible to it. I’d like to see if it’s worth the upgrade to an iPhone 6.

Anyways, I hope you’ll enjoy the update!

Edit : Oh, and if you’re looking for an extensive changelog of the update, Jason Snell of Six Colors got you covered.

Apple announces its new iPad and Mac lineup

Thu, 16 Oct 2014 21:37:00 +0000

Even if Apple spoiled its event the day before it has happened, it has still a few things to announces. More precisely, it has 5 products to announce: (Y)OSX Yosemite, iOS 8.1, new iPads, iMac Retina and the Mac Mini.

OSX Yosemite

Not new things to see since WWDC (except iTunes 12), but Craig Federighi took some time to recap all the new features that was introduced this summer. If you want more details, you can read John Siracusa’s extensive review on Ars Technica (be careful, it’s a 26,485 words behemoth) or see the new features on Apple’s website.

Of course, the features that I’m the most exited of are Handoff and Continuity, but, since I’m still not a Mac user (I’ll explain everything a bit later), I’m not very concerned by those.

iOS 8.1

iOS 8.1 brings Apple Pay in U.S (we’re still waiting for a release date in Europe, where most point of sale are equipped with NFC readers), iCloud Photos in beta, SMS Handoff with iPad (yayyy!) and should squash some bugs (Which ones? I don’t know yet). It’s coming on Monday.

The new iPad Air and iPad Mini

Not very much to say about them; iPad Air 2 is lighter and thinner than iPad Air 1, rocks an A8X processor who increases dramatically the graphics performance compared to the A7 processor (which was already an huge improvement), and has Touch ID. That’s pretty much all. And Apple spoke about the iPad Mini 3 (I have the Mini 2) for about 10 seconds, which, I think, is still less than when they spoke about the last iPod Touch update. It’s quite a shame. You can pre-order them on Friday 17 (tomorrow), and they are shipping next week.

iMac Retina

As it was expected, Apple released a retina version of its iMac, with a stunning 5K display (as if 4K wasn’t enough). Since Intel is behind the schedule to deliver its new Broadwell processors, there is not much improvements under the hood except for the screen : the “classic” iMacs weren’t updated, as the entire lineup of Macbooks (I’m desperately waiting for a Broadwell Macbook, either the Pro or the Air, to make the switch to MacOS.)

Mac Mini

After two years, the Mac Mini is given an Haswell refresh and a price drop. This is a interesting option for a light desktop computer (to replace my parents’ 2006 iMac, which is stuck on Snow Leopard, and takes about 5 minutes to boot.)

Facebook Introduces Safety Check

Thu, 16 Oct 2014 19:48:00 +0000

Facebook :

In times of disaster or crisis, people turn to Facebook to check on loved ones and get updates. It is in these moments that communication is most critical both for people in the affected areas and for their friends and families anxious for news. We want to provide a helpful tool that people can use when major disasters strike, so we’ve created Safety Check – a simple and easy way to say you’re safe and check on others.

This is actually a great idea, being able to check for family members or close friends in case of natural disasters or huge crisis (if there’s still Internet, of course.)

The "Double Irish" might be over

Thu, 16 Oct 2014 18:51:00 +0000

Re/Code :

“Ireland’s plan to close a “Double Irish” tax loophole could cost U.S. companies including Apple and Google billions of dollars, although a new break and pressure to tackle tax avoidance elsewhere means they are unlikely to decamp.”

Finally, the E.U is taking some actions against tax evasion taking place in Ireland, Luxembourg and Netherlands. It’s time to end the economical disadvantaged induced by those taxes policies, whether it’s the" Double Irish" or the “Dutch Sandwich”, or anything else.

For those unfamiliar with this topic, those countries took their taxes policies as a mean to increase their economical competitiveness, as tax heavens countries like Bermudas do (but with higher tax rates, and with better transparency rules to not be classified as tax heavens). In a globalized world, this should not poses huge problems, individual States should be able to take some actions against tax evasions. This is what’s happening in the U.S, where the government can take some actions to avoid those tax evasions for domestic profits (they have a bit less power concerning international profits). But being in the E.U, and more importantly, in the Schengen and/or the Euro zone changes everything, because there is almost absolutely no borders between it’s members (which can be a wonderful thing, trust me), and you can operate freely in France or Germany with a company based in Ireland or Luxembourg (This is what Microsoft or Amazon do). This means that, when I buy something on Amazon.fr, the transaction is made between me and Amazon EU S.A.R.L. in Luxembourg, and not Amazon France. So, the French state doesn’t see any of my money (except the VAT, 20%), even if I made the purchase in France. The only purpose of Amazon France is to operate its warehouses and pay its workers. Also, if a French ad firm want to buy some ad spots on Google, it make a deal with Google Ireland, and not Google France.

I could write something more about why (moderate) taxation is important, but you have just to understand that the trucks that Amazon uses to deliver its goods, and all the infrastructures, were paid with theses taxes. Everyone has to contribute to the effort, and forgetting this is something that I cannot understand.

Anyways, back to the point :

“The question is where do you go to? There’s nowhere else in the European Union. It’s just getting too hot,” said George Bull, head of the tax practice at advisory group Baker Tilly.

This proves that there was only a few countries that was making this tax evasion possible, and not that there was only a handful of countries with very high taxes. But this also proves that the EU is a mess when it comes to imposition¹. And this have to change.

And, to conclude all of this :

The impact on companies’ profits will be reduced by a new scheme announced by Noonan under which profits linked to the exploitation of patents will attract lower tax rates. Britain already has such a scheme, under which profits linked even tangentially to a patent can face a tax rate of 10 percent.

If you were seeking a new loophole, the Irish government has your back!

See this table on “Adjusted top statutory tax rate on corporate income, 1995–2014”, Eurostat ↩︎

This Is The Nexus Player

Wed, 15 Oct 2014 21:56:00 +0000

The Verge :

The Nexus Player takes the shape of a sleek-looking puck around 4.7-inches in diameter. It comes with a tiny remote with just a few buttons on it, one of which allows you to enable a microphone to perform a voice search — it’s actually a lot like what you’d find with Amazon’s Fire TV. The box includes 8GB of internal storage and is able to stream over 802.11ac Wi-Fi, which is the newest standard out there right now. The Nexus Player also supports a game controller, but it’s sold separately.

This seems to be a very cool gadget : the Android TV interface is sleek, and the Player looks great. BUT, I think it misses two features :

An Ethernet Port : Because, if you’re trying to stream a full 1080p movie on Wi-Fi, it’s just gonna choke to death, and, even if 802.11ac is better, not everyone has a router compatible with it.

An Audio-Out jack port on the remote : I discovered this feature with the Roku 3, and it’s great : I’m able to watch everything I want (action movie with big explosions for example) at the hour I want, without annoying people whit the sound.

So, maybe I don’t need an Nexus Player. Maybe I just need a Roku 3 with Android TV (and a better Plex app, which Android TV has, not Roku).

Google Announces The Nexus 6

Wed, 15 Oct 2014 07:26:00 +0000

The Verge :

Google is today introducing the Nexus 6, the latest in its line of smartphones designed to show off the capabilities of new Android releases. This is the biggest Nexus phone that Google has released yet, adopting a 6-inch display — bigger than both the iPhone 6 Plus and Galaxy Note 4. Like the Note 4, Google’s Nexus 6 also uses a Quad HD display, which means that text and images on the phone should still be really sharp, despite its large size. The phone has a Snapdragon 805 processor, a 13-megapixel rear camera, a 2-megapixel front camera, a 3220 mAh battery, and two front-facing speakers. It can include either 32 or 64GB of internal storage and comes in either blue or white.

This is comically huge. The iPhone 6 plus and Galaxy Note are, I think, the biggest form factor you could use on a phone, but this? Come on Google! 5.92 inches?!

And 649$ too. I’m very happy to have bought a 349$/€ Nexus 5, but those 2 things makes the Nexus 6 a deal breaker for me.

But those 2 things take apart, this is a beautiful phone, if you remove the 2 HORRIBLE front speaker (great feature, but so badly implemented, it looks like two scars on the glass).

Apple accidentally reveals iPad Air 2 and iPad mini 3

Wed, 15 Oct 2014 07:23:00 +0000

The Verge :

Apple has made an unusual blunder ahead of its Thursday press event; the company has accidentally revealed both of its new iPads: the iPad Air 2 and iPad mini 3. An official user guide for iOS 8 in the iTunes Store has apparently had its screenshots updated ahead of schedule; both new iPads are pictured, and the images reveal each will have a Touch ID fingerprint sensor. Aside from the addition of Touch ID, it seems these latest iPads will look nearly identical to their predecessors. 9to5Mac first spotted the premature iPad screenshots.

Oops.

Illiad gives up bid for T-Mobile US

Wed, 15 Oct 2014 07:00:00 +0000

Re/Code :

French carrier Iliad said Monday it is giving up on plans to acquire T-Mobile US, following discussions with majority owner Deutsche Telekom and the T-Mobile US board.

It’s not really a surprise, if you look at the size of Illiad, and the will of it’s main shareholder, Xavier Niel, to not run into a huge debt. But it’s not a good news for U.S. citizens, if you look at the mobile subscriptions prices and all the Comcast bulls**t. Thanks to Free (the name of Illiad’s ISP and mobile carrier), who shoke the French market 2 years ago, I pay only 25€ (30 $US) for unlimited voice and text, 5GB of LTE and with no engagement. How much Verizon/AT&T charges for this?

7 million Dropbox accounts in the wild

Tue, 14 Oct 2014 07:32:00 +0000

Ars Technica :

Popular online locker service Dropbox appears to have been hacked. A series of posts have been made to Pastebin purporting to contain login credentials for hundreds of Dropbox accounts, with the poster claiming that altogether 6,937,081 account credentials have been compromised

And this is the Dropbox answer:

Dropbox has not been hacked. These usernames and passwords were unfortunately stolen from other services and used in attempts to log in to Dropbox accounts. We’d previously detected these attacks and the vast majority of the passwords posted have been expired for some time now. All other remaining passwords have been expired as well.

While I’m changing my password, I think this is the right time for you to activate the two-factor authentication and subscribe to Have I’ve Been Pwned? website. This very useful tool automatically checks websites like Pastebin and others, where the results of those hacks are often published, and send you an email if any of the logins that you have and registered to Have I’ve been pwned are in those leaks. Very useful.

So, be careful with any third party service. And check 1Password or Dashlane too.

Lawgier.net

Mon, 13 Oct 2014 07:33:00 +0000

fma16.com become lawgier.net. “Fma16 blog” is not original enough for a blog name. I needed a pun. So I made an awful one. But I like it.

Why Lawgier?

2 things :

I study law.
In French, the first syllable of my name is pronounced like law : it’s kinda like law-ge-ey with a very flat accent.
THAT PUN IS AWESOME!!! (And that’s make 3)

So, that’s it. I’ve created a new Twitter account for it, so be sure to check it out : @lawgier.

The App That Holds iOS Back

Mon, 13 Oct 2014 07:06:00 +0000

Bradley Chambers :

When the iPhone was released in 2007 with a “desktop class” web browser, it was widely praised. You could view entire websites on a mobile device. Coming from the days of WAP optimized sites on my Motorola Q, this was incredible. If we fast forward to 2014, Mobile Safari has become the app that is holding the iPad back from becoming a fully featured laptop replacement for a lot of people. While Mobile Safari is fast and loads website reasonably well, it cannot upload and download files. It can upload pictures, but that is it. Mobile Safari needs a way to upload and download any type of file into iCloud Drive (a mirrored from the Mac downloads folder).

Android does this almost since it’s beginning, and, even if it’s implementation can be messy, it is very helpful. This should be on the top list of iOS development (after bugs squashing, of course).

GT Ask Court To Close Its Apple Plant

Fri, 10 Oct 2014 07:37:00 +0000

Re/Code delivers some new information about the GT Bankruptcy story:

Sapphire crystal maker GT Advanced Technologies, which filed for Chapter 11 bankruptcy protection earlier this week, has asked the court’s permission to “wind down” operations at its manufacturing plant, less than a year after announcing a high-profile deal to supply the material to Apple.

I think that bury even more the possibility of seeing some GT Advanced sapphire screens on future iPhones. But we don’t know many things about this bankruptcy filling (GT seems to go to great lengths to comply to a confidentiality agreement that it made with Apple).

Why GT Advanced filed for bankruptcy

Thu, 09 Oct 2014 07:42:00 +0000

Daisuke Wakabayashi, for the WSJ:

A hint of troubles at GT came last month, when Apple said it wouldn’t use sapphire screens in its new iPhones, contrary to what many observers expected. Apple added to GT’s financial pressures by not making a final $139 million prepayment loan because GT hadn’t met the technical milestones laid out by the company, the people familiar with the matter said. GT had said earlier that it expected Apple to make that payment by the end of October.

This gives a great insight about why GT filed for bankruptcy a few days ago. This information was completely unexpected for a lot of people, including me, after the news of Apple investing heavily in their sapphire production for the Apple Watch (only rumored at this time) and the iPhone :

“To be clear, we did not see this coming,” Pavel Molchanov, an analyst with Raymond James & Associates, wrote in a note Monday. “We don’t think anyone else did either.”

That’s the best explanation for the absence of sapphire screens on the iPhone 6, a feature that was widely expected but not delivered by Apple at it’s September press event. But also, it implies some serious threats about the presence of a sapphire screen in an iPhone 6s(?) at all, since it putted GT in a very delicate situation.

But we don’t know of what the future is made of.

(Thanks to John Gruber for the link)

“Your Argument Is Invalid”

Thu, 09 Oct 2014 07:39:00 +0000

The great Jason Snell, about Apple and Google decision to encrypt user’s phones:

Law enforcement officials usually play on our fears whenever their powers are limited, but those limitations are what keep our society from being a police state. The Supreme Court’s ruling in Miranda v. Arizona in 1966 led to catastrophic predictions that many criminals would go free and society would be harmed if all arrested people were informed of their rights. Didn’t happen.

Those argument against privacy always seem ridiculous for me. The U.S government, as the French one and many others, seems to forget that their goal is to protect theirs citizen, and not to know everything, and, to reach it, they can’t do whatever they want to. And privacy is a price they shouldn’t afford.

Anyways, Apple’s decision was the good one, but, the response of the government official may be just smoke and mirrors. This is what Micah Lee, from The Intercept, is saying, remembering us that Apple still have the Golden Keys for iCloud backups:

But despite these nods to privacy-conscious consumers, Apple still strongly encourages all its users to sign up for and use iCloud, the internet syncing and storage service where Apple has the capability to unlock key data like backups, documents, contacts, and calendar information in response to a government demand.

iCloud is also used to sync photos, as a slew of celebrities learned in recent weeks when hackers reaped nude photos from the Apple service.

So, Apple is moving in the right direction, but the Patriot Act may be in the way.

Hello, world!

Thu, 09 Oct 2014 07:04:00 +0000

OK, so, this is not a very original way to start a blog about technology, but I’m not a very original guy so, why not? Those are the first words of the Fma16 blog, which, indeed, is not a very original name for a blog, as would be Hypercritical, Daring Fireball, or the wonderful Liss Is More (kudos to Casey Liss for this one, very funny).

What is this?

This is a blog mainly about technology, which is not something that I work in, but is something that I love. I’m mostly interested in Apple things, but I’m still a PC guy, and an old Android user (I have a Nexus 5 in one of my drawer) so I’m not very biased (I know both sides).

Who is this guy?

I might need to introduce myself to the many people who might not know about me. My name is Paul-Henri Laugier, I study Law at the Aix-Marseille Université, I’m a technology enthusiast (an another term for geek) and I’m French.

Yep, I’m French. So, what does it implies?

It implies that English is not my native language, and, since I live in France, is not my main language. So, you might frequently stumble on grammar and vocabulary errors, even if I try my best to avoid them. So, it also means that if you see one of those, giving me feedback would be very much appreciated. I try to proofread my work, but it might be not enough. Also, I might also publish some bytes of text in French, but I’ll try my best to translate it in English, or to keep them separated from the main feed.

Why doing this?

Anyways, since I’m here, I might also try to explain my motivations for writing there. Those are quite simple:

I want to improve my writing level in English. This blog is a way to force me to write something, and also to make mistakes, and to learn from these.
I want to say what I think in more than 160 characters. Twitter is a great way to find and publish information, but sometimes, could be very restricted. Here, I’m free to do what I want, and say what I need to say.
I want to learn Markdown. Markdown is a great language for blogging and to write on the web, and is used on website like GitHub and many more. Since I’m curious, I though that learning it would be very cool (and way simpler than learning to code… I wish I could find some time to learn C, or Java, or Swift, or… . This post was written in Markup with MarkdownPad.
I want to play with blogging software. I’ll talk about this more extensively in an another post.
Others motivations that I’ve didn’t found yet.

I’ll not post here everyday, but just when I’ve found something worth talking about, or that I’ll like to talk about (which is not the same ;) ).

You can subscribe to the RSS and also follow me on Twitter if you like to be keep updated about the blog.

I’ll also make some tweaks to the website during the week, this is still a work in progress!

Have a nice day,

Paul-Henri