Alexis C. Madrigal, « How Netflix Reverse Engineered Hollywood », The Atlantic, 2 janvier 2014, disponible en ligne.

« They [Netflix] even offered a $1 million prize to the team that could design an algorithm that would improve the company’s ability to predict how many stars users would give movies. It took years to improve the algorithm by a mere 10 percent.
The prize was awarded in 2009, but Netflix never actually incorporated the new models. That's in part because of the work required, but also because Netflix had decided to “go beyond the 5 stars,” which is where the personalized genres come in.

The human language of the genres helps people identify with the recommendations. “Predicting something is 3.2 stars is kind of fun if you have a engineering sensibility, but it would be more useful to talk about dysfunctional families and viral plagues. We wanted to put in more language,” Yellin said. “We wanted to highlight our personalization because we pride ourselves on putting the right title in front of the right person at the right time.”
And nothing highlights their personalization like throwing you a very, very specific altgenre. »

Duong Senh, « Rotten-Tomatoes.com », AsianConnections.com, 1999, disponible en ligne.

« I had the idea a while back, but what eventually inspired me to create Rotten Tomatoes is my favorite all-time actor, Jackie Chan. Every kid has their hero(es), and one of my heroes is Jackie Chan. Since I discovered him in 1988, I would beg (and pay) my older brother to take me to Chinatown, San Francisco - a two hour drive from Sacramento - just to see a Jackie Chan movie. When "Rumble in the Bronx" eventually opened in the States, I wanted so much to support my hero that I went to see the movie every week until it was out of my local theatre - which I think ended up being about five to six weeks. On the day the movie opened, I went to my nearest magazine stands, and read up on all the reviews from every paper I can get my hands on. I would follow that by going through the web, searching for the nation's papers for reviews. Every time a Jackie Chan movie opened, I would repeat this process. After four movies - "Rumble in the Bronx," "Supercop," "First Strike," and "Operation Condor" - I had bookmarked on my web browser a collection of the nation's papers. When "Rush Hour" was about to be released, I had already spent a year as creative director for Design Reactor, an interactive web design studio that I helped started. By then, I've learned a thing or two about making web pages. I wanted to launch Rotten Tomatoes on the opening date of "Rush Hour," Jackie Chan's first american produced film since "The Protector" (1985).

"Rush Hour" was originally scheduled to be released in August 1988, so I started working on the site in early summer during my free time. I must have spent a couple weeks just thinking about the name. Originally, I just wanted to pay tribune to one of my favorite shows on TV, Siskel & Ebert, by calling the site either "Thumbs Up" or "Thumbs Down." Unfortunately - or fortunately for legal reasons - every domain name with the word "thumbs" was taken. Popular names, and variations of them, like "critics," "moviereviews," and "reviews" were also taken, so I tried my best to come up with an obscure name. I thought about a very bizarre movie that I've seen years ago called "Leolo" in which a woman was impregnated by a tomato. I thought the quality of a tomato would be a great rating system - rotten, maggot-infested tomatoes for bad movies, fresh, juicy tomatoes for good movies. The domain name "rotten" was taken; "tomato" and "tomatoes" were also taken; "rotten-tomatoes" was not. So "Rotten Tomatoes" it is.

As far as the look and format of the site goes, I wanted the front page to spoof grocery ads from Lucky. Instead of fresh, sparkling tomatoes, I would have rotten, maggot-infested tomatoes; instead of the slogan "Freshness First," I would use "Freshness Last." In the original design released around August of '98, it did look like an ad from Lucky, but it was kinda boring to have the same picture on the front-page everyday. That's been replaced with a movie collage that changes weekly. I wanted the review pages to look like a movie ad from a newspaper. The big influence here was a full-page ad I saw for "In the Line of Fire" when it was released in 1993 (funny how these things come up when you need them). The ad had what must have been over fifty something blurbs from critics. It's the first time I've seen so many blurbs on an ad!

With the name, format, and look finalized, and the pages coded, I was ready to launch the site on August 20th, but to my disappointment, "Rush Hour" was pushed back to September 18th. I launched it anyway on August 18th because the site was ready, and I was anxious to do a reviews page. The first reviews page on Rotten Tomatoes was "Your Friends and Neighbors," and the second one was "Blade." »

Jeremy Narby, Le Serpent cosmique, l'ADN et les origines du savoir [1995], Genève, Georg, 2012, p. 102-103.

« Les enzymes de lecture ne lisent que les passages de l'ADN qui codent pour la construction de protéines et d'enzymes. Ces segments, appelées "gènes", représentent seulement 3 % du génôme humain. Les 97 % restants ne sont jamais lus ; leur utilité demeure mystérieuse.

Les chercheurs ont trouvé, éparpillées dans ces parties non-codantes du texte, de nombreuses séquences sans queue ni tête, qui se répètent inlassablement, et même des palindromes, c'est-à-dire des mots ou des phrases qui peuvent être lus dans un sens ou dans l'autre. Ils ont appelé ce charabia apparent, qui constitue la plus grande partie du génome, junk DNA – de l'ADN camelote.

Dans cette "camelote", on trouve, par exemple, des dizaines de milliers de passages comme celui-ci : ACACACACACACACACACACACACACA… Il existe même une séquence, longue de trois cents lettres, qui est répétée un demi-million de fois en tout. L'ensemble des diverses répétitions occupe environ un tiers du génôme. Leur sens est à ce jour inconnu.

Les biologistes Chris Calladine et Horace Drew résument ainsi la situation : "La plus grande partie de l'ADN dans notre corps fait des choses que nous ne comprenons pas pour l'instant".

Dispersés dans cet océan de non-sens, les gènes représentent une sorte de terre ferme où le langage de l'ADN devient compréhensible : tous les mots ont trois lettres, et comme l'alphabet de l'ADN dispose de quatre caractères, il y a (4 x 4 x 4 =) soixante-quatre mots possibles. Les soixante-quatre mots du code génétique possèdent tous un sens, et correspondent soit à un des vingt acides aminés utilisés dans la construction de protéines, soit à l'un des deux signes de ponctuation ("start", "stop"). Il y a donc vingt-deux sens possibles pour soixante-quatre mots. Cette redondance a fait dire aux chercheurs que le code génétique était "dégénéré". En fait, il s'agit simplement d'un langage riche en synonymes. C'est un peu comme une langue où des mots aussi différents que "jaguar" et "panier" auraient régulièrement le même sens.

La réalité s'avère encore plus complexe lorsqu'on va dans le détail. Ainsi, à l'intérieur même des gènes, il existe de nombreux segments non-codants, appelés "introns". Aussitôt transcrits par les enzymes de lecture, ces derniers sont éliminés du message génétique par des enzymes de rédaction. Celles-ci découpent les introns avec une précision atomique et raccordent les passages réellement codants, appelés "exons". Certains gènes contiennent jusqu'à 98 % d'introns – ce qui signifie qu'ils ne comportent que deux pour cent d'information génétique. Le rôle de ces introns demeure mystérieux.

La proportion d'introns et d'exons dans le génôme humain n'est pas encore connue, car pour l'instant, seuls quatre mille gènes sont répertoriés, sur un total qui varie, selon les estimations, de cent mille à quatre cent mille. Autrement dit, on ne connaît, aujourd'hui, guère plus d'un pour cent de nos propres gènes.

Au sein de l'ADN, les passages "camelote" alternent donc avec les gènes et, à l'intérieur de ces derniers, les introns s'entremêlent aux exons qui eux-mêmes sont exprimés en un langage où presque chaque mot possède des synonymes.

Au niveau de son contenu et de sa forme, l'ADN se présente comme un langage doublement double qui s'enroule autour de lui-même.

Tout comme le langage double et entrelacé des esprits de la nature. »