After much pestering from Lerouge, I switched my #search habits from #Google to ►https://duckduckgo.com - #privacy and Google’s increasingly annoying filter #bubble convinced me : the effect of wearing blinders outweighs the benefits of #personalization.
Une belle collection des logiciels alternatives contre #PRISM et la surveillance digitale se trouve chez ►http://prism-break.org (de l’EFF)
Utiliser Google search sans les inconvénients de PRISM
►https://www.startpage.com/eng/prism-program-exposed.html
Programs like PRISM undermine our Privacy, disrupt faith in governments, and are a danger to the free Internet.
StartPage and its sister search engine Ixquick have in their 14-year history never provided a single byte of user data to the US government, or any other government or agency. Not under PRISM, nor under any other program in the US, nor under any program anywhere in the world. We are not like Yahoo, Facebook, Google, Apple, Skype, or the other US companies who got caught up in the web of PRISM surveillance.
Here’s how we are different:
StartPage does not store any user data. We make this perfectly clear to everyone, including any governmental agencies. We do not record the IP addresses of our users and we don’t use tracking cookies, so there is literally no data about you on our servers to access. Since we don’t even know who our customers are, we can’t share anything with Big Brother. In fact, we’ve never gotten even a single request from a governmental authority to supply user data in the fourteen years we’ve been in business.
StartPage uses encryption (HTTPS) by default. Encryption prevents snooping. Your searches are encrypted, so others can’t “tap” the Internet connection to snoop what you’re searching for. This combination of not storing data together with using strong encryption for the connections is key in protecting your Privacy.
Our company is based in The Netherlands, Europe. US jurisdiction does not apply to us, at least not directly. Any request or demand from ANY government (including the US) to deliver user data, will be thoroughly checked by our lawyers, and we will not comply unless the law which actually applies to us would undeniably require it from us. And even in that hypothetical situation, we refer to our first point; we don’t even have any user data to give. We will never cooperate with voluntary spying programs like PRISM.
StartPage cannot be forced to start spying. Given the strong protection of the Right to Privacy in Europe , European governments cannot just start forcing service providers like us to implement a blanket spying program on their users. And if that ever changed, we would fight this to the end.
Bon, c’est leur page de publicité, mais je fais confiance à nos voisins à cause de leur merveilleux #fromage.
Ce genre de services me pose un problème de fond : il traite le symptôme sans en traiter l’origine.
Le Web est un modèle centralisé dans lequel le serveur stocke beaucoup trop d’informations sur l’usager. Pour cette seule raison, c’est l’instrument idéal de la mise sous surveillance globale de la population. C’est donc lui dont il faut se débarasser, une bonne fois pour toutes.
Les #Google Maps ne peuvent tuer l’espace public - Cyborgology
http://thesocietypages.org/cyborgology/2013/06/05/google-maps-cant-kill-public-space-a-belated-reply-to-evgeny-morozo
David Banks répond à la tribune d’Evgeny Morozov dans Slate - ►http://www.slate.com/articles/technology/future_tense/2013/05/google_maps_personalization_will_hurt_public_space_and_engagement.html . Si les cartes sont toujours politiques, le processus de sélection a toujours été à l’origine des cartes. Si Banks reconnait que l’imprévisibilité est nécessaire dans l’espace urbain, il soutient que Google ne fait pas assez pour que la personnalisation de ses cartes lui permette de s’intégrer à nos vies (...)
Même problème et même solution que pour toute autre source d’information : don’t hate the media, be the #media ! Hier et aujourd’hui avec la #Toile, aujourd’hui avec #Openstreetmap les utilisateurs ont la possibilité de d’opter pour la liberté - reste à savoir si c’est ce qu’ils choisiront.
“Printing out the Internet”
A crowdsourced project to literally print out the entire internet.
http://printingtheinternet.tumblr.com
~~ In memory of #Aaron_Swartz ~~
LABOR, #UbuWeb and #Kenneth_Goldsmith invite you to participate in the first-ever attempt to print out the entire #internet.
http://labor.org.mx
►http://ubu.com
http://en.wikipedia.org/wiki/Kenneth_Goldsmith
The idea is simple: print out as much as of the web as you want — be it one sheet or a truckload — send it to Mexico City, and we’ll display it in the gallery for the duration of the exhibition, which runs from July 26 to August 30, 2013.
The process is entirely open: If it exists online and is printed out, it will be accepted. Every contributor will be listed as a participating artist in the show and will be listed on this Tumblr.
What you decide to print out is up to you — as long as it exists somewhere online, it’s in. We’re not looking for creative interpretations of the project. We don’t want objects. We just want shitloads of paper. We’re literally looking for folks to print out the entire internet. We have over 500 square meters of space to fill, with ceilings that are over 6 meters high.
There are many ways to go about this: you can act alone (print out your own blog, Gmail inbox or spam folder) or you could organize a group of friends to print out a particular corner of the internet, say, all of Wikipedia, the entire New York Times archive, every dossier leaked by Wikileaks for starters. The more the better.
Print out the internet. Post it to Mexico City.
Send your printouts to LABOR in Mexico City by July 15th:
LABOR
Francisco Ramírez #5
Col. Daniel Garza
Del. Miguel Hidalgo 11830
México D.F.
At the conclusion of the show, the entire archive will be recycled. No materials can be returned.
All queries and questions: printingouttheinternet@gmail.com
https://twitter.com/internetprint
[là il y a un lien Fcbk, mais j’arrive vraiment pas à le mettre]
@ubuweb
https://twitter.com/ubuweb/statuses/340851591269740545
What would it take to print out the Internet? 39,110,491,072 boxes of paper. We’re working on it.
The size of the World Wide Web (The Internet)
http://www.worldwidewebsize.com
Et de ferventes oppositions à ce projet #ubuesque :
Petition “Please don’t print the internet”
https://www.change.org/petitions/please-don-t-print-the-internet
A sustainable responsible approach to life requires that humans adopt a conservative approach to resource usage.
While it is appreciated that Mr Goldsmith plans to recycle the paper used in his art exhibit, reduction of usage is more important than post-usage recycling.
Chemicals harmful to the environment are used in the production of paper, and the recycling of paper, not to mention the amount of plastic and ink waste from toner cartridges, ink cartridges and other non-paper related supplies which may pollute the landfill.
“What would it take to print out the Internet?”
http://www.techhive.com/article/2040466/what-would-it-take-to-print-out-the-internet-.html
Let’s assume that the each Web page can be printed out on a single sheet of 8.5 x 11 paper. Granted, some pages are larger than others, and “infinitely scrolling” pages like Twitter or Google images completely break the model. But for all of the modern, formatted pages with dozens of images that scroll on and on, let’s assume that there just as many older pages with archived content, buried in the back corners of the Web. Waving our magic wand, we now have 4.73 billion sheets of paper.
Burj KhalifaWikipediaThe Burj Khalifa, Earth’s tallest building.
So how big is that? Well, each ream of paper—500 sheets—is about 5.2 centimeters thick, with one sheet equaling 0.0104 cm thick. Multiply that by 4.73 billion, and we get a stack of paper 491,920 meters high, or 1.61 million feet, or 305.67 miles.
That’s about 55.7 times the height of Mt. Everest—or, because it’s my favorite building—582.2 times the height of the Burj Khalifa in Dubai, the world’s tallest building.
But that’s also a very narrow stack, and it certainly wouldn’t stay upright in a stiff breeze. In fact, if it tipped, it would fall nearly the distance from Los Angeles to San Francisco, as the crow flies.That’s where Goldsmith’s storage space comes in.
Actually, if all our assumptions hold true (and there’s a lot of wiggle room built in) Goldsmith’s storage space isn’t nearly enough to hold the Internet. A 500 square meter warehouse with six-meter ceilings equals 3,000 cubic meters. An 8.5 x 11 sheet of paper works out to 216 mm x 279 mm (or 0.216 m x 0.279 m); multiply that by our 49,192-meter stack, and we get 29,645 cubic meters, without gaps. Not even close.
(...)
If you begin factoring in email into the definition of “Internet,” then the solution really blows up.
In fact, Royal Pingdom estimated that in 2012, 144 billion emails per day were sent over the Internet. That’s 52,560,000,000,000 emails per year. Assuming each email represents a piece of paper, and adding back our 4.5 billion Web pages just for fun, we get a stack 52,564,500,000,000 sheets high. That’s a stack 5,466,708,000 meters high, or 117 billion feet, or 3,396,850 miles. That’s 14 separate stacks of paper from the Earth to the Moon, or 11,634,300,000 cubic feet, just for one year of email, incluidng the total amount of the Web’s content.
Assuming our math is correct (and, knowing the Internet, you’ll check it) we think we can safely assume that Goldsmith doesn’t have a clue as to the size of the Internet.
And what would it cost?
The United States Postal Service offers a handy international box option: 5.5 inches deep, and sized for standard 8.5 x 11 paper. By my calculations, the box will hold 1,344 pages.
To satisfy Goldsmith’s request ( a year’s worth of email, plus the entire Web) would require 39,110,491,072 separate boxes. At USPS rates, the total cost would be $3.126 trillion, excluding printing costs.
Don’t bother.
@internetprint
https://twitter.com/internetprint/status/346621697614548994
#printingtheinternet is about the fear of information & the terror of facing the monstrous amount of data we mindlessly produce every day.
Mes cartes ou les vôtres ? - Slate.com
►http://www.slate.com/articles/technology/future_tense/2013/05/google_maps_personalization_will_hurt_public_space_and_engagement.html
Evgeny Morozov revient sur la personnalisation des Google Maps annoncée récemment par Google. Annoncent-elles la fin de l’espace public ? La meilleure façon de conserver la publicité comme pilier de son activité, pour Google, « est de nous transformer en créatures hautement prévisibles en limitant artificiellement nos choix ». Mais l’enjeu n’est pas que sur les cartes, avec les voitures autonomes et les lunettes de réalité augmenté, Google s’apprête à modifier profondément la façon dont nous faisons (...)
Qu’y a-t-il de bon sur YouTube ce soir ? — Pegleg
http://pegleg.it
Pegleg crowd-sources full-length movies from #YouTube
(bon, j’ai #free, donc c’est mort pour moi… mais tu vois l’idée)
et pendant que #youtube diffuse tout ça :
#BitTorrent Site isoHunt Demands Jury Trial | Wired
http://www.wired.com/threatlevel/2013/04/isohunt-demands-trial
The appeals court had ruled that isoHunt, and Fung’s related sites, TorrenTBox and Podtropolis, unlawfully pointed the way to free movies, music, videogames and software that were copyrighted and not authorized for Fung to help distribute. The court said Fung had so-called “red-flag” knowledge of the infringement, and didn’t take enough steps to rectify it.
Rothken demanded a trial, saying Fung’s activities are no different than Google, for example, which also hosts links to infringing material.
Le racisme empoisonne la diffusion de publicité en ligne - Technology Review
http://www.technologyreview.com/view/510646/racism-is-poisoning-online-ad-delivery-says-harvard-professor
Pour Latanya Sweeney de l’université d’Harvard, la discrimination raciale s’est emparée de la publicité en ligne. En entrant son prénom dans Google (qui semble connoté à la communauté noire) des #publicités pour un service de renseignement en ligne lui propose de savoir si elle a été arrêtée, ce qui semble suggérer qu’elle a un casier judiciaire. Les prénoms les plus corrélés à la communauté noire américaine tendent à servir cette publicité 25 % de fois plus souvent que les prénoms attachés à un public blanc, (...)
Facebook’s Graph Search : Kiss Your Privacy Goodbye
http://slashdot.org/topic/cloud/facebooks-graph-search-kiss-your-privacy-goodbye
Over the past couple of weeks, Facebook has made a lot of hoopla about its new “Graph Search.” Regardless of one’s personal feelings about Facebook itself, the social network is deeply baked into the fabric of the Web—so whenever they introduce anything new, it has the potential to greatly impact Web developers.
Privacy advocates are also concerned about Facebook’s new search tool. What does it really do? How does it compare to sites like Google with regard to tracking personal information.
Let’s get under Graph Search’s hood and see whether those privacy concerns are justified.
#Sphinx in action | Ivinco
http://www.ivinco.com/blog/tag/sphinx-in-action
Top related queries
http://www.ivinco.com/blog/sphinx-in-action-top-related-queries
Fuzzy matching and 2nd pass query
http://www.ivinco.com/blog/sphinx-in-action-fuzzy-matching-and-2nd-pass-query
How Sphinx handles text during indexing
http://www.ivinco.com/blog/sphinx-in-action-how-sphinx-handles-text-during-indexing
Autocomplete
http://www.ivinco.com/blog/sphinx-in-action-autocomplete
It all starts with indexing
http://www.ivinco.com/blog/sphinx-in-action-it-all-starts-with-indexing
Did you mean … ?
http://www.ivinco.com/blog/sphinx-in-action-did-you-mean
Good and bad in Sphinx real time indexes
http://www.ivinco.com/blog/sphinx-in-action-good-and-bad-in-sphinx-real-time-indexes
où je constate que j’ai pas fait assez attention au risque de collisions avec #CRC32 :
http://www.quora.com/Hashing/What-is-the-probability-of-having-CRC32-collision-on-simple-strings
http://stackoverflow.com/questions/4940316/sphinx-and-guids
=> il faudrait compiler sphinx avec enable-id64, cf http://seenthis.net/messages/101256 ; pour l’instant je n’ai pas réussi, et j’ai adopté une autre stratégie de numérotation
#documentation
DuckDuckGo Architecture - 1 Million Deep Searches a Day and Growing
http://highscalability.com/blog/2013/1/28/duckduckgo-architecture-1-million-deep-searches-a-day-and-gr.html
This is an interview with Gabriel Weinberg, founder of Duck Duck Go and general all around startup guru, on what DDG’s architecture looks like in 2012.
Innovative search engine upstart DuckDuckGo had 30 million searches in February 2012 and averages over 1 million searches a day. It’s …
http://farm9.staticflickr.com/8227/8424385194_15feca732d_o.png
Source: High Scalability - High Scalability - Todd Hoff
Google vs. the press : avoiding the lose-lose scenario | Monday Note
http://www.mondaynote.com/2013/01/20/google-vs-the-press-avoiding-the-lose-lose-scenario
Argument de Filloux : si Google ne donne pas de fric aux journaux, il va politiquement souffrir, des années durant — et si les journaux s’entêtent, ils n’auront rien ; j’avoue que je ne suis pas très convaincu par la première partie de l’argument. Par ailleurs, s’il doit se résoudre à en arriver là, Google aura plus vite fait de s’inspirer de Niel, et de présenter sa dépense comme un investissement…
#google #presse #fiscalité
Shodan Search Engine Project Enumerates Internet-Facing Critical Infrastructure Devices
https://threatpost.com/en_us/blogs/shodan-search-engine-project-enumerates-internet-facing-critical-infrast
The duo, Bob Radvanovsky and Jacob Brodsky of consultancy InfraCritical, have with some help from the Department of Homeland Security pared down an initial list of 500,000 devices to 7,200, many of which contain online login interfaces with little more than a default password standing between an attacker and potential havoc. DHS has done outreach to the affected asset owners, yet these tides turn slowly and progress has been slow in remedying many of those weaknesses
The press, Google, its algorithm, their scale | Monday Note
http://www.mondaynote.com/2012/11/04/the-press-google-its-algorithm-their-scale
for Google, the most data-driven company in the world, having newspapers articles in its search system is no more than small cool stuff.
quelques #chiffres qui indiquent assez clairement que ce n’est pas la #presse qui fait la force des algorithmes de #google
#algorithme
On a longtemps cru que le Pagerank ...
http://affordance.typepad.com//mon_weblog/2012/11/on-a-longtemps-cru-que-le-pagerank.html
Europe Presses Google to Change Privacy Policy - NYTimes.com
http://www.nytimes.com/2012/10/17/business/global/17iht-google17.html
PARIS — European privacy regulators on Tuesday warned the company to clarify those issues — or risk fines or other penalties by early next year.
http://www.cnil.fr/nc/la-cnil/actualite/article/article/regles-de-confidentialite-de-google-une-information-incomplete-et-une-combinais
http://www.franceculture.fr/2012-10-16-la-cnil-vers-un-conflit-avec-google
http://www.franceculture.fr/player/export-reecouter?content=4519689
The letter to Mr. Page is only the latest addition to a growing list of regulatory headaches for Google. Antitrust officials at the European Commission are investigating whether Google has used its search engine to favor its own services and through preferential rankings to put competitors at a disadvantage. A similar inquiry is under way at the U.S. Federal Trade Commission.
Google Product Search To Become Google Shopping, Use Pay-To-Play Model
http://searchengineland.com/google-product-search-to-become-google-shopping-use-pay-to-play-mo
Google Product Search is getting a new name, Google Shopping, and a new business model where only merchants that pay will be listed. It’s the first time Google will decommission a search product that previously listed companies for free. The company says the change will improve the searcher experience, but it will also likely raise new worries that Google may further cut free listings elsewhere.
(En même temps, je l’avais jamais remarqué, cet onglet de Google.)
The Terrifying Ways #Google Is Destroying Your #Privacy | | AlterNet
http://www.alternet.org/story/155479/the_terrifying_ways_google_is_destroying_your_privacy?page=entire
(rien de spécial) via @opironet
Une liste d’outils pour les journalistes, blogueurs (et pour toi aussi)
http://www.newsressources.com/liste-des-outils-utiles-pour-journalistes-blogueurs
Vous trouverez ci-dessous une liste des quelques sites indispensables à connaître.
Elle ne demande qu’à être enrichie, annotée, commentée via un Google docs ouvert à vos contributions.
Cette page est actualisée à peu près chaque semaine – Dernière mise à jour jeudi 8 mars 2012.
Dedans, beaucoup de cloud (des outils accessibles via Internet) et quelques liens bien intéressants dont l’outil de recherche Creative Commons :
http://search.creativecommons.org
Sphinx : http://sphinxsearch.com
“Sphinx is an open source full text search server, designed from the ground up with performance, relevance (aka search quality), and integration simplicity in mind. It’s written in C++ and works on Linux (RedHat, Ubuntu, etc), Windows, MacOS, Solaris, FreeBSD, and a few other systems.
Sphinx lets you either batch index and search data stored in an SQL database, NoSQL storage, or just files quickly and easily — or index and search data on the fly, working with Sphinx pretty much as with a database server.”
Le projet #Tatoeba était aussi passé à #Sphinx (http://blog.tatoeba.org/2010/04/switching-from-lucene-to-sphinx.html) indiquant que la vitesse d’indexation était plus rapide :
«Now everything is soooo fast, it’s awesome. Besides, indexing with Sphinx only takes 30-60 seconds (compared to 15-20 minutes with our 3 year-old Lucene code). So we can afford to index much more often.»
T’inquiète, c’est pour le boulot.
http://www.alexa.com/topsites/countries/SA
Top Sites in Saudi Arabia
The top 500 sites in Saudi Arabia.
Energy.gov: Where information goes to die
http://www.thebulletin.org/web-edition/columnists/dawn-stover/energygov-where-information-goes-to-die
Wait a minute — 22 results? For a decades-long research effort that cost $10 billion and generated more than three million documents exploring one of the most contentious issues of our time?
Larry Page’s “Beautiful” New Google | PandoDaily
http://pandodaily.com/2012/01/25/larry-pages-beautiful-new-google
As someone who goes to Google more times a day than any other site, I hope I’m wrong about how deep this change in philosophy is among Google’s leadership. I already have two companies who make sense of my world for me: Twitter and Facebook. I don’t need a third. I need a clean, reliable search engine and email service I can trust.
J’ai activé DuckDuckGo comme moteur de recherche par défaut pour voir. Une petite comparaison des résultats avec ceux de Google montre que le biais "social" de Google est déjà très fort...
Si je puis me permettre de conseiller un moteur de recherche qui à mon sens est meilleur que Google et Duckduckgo, c’est sans aucun doute seeks.fr
Meilleur que Google car les résultats sont au moins aussi pertinents et la « recherche sociale » est également présente à ceci près qu’elle est sur le réseau Seeks totalement anonyme.
Meilleur que Duckduckgo car Seeks ne se contente pas d’aller chercher ses résultats sur les moteurs de recherche existants comme le fait DDG, mais il fait de l’indexation par la curation. Les résultats les mieux classés sont ceux sur lesquels les utilisateurs auront cliqué le plus souvent.
Le projet a en plus le mérite d’être 100% français.
Plus d’infos sur http://seeks-project.info/wiki/index.php/FAQ_fr#Qu.27est-ce_que_Seeks_.3F
2011: Piracy Wars and Internet Censorship | TorrentFreak
http://torrentfreak.com/piracy-wars-and-internet-censorship-111231
January
After pressure from the entertainment industries, Google started to censor piracy-related keywords from its instant and autocomplete services. Keywords such as ‘torrent,’ ‘BitTorrent’ and ‘RapidShare’ were excluded from the start, and later in the year Google added a wide range of new terms including ‘The Pirate Bay.’
Google pirate bayDespite these efforts the RIAA remained dissatisfied, patronizing the search engine with a could-do-much-better “Report Card” in December.
ArXiv at 20 (Nature)
►http://www.nature.com/nature/journal/v476/n7359/full/476145a.html
Configuring scholarly communication infrastructure for the next generation of researchers requires getting into the heads of current undergraduates and graduate students. Their life experience is of immediate online availability and global search engines, and they arrive imbued with the social-network mentality of sharing links, photos, videos and status updates. Yet, my own informal survey of graduate students reveals information-gathering techniques familiar to most older scientists. Students still follow citation trees, search by keywords and consult with peers and mentors, with the latter as important as ever for weeding out unreliable sources. Students also say that they search preferentially for open-access resources when working from home, because accessing subscription-based journals, even when available through an institutional proxy, can be frustratingly painful. (...) Source: Nature
The Great Tech War Of 2012 | Fast Company
http://www.fastcompany.com/magazine/160/tech-wars-2012-amazon-apple-google-facebook
#amazon #apple #google #facebook #tech_companies
via @opironet
Et pour les récents développements cf. http://www.fastcompany.com/1788728/the-great-tech-war-of-2012-ongoing-skirmishes
Google investigated for tax evasion - Shifted money off shore | TechEye
http://news.techeye.net/business/google-investigated-for-tax-evasion#utm_source=thetecheye
According to the Sydney Morning Herald, Google had cut its taxes by $3.1 billion in the three years prior using a technique that moves most of its foreign profits through Ireland and the Netherlands to Bermuda.
Amongst the financial geniuses, who bought us the world economic crisis, Google cut its worldwide tax bill by about $1 billion a year using a pair of strategies called the “Double Irish” and “Dutch Sandwich”. These move profits through units in Ireland, the Netherlands and Bermuda.
This meant that it could report an effective tax rate of 18.8 percent in the second quarter, less than half the average combined US and state statutory rate of 39.2 percent.