Newer posts are loading.
You are at the newest post.
Click here to check if anything new just came in.

January 16 2014

Court prods FCC in unexpected direction in this week’s Verizon ruling

A court ruling this past Tuesday on FCC “network neutrality” regulation closes and opens a few paths in a three-way chess game that has been going on for years between the US District Court of Appeals, the FCC, and the major Internet server providers. (Four-way if you include Congress, and five-way if you include big Internet users such as Google — so, our chess game is coming closer to Chinese Checkers at this point.)

A lot of bloggers, and even news headlines, careened into histrionics (“Net neutrality is dead. Bow to Comcast and Verizon, your overlords“). The Free Press, although oversimplifying the impact, did correctly link the ruling to what they and many other network neutrality supporters consider the original sin of FCC rulings: eviscerating the common carrier regulation of broadband providers.

Even better, many commenters noted the ambiguities and double messages in the ruling. Unlike a famous earlier ruling on Comcast regulation, this week’s court ruling spends a good deal of time affirming the FCC’s right to regulate Internet providers. Notably, pp. 35-36 essentially confirm the value and validity of network neutrality (in the form of promoting innovation at the edges by placing no restraints on transmissions).

Let’s go over the efforts of Congress and the FCC to promote competition in Internet service, leading up to Tuesday’s ruling.

Two mandates handed down from the 20th century: Computer II and the Telecom Act of 1996

The major goal of the landmark 1996 Act was to promote competition. The market for Internet service was stunningly different in 1996 from what it is in the US today. There was a lot of competition — but not much reach and not much bandwidth. Many people still lacked Internet access, and those who got it from home dialed into an ISP, often a mom-and-pop operation. Few people could get Internet access over their cable TV network. More competition would presumably lead to more and faster Internet connections.

Although idealists (like me) looked forward to a teeming ecosystem of multiple ISPs, Congress was perhaps more realistic in expecting an oligopoly with a few big players in each geographic market (three companies was considered a good number for robust competition). Many expectations were placed on incumbents: there were seven “baby Bells” that came out of the breakup of the old AT&T, and although each occupied a separate geographic market, observers hoped they would enter each other’s markets. Cable companies were in most markets as well. Somehow, from all this raw material, new services were supposed to arise.

The law established interconnection points that the baby Bells had to provide to competitors. Theoretically, smaller companies could have exploited those points to find a market niche, but that hardly went anywhere (and many observers blamed the Bells for stymying competition). The seven baby Bells quickly recombined to make three (Verizon, CenturyLink, and a new AT&T), who competed for mobile phones but refrained from competing with landlines in each other’s regions. In many areas, a baby Bell and a cable company formed a duopoly.

Much of the country has enjoyed continuous Internet access (no dial-up) with increasing bandwidth, but many observers think the age of fiber expansion is over at both Verizon and AT&T. While the US remains far behind other developed countries in bandwidth.

Internet regulation (or lack thereof) goes back to 1966 with a series of “Computer inquiries” from the FCC. These have been universally praised for allowing the Internet to arise and spread, simply by announcing that the FCC would not regulate it. Computer II, in particular, distinguished the service offered by telephone companies over the line from the data service running through modems on either side. The Telecom Act enshrined this difference by defining “information services” that were separate from the “telecommunication services” that the FCC had long regulated as common carriers.

Telecommunication services (regulated under Title II of the law) have to provide equal, non-discriminitory access to all users. Information services do not. Clearly, companies will go to extreme lengths to evade being labeled a telecommunication service.

The big divide: cable versus baby Bell

Although we hear a lot about “digital divides” between urban and rural areas, rich and poor, white and minority (luckily decreasing), the divide I’m going to talk about here is a regulatory one. Cable companies are not common carriers; they have always been regulated differently. Local communities can require certain services (such as community and educational channels), but the cable companies are definitively free of the burdens of Title II.

Thanks to the Telecom Act, however, cable companies and telecom companies have come to look more and more alike. They all provide voice calls; they all provide TV channels; they all provide Internet access; and, increasingly, they all provide movies on demand and other services. The key problem the FCC faces — not blamable on Congress, the US District Court, or anybody in particular — is that for historical reasons it imposes much heavier requirements on telecom companies than on functionally identical cable companies. Cable companies offer both Internet transport and content of their own, all over the same physical channel — and now, telecom companies do the same. Something’s gotta give: either Title II regulation has to be imposed on cable companies, or it has to be removed from the baby Bells.

We should note, for historical context, that a Republican administration replaced a Democratic one in 2000, and in 2001 Michael K. Powell was appointed FCC chair. He brought with him a profound faith in the free market as a spur to competition and innovation. When the FCC announced in 2002 that cable modem service was an information service, Powell wrote a justification that reads almost like an apology:

The Commission does not have unconstrained discretion to pick its preferred definition or classification, as some imply. The Commission must attempt to faithfully apply the statutory definition to a service, based on the nature of the service, including the technology used and its capabilities, and the nature of the interactive experience for the consumer…The Commission is not permitted to look at the consequences of different definitions and then choose the label that comports with its preferred regulatory treatment.

But that, of course, is exactly what they did in their inquiry. “Even if Computer II were to apply, however, we waive on our own motion the requirements of Computer II in situations where the cable operator additionally offers local exchange service. The Commission, on its own motion or on petition, may exercise its discretion to waive such requirements on the basis of good cause shown and where the particular facts would make strict compliance inconsistent with the public interest.” (paragraph 45)

I’d like to argue that it was inevitable for them to jump off on this side of the fence. They could hardly evade the reasoning in paragraph 43: “The Commission has never before applied Computer II to information services provided over cable facilities. Indeed, for more than 20 years, Computer II obligations have been applied exclusively to traditional wireline services and facilities.” Regarding the alternative they saw, “to find a telecommunications service inside every information service,” they say, “Such radical surgery is not required.” In short, the technical reality behind Internet connections was irrelevant to the policy dilemma. This FCC decision is often called Brand X, after a court ruling that upheld the decision after a challenge led by an ISP of that name.

By the way, it’s not fair to consider Powell a tool of large corporations, as some critics do. He was deeply committed to the principle of free markets, and articulated four “Internet freedoms” reminiscent of Richard M. Stallman’s four software freedoms.

The sin ascribed to the FCC by Free Press and other network neutrality supporters is actually an inescapable corollary to the cable decision. In 2005 — after Powell left — they decided that new lines and equipment rolled out by telecom companies would not be subject to the common carrier requirements that had been in place for some 70 years. The decision explicitly and repeatedly refers to their Brand X cable modem ruling. They claim the change will enhance competition rather than hurting it.

I think the FCC was hamstrung by the evolution of the Internet industry. The hoped-for ecosystem of small Internet competitors was stunted and scattered. Real competition existed only among the big incumbents, both in telecom and in cable. As we’ll see, this had a major impact on campaigns among Internet activists. As for the FCC, the decisions to free those companies from common carrier status stemmed from a hope that they’d put on their boxing gloves. And they did — but the punches were aimed at the FCC rather than each other.

Giving up on the substrate

Over the past few years, advocates for more competition and growth on the Internet have tacitly moved “up the stack,” complaining about ISP practices such as interfering with certain content and their plans to charge certain Internet sites for favorable treatment. For instance, Comcast was found to be secretly throttling traffic when users were downloading large files. When unmasked, Comcast claimed it was placing restrictions on downloads to be fair to all users; critics suggested it regarded the streaming downloads as competition for its own offerings since movies played a large part in the downloads.

One can imagine that, back in the 1990s, ISP practices like this would lead to an exodus by disgusted customers. Nowadays, there’s much less choice. Network neutrality advocates seem to be taking the battle to the software layer because achieving large-scale competition at lower layers seems unattainable. Indeed, real competition would require companies to compete more on the physical layer. Meanwhile, advocates for tiered service suggest it will lower costs and encourage competition.

The FCC is caught between an aroused community of network neutrality advocates and a powerful set of industries looking for ways to increase revenue. Occasionally, it tries to intervene. But the same argument the FCC makes for removing regulation, enthusiastically accepted by the industry, is bitterly opposed when used for exerting regulation. In each case, this argument is:

  • The action we’re taking will promote investment and new services by Internet companies, such as the social networks and content providers.
  • That innovation will stimulate demand by users for more bandwidth, along with a willingness to pay.
  • That in turn leads to more investment and innovation (such as more efficient codecs for multimedia content) in Internet infrastructure.

Comcast’s secret traffic stifling led to the first court battle. In its 2010 ruling, the DC district court basically told the FCC that it had tied its own hands by refusing to regulate the cable companies as common carriers. Cable modems fall in the cracks between the various categories regulated by the Telecom Act. The FCC can’t use Title II (common carrier status). Title III (broadcasting) doesn’t permit the kinds of regulation the FCC was trying to impose. When the FCC tries to cite its mandate to regulate pricing, the court tells it that it can regulate on the basic tier.

The court essentially looked through the Telecom Act for a clause that explicitly let the FCC regulate a practice that didn’t emerge until a decade after the Act was passed, and — unsurprisingly — didn’t find one. The core of the ruling might be found on page 16: “…the Commission must defend its exercise of ancillary authority on a case-by-case basis.”

It would seem like all the players and props were on stage for the final act of the network neutrality drama. But Tuesday’s court ruling showed that the endgame is not at hand. The bottom line is the same — the FCC cannot apply its anti-discrimination and anti-blocking rules; but, as I mentioned at the beginning of the article, the court offered its own sort of encouragement.

The court essentially used a duck test. They found that the FCC regulation looked like a common carrier obligation, so they rapped its knuckles for trying to force common carrier status on companies. Because the FCC had previously removed common carrier status from these companies, the court said it couldn’t impose such regulations now.

Verizon’s lawyers started by cutting and pasting Comcast’s objections to the FCC ruling, changing section 230(b) of the Telecom Act to section 706 and adding some other distracting objections of their own. The court didn’t buy the comparison, which leaves hope for those who want the FCC to rein in ISP business practices. The court even revises the appearance of its early ruling, saying a bit snarkily, “In Comcast, we held that the Commission had failed to cite any statutory authority that justified its order, not that Comcast had never impaired Internet traffic.”

Some network neutrality advocates have reacted to the decisions and rulings I’ve discussed (as Free Press does) by asking the FCC to reverse its 2005 decision that allowed telecom companies essentially to expand as much as they want without opening up to competition. This would encounter insurmountable hurdles because government agencies have to cite compelling reasons to change any decision they’ve made, and eagle-eyed courts hold them to that high standard.

Other people trace the problem to the 1996 Telecom Act, apparently already outdated by rapid changes in the industry. I don’t have to assess the likelihood of getting Congress to take on a major revision at this time in its history, or the likelihood of Internet activists getting the result they want.

Or maybe communities will pool their resources to create their own infrastructure, a particularly bold suggestion when you consider how much it costs to string fiber between cities.

Tuesday’s ruling did not close off the FCC’s right to regulate Internet services — in fact, I think it expanded possibilities beyond the place they seemed to stand following the Comcast decision. I am not sure the current debate over things such as blocking is productive. I think much bigger forces are in play, as I discussed in my article last week about Internet centralization. However, I’ll lay odds that few, if any, lawyers will lose business as a result of Tuesday’s decision.

September 02 2013

Y a pas que la NSA dans la vie, y aussi la DEA - ou The_Wire en vrai Drug Agents Use Vast Phone…

Y a pas que la #NSA dans la vie, y aussi la #DEA - ou #The_Wire en vrai

Drug Agents Use Vast Phone Trove, Eclipsing N.S.A.’s - NYTimes.com
http://www.nytimes.com/2013/09/02/us/drug-agents-use-vast-phone-trove-eclipsing-nsas.html?pagewanted=all

The #Hemisphere Project, a partnership between federal and local drug officials and AT&T that has not previously been reported, involves an extremely close association between the government and the telecommunications giant.

The government pays AT&T to place its employees in drug-fighting units around the country. Those employees sit alongside Drug Enforcement Administration agents and local detectives and supply them with the phone data from as far back as 1987.

#telecoms #surveillance

#Verizon declined to comment

*L'accord Verizon-Vodafone à $130 milliards attendu ce lundi* NEW YORK/LONDRES, 2 septembre…

L’accord Verizon-Vodafone à $130 milliards attendu ce lundi
NEW YORK/LONDRES, 2 septembre (Reuters)

#Verizon Communications et #Vodafone Group se préparent à annoncer ce lundi un accord à 130 milliards de dollars (98 milliards d’euros) assurant à l’opérateur américain le contrôle total de sa filiale mobile Verizon Wireless, si les conseils d’administration des deux groupes donnent leur feu vert, a-t-on appris de sources proches du dossier.

Vodafone a annoncé dimanche soir dans un communiqué être en discussions avancées avec Verizon sur la vente de sa participation de 45% dans Verizon Wireless pour 130 milliards de dollars, payables en numéraires et en actions.

Rien ne permet d’assurer qu’un accord sera bien conclu, a-t-il ajouté.
« Une nouvelle annonce interviendra dès que possible », a-t-il simplement précisé.

(...)

Le montant évoqué ferait de ce rachat la troisième opération de fusion-acquisition de tous les temps après l’OPA de 203 milliards de dollars de Vodafone sur Mannesmann en 1999 et celle de 181 milliards de Time Warner sur AOL l’année suivante, en plein boom de la « nouvelle économie ».

L’acquisition couronnerait enfin les efforts de Verizon pour contrôler 100% de Verizon Wireless, le premier réseau de téléphonie mobile aux Etats-Unis, et marquerait la sortie de Vodafone du marché américain.

Le géant britannique des télécoms détient 45% de Verizon Wireless depuis la création de la coentreprise en 2000.

(...)

Verizon Communications ayant déjà le contrôle opérationnel de Verizon Wireless, le rachat des parts de Vodafone n’entraînerait pas de changement pour ses clients, mais la puissance de frappe supplémentaire du groupe américain pourrait l’aider à améliorer ses services.

L’opération laissera Vodafone, le deuxième opérateur de téléphonie mobile dans le monde, avec des actifs en Europe et dans des marchés émergents comme l’Inde, la Turquie et l’Afrique.

(...)

Le secteur des télécoms a été marqué par une série de fusions-acquisitions ces dernières années, la plus récente en date étant la prise de contrôle par le japonais SoftBank de Sprint Nextel, le troisième opérateur américain de téléphonie mobile, pour 21,6 milliards de dollars.

(...)

En 2004 déjà, Vodafone envisageait de se désengager de Verizon Wireless pour jeter son dévolu sur l’activité de téléphonie mobile d’AT&T, finalement acquise par Cingular.

Regroupement attendu au sein de la #silicon_army | #telecoms #tuyaux #câbles

Sachant que :

Le #renseignement [est] une guerre sous-marine
Le Monde | Article paru dans l’édition du 29.08.13
Cf. http://seenthis.net/messages/167232
http://seenthis.net/messages/155019

99 % du trafic intercontinental, Internet comme téléphone, transite aujourd’hui sous les océans

#internet_traffic

vous pouvez être sûr que les paquets d’informations qui transporteront votre requête passeront à un moment ou à un autre par l’un de ces « tuyaux » de l’Internet, propriété de géants comme Vodafone, Verizon ou Orange.

Comme le remarque l’organisme de recherche CorpWatch, le matériel [de tapping des #câbles_sous-marins] décrit par M. Snowden semble correspondre en tout point à celui que fournit l’entreprise américaine Glimmerglass, le CyberSweep

August 16 2013

Verizon : Get Exposed for Spying, Win $1 Billion ! | emptywheel

Verizon : Get Exposed for Spying, Win $1 Billion ! | emptywheel
http://www.emptywheel.net/2013/08/16/verizon-get-exposed-for-spying-win-1-billion

#Verizon récompensée par le gouvernement étasunien pour ses bons et déloyaux services.

Congratulations to Verizon!

Just a few months after being exposed for providing all its American customer records to the government, it just won part of a $10 billion contract to provide cloud storage for the Department of Interior that may be worth as much as $1 billion.

The U.S. Department of the Interior has selected Verizon to participate in a $10 billion, 10-year contract to provide cloud and hosting services. This is potentially one of Verizon’s largest federal cloud contracts to date.

Verizon is one of 10 companies that will compete to offer cloud-based storage, secure file transfer, virtual machine, and database, Web, and development and test environment hosting services. The company is also one of four selected to offer SAP application hosting services.

Each of the 10 agreements awarded under the Foundation Cloud Hosting Services contract has a potential maximum value of $1 billion.

Don’t worry. I’m sure the spying had nothing to do with Verizon winning this huge contract.

But I’m sure it will make Verizon much less interested in pushing the government to roll back the spying.

#surveillance #bras_d'honneur

June 10 2013

Codename Prism, Kroes zur Netzneutralität, WLAN-Haftung im Bundestag

Der US-Geheimdienst NSA hat heimlich ein umfangreiches Abhörprogramm für Google, Facebook, Apple & Co. aufgebaut, Digitalkommissarin Kroes will die Netzneutralität europarechtlich verankern, im Bundestag stellt sich die Koalition gegen entschärfte Störerhaftung. Außerdem im Wochenrückblick: Hadopi offenbar vor dem Aus, BGH zur GPS-Überwachung, SPD-Initiative gegen Routerzwang, Prozess um Apples E-Book-Preise.

Codename Prism: Aufregung über Daten-Überwachung durch die NSA

Der US-Geheimdienst NSA soll seit 2007 unter dem Codenamen „Prism” ein umfangreiches Abhörprogramm des Datenverkehrs aufgebaut haben. Das berichten der Guardian und die Washington Post. Angeblich hat die NSA Zugriff auf Daten von US-Internetriesen wie Google, Facebook, Microsoft und Apple. So würden in Rechenzentren der NSA gewaltige Datenmengen liegen – E-Mails, Fotos, gespeicherte und übertragene Dateien. Die genannten Unternehmen bestreiten, US-Behörden „direkten Zugriff” auf ihre Server einzuräumen. Schon am Mittwoch wurde ein Gerichtsbeschluss bekannt, durch den der US-Provider Verizon verpflichtet wurde, sämtliche Verbindungsdaten von Kunden an die NSA herauszugeben. Der Bundesdatenschutzbeauftragte Peter Schaar hat die Bundesregierung aufgefordert, bei der US-Regierung in Erfahrung zu bringen, ob und inwieweit auch deutsche Nutzer von „Prism” betroffen sind. US-Präsident Obama hält die Maßnahmen zur Abwehr von Terrorgefahren für geboten.
Zum Bericht bei Spiegel Online.
Einschätzung von RA Thomas Stadler (internet-law.de).

Kroes will Netzneutralität im EU-Recht verankern

Die EU-Kommissarin für die Digitale Agenda Neelie Kroes will die Netzneutralität europarechtlich verankern. Kroes fordert einen „garantierten Zugang zum vollen und offenen Internet ohne jede Blockade oder Drosselung konkurrierender Dienste”. So sollen etwa Provider keine Telefoniedienste konkurrierender Anbieter mehr behindern dürfen. Die EU will jedoch Tarifmodelle für verschiedene Surfgeschwindigkeiten nicht verhindern, so Kroes. Geplant ist auch, die Roaming-Gebühren für Mobilfunk innerhalb der EU abzuschaffen. Kroes will entsprechende Gesetzesvorschläge im Sommer vorlegen.
Zur Meldung bei Spiegel Online.

WLAN-Haftung: CDU und FDP gegen Privilegierung für offene Netzwerke

CDU und FDP haben sich im Bundestag gegen einen Privilegierungstatbestand für Betreiber offener Drahtlosnetzwerke ausgesprochen. Die SPD-Fraktion ist vergangene Woche mit einem entsprechenden Antrag im Wirtschaftsausschuss des Bundestages gescheitert. Ziel war es, die Haftung von Betreibern offener Drahtlosnetzwerke zu beschränken. Damit sollte nach Ansicht der SPD-Fraktion für Hotspot-Betreiber – etwa in Hotels oder auf öffentlichen Plätzen – Rechtssicherheit geschaffen werden. Der IT-Branchenverband Bitkom konstatiert: Die „uneinheitliche Rechtsprechung und die Gefahr von Abmahnungen haben in der Vergangenheit bereits viele Betreiber eines Hotspots veranlasst, ihre Angebote einzustellen”.
Zum Bericht bei golem.de.

Three Strikes in Frankreich: Hadopi vor dem Aus?

Frankreichs umstrittene Warnbehörde bei Urheberrechtsverstößen Hadopi und mit ihr das Three-Strikes-Modell steht offenbar vor dem Aus. Das hat die Ministerin Fleur Pellerin verkündet, die im französischen Kabinett für digitale Fragen zuständig ist. Die Regierung wolle von Netzsperren als mögliche Sanktionsmaßnahme für Urheberrechtsverletzungen absehen: Den Zugang zum Internet zu sperren sei, als wolle man den Wasserzugang kappen, so Pellerin. Ohnehin hat Hadopi seit ihrem Bestehen nur Warnungen verschickt, nicht aber die Sperrung von Internetzugängen veranlasst. Denkbar sei aber, von Netzsperren auf Geldbußen umzuschwenken.
Zum Bericht bei Süddeutsche Online.

BGH: GPS-Überwachung am Auto strafbar

Die Überwachung von Personen mittels an Fahrzeugen angebrachter GPS-Empfänger ist grundsätzlich strafbar. Das hat der Bundesgerichtshof entschieden. Angeklagt waren Betreiber einer Detektei, die im Auftrag von Privatpersonen Details über das Privatleben der zu überwachenden Zielpersonen in Erfahrung bringen sollten. Dafür brachten die Angeklagten GPS-Empfänger zur Ortung an den Pkw der Zielpersonen an. Das Landgericht Mannheim hatte die Angeklagten wegen des unbefugten Erhebens von Daten gegen Entgelt verurteilt (strafbar gemäß Paragrafen 4443 Abs. 2 Nr. 1 Bundesdatenschutzgesetz). Der Bundesgerichtshof hat die grundsätzliche Strafbarkeit des Verhaltens bestätigt; dennoch verwies er die Sache an das Landgericht zurück: So habe das Landgericht keine ausreichenden Feststellungen zu einem möglichen „starken berechtigten Interesse an dieser Datenerhebung” getroffen. Ausnahmsweise sei „etwa in notwehrähnlichen Situationen” das Merkmal des unbefugten Handelns zu verneinen. Das Urteil liegt noch nicht im Volltext vor.
Zur Meldung bei Juris.

Hamburger SPD startet Initiative gegen Routerzwang

Die Hamburger SPD-Fraktion hat eine Initiative gegen den Routerzwang der Internetprovider eingesetzt. Ansinnen der Initiative ist es, den Routerzwang gesetzlich zu unterbinden. Bietet ein Provider seinen Zugang nur unter Routerzwang an, besteht für Kunden keine Möglichkeit, einen anderen als den vom Provider gestellten Router zu verwenden. Der Hamburger SPD-Fraktionsangehörge Hansjörg Schmidt vergleicht den Gerätezwang damit, dass „plötzlich weniger Wasser aus dem Hahn kommt” nur weil man „den Duschkopf der ‚falschen’ Firma” habe. Grund: Die Provider können auf diese Weise das Onlineangebot mitbestimmen, das die Kunden nutzen, so Schmidt. Die Bundesnetzagentur sieht für ein Tätigwerden gegen den Routerzwang keine rechtliche Handhabe.
Zum Bericht bei golem.de.

E-Book-Preise: Prozess gegen Apple hat begonnen

Der Prozess gegen Apple die um angebliche Wettbewerbsverzerrung im E-Book-Markt in den USA hat begonnen. Kläger ist das US-Justizministerium. Es wirft Apple vor, durch das sogenannte Agenturmodell höhere Preise für E-Books lanciert zu haben. Vor der Einführung des iPad im Jahr 2010 hatte Amazon die Position des Marktführers im E-Books-Segment inne und zahlte den Verlagen Großhandelspreise für E-Books. Durch den Wechsel zum Agenturmodell können Verlage die Preise selbst festlegen und verpflichten sich zugleich, die E-Books anderswo nicht günstiger anzubieten – weshalb auch Amazon die Preise erhöhen musste. Apple bestreitet die Vorwürfe, für Preiserhöhungen verantwortlich zu sein.
Zur Meldung bei heise.

 

Lizenz dieses Artikels: CC BY-NC-SA.

October 18 2012

Four short links: 18 October 2012

  1. Let’s Pool Our Medical Data (TED) — John Wilbanks (of Science Commons fame) gives a strong talk for creating an open, massive, mine-able database of data about health and genomics from many sources. Money quote: Facebook would never make a change to something as important as an advertising with a sample size as small as a Phase 3 clinical trial.
  2. Verizon Sells App Use, Browsing Habits, Location (CNet) — Verizon Wireless has begun selling information about its customers’ geographical locations, app usage, and Web browsing activities, a move that raises privacy questions and could brush up against federal wiretapping law. To Verizon, even when you do pay for it, you’re still the product. Carriers: they’re like graverobbing organ harvesters but without the strict ethical standards.
  3. IBM Watson About to Launch in Medicine (Fast Company) — This fall, after six months of teaching their treatment guidelines to Watson, the doctors at Sloan-Kettering will begin testing the IBM machine on real patients. [...] On the screen, a colorful globe spins. In a few seconds, Watson offers three possible courses of chemotherapy, charted as bars with varying levels of confidence–one choice above 90% and two above 80%. “Watson doesn’t give you the answer,” Kris says. “It gives you a range of answers.” Then it’s up to [the doctor] to make the call. (via Reddit)
  4. Robot Kills Weeds With 98% AccuracyDuring tests, this automated system gathered over a million images as it moved through the fields. Its Computer Vision System was able to detect and segment individual plants – even those that were touching each other – with 98% accuracy.

December 08 2011

Commerce Weekly: Verizon drops Google Wallet

Here's what caught my attention in the commerce space this week.

Verizon to Google: Leave your Wallet at home

VerizonVerizon's decision not to support Google Wallet in a new NFC-enabled phone renewed concerns that the mobile payments landscape is in for a long turf battle. Samsung's Galaxy Nexus, an Android-powered phone in all other respects, won't support Google Wallet on Verizon because, the carrier says, Google Wallet differs from other apps in that it interacts not only with the operating system but also with "a new and proprietary hardware element in our phones" — presumably, the NFC chip.

Few thought that was the whole story: the consensus among observers was that Verizon won't ship Google Wallet because it's one of the founding partners of Isis, a competing mobile-wallet solution. Isis isn't on any phones yet, but it's planning trials in Austin and Salt Lake City later this year. Verizon teamed up with fellow telecoms AT&T and T-Mobile, along with Barclaycard US and Discover Financial Services to launch the Isis effort last autumn. Since then, Isis has signed agreements with the other major credit card services to collaborate on development.

ZDNet's James Kendrick said it's not just about Isis, but about how much Google should pay to reach a new, large pool of customers: "Google will have to pay Verizon to play." And besides, Kendrick wrote, Google has a deal with Sprint right now — though it's going to be a long, cold winter for Google if that agreement keeps it off the other major carriers.

This fragmentation is likely to be the case for a while, noted Rebecca Greenfield at The Atlantic. Mobile wallets are probably fine for enthusiasts and early adopters, but mainstream shoppers won't take it too seriously until they know they can use their mobile wallets in most of the places they go. "[W]e won't leave our wallets at home until we get a cord-cutting equivalent," Greenfield wrote. "For now, users either have to load a smorgasbord of mobile payment apps, or settle for the current half-hearted solutions."


X.commerce harnesses the technologies of eBay, PayPal and Magento to create the first end-to-end multi-channel commerce technology platform. Our vision is to enable merchants of every size, service providers and developers to thrive in a marketplace where in-store, online, mobile and social selling are all mission critical to business success. Learn more at x.com.

Crooks take swipe-and-pay literally

In a reminder that unattended payment points — such as pay-at-the-pump terminals, ATMs, and self-serve check-out lanes — are weak links in the payments security chain, shoppers at Lucky supermarkets in Northern California appear to be the victims of criminals who tampered with card readers to skim debit card and PIN data. As of Tuesday, police in Petaluma, Calif., had 112 reports from customers of the local Lucky who reported unauthorized withdrawals from their accounts.

SaveMart, Lucky's parent corporation based in Modesto, Calif., said on its website that it discovered breaches back on November 23 and replaced compromised card readers at 20 stores. Unfortunately, they missed a few stores, including the one in Petaluma.

It's not clear yet what method was used to skim the data and the PINs. BankInfoSecurity.com reports that there are several ways to capture the PIN, including compromising the PIN pad hardware inside the box. In that case, it's also possible that Bluetooth technology could be used to transmit data to a laptop in a car parked close outside. Michael's craft stores were hit by a similar breach last May, causing that company to replace 7,200 PIN pads.

"Criminals realize that retailers are understaffed to the point that swapping out a [point-of-sale terminal] will go unnoticed," McAfee consultant Robert Siciliano told BankInfoSecurity. "Once they determine the make and model of an easily swappable device, they target a chain they can easily comprise."

Starbucks succeeds with payments, moves on to augmented reality

Starbucks said it has processed more than 26 million transactions on its mobile app since launching it last January. The novelty effect appears not to have worn off: in the first nine weeks of the program, there were three million transactions. For the nine-week period starting in October, there were twice that number. The Starbucks app is a nice example of what's possible with lightweight payments when you have complete control of a closed-loop system. Customers can load up a Starbucks card by credit card online or at a store (with cash or credit). They can draw off that card's credit by clicking a button on the mobile app, which displays a barcode that Starbucks' cashiers scan to debit the card.

But payment is just one part of Starbucks' mobile strategy. In November, it introduced Cup Magic, an augmented-reality application that lets users interact with characters on its red holiday cups and on displays in its stores. After launching the associated iOS or Android app, you find drawings of the characters and view them with your phone's camera. The app identifies shapes in the characters and launches simple interactive animations, like snowflakes falling and the characters playing.

When I first read the release, I thought it was yet another way for customers to engage with their phones rather than anyone else in the store. But when my daughter and I went to a nearby Starbucks to try it out, just the opposite happened: a crowd of curious customers gathered around to see what we were laughing at. Some downloaded the app right away and began doing the same. The two cashiers, who were unaware of the app or the secret behind the character drawings, demanded to know what we were all doing. When we explained, they agreed it was pretty cool and helped us locate the other characters in the store. As we drove home with her hot chocolate, my daughter explained to me how each of those people in the store would probably go home and tell a few other people about what they saw at Starbucks this evening. I smiled and thought to myself: a viral marketer's dream.

Got news?

News tips and suggestions are always welcome, so please send them along.


If you're interested in learning more about the commerce space, check out DevZone on x.com, a collaboration between O'Reilly and X.commerce.


Related:

January 13 2011

ePayments Week: Square rounds up

Here are a few stories in payment news that caught my eye this week.

Should Square lose the square?

SquareSquare, the mobile payment system that lets anyone take credit-card payments with their iPhone, said it has raised another $27.5 million in a round that funded the company at about $240 million. The startup has been watched closely since day one, partly because Twitter co-founder Jack Dorsey is a founder here, and partly because of the simple square block that snaps into the iPhone's audio jack and turns it into a credit card terminal.

But Business Insider has some unexpected advice for the start-up: lose the dongle. Although Square's square is an iconic attention grabber — part of its system that is, in the words of COO Keith Rabois, "an intuitive product that anyone can use as easily as a microwave" — it also makes the Square system dependent on credit cards. Since Silicon Valley startups are "about disrupting dumb, slow incumbents," Business Insider wants Square to dump the credit cards and turn to a more 21st-century way of transferring money, like Bump's PayPal API.

Meanwhile Intuit, which also has a card reader that attaches to the iPhone 4, has caved to peer pressure and decided to offer a plan that gives its GoPayment hardware away for free, just like Square does. Last summer Intuit teamed up with iPhone peripheral maker Mophie to offer its Complete Credit Card Solution for $179. The new plan makes up the difference by taking a bigger cut from credit card transactions (2.7 percent instead of 1.7 percent, which is still lower than Square's 2.75 percent).

Are you using Square or GoPayment? I'd like to hear your experience. Drop me a note.



PayPal's mobile strategy gets a nod


JJ Hornblass, who blogs at BankInnovation.net, gave his Best Mobile Strategy award to PayPal for acting on the conviction (sooner or more assertively than most payment providers) that "the payments battlefield shifted irrevocably to the mobile front in 2010." Given that PayPal has been experimenting with mobile payments since 2005, one suspects the company actually made the mobile shift a few years back.

Now PayPal appears to be waiting for its customers to catch up. Hornblass links to a BNET video from a mobile conference in October 2010 with PayPal's Osama Bedier saying that PayPal expects to process $500 million in mobile payments for 2010, up from only $24 million in 2008. Bedier says the time is right with smart phones, apps, and customers all finally reaching critical mass.

While they may be a leader in mobile payment strategy, PayPal still has plenty of runway. Fellow Bank Innovation blogger Nitin Gupta says that most PayPal customers make only one transaction per year, leaving PayPal's transaction volume far behind payment champs Visa and MasterCard (according to this chart from Silicon Alley Insider).

Verizon expands iOS market

Verizon's long-awaited announcement that it would offer the iPhone can hardly be categorized as news, given how long it's been anticipated. But it did create opportunities for reflection on how this would further spur the growth of mobile commerce. Mobile-Financial.com pointed out that between Verizon and AT&T 180 million subscribers in the US will soon have access to the iPhone, and many pundits (particularly those in the Northeast US, where Verizon's network is strong and AT&T's weak) relished the chance to switch carriers so they could actually use the device to successfully make calls. New York Times tech columnist David Pogue even offered a tip to pay off early-cancellation fees: sell your AT&T iPhone to Gazelle.com, which pays for gadgets in order to keep them out of landfills.

Got news?

Suggestions are always welcome, so please send tips or news here.


September 13 2010

The distinctions and controversies of net neutrality

Network neutrality draws the kind of zealous attention that activists
like me have been hoping for years we'd get for Internet policy. I
don't think any cyber-issue has ever excited as much interest among
the general public. Nor, unfortunately, has any been described in so
many confusing and inconsistent ways.

My latest policy foray has been a small attempt to fix some of that.
I'm inviting everyone with an interest in network neutrality to view
and edit a wiki at the O'Reilly Commons site: Network Neutrality: Distinctions and Controversies.

Check the table of contents to get a sense of why polemics over
"network neutrality" get so tangled up. The term can mean at least
half a dozen different things, and there is just as much variety in
the practices that it's supposed to stop. So far as I know, this wiki
is the first disciplined attempt to distinguish the various threads.

I started the wiki because I think we need it. Just over the past few
weeks we've been treated to news coverage of href="http://www.scribd.com/doc/35599242/Verizon-Google-Legislative-Framework-Proposal">a
joint proposal from Google and Verizon, which I found href="http://radar.oreilly.com/2010/08/what-i-get-and-dont-get-about.html">muddled
in ways that show why we need a finer understanding of the many topics
involved. The FCC has released a href="http://www.fcc.gov/Daily_Releases/Daily_Business/2010/db0901/DOC-301262A1.pdf">request
for comments that shows they're trying to hone in on the
distinctions. And a recent href="http://www.praxagora.com/andyo/ar/network_neutrality_fear_level.html">article
where I made an initial stab at dissecting the arguments was well
received and summarized in href="http://www.forbes.com/2010/09/02/internet-google-verizon-technology-net-neutrality.html">Forbes
online.

The next step is up to you. The wiki just digs down one or two levels,
and can benefit from lots of alternative viewpoints. It should also
have more references. The href="http://commons.oreilly.com/wiki/index.php/Uploading/contribution_Policy">license
on the wiki is pretty permissive about copying and reuse, and I'll
be happy to donate my contributions to any other worthy project. I ask
you to help me moderate it to cut down on the all-too-common
rhetorical tricks that muddy the issues instead of delineating them. I
know that many philosophical positions and intriguing proposals are
hard to fit within the existing format of the wiki, and I think those
positions and proposals need to be outlined elsewhere. Let's try to
make the wiki rich but brief.



Related:


August 30 2010

The network neutrality debate: It all depends on what you fear

The debate over network neutrality, which saw a spike in news reporting two weeks ago over the joint Google/Verizon proposal, confuses a lot of laypeople -- and not just laypeople -- because of all the different levels on which it's being argued and the opposing ways language is used by different participants. I discuss the use of loaded language such as "censorship" and "innovation" in an article on my website, which is also being published today in the American Reporter.

In that article, I suggest three levels into which one can roughly
divide the arguments made on each side. Competition is the
basis for the controversy, but both sides like to extend it to issues
of censorship. Most subtle, and most important, are questions
of creativity, usually aired as concerns about "innovation"
or a recently invented term, "generativity."

Each of these concepts appears different to those debating them
because they also apply at multiple levels of networking. For
instance, some people are most concerned with competition, investment,
and innovation at the application level (Google, Yahoo!, and so on)
while others worry more about the lower levels of networking
infrastructure: who will supply fiber and routers to make the Internet
work? I end the article with a bit about motivations and sources for content on the Internet.

You can find the full piece here.

Related:

August 11 2010

What I get and don't get about the Google/Verizon proposal

Nobody knew for a long time what Google and Verizon were cooking up on
the network neutrality front, and after the release of their brief,
two-page roadmap (posted href="http://www.scribd.com/doc/35599242/Verizon-Google-Legislative-Framework-Proposal">On
Scribd as a PDF, among other places) nobody still knows. All the
usual Internet observers have had their say, and in general the
assessment is negative.

My first reaction was to ignore the whole thing, mainly because the
language of the agreement didn't match any Internet activity I could
recognize. Some of the false notes struck:

  • The Consumer Protections section keeps using the term "lawful" as if
    there was a regulatory regime on the Internet. Not even the people
    regularly accused of trying to extend government control over the
    Internet (ICANN, WSIS, and the ITU) believe they can define what's
    lawful and make people stick to it.

    If I can send and receive only lawful content, who do Google and
    Verizon think can stop me from exchanging child pornography or
    instructions to blow up buildings? What, in turn, distinguishes lawful
    applications and services from unlawful ones (outside of Saudi Arabia
    and the United Arab Emirates)?

    Deduction: This passage represents no meaningful or enforceable rules,
    but is thrown in to make regulators feel there's a policy where in
    fact there is none.

  • The Transparency section strews around nice, general statements no one
    could complain about--don't we all want our services to tell us what
    they're doing?--but the admonitions are too general to interpret or
    apply.

    For instance, Apple is adamant about its right to determine what apps
    are available to iPhone and iPad buyers. Is that transparency?
    Apparently not, because every Apple developer gnaws his fingernails
    waiting to hear whether and when his app will be accepted into the App
    Store. But I don't see language in the Google/Verizon transparency
    section that covers the App Store at all. They might well say it's not
    networking issue.

    Fine, let's turn to networking. The carriers maintain that they need
    flexibility and a certain degree of secrecy to combat abuses such as
    spam; see for instance my blog href="http://www.oreillynet.com/onlamp/blog/2008/04/consider_the_economics_in_netw.html">Consider
    the economics in network neutrality. Squaring this complex
    issue--which is covered by the Google/Verizon in the next item on
    Network Management--with transparency is a dilemma.

    Deduction: we can all say we're transparent and feel good, but life is
    too complex for authorities to be totally transparent about what
    they're transparent about.

  • The worst passage in my view is the one in the Regulatory Authority
    section assigning authority to the FCC for "broadband." That
    ill-defined term, used far too much in Washington, tends to appear in
    the context of universal service. One can regulate broadband by such
    things as providing incentives to build more networks, but the
    Regulatory Authority section sidesteps the more basic questions of who
    gets to regulate the building, interconnecting, and routing through
    networks.

    Deduction: Google and Verizon put this in to encourage the government
    to continue pouring money into the current telcos and cable companies
    so they can build more high-speed networks, but its effect on
    regulation is nil.

Not too inspiring on first impression, but because so many other
people raised such a brouhaha over the Google/Verizon announcement, I
decided to think about it a bit more. And I actually ended up feeling
good about one aspect. The proposal is really a big concession to the
network neutrality advocates. I had been feeling sour about proposals
for network neutrality because, as nice as they sound in the abstract,
the devil is in the details. Network management for spam and other
attacks provides one example.

But the Google/Verizon announcement explicitly denounces
discrimination and mandates adherence to Internet standards. (Of
course, some Internet standards govern discrimination.) It seems to me
that, after this announcement, no network provider can weep and wring
its hands and claim that it would be unable to do business on a
non-discriminatory basis. And network neutrality advocates can cite
this document for support.

But as others have pointed out, the concession granted in the
"Non-Discrimination Requirement" section is ripped away by the
"Additional Online Services" section to "traffic prioritization." This
makes it clear that the "services" offered in that section reach deep
into the network infrastructure where they can conflict directly with
public Internet service. Unless someone acknowledges the contradiction
between the two sections and resolves it in a logical manner, this
document becomes effectively unusable.

What about the other pesky little exemption in the proposal--wireless
networks? Certainly, a lot of computing is moving to mobile devices.
But wireless networks really are special. Not only are they hampered
by real limits on traffic--the networks being shared and having
limited spectrum--but users have limited tolerance for unwanted
content and for fidgeting around with their devices. They don't want
to perform sophisticated control over transmission over content; they
need someone to do it for them.

Anyway, fiber is always going to provide higher bandwidth than
wireless spectrum. So I don't believe wireless will become
dominant. It will be a extremely valuable companion to us as we walk
through the day, saving data about ourselves and getting information
about our environment, but plenty of serious work will go on over the
open Internet.

So in short, I disdain the Google/Verizon agreement from an editor's
point of view but don't mind it as a user. In general, I have nothing
against parties in a dispute (here, the telephone companies who want
to shape traffic and the Internet sites who don't want them to)
conducting talks to break down rigid policy positions and arrive at
compromises. The Google/Verizon talks are fraught with implications,
of course, because Google is a wireless provider and Verizon
distributes lots of phones with Google's software and services. So I
take the announcement as just one stake in the ground along a large
frontier. I don't see the proposal being adopted in any regulatory
context--it's too vague and limited--but it's interesting for what it
says about Google and Verizon.

Older posts are this way If this message doesn't go away, click anywhere on the page to continue loading posts.
Could not load more posts
Maybe Soup is currently being updated? I'll try again automatically in a few seconds...
Just a second, loading more posts...
You've reached the end.

Don't be the product, buy the product!

Schweinderl