From hero to zero in the blink of an eye: Cisco Shuts Down Flip

Posted on by

From today’s NYTimes.com.

It was one of the great tech start-up success stories of the last decade.

The Flip video camera, conceived by a few entrepreneurs in an office above Gump’s department store in San Francisco, went on sale in 2007, and quickly dominated the camcorder market.

The start-up sold two million of the pocket-size, easy-to-use cameras in the first two years. Then, in 2009, the founders cashed out and sold to Cisco Systems, the computer networking giant, for $590 million.

On Tuesday, Cisco announced it was shutting down its Flip video camera division.

Wow! I have a Flip. It’s a lovely gadget, and it came with quite elegant software. But I haven’t used it since I got an iPhone. Another illustration of the adage that the best camera is always the one you happen to have with you. It’s also a salutary lesson in how quickly this ecosystem can change.

Bet the guys who sold out to Cisco are laughing all the way to the bank.

James Gleick and the mystery of information

Posted on by

In today’s Observer there’s a conversation between me and James Gleick, whose book, The Information: A History, a Theory, a Flood has just been published in the UK.

Here’s a paradox: we live in an “information age” and yet information is a maddeningly elusive concept. We habitually confuse it with data, on the one hand, and with knowledge on the other. And yet it’s neither. There’s an arcane mathematical discipline called “information theory” that underpins all digital communications nowadays and yet resolutely disdains to make any connection between information and meaning. It would take a brave author to pursue such an elusive quarry. Or a foolhardy one.

James Gleick is an accomplished stalker of mysterious ideas. His first book, Chaos (1987), provided a compelling introduction to a new science of disorder, unpredictability and complex systems. His new book, The Information, is in the same tradition. It’s a learned, discursive, sometimes wayward exploration of a very complicated subject…

I had a nice email this morning from Chris Stewart, a reader in Australia, who had just seen the piece. It reminded him, he said of a limerick that did the rounds in late 1960s Information Science circles. “I have”, he writes, “no idea who wrote it and after quoting it for more than 40 years no one has claimed it …”.

“Shannon and Weaver and I
Have found it instructive to try
To measure sagacity
And channel capacity
With sigma p i log p i”

Which is a nice way of summarising Shannon’s formula for information as the measure of ‘unexpectedness’ of a message — H, as here:

“Weaver” refers to Warren Weaver who wrote a piece for Scientific American (“The Mathematics of Communication”, July 1949, p 11-15) explaining the significance of Shannon’s original paper, “A Mathematical Theory of Communication” which had been published in two issues of the Bell Systems Technical Journal in 1948. The book, The Mathematical Theory of Communication by Shannon and Weaver, was published in 1949. It consisted of Shannon’s journal articles plus Weaver’s more accessible explanation.

Maddeningly, I can’t find a copy of Weaver’s SciAm article online, though I’m sure it’s around somewhere. And the SciAm search engine denies all knowledge of Warren Weaver.

Still, apropos the ditty forwarded by Chris Stewart, it’s good to know that Limerick, Ireland’s fourth city, is located on the Shannon, which is Ireland’s largest river.

Amazon’s new Cloud Drive rains on everyone’s parade

Posted on by

This morning’s Observer column.

“Impetuosity and audacity,” wrote Machiavelli, “often achieve what ordinary means fail to achieve.” If you doubt that, may I propose a visit to the upper echelons of Apple, Google and Sony, where steam might be observed venting from every orifice of senior executives? If you do undertake such a visit, do not under any circumstances mention the word “Amazon”.

The proximate cause of all this corporate spleen is the launch last week of Amazon’s Cloud Drive service. At first sight, it seems straightforward: it looks like a digital locker in which one may for a fee securely store one’s digital assets in the internet ‘cloud’. “Anything digital, securely stored,” runs the blurb, “available anywhere.” The first 5GB of storage is free, with more available at an annual cost of a dollar per gigabyte. Upload files to your "cloud drive", where they are stored online and from where they can be accessed by any device that you own.

So far, so innocuous. It’s not the online storage business that has Apple, Google & Co spitting feathers, but the Amazon CloudPlayer which goes along with the digital locker…

The Master Switch

Posted on by

My review of Tim Wu’s new book.

At the heart of this fascinating book is one of the central questions of our age – rendered more urgent by recent events in the Arab world. The question is this: is the internet a revolutionary innovation, something that will overthrow the established order? Or will it turn out to have been just an unruly technology that the ancien regime will eventually capture and subdue?

Faced with the upheavals triggered by the network so far in economics, social life and politics, most people would probably say that the internet is indeed sui generis. But Professor Wu is not so sure, and therein lies the importance of his book. If the internet does indeed succeed in escaping the controlling embrace of corporations or governments, he argues, then it will be a historic first. For every other modern communications technology – telephone, radio, cinema and TV – has eventually succumbed to these forces…

The trouble with Moleskine notebooks…

Posted on by

Two pictures that tell both sides of the story.

EXHIBIT A

EXHIBIT B: overleaf

Of course you will object that I’m foolish to insist on using a fountain pen. Why not use a pencil like any self-respecting author? My answer is that expensive notebooks (and Moleskines are damned expensive) ought to have really good quality paper.

Federated social networking

Posted on by

There’s a useful piece on the Electronic Frontier Foundation’s site about federated networking, seen as a way of counteracting the centralising power of outfits like Facebook.

To understand how federated social networking would be an improvement, we should understand how online social networking essentially works today. Right now, when you sign up for Facebook, you get a Facebook profile, which is a collection of data about you that lives on Facebook's servers. You can add words and pictures to your Facebook profile, and your Facebook profile can have a variety of relationships — it can be friends with other Facebook profiles, it can be a ‘fan’ of another Facebook page, or ‘like’ a web page containing a Facebook widget. Crucially, if you want to interact meaningfully with anyone else’s Facebook profile or any application offered on the Facebook platform, you have to sign up with Facebook and conduct your online social networking on Facebook’s servers, and according to Facebook’s rules and preferences. (You can replace “Facebook” with “Orkut,” “LinkedIn,” “Twitter,” and essentially tell the same story.)

We’ve all watched the dark side of this arrangement unfold, building a sad catalog of the consequences of turning over data to a social networking company. The social networking company might cause you to overshare information that you don’t want shared, or might disclose your information to advertisers or the government, harming your privacy. And conversely, the company may force you to undershare by deleting your profile, or censoring information that you want to see make it out into the world, ultimately curbing your freedom of expression online. And because the company may do this, governments might attempt to require them to do it, sometimes even without asking or informing the end-user.

How does it work?

To join a federated social network, you’ll be able to choose from an array of “profile providers,” just like you can choose an email provider. You will even be able to set up your own server and provide your social networking profile yourself. And in a federated social network, any profile can talk to another profile — even if it’s on a different server.

Imagine the Web as an open sea. To use Facebook, you have to immigrate to Facebook Island and get a Facebook House, in a land with a single ruler. But the distributed social networks being developed now will allow you to choose from many islands, connected to one another by bridges, and you can even have the option of building your own island and your own bridges.
Why is this important?

Why does this matter?

The beauty of the Internet so far is that its greatest ideas tend to put as much control as possible in the hands of individual users. And online social networking is a powerful tool for the many who want a service that compiles all the digital stuff shared by family, friends, and colleagues. But so far, social networking has grown in a way that concentrates control over that information — status posts, photos, and even your relationships themselves — with individual companies.

Distributed social networks represent a model that can plausibly return control and choice to the hands of the Internet user. If this seems mundane, consider that informed citizens worldwide are using online social networking tools to share vital information about how to improve their communities and their governments — and that in some places, the consequences if you’re discovered to be doing so are arrest, torture, or imprisonment. With more user control, diversity, and innovation, individuals speaking out under oppressive governments could conduct activism on social networking sites while also having a choice of services and providers that may be better equipped to protect their security and anonymity.

Eye-Fi

Posted on by

Ten years ago if you said that I would like to have an Internet-connected camera I’d have said you were nuts. But acquiring an iPhone changed my view: I’ve found it really useful to be able to upload pictures from anywhere at any time without having to be tethered to a computer. The iPhone camera isn’t great, but as the man said the best camera is always the one you have with you, and I always have the phone. But it’d be nice to be able to have instant uploads from a better camera.

Enter Eye-Fi, an SD card which can talk to a wireless network from inside your camera. I bought one from Amazon for just under £70 — which is expensive for an SD card, but what the hell. You install some software on PC or laptop, register with Eye-Fi, put the card into your camera and — Bingo! Images are automatically uploaded. You can link your account to other services like Flickr and Facebook. And the card can do geolocation based on Wi-Fi network location.

Sounds too good to be true? In a way, it is. The system works fine, but uploads are slow unless one constrains the size and quality of the images. For shooting and uploading web-friendly jpegs it works fine: in fact it might be a good way of getting stuff in near-real-time onto Flickr. But you can’t use the full range of image quality and size available on a decent camera. So it’s got its applications, but it looks as though the iPhone camera will still find plenty of use.

It’s also got lots of embarrassing potential. Suppose, for example, you were careless with the upload settings: you might find that a set of, er, intimate pictures were attracting an admiring audience to your Flickr account or Facebook page. And, then of course, there’s this.

Paul Baran RIP

Posted on by

Paul Baran, the engineer who first thought of packet-switching (Donald Davies independently came up with the same idea later) has died at the age of 84.

Baran was one of the most entertaining and intriguing figures I came across when I was researching my history of the Internet way back in the 1990s. The story of how he came up with the idea — and of his hilarious experiences with AT&T — is told in Chapter 6. Essentially, AT&T’s position was: “this packet-switching stuff couldn’t work, but even if it did we wouldn’t allow it”. After he’d submitted his proposal for a packet-switched network to the Pentagon, Baran realised that the contract to build the pilot network would go to an agency staffed mainly by ex-AT&T engineers, concluded that they would make sure that it didn’t work and — rather than have them strangle his baby at birth — withdrew the proposal. It’s the kind of story that one couldn’t make up. And yet it happened.

Another Times, another paywall

Posted on by

Today the New York Times disappears behind a paywall. Introductory pricing is 99c for first four weeks, but the non-discounted rates seem steep, even for such a good journalistic product. After reading 20 articles over 4 weeks, you hit the wall. Then you must choose between: $15/month for web viewing + smartphone; $20/month for web access + app on a tablet; or $35/month for accessing the NYTimes on all devices

Lots of commentary around on this. I liked Frederic Filloux’s analysis which includes the observation that the NYT paywall pricing “is like the French tax system: expensive, utterly complicated, disconnected from the reality and designed to be bypassed”.

In another post, Filloux explores an intriguing option: that the Times continues to print its blockbuster Sunday Edition (which makes tons of money), while going online-only for the rest of the week. His conclusion is that this could result in annual revenue of $1 billion compared with the $1.5 billion the Times has been earning to date.