Clipperhouse

Dave Winer laments that mobile apps are not universally addressable via a locator like a URL.

Actually, they are, it’s just underexploited. For example, this little radio app allows me to deep-link specific radio stations as tiles on my (Win Phone 7) home screen.

Guess what? The tile has a URL, pointed to a specific place in the app. Now, I haven’t dug deeply enough to know about the security boundaries, but it seems to me it would be a small step to allow the app to register a namespace on the phone (app://appid) and expose endpoints to any other app.

This seems like a nice way to bridge the gap between the advantages of phone and web. iPhone allows protocol handlers which are the same idea, putting the namespace at the front of the URL.

(Of course, URLs would need to be managed just like public, web URLs — you’d want them to be stable, and deliberate about what’s exposed.)

A future of on-device linking may be closer than we think.

The Wall Street Journal restricts access to some stories – as when it will show you the first paragraph or two as a “Subscribe Content Preview”.

There’s an easy way to get around this if you want the whole story. (I am using the Chrome browser for this, should be similar in others.)

1. Select the story headline, and right click to “Search Google for…”

2. Hopefully the desired story will be at the top of the search results, under News…

3. Click it!

I assume this works because the WSJ likes referrals from Google, offering a better preview for potentially new subscribers. The NYT has a similar “porous” model.

It’s easy to see the web as sets of advances that come in waves. The first was its emergence as a consumer (and corporate) phenomenon, starting with Netscape and continuing through Internet Explorer 6 (which, we often forget, was the first “legitimate” browser for many businesses).

The second wave was Ajax and broadband. This combination allowed web sites to act less like pages and more like applications — responsive enough for things like email. Javascript came of age.

Even at that second milestone, however, the web as an application platform was still behind most desktop apps. I began my professional career on Filemaker, and its capabilities weren’t matched by web sites for a decade or more.

The main limitation, in my opinion, was markup + styles as the interface language. HTML and CSS are a complex of interacting and highly dependent rules. Any change to the DOM requires a recalculation of the layout. Small operations compel the browser runtime to do work that is disproportionate to the size of the change.

(Of course there are optimizations to be had, and browsers are amazing from this perspective.)

We are experiencing the third wave now, which I will call “the binary web”. Three technologies in particular — Canvas, WebGL and sockets — are the next big step forward. What these have in common is the ability to do truly granular operations.

These technologies replace a world of markup and bytes with one of pixels and bits. WebGL and Canvas allow pixels to be changed without reflowing a document; witness the frame rates of web pages that use these technologies.

Similarly, web sockets free us from the expense of the HTTP request-response cycle; the data-to-overhead ratio is greatly improved. (And the “server push” promised in the 90′s finally arrives.)

In other words, these technologies make the web computationally cheaper, perhaps by an order of magnitude. It’s a matter of economics more than technology. Think of it as Moore’s Law for the browser.

comments on Hacker News

In recent comments on Hacker News, I suggested, with some snark, that the definition of net neutrality is slippery and ad-hoc, and therefore exploitable. I was promptly voted down.

In this case, an ISP in Canada has introduced data caps – a reasonable if undesirable thing – and then went on to make their own video service exempt from counting against these caps.

Shady? Sure. Does it violate the suggested definitions of neutrality? Not really. No packets are being selectively blocked. The ISP is essentially offering free shipping for their own bits. It’s fine not to like it, but it’s not fine to think we can legislate it away.

An essential characteristic of laws is that they are gameable. They are not algorithms against which we can submit test cases and get an objective result. Their definitions are non-deterministic, and are interpreted by humans.

As such, businesses should be expected to exploit them to their own ends. In a highly regulated environment, business advantage comes not from better products, but from greater legal agility. Laws beg the market to game them.

This advantages incumbents more than anything. They can build the legal and technical expertise to influence and exploit the legal system – and this forms a wall that upstarts struggle to climb. It is not a coincidence that the telecom industry is highly regulated, and at the same time uncompetitive.

Thoughtful geeks recognize that complexity and ambiguity are the enemies of secure, reliable code. More moving parts means more unexpected outcomes.

It is natural for us to think that if we put together the correct legal code, that the problem will be fixed.  But the law is not Turing-complete. It is a fountain of exceptions.

Discuss on HN

I haven’t tried Google+ (pronounced ploo, the French way), but I note that there hasn’t been the dismissive pile-on that greeted Buzz and Wave. My meta-analysis is that they might be on to something.

It’s been noted many times that Google has a social graph, even if they haven’t presented it as such. They know who your friends are via Gmail, and they know what you are interested in via Search and its many tentacles.

Assuming they’ve created an understandable, well-implemented product, the one advantage they have over Facebook is that you are on Google anyway, and not for social reasons. They can overlay the social stuff in a place you will already be.

Simply by going about your business, you are reminded of your friends’ activities and are generating a stream of your own.

People go to Facebook because they want to be social – a deliberate step. With Google, you are already there.

There’s an adage that nothing will kill a bad product faster than good marketing. If you combine a good product with Google’s inherently massive market, then “not bad” might suffice.

–

It’s worth noting that Google hasn’t committed to making plus about “personal” or “business”. Facebook and LinkedIn position themselves on opposite ends of this spectrum.

Google’s products span both business and pleasure. Your inbox probably has both types of communication in it. Gmail itself shape-shifts between the two – in the form of Google Apps.

This blurring of lines may be +’s sleeper advantage.

It may be intemperate of me, but why do geeks respond in such a…Pavlovian way when given the opportunity to assume another person’s stupidity? The latest exhibit is Gizmodo, claiming that the New York Times has implemented their paywall poorly.

The paywall’s prima-facie purpose is to limit your ability to read articles past 20 a month, in the hope of converting you to a paid subscriber.

In truth, the NYT has very little ability to tell who you are, and therefore to count how many articles you’ve read. If you happen to log in, sure, but then all you have to do is…log out. You can clear cookies, you can open up an incognito window, you can switch browsers or computers. Or, as Gizmodo points out, you can change the URL.

One might conclude that the NYT’s strategy does not depend on enforcing the 20-article limit. It depends, instead, on identifying customers who are likely to pay. This is called segmenting.

(As an exercise for the reader, research why airline tickets with Saturday-night stays are cheaper.)

You and I, who are willing to make a small effort to anonymize ourselves, are not likely to pay. The NYT knows this. So they don’t waste effort on faux-security.

Those who will pay are those who can’t be bothered to worry about such things. They are not interested in changing URL’s or browsers. And the NYT need only convert a small percentage of them to make the exercise worthwhile.

By “circumventing” the paywall, we simply identify ourselves as unlikely customers. Perhaps it is working as intended.

Every business, if it’s smart, should realize that the Internet has the ability to drop out its pricing floor, or more precisely, to eliminate the scarcity on which price depends. This is what is usually meant by the catchphrase “disruption.”

What often happens, subtly, is that the locus of scarcity is reversed – in the consumer’s favor.

Consider information businesses, like newspapers, classified ads, Bloomberg, and even employment agencies. It used to be that the information they supplied was scarce. Thus a consumer might reasonably pay for access for such “rare” goods.

As information has become ubiquitous, its scarcity obviously falls away, and with it the pricing power of gatekeepers. But note the complementary phenomenon…

This information abundance has made consumers’ attention proportionally more scarce. The market has flipped.

The ability to reach smart, desirable people has become much more difficult. This is a common complaint among advertisers, and perhaps it’s not new.

What they should understand is that consumers’ attention is now a very liquid market. Good deals are scarce, and by “good deal” I mean underpriced eyeballs.

Advertisers might (and do) try ever more-invasive techniques. It can work tactically. But those eyeballs have choices, and invasive ads will turn many away.

Inevitably, the eyeballs will become more expensive for advertisers. They’ll need to pony up – either in $$, or in the creation of valuable content.

Consumers may gripe about advertisers’ desperate moves, but it should be understood as a sign of the consumer’s strength. They are an abundance of suitors.