Browsers should have been cars. Instead they’re shopping carts.

netscape-navigator-icon-199x300Back in 1995, one of my wife’s sisters became one of the first executives at a hot new startup called Netscape. We wore Netscape t-shirts, used Netscape’s browser, and paid close attention to what was happening in Netscape’s space, which was the entire Web.

One of the first things to happen on that Web was e-commerce: Amazon, eBay, and the rest of them. All of them had their own shopping carts, and systems for managing sales and service. Which was fine, except for one dysfunctional pre-Web retail default assumption that my wife spotted one day. She wasn’t meaning to spot it, but rather exposed it with a naive but deeply profound question:

Why can’t I take my shopping cart from one site to another?

What she saw was that the shopping cart should be the customer’s, and not each retailer’s alone.

She wanted to drive, and not just be driven. That was what the first browsers were promised. That’s why the steering wheel was the emblem for Netscape’s browser.

Nineteen years later, that hasn’t happened. You no longer drive your browser. On the commercial side of the Web, you are driven as a passenger, with little control over what the driver does, even if the driver (actually a robotic process — an algorithm) wants to help.

Take, for example, the matter of location. The Internet has no location, and that’s one of its graces. But many sites and services go out of their way to notice what IP address you appear to be arriving from, and customize their pages for you, based on that location. While that might sound innocent enough, and well-intended, it also fails to know the driver’s true intentions — or even that they wish to drive.

Our intentions — our will to drive — matters far more to each of us than whatever a website guesses about us, especially if the guessing is wrong, which it is, a great deal of the time.

Last week I happened to be in New York when a friend in Toronto and I were both looking up the same thing on Google while we talked over Skype. We were unable to see the same thing, or anything close, on Google, because the engine insisted on giving us both localized results, which neither of us wanted. We could change our locations, but not to no location at all. In other words, it wouldn’t let us drive. We could only be driven.

Right now I’m in Paris, and cannot get Google to let me look at Google.com (presumably google.us), Google.uk or Google.anywhere other than France. At least not on its Web page. (If I use the location bar as a place to search, it gives me google.com results, for some non-obvious reason.)

After reading Larry Magid’s latest in Huffpo, about the iPhone 5, which says this…

Gazelle.com is paying $240 for an iPhone 4s in good condition, which is $41 more than the cost of a subsidized iPhone 5. If you buy a new iPhone from Sprint they’ll buy back your iPhone 4s for $235. Trouble is, if you bought a 4s it’s probably still under contract. Sprint is paying $125 for an 8 GB iPhone 4 and Gazelle is paying $145 for a 16 GB iPhone 4 which means that it you can get the $199 upgrade price, your out of pocket could be as little as $54.

… I wondered what BestBuy might give me for my 16GB iPhone 4. But when I go to http://bestbuy.com, the company gives me a page in French. I guess that’s okay, but it’s still annoying. (So is seeing that I can’t get a trade-in price without visiting a store.)

Back in the search world, I’ve been looking for a prepaid wireless internet access strategy to get data at sane prices in the next few countries I visit. A search for “prepaid wireless internet access” on google.fr gets me lots of ads in French, some of which might be more interesting if I knew French as well as I know English, but I doubt it. The “I’m feeling lucky” result is a faux-useful SEO-elevated page with the same title as the search query. The rest of the first page results are useless as well. (I did eventually find a useful site for my UK visit the week after next, but I’ll save that for another post.)

To describe what the Web has become, two metaphors come to mind.

The first is a train system that mostly runs between commercial destinations. In a surreal way, you are transported from one destination to another near-instantly (or at the speed of a page loading ads and cookies along with whatever it was you went there for), and are trapped at every destination in a cabin with a view only of what the destination wants you to see. The cabin is co-occupied by dozens or hundreds of conductors at any given time, all competing for your attention and telling you something they hope will make you buy something or visit other sites. In the parlance of professionals on the supply side of this system, what you get here is an “experience” that they “deliver.” To an increasing degree this experience is personalized, and for every person it’s different. If you looked at pants a few sites back, you might see ads for pants, or whatever it is that the system thinks you might want to buy, whether you’re in a buying mood or not at the time. (And most of the time you’re not, but they don’t care about that.)

Google once aspired to give us access to “all the world’s information”, which suggests a library. But the library-building job is now up to Archive.org. Instead, Google now personalizes the living shit out of its search results. One reason, of course, is to give us better search results. But the other is to maximize the likelihood that we’ll click on an ad. But neither is served well by whatever it is that Google thinks it knows about us. Nor will it ever be, so long as we are driven, rather than driving.

I think what’s happened in recent years is that users searching for stuff have been stampeded by sellers searching for users. I know Googlers will bristle at that characterization, but that’s what it appears to have become, way too much of the time.

But that’s not the main problem. The main problem is that browsers are antique vehicles.

See, we need to drive, and browsers aren’t cars. They’re shopping carts that shape-shift with every site we visit. They are optimized for being inside websites, not for driving outside them, or between them. In fact, we can hardly imagine the Net or the Web as a space that’s larger than the sites in it. But we need to do that if we’re going to start designing means of self-transport that transcend the limitations of browsing and browsers.

Think about what it means to drive.  The cabin, steering wheel, pedals, controls, engine, tires and chassis of a car are all controlled by you. The world through which you move is outside, not inside. Even in malls, you park outside the stores. The stores do not intrude inside your personal space. Driving is no less personal and no less masterfully yours when you ride a bike or a motorcycle, or pilot a plane. Those are all personal vehicles too. A browser should have been like one of those, and that was kind of the idea back in the early days when we talked about “surfing” and the “information highway.” But it didn’t turn out that way. Instead browsers became shopping carts that get fresh skins at every website.

We need a new vehicle. One that’s ours.

The smartphone would be ideal if it wasn’t also a phone. But that’s what it is. With few exceptions, we rent smartphones from phone companies and equipment makers, which collude to sentence us to “plans” that last for two years at a run.

I had some hope for Android., but that hope is fading now. Although supporting general purpose hardware and software was one of Google’s basic ideas behind Android, that’s not how it’s turning out. Android in most cases is an embedded operating system on a special purpose device. In the most familiar U.S. cases (AT&T’s, Sprint’s, T-Mobile’s and Verizon’s) the most special purpose of that device is locking you to a plan and soaking you for some quantity of minutes, texts and GB of data, whether you use the full amounts or not, and then punishing you for going over. They play an asymmetrical knowledge game with you, where they can monitor your every move, and all your usage, while you can barely do the same, if at all.

So we have a long way to go before mobile phones become the equivalent of a car, a bicycle, a motorcycle or a small plane. I don’t think there is an evolutionary path to the Net’s equivalent of a car that starts with a smartphone. Unless it’s not a phone first and a computing/communication device second.

The personal computing and communications revolution is thirty years old now, if we date it from the first IBM PC.  And right now we’re stuck, mostly because we think having the Web “personalized” is the same thing as having a personal vehicle. And because we think having a smartphone makes us independent. Neither is true. That’s why we won’t make progress past those problems until we start thinking and inventing outside their old boxes.



10 responses to “Browsers should have been cars. Instead they’re shopping carts.”

  1. I’ve been looking at Ting, a Sprint reseller brought to us by the folks at Tucows. Pay for what you use, no contract. I’m not sure it’s what I need, at least for now, because I do OK with a Cricket non-smart phone. Tucows also fathered Hover, a domain registrar dedicated to “no bullshit”, a refreshing change from the usual upselling frenzy.

  2. I’ve always thought that Microsoft must be over charging their banner clients if I”m getting UK Fish and Chip dining offers in Bangkok. I even know a Kahuna in MS ad sales regionally and when I informed him on Facebook the silence that ensued had that feel of unwelcome honesty when everybody gets along just fine by lying through their teeth.

    I might be wrong on that last point but the imagination runs wild if no reply is given.

  3. Another side to Google’s cultural imperialism that we feel especially in the UK. Google assumes we want to see English language results and that basically means 75% US, 25% UK. Some times it feels like the non-English language web simply doesn’t exist and that the US viewpoint is all there is.

  4. Kevin,

    I’m a very satisfied Ting and Hover customer, and am close personally with a number of people who work there. They love customers and have the lowest BS level of any company I know. They also learn well, which also matters, a lot.

  5. If you ask me, tackling GEO-IP issues are the next frontier for search engines and browser, without a doubt. They don’t resolve for “people”, they resolve for machines logging in from specific locations. I manage SEO at a large tech company and managing country and language rendering is a very, very difficult task when you’re doing it globally. Google will get it, I’m confident. They’re just not there yet.

  6. Thanks, Tommy.

    As a browser and search engine user, I’m glad you’re optimistic about the future for GEO-IP. As an independent human being looking to operate independently on the Net and the Web, I am also looking for a vehicle that gives me a new and better way to drive. As I say above, something that is like a car rather than a shopping cart.

    I am also looking for systems that give entities of all kinds, including companies, better ways than websites to present their goods and services, and to engage others. APIs are good step in that direction. Through APIs companies can expose their competencies in ways that can be engaged. So, for that matter, can you and me. Driving and engaging can be done through intentional logic by both parties that does not require either a search engine or a website in the middle.

    But the scenarios of possibility can’t be seen at the level of browsers, search engines and websites, where we’re been living on the Web since 1995. We need to think and build above that level, where the whole environment is a Cambrian Explosion of Everything, and everybody and everything can have a cloud and an operating system of their own.

    And, since I’ve been thinking (and, as much as possible working) with outliners since the 1980s, I also believe that Dave Winer is onto something with his ground-making work with OPML, comments (which can be comments and blog posts at once, among other things), and much more.

    All the developments I’ve linked in these last two paragraphs exceed the dimensions of the old browser-website-search engine conceptual box. They also give me faith that we won’t stay stuck where we’ve been for the last 17 years.

  7. Doc, I have experienced the very low BS level at Hover. Perhaps there should be a metric . . . the only thing approaching BS is “share this with your friends”,which is really only natural.

    Are you using Dave Winer’s comments/posts/comments? As usual for him, a very cool idea that gives power back to us. I wish I had a bit more technical chops so I could try it out myself.

  8. Mozilla makes happy noises about privacy, but EFF has been pointing out the problem of browser fingerprinting for quite a while, and we haven’t seen much interest from Mozilla in addressing the problem. http://zgp.org/~dmarti/www/firefox-secret-shame/

    This is a really good idea there, but not much action in a while: https://bugzilla.mozilla.org/show_bug.cgi?id=565965

  9. […] Browsers should have been cars. Instead they’re shopping carts.: Google once aspired to give us access to “all the world’s information”, which suggests a library. But the library-building job is now up to Archive.org. Instead, Google now personalizes the living shit out of its search results. One reason, of course, is to give us better search results. But the other is to maximize the likelihood that we’ll click on an ad. But neither is served well by whatever it is that Google thinks it knows about us. Nor will it ever be, so long as we are driven, rather than driving. […]

Leave a Reply

Your email address will not be published. Required fields are marked *