Did Surveillance Capitalism Predate the World Wide Web?
Analysing the very early Web to determine whether the purpose was a product of the mechanism, or the mechanism was a product of the purpose.
Mouse click position-tracking was conceived in the early 1990s, even before a Web browser was available for any mainstream operating system. You'd think a browser provider might finalise compatibility with Windows, Mac or DOS as a first step. But apparently, devising a pixel-accurate mouse point logger was more urgent.
In her chillingly observant book The Age of Surveillance Capitalism, Shoshana Zuboff cited Google as the original surveillance capitalist. And that certainly looks at first glance like just another piece of on-point observation.
But if you trace back the roots of the things for which Google ultimately became infamous, you'll notice a glaring conundrum. To coin a phrase, the cart came before the horse. There's a period in the early to mid 1990s, during which the fledgling tools of surveillance capitalism are being developed, but there is no visible use for them. Historical parallels to this are exceedingly rare if existent at all. Logically, the need comes before the invention. So why does the early Web repeatedly produce the invention before the need arises?
It's assumed that capitalism is the end and surveillance is the means. But is it, and was it always, the other way around?
Quote
Surveillance-based advertising is more effective than contextual advertising, but not so much more effective that it's worth the insane level of hassle and the hundreds of $billions it's cost to implement. At some point we have to notice that wall-to-wall surveillance, considered solely as a consumer marketing device, runs at a wildly unacceptable loss versus simple, cheap and effective contextual marketing, and ask ourselves what the surveillance component is really there for.
- Backlit
SETTING THE SCENE
Much like the never-ending line of Brits who at one point or another claimed to have seen the Beatles at The Cavern in Liverpool, most of the people who tell you they were "browsing the World Wide Web in 1992" can safely be categorised as fantasists.
For a start, there were almost no websites beyond the academic havens implementing and supporting the launch. Even by mid spring 1993, worldwide, the total of individual WWW properties had only just crept into triple digits. Imagine that. Just over a hundred websites in the entire world. The first graphically-capable Web browser - Mosaic - was barely post-launch. And no operating system of the period came with any kind of Web connectivity software. You had to install the necessaries yourself - which could be far from simple in Windows 3.1 or DOS. Not that Mosaic could in any case be used within a Microsoft OS - or indeed on a Mac - at that time. It was strictly Unix-only until latter '93.
A year or so later in summer 1994, the commercial Web was just undergoing birth. Here in the UK, there were only 18 .co.uk domains in use, including the ISPs. The volume of UK Internet users had risen exponentially from the meagre total of around 2,000 at the start of the year. But the figure still stood short of five digits. And the lists people used to surf the Web [circa July 1994] in the absence of a capable search engine were still heavily weighted towards the predominant educational networks, "Webfathers" and digital evangelists. In short, if you were a serious tech bod, you pored over a wealth of formative and evolving HTML guides; if you wanted to look at cat pictures you went to a shop and bought a book.
"Despite local protests from disgruntled land owners, the latest extension of the Information Super Highway has successfully reached the city gates of Ely." - Guitairst Magazine introduces its email address (but no website), East England, March 1995.
Before summer '94, the commercial Web did not, by any credible definition, exist.
ISMAP - TRACKING 1993 STYLE
How strange, then, that more than a year prior, in mid spring 1993, Mosaic browser co-creator Marc Andreesson should announce a remarkably Google-esque piece of surveillanceware, which would subsequently become a staple of late 1990s ad banner tracking.
It was an HTML attribute called ISMAP. And despite its deeply dodgy reputation in current tech circles, it's still recognised by today's mainstream browsers. The attribute would be inserted into the HTML code for an image on a Web page. And then, when the visitor clicked on the image, ISMAP would tell the server exactly where, in the image, the user had clicked. Did they click the dismiss cross? Click on the female face? Click on the male face? Did they click to try to enlarge the integrated caption? ISMAP would find out and relay the info to the site administration.
<table border="0" cellpadding="0" cellspacing="0" width="100%"> <tr> <td width="100%" align="center"> <a href="http://ad.linkexchange.com/09/X329001/gotoad.map" target="_top"> <img width="440" height="40" border="1" ismap alt="LinkExchange" src="http://ad.linkexchange.com/09/X329001/logoshowad?free"></a><br> <font size="1"> <a href="http://www.linkexchange.com/" target="_top">LinkExchange Member</a> </font> <br> </td> </tr> </table>
Actual display ad code from the late 1990s, with the ISMAP attribute highlighted.
This Mosaic-authored technology could admittedly be used outside of ad-click tracking. Well, it could if anyone was insane enough to put images into a Web page in the days when a 14.4K dial-up modem was considered fast. But every surveillance tool Silicon Valley has ever introduced has come in with a palatable pretext. And even assuming there was some lunatic fringe of bandwidth-torture enthusiasts who have since been mysteriously airbrushed from Web history, the idea that a browser provider would focus on mouse-position click-tracking before implementing basic page layout and formatting staples is truly bizarre.
In spring '93, you couldn't put your image into an elementary grid or give it a simple border - if you could get the bloody thing to load at all. In fact, the actual HTML tag required to embed an image into a page had only been proposed mere weeks earlier. Yet here was Mosaic announcing a means to set a sophisticated tracking property on the rich media that no one was yet bothering to incorporate. At the time, Mosaic wouldn't even apply basic bold formatting to text headings. This was, and still is, a mind-blowing example of skewed priority, which perfectly aligns with the behaviour of today's surveillance giants.
NETSCAPE: 21st CENTURY TRACKING IN UMM... 1994
The Netscape team were asked to develop tracking cookies not by a consortium of online shops, but by Vinton Cerf - a man with direct connections to military and government agencies and a long background in intelligence computing. Indeed, Cerf's name runs across the whole story of the Web's beginnings like Brighton through a stick of rock. Whilst cookies were billed as an online shopping aid, they were requested by Cerf on behalf of the telecomms company MCI, and their initial use was not to persist a shopping cart, but to track return visits to websites.
Netscape is a central focus in the timeline of online surveillance. Based in Mountain View, precisely where Google would subsequently grow its stalking empire, Netscape was amazingly synergistic with Google's path. So much so, that the historical progression from the pre-Netscape Mosaic, through Netscape, Mozilla and Firefox to Chrome, looks like a single company's journey. Same ideology. Same dark patterns. Same agenda. Same pathological obsession with amassing human behavioural data, regardless of whether any remote facet of commerce had ever expressed an interest in it. Same penchant for unilaterally dictating technological standards... I genuinely could go on.
Netscape - with Mosaic's Andreesson holding the helm - was much more than just proto-Google. It was as if both operations (as well as Mosaic before them) were working to a master template created by a higher power.
HALF PLAN, HALF BISCUIT
Even before Version 1, released in 1994, Netscape had begun a quest to incorporate a now globally infamous, but at the time calculatedly hushed technology called... Ready?... Persistent Clientstate HyperText Transfer Protocol Cookies. AKA HTTP Cookies. AKA cookies. From the very start, and before any other purpose, cookies were used to recognise website visitors upon their return. And the browser's interface provided no delete option.
The Netscape browser also necessarily stored the user's browsing history, and to the obvious frustration of users, offered no UI-resident means to delete that either. The History dialogue featured three buttons: Go to [the selected Web page], Create bookmark, and Close [the dialogue]. There was no option to clear. If you wanted to do that you'd have to pitch into the application's program folder and zap a file called Netscape.hst. Not a strenuous task once you knew how. But the average user didn't know how. That was the problem.
Both of these core tracking tenets were designed to preserve usage data without consent - or even knowledge in the case of cookies. And there's a stark sense that plans to surveil the public were firmly in place before online retailers had even set up shop - let alone asked for analytical resources.
That the request for cookie-type technology should have come from a government-associated network communications specialist, working for a major comms company, makes the whole picture mightily suspicious. Even if Vinton Cerf really did care so thoughtfully about a bunch of online retailers who didn't yet exist, one has to ask the question: why?
And when you add in the lack of public transparency, it becomes more suspicious still. Netscape knew the public would be outraged at being spied on. But if cookies were not there to please users, and there had initially been no request for them from retailers (most of whom didn't know what the WWW was at that point), then for whose benefit, exactly, were cookies created?
In February 1996, Tim Jackson dramatically exposed the cookie to the general public for a Financial Times feature. The article was captured by Wayback but has since been rendered inaccessible; search results are overwhelmed with propaganda pumps mentioning it but not linking to it... Looks like a typical organised burial. But I dug deep and found it here - scroll down to "This Bug In Your PC Is A Smart Cookie by Tim Jackson" .
It's still a good read today, and it's noticeably not a tirade of hyperventilation. It's presented with calm balance, and yet it whipped up a public furore, instantly prompting the Web powers of the day to pretend they hadn't realised what was going on and draft a PR-aware framework for the use of cookies. A sham, as indeed all PR exercises are. But confirmation that the WWW's culture of shrouded surveillance would not be shrugged off by the public.
JAVASCRIPT - THE RESOURCE THAT NO ONE BUT THE TECH INDUSTRY ASKED FOR
But by the time of the Jackson article, Netscape had already introduced an immeasurably bigger threat to Web-users' privacy. Mapped out in May '95 under the working title Mocha, and originally released as LiveScript, Netscape's new, active technology promised to take mass surveillance to warp-level. Subsequently launched with much more fanfare as JavaScript, in a publicity pact with Sun Microsystems and their Java product in December '95, Netscape's browser-resident scripting resource was a gift-in-waiting for data-gathering surveillance lords.
JavaScript could give website admins control of not just the server that processed requests from the visitor's browser, but of the visitor's browser window itself. And once again, among the general public and early site authors, the idea didn't appear to have any pre-existing demand.
Most websites at the time were simply collections of static files, and this remained the case well into the latter part of the decade. Analysing 40 fully preserved websites from an unmanipulated 1998 snapshot, I found only marginal use of JavaScript. Just 9 of the sites used any JavaScript at all, and in those cases the language routinely appeared in very trivial measure. Rarely more than a few lines in a page. It was mainly people adding live dates to their headers or footers, adding a defaultStatus
declaration to put a slogan into the browser's status bar, or using document.write()
insertions inline instead of simply entering HTML. For example:
<b><script language="JavaScript"> <!-- document.write(conversion(53900)) // --> </script> Euro</b>
Notably, the script in the rather typical example above is entirely contained within a single HTML bold tag! And throughout the analysis there was a definite sense that, even a good couple of years after JavaScript's widespread intro, Web admins working outside of the tech industry's nascent, part-formed surveillance effort didn't really know what to do with it. Hardly what you'd describe as "public demand".
So why bother to wedge browser-resident scripting capability into Netscape at all, when most sites are intentionally static, and those that aren't have capable and far more flexible scripting routines on the server? No one's ever given a compelling explanation. But Netscape's launch spiel gave away the true answer to anyone who could read between the lines. JavaScript, said its inaugurating entity:
"allows the page designer to access events such as startups, exits, and user mouse clicks."
But that's not designer talk. It's datamonger talk. At the beginning of 1995 there was no way for a would-be intelligence gatherer to take control of a site visitor's page window and monitor their activity in depth. For that, there had to be a scripting handler in the browser itself. That's what JavaScript brought to the table. And since, at inception, there was clearly close to zero demand for browser-resident scripting among ordinary website admins, at face value the investment in JavaScript makes little sense.
Until you discover that even before inception, there was a comprehensive list of 28 parties who were interested in the concept to the point of pledging to formally endorse it. All of them were tech corporations, doubtless salivating at the thought of the wealth of user data that could be extracted through it. The dragon's den of endorsees - the Big Tech of the time, if you like, included:
- America Online
- Apple
- Architext Software
- Attachmate Corporation
- AT&T
- Borland International
- Brio Technology
- Computer Associates
- Digital Equipment Corporation
- Hewlett Packard
- Iconovex Corporation
- Illustra Information Technologies
- Informix Software
- Intuit
- Macromedia
- Metrowerks
- Novell
- Oracle
- Paper Software
- Precept Software
- RAD Technologies
- The Santa Cruz Operation
- Silicon Graphics
- Spider Technologies
- Sybase
- Toshiba
- Verity
- Vermeer Technologies
Have you ever seen such a long queue of profiteering money-grabbers jostling for the chance to give something away to the public for free?
And that question touches on one of the key characteristics of Web pioneers like Netscape. They were, as far as most users were concerned, working for nothing. And to coin a much later adage, if it's free, you're the product. The bit that's difficult to reconcile is: if surveillance capitalism did not yet have any chartable sources of revenue, then who, exactly, was the customer?
"WEB STANDARDS UNILATERALISM" - THE LINK BETWEEN NETSCAPE AND GOOGLE
In late May 1994, Tim Berners-Lee and CERN - the founding architects of the World Wide Web - held a conference near Geneva. Known as WWW94, the conference is now remembered as a historical landmark during which Berners-Lee proposed the formation of a consortium which would set and arbitrate on standards for the Web. The consortium we now know as W3C, and which - whether or not Berners-Lee's original intention was to steep the body in altruism - eventually transmogrified into nil but a collection of the world's most evil corporations imposing non-guidelines on themselves.
One of the conference's most notable causes of brow-furrowing, was that the team preparing to launch the decade's most important browser - Netscape - did not attend. Already Internet-famous for their raging success with Mosaic, the team were among the most obvious candidates to desire influence on the course of the World Wide Web.
But they had a bigger plan. As far as they were concerned, the Web was not going to be governed by Berners-Lee's consortium. It was going to be governed by them. And they were not going to argue with anybody. They were going to take an irresistible package to the public and let the public do the arguing for them.
Netscape relentlessly set their own Web standards, using the public vote to force everyone else (including the ever-begruding Microsoft) to adopt them, and leaving the W3C no option but to ratify the conventions. Most of the HTML tags and attributes that people today credit to Berners-Lee were actually introduced by Netscape as unilateral additions to the Navigator browser.
Table structures. Page backgrounds - yep, those most-often cringeworthy tiled-image backdrops on 1990s websites were Netscape's doing. Text and block alignment attributes. Alignment, width, height and border attributes for images. The <CENTER>
tag. The <FONT>
tag with text size adjustment. And as already discussed in the JavaScript coverage, the <SCRIPT>
tag. This is only a tiny subset of what Netscape and to a lesser extent Mosaic added to Berners-Lee's very basic complement of original HTML designations.
It was a devastatingly powerful strategy. Give the public something they like, infused with something that serves your own agenda, and you have control of the landscape. Not only that, but if competitors don't follow your lead, they die, and you effectively become a monopoly. Ask the average person today to cite the epitome of that, and they'll probably name Google. But Netscape did exactly the same thing, before Google came into being.
Only by using its now legendary strategy of embrace, extend and extinguish (fused, or course, with massive market advantages), could Microsoft ultimately beat Netscape in the first browser war. And in spirit or kin if not in person, Netscape got its own back. First through the rise of Firefox - in a campaign masterminded and heavily bankrolled by Google. And then, after Mozilla handed on the baton to Google itself, through the battering that Google Chrome finally gave Internet Explorer.
Surprisingly, for me at least, one thing stands largely unspoken in the journey from Mosaic, through Netscape and Firefox to Chrome. We were never asked to pay for any of them. And it seemed that however rocky the road got for the vendors, that never changed. Never looked like changing. The survival mattered, but not, apparently, the money.
And the highly uncorporate disregard for money appeared to be catching. Imagine you're AOL, who bought out Netscape Navigator in 1999. You've basically chucked $10 billion at Netscape, for a browser whose fortunes are in dramatic decline, and which has just opened its source code to the world so you could in any case have it for free. And then, a few years later when you're left with a completely dead duck, you courteously pay $2 million to set up a foundation for the product that's lost you enough money to feed 5 million people for a full year. That's not corporate business. It's a Mickey Pearce sketch.
Then as soon as AOL stops doling out free money to Mozza, Google starts. There was clearly someone, somewhere, for whom the survival of this endlessly loss-making, but surveillance-tactic-defining browser was an imperative at any price.
THE REAL ONSET OF SURVEILLANCE CAPITALISM
Zuboff stressed in The Age of Surveillance Capitalism that the associated concentrations of wealth, knowledge and power have been unprecented in human history.
But those unprecedented traits were heralded by an equally unprecedented anomaly that came before the online advertising industry established itself. Namely, the tide of companies investing a fortune in time, effort and labour, then giving away the fruits for free. To those of us who recall a time before the WWW existed, this didn't conform to basic logic. Later, the practice was explained by advertising. Like:
"Yeh, it's okay, 'cos what happens is everything gets given away for nothing and the providers put ads on their sites and that's how they get paid. And everyone lives happily ever after.
Fair enough... -ish. But what happens before there are online ads? In the story of the early Web, you see commercial enterprise giving away phenomenally expensive products for free, with no apparent means to even recover the outlay, let alone make a profit. And Google's arrival encapsulated that puzzling phenomenon. The company took so long to consider and unveil a monetisation plan that commentators firmly believed its disappearance was imminent.
Today, we expect cybertech businesses to run for years with no business model. It's been shown to work in the realm of cybertech, and it's become the done thing. But where did that idea come from? It was totally alien in the 1980s. Companies would incur costs setting up for production or funding publicity, but they would not complete the product and then distribute it to basically anyone who wanted it free of charge.
And you can say the Web slashed distribution costs, or that digital, non-physical products had no material overhead to fund. That, indeed, is the narrative we've heard from the mouthpieces of the Web's "free culture".
But what would you get for your £750 when you ordered Sage Financial Controller in 1986? A packet of floppy disks and a book. That is not seven hundred and fifty quid's worth of material value. Before the WWW, software cost money. A lot of money. And the material value plus the distribution cost did not account for more than a micro-fraction of what was being spent. Suddenly, the Web arrives, and out of the blue there's this new breed of software manufacturer that doesn't want the contents of your wallet. And "Oh okay then, if you're an enterprise customer you can pay us if you like", but the overwhelming majority of users are not shelling out a sausage.
I believe this new trend was a key marque of the pre-capitalist phase of the surveillance machine. Surveillance capitalism, I would suggest, does not begin when Google finally decides it's going to sell ad space. It begins when businesses begin investing months, years of their time and money in things they know they're never going to charge for. And that comes well before Google in the grand thread of digital history.
As we know, the roots of the Internet and subsequently the World Wide Web lay in state-funded academia. And that makes it extremely difficult to, if you like, "detect the join". The point at which "free culture" progressed from places where you did expect it - like educational environments, to places where you didn't expect it - like major IT companies. Indeed, there "is" no perceptible join between the free browser Mosaic - technically an academic product, and Netscape - technically a commercial product, developed by the same people.
Without the blur of those long-running academic roots sitting parallel to industry for years on end, we would see a defineable point at which corporations very suddenly came up with the idea of free stuff. Which necessarily turns the consumer into a product, subjugates them to corporate agenda, and monetises them, as opposed to the software they're using.
I have a million more things to say about this, but unfortunately no more time to say them today. And there are so many more dots to join. Please feel free to join some of those dots, along with some I'm sure I haven't even noticed.
Netscape personified the smiley-faced evil that came to characterise Big Tech. It built both the foundations and the scaffold for surveillance capitalism. The Navigator browser brought in cookies, front-end scripting, indelible browsing histories, and unilateral "Web Standards" definitions that overpowered academic democracy. Its legacy, Firefox, insidiously advanced online stalking far beyond any known boundary, all while fabricating a completely false and ficticious commitment to privacy. Netscape's cocktail of shark-like privacy violation beneath a baitbox of prettification is still the backbone of corporate surveillance today.
IN CONCLUSION
In the end, maybe the excesses and gluttony of the 1980s taught a power-hungry clique that the route to controlling public behaviour would not be through the gun, the police baton or the water-cannon, but through consumerism. And maybe the World Wide Web was not the gift of academic altruism that got corrupted by a bunch of rogue hypercapitalists; but a calculated, theatrical and above all successful bid to Big Brother the entire globe into resigned, unquestioning, zombie-like compliance with the master's plan.