Web and flow
If we consider the way most transformative technologies develop, their evolution follows a pattern that has been characterised among other ways by the Gartner hype cycle which “claims to provide a graphical and conceptual presentation of the maturity of emerging technologies through five phases”. In short, technologies go through the phases of hype, disillusionment, enlightenment then productivity.
Successful technologies emerge, gain traction, and with some readjustment of expectations after that initial hype, find their place in a wider world. And then they plateau, becoming an increasingly comfortable, rather boring part of the landscape.
Mainframe computing is perhaps the archetype here. With its origins 70 years ago, most of us might think mainframes have long passed into history, but there are 10,000 mainframes still in use around the world, at just about every major financial institution, and it’s a $3Bn market still growing around 5% per annum.
Ebb and flow
The Web has, so far, followed a rather different path. First, its longevity is astonishing. It has been more or less 30 years since it started to gain some real traction if, not among broad consumer markets, at least among people who used computers a lot in the early 1990s.
The Web has also adapted to changes in computing technologies and whole paradigms
- to the transition from incredibly slow dialup networks to today’s high speed networks–a common 14Kbps modem connection of the early to mid 1990s is .0001% of the speed of a not uncommon 1Gbps connection today)
- to multiple changes in interaction paradigms–from largely monochromatic, low resolution desktop computers, through to super high resolution large screen laptops, smart phones, tablets. There’s a good chance your Smart TV’s UI is web technology based (as are the digital menu boards in many fast food restaurants, the touchscreen interfaces used by astronauts in SpaceX’s Crew Dragon Capsule)–ok you get the picture
- It’s not only adapted to, it’s brought about the transformation of just about every industry and created whole new ones.
Today though I want to focus on how the technologies of the Web, specifically the front end technologies have evolved and continued to evolve across the last 3 decades. And how perhaps this decades-long run may be threatened.
I feel there’s a sort of pattern in the evolution of the Web that has repeated several times. The Web ebbs and flows, if you will, something like this:
After a surge of innovation, while competing browsers look to innovate the Web platform and gain traction and share (early to late 1990s Netscape versus IE, 2005-2010 Safari/Firefox/Chrome, 2016-?) we see the emergence of a browser (Internet Explorer, Chrome–and Safari on iOS, an important point we’ll come back to), which having gained dominant market share, reduces their investment in innovation, leading to the relative stagnation in the development of the underlying technologies of the web.
The most recent cycle is somewhat different in some ways, but follows the pattern in others.
- It’s different because despite the increasing dominance, not only as a browser (Chrome) on desktop and Android, but as an engine (Blink) that saw Microsoft, and Opera, which had both continued to develop their own browser engines (Opera until 2013, Microsoft until 2018), Google (the main but by no means exclusive driver of the Blink engine) has by no means coasted on the continued innovation of the web platform–if anything that innovation has accelerated.
- But it’s similar since on one of the most significant computing platforms, iOS, innovation of the Web platform has been far less dynamic, due to Webkit’s exclusivity (more on this in a moment) on the iOS platform.
That’s not to say the Web on iOS has stagnated in the way we saw in earlier cycles–only this week Apple announced a huge range of new technologies in Safari. But while by some measures iOS accounts for less than a third of all global smartphone marketshare (and Android over two thirds), in the most wealthy nations and regions of the world, iOS market share is far greater (in Japan it’s 66%, in the US it’s nearly 60%).
But what does the relevance operating system market share have to browsers and Web technology innovation?
lowest common denominator
It’s a surprisingly little known fact, even among developers, that on iOS, no matter what browser you choose to use, you are using the same browser engine–Apple’s Webkit. Apple simply does not allow alternative browser engines. And so Apple exerts an outsized impact on the overall development of the Web platform, because their platform represents not so much the lowest common denominator for web developers to target, but the most sophisticated possible version of the Web platform for many users. If Safari does not support a Web platform feature, then its value to developers is significantly curtailed.
Features like Web BlueTooth, and WebNFC, which would allow developers to create whole new classes of application using web technologies, (Web Bluetooth has, according to Can I Use, 70% global support, that is 70% of all web users in the world have access to Web Bluetooth in their browser). But the absence of support in any browser on iOS, due to Apple’s requirement that WebKit be the only browser engine on the iOS platform, coupled with their decision–they argue for security and privacy reasons–to not support Web Bluetooth, means the use of this API is far less attractive for companies and developers. Have you ever used it?
The Web’s darkest days
This brings to mind perhaps the darkest days in my experience of the Web, the early 2000s, where Internet Explorer had nearly 100% browser share on Windows, and Windows accounted for over 90% of all operating systems used in the World. And after the release of Internet Explorer 6, in 2001, it was not significantly updated until late 2006–5 years of very little innovation, which meant, too, 5 years of very little innovation in the mainstream web platform. Developers at the time genuinely feared for the viability of the platform.
In a way it is something of a miracle the Web survived–but a couple of very important things happened in those darkest days–in particular, two browsers happened.
Safari 1.0
In January 2003, Apple announced their own browser, Safari. Mac OS, while an increasingly niche operating system, had since late 1999 featured the best browser at the time–Microsoft’s Internet Explorer 5 for the Mac, based on a completely different browser engine from the Windows version, and driven by a strong commitment to supporting Web standards. Without it and innovations like DOCTYPE switching that it first implemented the chances of the Web’s emergence from the late 1990s as a viable platform are, I’d argue, slim.
But Safari, based on Apple’s ‘Webkit’ fork of the long running KHTML browser engine, brought at least a small ray of hope at a time when the evolution of the Web felt moribund. At the time, the iPhone was little more than a glimmer in Steve Job’s eye, and little did we know how impactful Safari was going to be for the Web’s rejuvenation. Safari had a focus on excellent support for CSS and other Web standards, and over the next half decade or so introduced many of the features of CSS, and the HTML5 canvas that rejuvenated the Web.
Firefox
For much of the 1990s, the Web meant Netscape Navigator, the first successful commercial web browser. Netscape innovated on the web’s core technologies introducing ground breaking features like images and styled text–(no seriously, these weren’t part of the Web for its first few years). But Microsoft responded to the sudden threat to their foundation, Windows, by investing enormous energy and resources into Internet Explorer, and by the late 1990s, had won the ‘browser wars’.
Netscape in 1998 released the source code for their browser under an open source license, but for years subsequently, the browser remained essentially dormant and became increasingly irrelevant (market share falling to 5% by 2001).
Dave Hyatt (subsequently heavily involved with Safari), Joe Hewitt (the developer of Firebug, the origin of today’s browser developer tools), and Blake Ross initiated the project that became Firefox in 2002. While it may seem in hindsight its arrival was swift and its impact significant from the start, its market share didn’t hit 10% until late 2006, nor did it even leave beta until right at the end of 2004. It, like Safari, focused on supporting a standards based Web, and, being available on Windows, unlike Safari (which actually did for a period from 2007 ’til 2012 have a Windows version) represented a genuine challenge to Internet Explorer.
And the emergence of these challengers to the dominant, moribund Internet Explorer sparked a second wave of innovation of the Web platform the one that created a foundation for modern web applications–HTML5, CSS3, and the modern DOM.
Today’s Web, and tomorrow’s?
But seriously now. How can the Web today be similar to those dark days of the early 2000s? After all, while Chrome has a significant majority of the bowser market share, it’s less than Microsoft’s utterly dominant share of those years.
And the Chrome engine has continued to innovate in CSS, the DOM, JavaScript support, the networking stack, and elsewhere in the Web platform.
The challenge is, as we saw a little earlier, that if you wish to address the market of the most affluent consumers in the most affluent economies in the World (which is what the majority of companies will want to do), you need to reach them on iOS. If you are a government or not for profit with the obligation to reach everyone you need to reach many of them on iOS.
Which means either native iOS applications, or Webkit. And because of Apple’s decisions to–
- disallow other browser engines on their platform
- not support many standardised features of the Web platform well supported almost everywhere else
–in a great many cases, native iOS applications are the only option.
But even when your application’s functionality can be implemented in Webkit, making those applications first class citizens on iOS, deeply integrated into the user’s experience of the platform, using widely supported, standardised web technologies that we loosely call “Progressive Web Apps” isn’t possible.
On the Desktop, computing is the Web
On the desktop, computing is the Web now. More than 60% of the time users spend on their desktop and laptop computers is spent in a browser. The entire SAAS category of enterprise software exists because of the Web. Whole categories of computing no one would have imagined being feasible in the browser a decade ago are routine now.
And this brings us back to where we started–the trajectory of the Web, its evolution and ongoing development is unlike that of almost any other technology. It hasn’t followed the trajectory of becoming a boring part of the landscape (and ultimately fading into a potentially still important obscurity, like mainframe computers).
Rather it has, in part, driven and adapted to changes in computing across those decades–from desktop to laptop to touch based smart phones and tablets.
But…
But the Web does now face a challenge unlike any it’s faced since the late 1990s–on perhaps the most significant computing platform, iOS, its evolution is now dependent on a single party, Apple.
This is something that competition and consumer regulators around the world are recognising, with the UK’s Competition and Markets Authority releasing the Mobile ecosystems market study interim report. This found that
On iOS devices, Apple bans the use of alternative browser engines – this means that Apple has a monopoly over the supply of browser engines on iOS. It also chooses not to implement – or substantially delays – a wide range of features in its browser engine. This restriction has 2 main effects:
limiting rival browsers’ ability to differentiate themselves from Safari on factors such as speed and functionality, meaning that Safari faces less competition from other browsers than it otherwise could do; and
limiting the functionality of web apps – which could be an alternative to native apps as a means for mobile device users to access online content – and thereby limits the constraint from web apps on native apps. We have not seen compelling evidence that suggests Apple’s ban on alternative browser engines is justified on security grounds.
Competition and Markets Authority: Mobile ecosystems market study interim report
Regulators in the US, EU and elsewhere are also showing an interest in what the CMA described as a “ban” on “the use of alternative browser engines”.
My instinct is that if on iOS Webkit continues to be the solely permitted browser engine, the Web’s three decade long run of ongoing rejuvenation will slowly come to an end. It won’t of course go away, just, like so many other technologies before, plateau and become an increasingly boring part of the furniture–useful, but not exciting.
And I believe, for many reasons that will have to wait for another day, that would be a tragic loss.
More to come…
This is a big complex issue, on which reasonable people have reasonable differences of opinion. For example there are genuine discussion to be had around the privacy and security implications of at least some Web APIs like Bluetooth, that Mozilla has in addition to Apple expressed concerns about.
I do want to return to some of those issues in the future, but this piece got quite long and involved enough already.
In 2022 we have a whole series of events for Front End Developers
Across 2022 Web Directions is presenting our series of online conferences for front end designers and developers. Focussed deep dives, they go far beyond what you might expect from conference programs.
Priced individually from $195, or attend all 6, plus get access to our conference presentation platform Conffab for just $595, or $59 a month.
Great reading, every weekend.
We round up the best writing about the web and send it your way each Friday.
