It’s time for web 1.0 to die

Yes, this is going to be a controversial topic to write on and yes, I have a history of terribly messing up controversial topics. But I’m going to make an effort to keep my thoughts clear and justifiable. Without sounding too academic, let me first present my thesis:

The original internet and World Wide Web was one of the greatest technological innovations of human history. It has an interesting history fundamentally linked to many other developments in the past few decades. Any self-respecting computer geek should take a few hours and read about the start and growth of the net. However, in the 21st century, the old web is no longer enough. It is time for us to accept that the old days of plain HTML and static links are just that: old days. I’m going to show that important developments have made it necessary for us to look at the web in a new light and we must actively embrace these changes so that we can have a hope of shaping them into something as monumental as the original web.

Clarification: I’ve received my comments that made me realize that I should clarify what I mean by web 1.0. For the rest of this post I’ll take it to mean the web in terms of mostly human-generated, static pages with mostly static content. By contrast web 2.0 can be defined to be mostly auto-generated, constantly changing content where the main human role is to create the content, not manage its organization or delivery.

Trip down memory lane

In many ways the start of the internet can be traced back to Douglas Engelbart’s famous Mother of All Demos in 1968. This demo showed of the revolutionary NLS (oNLine System) which was years ahead of its time and presented important concepts such as voice and video conferencing, complex document formats and hyperlinks.

Fast forward a few decades and the internet was slightly more mainstream by the early 1990’s. Though the internet had implemented Engelbart’s vision to some extent, it was actually quite primitive by the standards of the demo. Hyperlinks were simple one-way roads instead of the ubiquitous and powerful cross-linking system that was part of the demo. Rich media communication was still years away. However, the key parts were in place. The early web was a rather simple place to live in. Mostly text, followed later by simple graphics. Pages were static and more often than not crafted by hand. For much of the 1990s the web bore the signs of the academic and somewhat austere and information dense culture that had spawned it.

That web is not the web of today. With low bandwidth, information density needs to be maximized, presentation is often an afterthought. While there are a small number of producers and a large number of viewers, it’s fairly simple to do things by hand. The drastic change of the internet in the past few years can be attributed to two main causes:

  1. The growth of scripting languages and applications built on them have allowed a massive explosion in the number of content producers. It’s possible to build a fairly high quality website without writing a single line of code.
  2. Bandwidth, storage and processing prices have fallen, opening the door for rich media applications, which can be distributed across the Internet.

Dynamism’s Day

The game of web development has changed at a fundamental level. You can no longer ship a website to a customer and expect it to stay the same way for months at a time. Dynamic content is the order of the day. Blogs are no longer creations of random individuals with too much time on their hands. Instead, if you want to have a viable web presence, you’d better have a blog and you’d better update it regularly. This goes for individuals as well as corporations (especially those whose business is the web). This in turns means that hand-rolling a website is impractical at best and downright stupid at worst. A recent article proclaimed that Dreamweaver is Dying, and I have to grudgingly agree. Over the past two years, I tried maintaining an old style website by hand with mostly static content, but it’s simply not worth it. I would rather spend 5 minutes working on a new post than fiddling around with HTML and CSS to get things to look right. And don’t even get me started with keeping links and navbars up to date. Dreamweaver and similar tools help. A lot. But they’re not enough.

Web 2.0 is showing us an important fact that we must not ignore: content creators do not need to be programmers as well. What that means for you and me is that if you want to create a content-focused site (and that’s what most of us want really) then an automated content management system like WordPress or Drupal combined with good themes and widgets should be our first pick. Only if the need for customization becomes overbearing (ie. you need to forge a brand image with a custom logo and theme) should you consider diving into the code.

For web developers, this means that you should pick a content platform or two and learn it back to front and inside out. Try porting some previous designs and suggest that customers take a more active role in creating their web presence. The web’s hallmark is that it is by definition a bidirectional medium. If anyone wants to be successful on the web, that bidirectionalim must be respected and utilized.

To generate the best content, you want to have good tools. Unfortunately HTML and CSS simply aren’t good tools for writers, journalists or artists. They want, no, they need WYSIWYG editors where they can see what their content will look like without worrying about the layers beneath. Cheap bandwidth means you can now use heavier technologies like Flash to build better looking tools. Cheap storage and processing power means that you can generate the webpages on the fly while keeping the actual content neatly stored in a backend database. HTML and CSS aren’t the core technologies of the web anymore; they’re a thin veneer that hides the raw power underneath and gives everyone a simpler, focused view of what they need to see.

Let’s face it, HTML is ugly. It’s angle-bracket hell and you know it!! No human should have to write that sort of thing by hand. CSS is better, but not by much. A lot of people complain that autogenerators create really bad, really redundant HTML and CSS. That’s true and I thought that was a bad thing and I hand crafted my code until not too long ago. But the fact is: no one cares!! No one is ever going to really read your generated code except browsers and they’re mostly pretty tolerant of what they’re fed as long as it’s not downright wrong. The massive boost in productiveness far outweighs any aesthetic qualms you may have. Sure you still need to do some amount of tweaking to get things just right, but it’s very easy to get good enough without doing any tweaking at all.

On another note, the database + dynamic code creates far better presentation/data separation than HTML/CSS ever could. You had to be really disciplined to not mix presentation consideration into your supposedly semantically structured HTML and you never got it quite right. Admit it, you know you bent the rules. But with web 2.0, such separation is natural. Theming is at the heart of almost every major CMS out there and the web is a better place thanks to it.

But wait, there’s more…

CMS’s are just the tip of the iceberg. Web 2.0 is becoming an application platform, much closer to Engelbart’s vision than to Web 1.0’s “information highway”. It’s not just content creation and delivery, it’s active interaction that is getting the spotlight. And HTML/CSS utterly fails at this. Text and graphics are ok, sound and video are good, but active interaction is even better. Web apps may not be as feature-rich as their desktop equivalents, but they’re not far behind either. I use Google Docs almost as much as I use Word and web-based IDEs are starting to become a reality. Take a look at the SproutCore and Cappuccino web frameworks for some examples of what’s on the horizon.

The limitations of Web 1.0 have spawned the developments of multiple ways around them. The web is a now a mix of different document formats and PDFs are gradually becoming the document interchange format of choice for many organizations and companies. Streaming media has proved to be a much more popular alternative than simply offering up files for download. The dynamic web is a much faster and more interesting place than 1.0 could ever be.

We’re only starting to explore the web as a core component of personal computing. Cloud computing is still a very nascent technology, but one that looks like will it progress by leaps and bounds in the years to come. Amazon S3 lets you store practically unlimited amounts of data for really cheap making it possible for any technologically savvy individual or group to roll their own webapp without investing in massive computing resources first. Why buy a external hard drive when you can pay a small monthly free for crash-proof, access-anywhere storage? (If you trust them with your data that is, but that’s for another post…

Web 1.0 is simple and for the past two decades it has served the needs of human society admirably. But we need more now. There is a massive amount of computing power in the world today, but we can’t use it properly if we stick to old fashioned HTML. The internet is no longer a web of statically linked pages. It’s a complex network of rapidly changing web applications interfacing with each other on a number of levels. It’s more like a growing, vibrant ecosystem than it is a spider-web. Web 2.0 is alive.

In the near future…

We must learn to start treating the web like a vast collection of interacting programs and not just as a simple file hierarchy. The author of the Dreamweaver is dying article says (quite rightly):

In the relatively near future every website will be a dynamically-generated web application and all of today’s sites built on multiple static pages will be ripped out and replaced.

I think that is pretty close to the truth. Sure there will be still be some holdovers, but any serious website will have no choice but to become a distributed webapp: part of it running on a server (or server farm) and part of it running as a scripted application in your browser. Sure you’ll still need a grasp of HTML/CSS and their descendants (for a while at least), but you’ll use them in much the same you people do baking nowadays: you make special treats now and then, but you’d hardly ever bake your own bread.

Web 1.0 is dying. It’s passing can be painful, but it doesn’t have to be. Start using content management systems. I would recommend open source systems like Drupal, WordPress and Joomla. If you’re a developer, learn PHP and JavaScript while keeping your HTML/CSS knowledge in fair shape. If you’re building a web application or framework, make it easier for non-browser clients like other webapps to access data and functionality. The easier it is to use your servive, the more people will use it. Twitter and Google Maps seem to do this quite well. Don’t worry about supporting every version or every browser ever built. Pick 2 or 3 modern browsers and make sure versions released in the last year or two work well. Above all, don’t do anything to turn away the early adopters.

It’s time for web 1.0 to die an honorable death. It’s time for the rest of us to move on.

Advertisements

Published by

Shrutarshi Basu

Programmer, writer and engineer, currently working out of Cornell University in Ithaca, New York.

2 thoughts on “It’s time for web 1.0 to die”

  1. I agree that CMS’es and web apps are the “new coming thing” but I have no idea what your argument here is. The technologies you mentioned are not replacements of HTML, they are additional tools on top of it. It is a somewhat useful tip to say “don’t try to do everything with HTML coding, instead try solving your problems with these additional tools.” However, that certainly doesn’t mean that web 1.0 (whatever that is) is dead, it only means that the “old standards” are more future-proof because they are modular and adaptive.

  2. I agree that the new replacements do not replace HTML, but they allow us to ignore it, in much the same way that UI builders let us ignore the fact that there must actually be code that creates the UI. It’s similar to how machine language is still the ‘actual’ way to program computers, but higher level languages let us ignore that The web apps gives us an increased level of abstraction that becomes necessary as we deal with larger amounts of rapidly changing content.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s