Browser innovation will happen. Google will drive it

I definitely ruffled some feathers with my last blog post, “The web sucks“.

The hardcore web developers are pretty upset. Those who have never developed for the iPhone, or are strong believers in standards and openness, tend to leave comments like, “you suck!” Really mature.

The more intelligent commenters seem to get my point. They do acknowledge that the web platform is innovating very slowly, and has quickly been eclipsed by native mobile apps.

I don’t know what the solution to the browser problem is. Maybe non-standardization isn’t the way to go, it’s just one option I propose.

I don’t think the web will die, but if the web platform loses any market share at all, Google stands to lose the most. That’s why I think they will lead innovation.

1. Google should drop support for IE6 across the board, most notably, in GMail. If they did this, imagine how much market share IE6 would lose, overnight! Even the slowest IT departments out there would no choice but to figure out an upgrade plan.

2. Google Chrome should release some non-standard APIs that make Google web apps really standout. You want to use GMail in all its glory? Then you have to use Chrome. Other browsers may or may not implement the APIs that Chrome creates. Developers may or may not use the Chrome only APIs to enhance their apps.

But people won’t complain about the non standard Chrome SDK because:

  1. It’s Google
  2. Google will open source it
  3. Other browsers will copy it
  4. The apps developed with it will be better than ever

And with that, a new era in browser competition and innovation will be born.


  1. Actually you point about browser API’s is interesting. I was proposing this idea to a colleague earlier in the week after he went to a Google Talk evening all about HTML5. Although I still think you’re wrong on file systems disappearing altogether, like the Mac Finder, the web does indeed suck. I’ve developed for iPhone, and would have no hesitation in never writing another line of HTML for the rest of my life. You can barely do anything with it, and when you have to, it’s stateless and a complete and utter pain to make pretty. Compared to Apple’s Interface Builder, styling for the web is a joke.

  2. While I like the idea of improving the web browser experience, I just don’t feel like fragmenting the capabilities of various browsers is at all a good idea. Isn’t fragmentation of standards as it is right now part of the reason why there’s such a desire to move to HTML5 over Flash, and to develop for the web in the first place, which is in part cross-platform compatibility, no? A move like this would only continue to foster the problems we have currently, I believe.

  3. Canvas + XHTML (and XML namespaces) + CSS + Javascript + a high-performance runtime = do any damn thing you please in a browser.Here’s the breakdown:Canvas – Super low-level 2D primitives allow all sorts of thing from a canvas implementation of flash to a canvas 3D engine. The possibilities are endless here.XHTML – I’ve been watching HTML5 evolve thinking that its kind of pointless if you think of HTML differently. First, I don’t think HTML should get "special treatment" in the browser. It should be treated as one XML namespace. A browser just understands a default way of rendering an HTML XML namespace. If developers want to do some different, fancier shit, they could mix-in their own custom XML namespaces, and add whatever arbitrary nodes they want to the DOM. For example, Posterous should be able to do something like:<html><body> <blog:title>Sachin’s Post</blog:title> <blog:content>blah blah blah I hate browsers blah blah blah</blog:content> <blog:author>Sachin</blog:author></body></html>which leads me to my next point…Javascript – We’ve got Canvas, XHTML, and custom XML namespace support, which, in theory means you can draw anything you damn well please on screen. Javascript is the code that makes this stuff move. There’s no reason you can’t build a "native" app with all of these primitives under the hood. You could also apply certain behaviors to various DOM nodes and distribute that globally with your XMLNS. What’s to prevent posterous from distributing their "blog" application with the markup above and something like:<html> <script language="javascript" href=""></script><body&gt; <blog:title>Sachin’s Post</blog:title> <blog:content>blah blah blah I hate browsers blah blah blah</blog:content> <blog:author>Sachin</blog:author></body></html>If somebody else on the global world-wides wanted to implement your behavior of blogs, they could just reference that js in their own code and your your xmlns.High performance run-time – You need all the shit above to run fast. Google is definitely doing that.Net net… I don’t think we need new tools or extensions from browser companies; rather they need to focus on making their existing stuff super fast and extensible in a way the other developers can write and extend the browser without requiring the installation of native code.

  4. ^^^ Aww come on… markdown or some kind of formatting support for comments eh? A bunch of shit I quoted was last.

  5. Web is already fragmented in the current state as browsers don’t even support most of the basics well. Standards are still in the game, even when people develop for mobile devices like iPhone and others. It’s just that User Interfaces and screen sizes of the different devices create difficulties for many so in one sense, it’s useful that there are manufacturer related SDK’s and guidelines that form new kind of standards for development for the specific devices.As mentioned in the comments of the previous blog post, al3x wrote about importance of APIs for web services since those make it possible to create new kinds of user interfaces for using content, instead of restricting it to the specific look and style. Mobile devices have made things different for web developers and many have realized how difficult it is to create things that would work on all devices and look like same thing. It’s probably not even how web was originally thought: there will always be different presentations of same information and standards are not making those differences disappear.(Offtopic, but still… you should offer even some basic abilities to format comments in some way. I understand that it’s kind of difficult to allow too wide amount of markup as it would make life of theme authors difficult but some basic things should be allowed.)

  6. About IE6 support, I remember reading that the market share of IE6 is extremely high in countries like China where large part of users use "modified" versions of Windows. Not sure about statistics, but thought that it could be interesting thing to note: are more than few reasons why large companies are still using IE6 and while many are doing work to migrate to new versions of browsers, large majority is still too slow about it. I surely understand that when business is tied to some specific customized software that requires IE6+ActiveX to work, it’s often painful process to move away from it. But, more or less, that shouldn’t restrict the whole web.Hopefully browsers will learn how to do new things in smart way, allowing to do new things via their own APIs while still working actively on supporting reasonably well defined base level specs that make things visible, regardless of the browser.

  7. Actually, I really agree with a lot of your post. I am not web developer but as you said in a previous post it really is all about the experience one has on the web. Google as a platform works well for me. The gui is minimalistic and not flashy but it gets the job done and integrates fairly seamlessly with its other applications.I really hope that Google does lead the way when it comes to innovation. I was actually hoping for that when Wave came along and was among the many that clamoured for an invite to see what the fuss was about. It certainly has its possibilities but needs to be developed further. However, many people are finding ways of making it work.What innovations would you like to see put forth into future browsers? I’m not interested as a developer but more as a consumer of data from the web.

Leave a Reply

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

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

Google photo

You are commenting using your Google 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 )

Connecting to %s