The browser wars are back on!

Remember the browser wars of the late nineties? Half the online population thought “Netscape” actually was the Internet and Microsoft was just starting to take the Internet seriously.

Of course, Internet Explorer emerged the victor and whether you’re in the camp that faults Microsoft’s heavy-handed tactics or the camp that recognizes Netscape’s failure to innovate, there’s probably some truth in both.

But now the war is back on. Except it’s not a browser war, it’s a “rendering engine” war.

What’s a rendering engine?
So, a browser is a program on your computer that you use to access websites. You probably know this and that may seem simple but Google went to the streets recently and found many people don’t grasp that concept.

But the browser program is fundamentally just the window, the menus and buttons, your bookmarks, etc. What’s really running the show is the part of the browser that actually displays or “renders” the website, and that is called the rendering engine.

Internet Explorer’s rendering engine is called “Trident.”

From Netscape’s ashes sprung a rendering engine called “Gecko.” That’s what’s inside Firefox for the rapidly growing population of Firefox users out there (me included).

So those are two major browsers, er, rendering engines, in use. But what about Apple’s Safari? And the new kid on the block, Google Chrome? Two more browsers? Kind of. They both actually use the same rendering engine: an open-source software project called “WebKit.”

Each of these rendering engines is updated from time to time so we have different versions and frankly, these different rendering engines make it very difficult to develop websites and software for the web!

They are all supposed to work in roughly the same way but they each have their quirks and nuances and it’s often difficult to get them all to behave the same way. Particularly when your goal is to work on some of the older versions.

Let’s get back on topic: The first browser war was about features. The rendering engine wars are about performance. For the sake of accessibility, however, let’s keep the rendering engine in mind but switch back to talking about browsers because rendering engines just aren’t what people use directly.

This performance war centers around online web applications like Google Maps, Apple’s Mobile Me service, Facebook and countless others. To provide the interactive interfaces you’ve come to expect from these types of websites, the rendering engine in your browser needs to be able to work very quickly on things like “DOM parsing” and “JavaScript execution,” technical terms for “making stuff happen.”

When the rendering engine can’t keep up, things tend to be a little slower to transition — you might see animation that isn’t as smooth as it should be or other hesitations that detract from the task you are trying to accomplish.

Given Firefox’s rapid growth in market share, its ability to be extremely competitive (if not always the fastest) when it comes to performance and its add-on ecosystem, it’s hard not to crown it the best all-around browser.

Chrome and Safari seem to hang on primarily because they both use the same rendering engine, which by all accounts is well written from a software engineering perspective (the same engine also powers the iPhone browser); plus they have some niche markets like Macintosh users and Google junkies.

And what about Internet Explorer? In my opinion, it’s got the worst rendering engine. It has the most quirks, the worst performance, the worst standards compliance and the most “versions” in use. Don’t get me wrong — we develop web software almost exclusively on Microsoft’s platform so I’m not a Microsoft hater. But I’m also not an apologist: Their browser’s rendering engine sucks! They like to say you have to use a slow-motion camera to see the difference between their browser loading a page and some “other” browser loading the same page. But it’s not about loading CNN.com today. It’s about loading an immersive and interactive “search-as-you-type CNN.com news interactive web application” tomorrow. And I don’t mean tomorrow next year. I mean tomorrow tomorrow.

So the good news is that there is renewed competition in this space and while that’s a royal pain for web developers like myself, it’s good news for the future of the interactive web.

Microsoft's Bing

Microsoft has launched its answer to Google in Bing, its replacement for Live Search. This latest go at a search platform has some interesting quirks, including instances of single results on the SERP (search engine results page) for keywords it’s certain have a single destination site.

On the homepage, Bing layers a pleasing image behind the search box, with your standard Images, News and other options down the left-hand column. The image has discoverable hotspots, which both fits the theme of searching and could be a very interesting way to integrate ads onto the most visited page at Bing.

Bing brings two unique features to the SERP: its Related Searches field and, above it, a useful info–type field that lists topics commonly associated with your search term. This pops up when your search item has easily identifiable subsections.

For example, search for “Subaru Legacy” and the field offers Reviews, Recalls, Dealers, Reference (a page scraped from Wikipedia), Videos and other options. It’s an interesting approach, and seems useful for generic searches. Not sure how many people perform short-string searches (when your search term has one or two words), though.

But, when you do a more natural long-string search, like “Subaru Legacy reviews,” the field doesn’t appear. Microsoft is Bing-ing big, though, investing around US$100MM in an ad campaign which includes, not surprisingly, a strong showing in Google AdWords.

Interested? Just Google “Bing” to try it yourself.

Search engines

The top search engines have collectively agreed on a solution to a problem that has long vexed web developers and web site owners.

Google, MSN, and Yahoo have announced support for the use of a "canonical" link tag as part of a page's markup to help identify duplicate content. The three have a combined search market share of over 90 percent, according to several independent rating surveys.

Search engines analyze web site content by programmatically crawling through each page. In an effort to provide better search results to consumers, the search engines will penalize a site for having duplicate content. However, if multiple different links such as:

http://www.example.com/Category/Item/12345

http://www.example.com/Item?id=12345

http://www.example.com/Item/some-item

...result in the same or similar content, then the site may be penalized by the search engine even though there may be a valid reason for this condition.

Now developers can use a link tag such as:

<link rel="canonical" href="http://www.example.com/Category/Item/12345" />

...on each of the duplicate results to alert the search engine that even though there may be the same content on several pages on the site, that the separate results are canonical to a single point on the site and should be treated as such.

This development is important to web professionals in that resources previously used to combat an erroneous duplicate content penalty can be shifted elsewhere.

Proper use of this new technique is one more tool for use in creating better search engine marketing results.