522-browser-trends

Despite the ongoing Firefox releases, the browser market has remained quiet during the summer months.

So here are the latest statistics. I’ve changed the table so Firefox 4, 5 and 6 are amalgamated into one; it makes little sense to analyze the separate figures since most of those users update their browsers as new versions appear:

BrowserJulyAugustchangerelative
IE 9.07.27%8.05%+0.78%+10.70%
IE 8.026.30%25.68%-0.62%-2.40%
IE 7.05.45%5.07%-0.38%-7.00%
IE 6.03.42%3.09%-0.33%-9.60%
Firefox 4.0+17.66%18.10%+0.44%+2.50%
Firefox 3.6-10.30%9.39%-0.91%-8.80%
Chrome22.17%23.17%+1.00%+4.50%
Safari5.15%5.18%+0.03%+0.60%
Opera1.66%1.67%+0.01%+0.60%
Others0.62%0.60%-0.02%-3.20%
IE (all)42.44%41.89%-0.55%-1.30%
Firefox (all)27.96%27.49%-0.47%-1.70%

The table shows market share estimates for desktop browsers. The ‘change’ column shows the absolute increase or decrease in market share. The ‘relative’ column indicates the proportional change, i.e. another 9.6% of IE6 users abandoned the browser last month. There are several caveats so I recommend you read How Browser Market Share is Calculated.

IE9 had another good month. Its progress is remains relatively sedate, but there are two solutions if Microsoft want massive adoption:

  1. Offer Windows 7, the hardware which runs it, installation, migration and training services to everyone. For free.
  2. Alternatively, release a version of IE9 which is compatible with XP. The other vendors support XP and still manage to offer fancy features such as hardware acceleration. And CSS3 text shadows.

IE’s overall drop has slowed a little this month, but I suspect that’s a statistical blip while business users enjoy a summer break.

Firefox 4/5/6 is rising but not at the pace Firefox 3/2/1 is falling. While the rapid releases are mostly good, users are becoming frustrated with add-on compatibility failures and memory usage problems on Mac OS. Mozilla is addressing the issues but they’re losing users who may never return.

There’s little to report for Opera and Safari. Both browsers made modest gains, but neither is setting the market alight.

That leaves us with Chrome. It’s the same story: usage continues to grow at 1% per month — sometimes more. If the current trend continues, Chrome will overtake Firefox in December 2011. It’s already occurred in the UK where Chrome has 23.41% lead over Firefox’s 21.75%.

Personally, I like Chrome and regularly recommend or install the browser; it’s fast, simple, stable and updates without fuss. However, I primarily use Firefox (on Windows 7) because it has a range of essential add-ons for power-surfing and development. I thought others would think the same but, having asked the question on Google+, it appears not. Developers are switching to Chrome in droves. Mozilla is losing the technical evangelists who once promoted Firefox.

Mobile Browser Usage

Desktop browsers account for 92.88% of web activity. The remaining 7.12% is mobile access and it’s evident more people are using their phones for general web browsing. The applications they primarily use are:

  1. Opera Mini/Mobile — 21.61% (down 0.46%)
  2. Android — 19.72% (up 1.55%)
  3. Nokia browser — 16.99% (down 0.11%)
  4. iPhone — 14.91% (down 0.19%)
  5. Blackberry — 11.64% (down 0.66%)

Note there are significant regional variations:

  • In the US and Canada, Android takes the top spot with 34.2% followed by the iPhone with 26.1%. Opera accounts for less than 4%.
  • The iPhone is most popular in Europe at 33.7% with Android second at 23.7%.
  • For Oceania, the iPhone has an almost monopolistic lead of 56.7%. Android is way behind at 19.4%.
  • It’s Asia, Africa and South America where Opera and less-expensive Nokia devices reign supreme.

Remember that these figures are collated from internet access — not sales trends. Users with an older mobile are less likely to use the web than those with the latest 3G handset. That said, in the developing world, users may not have access to a PC so mobile is the only option.

522-browser-trends

It’s increasingly difficult to keep track of the browser market. Chrome 12, Firefox 5 and Opera 11.5 were released last month. Some browsers auto-update, some don’t. Some vendors have lavish launch promotions, others don’t mention it.

The big news for July is that Chrome usage has passed 20% for the first time. Let’s examine the full StatCounter statistics in more detail…

BrowserMayJunechangerelative
IE 9.04.57%6.18%+1.61%+35.20%
IE 8.029.06%27.67%-1.39%-4.80%
IE 7.06.39%6.00%-0.39%-6.10%
IE 6.03.84%3.72%-0.12%-3.10%
Firefox 5.00.00%2.81%+2.81%n/a
Firefox 4.014.23%14.04%-0.19%-1.30%
Firefox 3.5+13.95%10.44%-3.51%-25.20%
Firefox 3.1-1.12%1.05%-0.07%-6.30%
Chrome19.38%20.67%+1.29%+6.70%
Safari5.01%5.07%+0.06%+1.20%
Opera1.83%1.74%-0.09%-4.90%
Others0.62%0.61%-0.01%-1.60%
IE (all)43.86%43.57%-0.29%-0.70%
Firefox (all)29.30%28.34%-0.96%-3.30%

This table shows market share estimates for desktop browsers. The ‘change’ column shows the absolute increase or decrease in market share. The ‘relative’ column indicates the proportional change, i.e. another 3.1% of IE6 users abandoned the browser last month (yay!) There are several caveats so I recommend you read How Browser Market Share is Calculated.

In June, Chrome 11 toppled Firefox 3.6 to become the world’s second most-used browser. Confusingly, the launch of Chrome 12 has split Google’s user base so Firefox 4.0 has now taken second place. Despite being available for little over a week, Firefox 5.0 has already gained 2.8% market share as Firefox 3.x and 4.0 users migrate.

However, there’s little good news for Mozilla. Firefox’s overall total dropped by almost 1% in June: three times worse than IE and one of the biggest falls the browser has ever experienced. There doesn’t appear to be a particular reason; Firefox 4 and 5 have been well-received but they haven’t halted Chrome’s progress. Perhaps the changes were too radical for some? Or did users investigate other options rather than upgrading?

IE9 has made good gains although IE8 remains the most popular browser version. IE6 and 7 continue to drop although the pace is slowing.

Opera also experienced a small drop. However, version 11.5 may be able to reverse that trend and there’s better news for the company in the mobile arena…

Mobile Browser Usage

According to StatCounter, desktop browsers account for 93.47% of web activity. Mobile browser usage grew by almost 1% last month to 6.53%. This may be a seasonal anomaly since it’s summer in much of the western world — net users may be out enjoying the sunshine (or drizzle for those of us in the UK).

Movements within the mobile browser market are quite unusual and possibly influenced by seasonal factors. Nokia may be experiencing business issues, but they will be pleased to discover that their (fairly basic) browser has overtaken Android and Safari on the iPhone. Opera has also made gains following the latest release of their mobile editions:

  1. Opera Mini/Mobile — 22.81% (up 1.00%)
  2. Nokia browser — 17.66% (up 1.16%)
  3. Android — 17.25% (up 0.24%)
  4. iPhone — 15.22% (down 1.49%)
  5. Blackberry — 11.98% (down 0.78%)

If you’ve not done so already, perhaps it’s time to consider how your business will be affected by the rapid rise of mobile platforms.

Mozilla’s Firefox 4 was released early Tuesday. The release comes nearly two years after Firefox 3.5 and three years after Firefox 3.

The web browsing landscape has changed significantly since then, with Google’s Chrome browser winning converts left and right, while mobile and tablet browsing gained new ground.

When Firefox first hit the scene in the early 2000s, Mozilla’s open-source browser was a refreshing change of pace for users and designers alike. It brought innovative features like tabbed browsing to the mainstream (although Opera did it first). It used add-ons to an extent that hadn’t been seen before.



Over the last few years, early adopters — once the core evangelists for the browser — shifted away from it. Those add-ons started to bog the program down. Meanwhile, the new layout engine of choice for web developers isn’t Gecko (which powers Firefox), but WebKit (which powers Apple Safari, Google Chrome, and the browsers for BlackBerry and Android).

Firefox 4 is an important release for Mozilla — perhaps the most important release since 1.0. The competition has never been so strong. We have been using the beta releases extensively and spent some time with the final release. So how does Firefox 4 stack up against the competition? Are the changes enough to keep current users from switching — and lure old users back?


Look and Feel


Mozilla first started talking about Firefox 4.0 in July 2009. The early screenshot previews — featuring tabs on top, a la Chrome — were a radical departure at the time.

Although the comparisons to Chrome are unavoidable, I think that Firefox 4 improves upon Google’s minimalist design.

Tabs are on top, but the browser window is still easily draggable. Users won’t make the mistake of dragging a tab rather than the full window. Moreover, cycling through tabs is more elegant and less cluttered than either Safari 5 or Chrome 10.

By default, Mozilla has changed the location of the home button. It also added a new bookmark bar. Fortunately, these components can be customized and removed (simply right click on them and hit “customize”). Like Chrome, Firefox eschews the the status bar on the bottom of the screen, only using it as an overlay when needed. This adds a few more pixels of space to the viewing window.

Firefox 4 includes an innovative new tab grouping feature known as Panorama. Panorama started life as Tab Candy, an experimental feature introduced by former Mozilla Creative Lead Aza Raskin. It creates different groups of tabs and lets you switch easily between them. Panorama is a great feature for power users, but anyone who don’t want to use a grouping system can ignore it and never know the difference.


Speed


Firefox used to be the fastest browser on the block. Over the years, the program has become bloated. Increasingly, the speed factor in web browsers is less about the rendering engine and more about the JavaScript engine.

Firefox 4 claims to be up to six times faster than its predecessor. In our tests, load times did seem about that fast — though Google Chrome 10 still seems to bring up pages more quickly.

The speed increases aren’t merely limited to page load times, however. Firefox 4 starts up significantly faster on my Mac (an iMac with a 2.8GHz i7 and 12GB of RAM running Mac OS X 10.6.7) than its predecessor. In fact, in a timed test, Firefox 4 launched from dock to default homepage at nearly the exact same speed as Google Chrome 10.


Performance, Memory Usage, Stability


As a full-time Mac user since 2007, I’ve long had a love/hate relationship with Firefox. Certain websites (particularly corporate backend systems) just work better in the browser than in Safari. But Firefox has never been particularly well tuned to Mac OS X machines. Firefox 3 was a significant improvement, but frankly, Firefox has remained a memory hog.

The biggest problem with Firefox versions of the past — and this is true of both Mac OS X and Windows releases — is that the program has the tendency to leak memory. This problem only gets worse on systems with lots of add-ons installed and can be made worse still by plugins like Flash.

Mozilla has said that Firefox 4 consumes less memory and is more stable. I wanted to see if this was true. Using the Activity Monitor in Mac OS X, I tracked the amount of real memory, CPU utilization and CPU threads in Firefox 4, Firefox 3.6.15, Safari 5.0.4 and Chrome 10.0.6.448.151 stable.

I tried to install the same number of add-ons or extensions to each browser. The goal was to re-create the average browsing session. I then opened a number of memory-hogging tabs, including Farmville and Hulu with video playing.

I tested the memory and CPU usage for each browser. Remember, your mileage may vary.

First, the good news — in my tests, Firefox 4 consumes less memory and CPU cycles than Firefox 3.6.15. When adding in Flash and other plugin usage to the total memory footprint, only Google Chrome 10 performs better.

The bad news — and this is really for all four browser variants tested — is that the overall usage is still fairly high. The big culprit here is Adobe Flash. Improvements have been made on this front in Windows and with certain graphics chipsets on the Mac (my Radeon HD 4850 unfortunately, is not included), but Flash is the greatest cause of browser performance and memory usage issu
es.

So if Firefox 3.6.x takes up a lot of memory on your system, the improvements in Firefox 4 might not be significantly better.

What is new is that Firefox 4 now segregates its regular browsing processes from so-called plugin processes. Previously, Firefox was the sole item to appear in the Mac OS X Activity Monitor. With Firefox 4, a “Firefox Plugin Process” appears as well.

So if Flash wasn’t running a game and playing back a video, that Plugin Process usage would be considerably less. Rather than relying on the browser to free up the memory (something Firefox is historically bad at doing), the plugin process can simply be freed up.

Moreover, if a plugin crashes, the browser can recover without taking down the entire session. Apple is doing something similar in Safari 5.0.x, which shows Flash Player as its own process. If Flash crashes, the rest of the browser can stay intact.

With Chrome, Google goes a step further and actually separates each tab into its own process. That makes it easy to shut down one tab and keep the rest of the session running. Chrome doesn’t separate Flash as its own entity; the browser uses its own sandboxed version of Flash Player.

It’s great that Mozilla has decided to split up the way Firefox uses memory. Recovering from crashes is less time consuming, and regular system memory can be reclaimed more quickly.

Since Firefox 4 Beta 8, I have found the browser to be very usable with few stability issues. The few issues that remained up until the final release — notably Netflix not wanting to work well on the Mac — have been resolved in Firefox 4. In the 24 hours I have been testing Firefox 4, I haven’t had the browser seize or crash. It’s rare that I don’t have to invoke “force quit” for Firefox 3.6.15, so this is a great sign.


Add-ons


Mozilla has restructured how its add-on system works. Add-on installation and browsing now takes place in a designated browser window, rather than a pop-up menu. This is much more easy on the eyes and makes finding and installing or removing add-ons and browser themes more fluid.

Most major Firefox add-ons now work with Firefox 4. Users may run into situations where an add-on is incompatible. But most developers seem to have answered the call. If your favorite add-on isn’t updated in the next week or two, it might be time to look for a replacement; it probably indicates developer abandonment.

Firefox is continuing to move towards lighter weight extensions like those for Google Chrome, Safari and Opera. These add-ons can be built using HTML, CSS and JavaScript and tend to use less memory and resources. They also tend to have less of an impact on overall browser performance and stability.

Still, at this stage, most major Firefox add-ons still use the traditional add-on API and require a browser restart when updated, installed or uninstalled.

I have long said that add-ons and extensions are Firefox’s greatest strength and its greatest weakness. The impact that the extensibility these add-ons added to the browser on overall user adoption cannot be understated. It’s equally true, however, that the performance impact some popular add-ons can have on the browser has hurt Firefox’s image as a whole.

Even with Chrome, users have to battle how many extensions are installed versus the performance impact on the browser. It’s a tough line to straddle between utility and performance. But from what I understand about the Firefox add-on APIs and toolkits, it is an area Mozilla has spent a lot of time working to make better.


Overall


So is Firefox 4 good enough to lure back old users and to keep existing users satisfied?

For me, the answer is yes. While I don’t anticipate using Firefox as my primary browser (I tend to use Safari), keeping Firefox running on my computer is no longer something I fear.

The new user interface is fresh and inviting. Panorama is something I could see using on a regular basis, and the memory and performance improvements live up the expectations.

Firefox fanatics are going to love it. Developers that test in multiple browsers are going to be very pleased. Still, I don’t know if being on par with the competition is enough to bring old users back.

I’m going to continue to use Firefox 4 more over the next few weeks. For me, that’s an important development. Since Google Chrome officially came to the Mac in December of 2009, I have used Firefox primarily only to access certain websites behind a VPN. It’s great to actually enjoy using the old girl again.

Let us know your thoughts about Firefox 4 in the comments.

More About: Browsers, chrome, Firefox, Firefox 4, Internet Explorer 9, mozilla, reviews, safari

For more Dev & Design coverage:

429-ie-50pc-share

In August 2010, Google Chrome exceeded 10% market share. Another milestone was achieved in September: Internet Explorer’s total usage dropped below 50% for the first time in over a decade.

There have been cheers throughout the web design and development community and the story has spread throughout the technical and mainstream press. However, it’s worth examining the StatCounter statistics in detail…

BrowserAugustSeptemberchangerelative
IE 9.0 beta0.00%0.09%+0.09%n/a
IE 8.029.40%29.38%-0.02%-0.10%
IE 7.013.91%12.98%-0.93%-6.70%
IE 6.08.02%7.42%-0.60%-7.50%
Firefox 4.0 beta0.00%0.26%+0.26%n/a
Firefox 3.5+28.03%28.33%+0.30%+1.10%
Firefox 3.0+2.60%2.48%-0.12%-4.60%
Chrome10.76%11.52%+0.76%+7.10%
Safari4.06%4.22%+0.16%+3.90%
Opera1.88%2.03%+0.15%+8.00%
Others1.34%1.38%+0.04%+3.00%
IE (all)51.33%49.87%-1.46%-2.80%
Firefox (all)30.63%31.07%+0.44%+1.40%

The ‘change’ column shows the absolute increase or decrease in market share. The ‘relative’ column indicates relative movements, i.e. 7.5% of IE6 users switched browser in the past month.

We can’t make too many assumptions from this data, but there’s one statistic reporters appear to have missed: IE8 usage has barely changed. The 0.02% drop is more than outweighed by the 0.09% gained by the IE9 beta release.

The most significant contributory factor for IE losses is migration from versions 6 and 7. Although a proportion of those users will have moved to IE8/9, a greater percentage has switched to an alternative such as Firefox or Chrome. There are several reasons why this has occurred:

  • IE8 is a capable browser but there are plenty of better options for IE6/7 users.
  • Good web applications are browser-agnostic or work on a variety of platforms. Legacy business applications are being updated and there’s less dependency on IE.
  • Microsoft and all other major vendors are backing HTML5. The rendering differences between browsers is smaller than ever and it rarely matters which application you use.
  • The majority of businesses use Windows XP and may have no intention of upgrading — especially during continued economic uncertainty. Yet IE9 is only available on Windows Vista/7. Why would a business continue to use an application which the vendor has (effectively) abandoned? It’s far cheaper and easier to install an alternative browser than upgrade the OS.

While we should be thankful for the drop in IE6/7 usage, 1 in 5 visitors continue to use the ancient browsers. Predictions of IE’s demise are premature and IE8 remains the world’s most-used browser.

It’s better news for the other vendors. Firefox, Safari and Opera all gained but Chrome remains the biggest winner. Google’s browser increases by nearly 1% every month and shows no sign of peaking. However, it’s about to face a stronger challenge from Firefox 4 and IE9.

422-xmarks-end

It’s been a bad month for online service closures. Bloglines, my favorite RSS aggregator, is closing this week and now Xmarks has reached the end of the road.

Xmarks is a free synchronization service which backs-up and replicates your bookmarks across Firefox, Chrome, Internet Explorer, and Safari. The product launched in October 2006 as a Firefox-only extension named Foxmarks. Additional browsers were added in 2009 and the company re-branded itself as Xmarks.

As well as bookmarks, the plug-in can also synchronize passwords, browser history, and open tabs. Early on, the company realized they were sitting on a gold mine of 100 million user-verified page links. They experimented with their own search engine but the system failed usability tests. Although it could provide spam-free categorized links, it couldn’t answer the specific questions people entered in typical search engines queries. However, the plug-in can append information to Google results to indicate the popularity of a link.

For me, Xmarks remains the best and most reliable bookmarks synchronization service. It’s invaluable if you’re using multiple PCs and it’s the only Firefox add-in I religiously install on every browser. It has 2 million users, supports 5 million desktops, and receives 3,000 new sign-ups every day.

Despite it’s success, co-founder and CTO Todd Agulnick explained they have struggled to monetize the service. The company has been for sale since early 2010, but no potential buyers have come forward. The business model may not have a viable future now that Microsoft, Mozilla, Google and Opera all offer free synchronization with their browsers. Unfortunately, few of these options are as good as Xmarks and none provide cross-browser functionality.

The company is sending emails to users and the Xmarks service will shut down at the end of 2010.

Do you use Xmarks? Will you be affected by its demise? Are you using a good alternative?

126-collective-noun

Web designers and developers usually have a selection of the most popular browsers installed on their PC. You don’t? Really? Why not install a few and give them a go.

Operating Systems allow you to set a default browser and we all have our favorites. Some love Firefox for its flexibility. Some swear by Chrome for its speed and clean interface. Others prefer Opera for its tools and features. Many Apple users love Safari’s OS integration. IE users like the browser because … erm, well, they have their reasons (and we have high hopes for IE9).

I started with Netscape 2, migrated to IE3, 4, 5, 5.5 and 6, then switched to Phoenix, Firebird and eventually Firefox. Although I had other browsers installed, I rarely used them for anything other web page testing.

However, in the past year or two I’ve noticed a change in my browser usage patterns. I now use whichever application is most practical — sometimes, it’s simply the icon closest to my cursor. There are a several reasons:

  1. The 5 main browsers are all good applications. You may prefer one over another, but none is perfect and even the worst is fine for general web surfing.
  2. Chrome and Safari may offer some amazing CSS3 effects but the gap between the browsers is smaller than it’s ever been. All of the top browsers offer decent rendering capabilities.
  3. It’s often practical to have two or more different browsers open, e.g. if you’re accessing work and private GMail accounts at the same time.

In most cases, though, I use whichever browser offers the best facilities for the task in hand. For fast browsing, I might use Chrome. On a netbook, I often use Opera for it’s speed, built-in email and turbo mode for slow connections. For storing bookmarks and web page development, it’s hard to beat Firefox. Finally, I still use IE for testing and a few specific corporate applications.

While I doubt many general Internet users flit between applications, it’s increasingly less likely for a power user to have monogamous relationship with a single browser. Then again, perhaps it’s just me — I’ve become a browser whore.

What do you think? Are you wedded to one browser or do you flirt with them all? Please vote on the SitePoint poll or leave your comments below…