Google announced further integration of YouTube and Chrome into its Google+ social network.

The search giant made the announcement in its official Google blog on Thursday, showing how it has continued its integration of YouTube into Google+.

Last summer, Google incorporated the ability to play YouTube videos in Google+ Hangouts. Now it takes that a step further, offering a YouTube icon on the top right of the Google+ interface that does a cool slide move when you mouse over it, asking “What would you like to play?”

Once you’ve entered the name of a video, topic or your favorite musician or band, a pop-up window appears, displaying a list of related videos that might interest you. I tried it by entering “Beatles,” and a Beatles video started playing in a dynamically re-sizing window, while offering more like it underneath.

Share a YouTube video and anyone in your chosen share circle has access to a related playlist, right from your post.

The company also announced a couple of new Google+ Chrome extensions — one is similar to a previously released version that lets you add a +1 to web pages, but now you can also share them with your circles right from its drop-down interface. The second extension lets you see your Google+ notifications right from your Chrome browser:

For those who don’t care to use the Chrome browser, Google says it’s offering a new version of Google Toolbar for Internet Explorer that also has these two new features on board.

More About: chrome, Google, YouTube

For more Dev & Design coverage:


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.

531-banishing-url-bar

Google wants Chrome to be a clean distraction-free browsing experience. They’re possibly about to take their most radical step yet. Interface minimalism will reach it’s ultimate zenith with the removal of the address bar.

Madness?

Perhaps. But Mozilla are considering the same UI move.

The idea has received an overwhelmingly negative response from technical users. However, before you reach for your soapbox, be aware that it’s only a proposal which may never see the light of day. If it does happen, it will almost certainly be an option and “compact view” might only be permitted on application tabs. When enabled, the user may have to double-click a tab to view the URL.

So why does Google think a 30-pixel gain is so important? It would provide an extra 5% of space on some tablet and netbook screens, but there are deeper reasons…

I use the address bar. You probably use it too. But many users don’t. Non-technical users rarely understand URLs; it’s plainly obvious when you observe them type www.whatever.com into Google’s search box. So why retain a feature few people use?

We should also consider how web use is changing. We know the browser is a separate application but it’s likely to evolve as operating system vendors attempt a more integrated approach. Icons, application tabs and pinned sites are just the start. The distinction between online and offline is already blurred and, within a few years, users won’t know or care where an application resides.

There’s also been a noticeable shift in internet marketing. While companies still promote their URL on advertising media, many now publish more memorable search keywords for Google or Facebook.

Finally, there are commercial incentives. Without the bar, users must resort to a search engine; they’ll aways see a page of results and revenue-paying adverts before reaching their destination.

But what about the drawbacks? If you can’t see the address bar, it’s more effort to enter a URL. If users really don’t want the bar, it can usually be hidden or they can switch to full-screen mode (F11 in most browsers).

Web developers also depend on the URL — especially when testing web applications or REST services. Removing the bar will make our lives more difficult.

Finally, without the address bar, it’s more difficult to ensure you’re on the correct site or check security settings. Those involved in phishing scams will be eagerly anticipating the UI change.

The idea makes me uncomfortable. Users may not understand URLs, but removing the bar won’t help them learn. I’m sure many car drivers don’t understand hydraulics but that’s not a reason to remove their brakes (OK — bad metaphor, but a web without URLs is not without danger).

I’m all for UI simplification, but this seems like a step too far. If it happens, Google should rename their browser: “Chrome-less” would be more apt.

What do you think? Should the address bar go? Could it be an option? Are the risks too great?

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.

518-chrome-11

Chrome 12 was released last week. You didn’t notice? Few people did. I hadn’t intended writing this article but a few people on Twitter convinced me otherwise (thanks @Mahen23). To start, let’s take a look at the usual list of improvements:

  • hardware-accelerated 3D CSS
  • the ability to analyze and delete Flash cookies within Chrome
  • a new safe browsing feature which protects against malicious file downloads
  • improved synchronization of browser settings
  • better screen reader support
  • new PDF save and print buttons
  • launch installed apps from the Omni-bar
  • 14 security holes plugged.

Chrome 12 also marks the end of an era: Gears has gone. Google Gears was launched in 2007 but development was abandoned a year later. The plug-in provided local data storage, JavaScript threading, desktop integration and geo-location but these have been superseded by standard HTML5 technologies.

Built-in JavaScript De-obfuscation

Merging and minifying JavaScript files has several benefits:

  1. Files, sizes and download times are reduced.
  2. Code processing speed can be improved.
  3. It hides your cutting-edge scripts from prying eyes.

Unfortunately, a minified script is impossible to debug. The code is an indecipherable mess contained on few lines which cannot have breakpoints set. Here’s an example from Google Analytics:

Chrome script debugger

Nasty. However, a quick right-click option will de-obfuscate the script into lovely readable source code:

Chrome script de-obfuscation

Built-in de-obfuscation is incredibly useful, although there are a couple of hitches:

  1. JavaScript minifiers often replace long function and variable names with shorter alternatives, e.g. MyLongFunctionName() becomes A(). De-obfuscation can never bring back the original names although you should be able to recognize patterns within your own code.
  2. Setting breakpoints on de-obfuscated code is more limited. Functions run in response to an event or timer can be analyzed. However, it’s not possible to break at code run when the page is loaded since the script has not been de-obfuscated at that point. Let’s hope the Chrome team address the issue in a future version.

For me, this is the most exciting development in Chrome. It may tempt you away from Firebug or Dragonfly when testing live code.

Have you discovered any great new features in Chrome 12?

Google has released a stable version of Chrome 11.

The new version, released Wednesday, brings bug fixes as well as some fascinating translation and speech-to-text features, GPU-accelerated 3D CSS and a simplified new icon.

Users can download Chrome 11 at the official Chrome page.

With the speech-to-text support, users will be able to click an icon and speak into the computer’s microphone, and Chrome 11 will transcribe the speech into text. Developers can add this feature to their website or web app.

This magic is made possible by the HTML5 speech input API, which you can also see in action at HTML5Rocks.com. Another nerdy implementation of the same feature can be seen in this Captain Kirk Bot.

Google Translate in Chrome 11 takes great advantage of the API, giving users the ability to translate spoken words into another language; users can both read and listen to translated speech.

As far as bug-squashing goes, Google shelled out a record $16,500 to individual developers who pitched in on taking the release from a beta to a stable version. The company paid between $500 and $3,000 for patching such vulnerabilities as corrupt node trees with mutation events and dangling pointers.

Google also gave special thanks to Apple Product Security team members miaubiz, kuzzcc, Sławomir Błażek, Drew Yao and Braden Thomas who helped take the browser to a less buggy stable release.

More About: chrome, chrome 11, Google

For more Dev & Design coverage:

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:




Google and Mozilla have both announced new browser initiatives that will allow users to opt out of having their activities tracked by online advertisers. These developments are at least partially in response to the “Do Not Track” lists proposed by the U.S. Federal Trade Commission.

In December, the FTC released a 122-page report [PDF] outlining the concept, which has been called a “Do Not Call” list for online behavioral advertising. Rather than make calls for legislation, the FTC has pushed for browser makers and advertisers to self-regulate.

Although targeting the same problem, Mozilla and Google are are approaching opt-out online behavioral advertising from different directions.


Firefox: Do Not Track HTTP Header


On Sunday, Mozilla formally announced its plans to build a do-not-track feature into future versions of Firefox. Alex Fowler, the global privacy and public policy leader at Mozilla, explained the proposed feature on his blog:

“When the feature is enabled and users turn it on, web sites will be told by Firefox that a user would like to opt out of OBA. We believe the header-based approach has the potential to be better for the web in the long run because it is a clearer and more universal opt-out mechanism than cookies or blacklists.”

Mozilla’s Sid Stamm has written his thoughts on the proposal and he explains why the HTTP header approach was chosen fro Firefox:

“Currently, to opt out of online behavioral advertisements, you have to get a site to set an opt-out cookie so they won’t track you. There are various web sites that help out (NAI, IAB UK) and there are Firefox add-ons (TACO, beef taco, etc.) that can streamline this process. But this is a bit of a hack; it’s nearly impossible to maintain a list of all the sites whose tracking people may want to opt-out from. It would be more attractive if there was one universal opt-out signal that would tell all sites you want to opt out.”

Instead, Stamm proposes the use of a HTTP header that is transmitted with every HTTP request and that lets ad networks know a user does not want to bee tracked.

This approach of using a Do-Not-Track HTTP header differs from some other opt-out online behavioral advertising solutions, which utilize either opt-out cookies or an opt-out registry. Michael Hanson from Mozilla Labs has posted a technical analysis of Mozilla’s proposal on his blog.

One advantage of using a header and not a cookie to carry opt-out information is that even if user clears his or her browser cache, the opt-out settings will still remain in place.

As The Wall Street Journal points out, however, for Mozilla’s tool to work, “tracking companies would need to agree to not monitor users who enable the do-not-track feature.” As of this writing, no companies have publicly agreed to participate. Mozilla will have to convince advertisers to comply with its header proposal for this idea to actually gain traction.


The Google Approach


Meanwhile, Google has released a new extension for Google Chrome called Keep My Opt-Outs. The Google Code page for Keep My Opt-Outs describes the extension as a way to “permanently [opt] your browser out of online ad personalization via cookies.”

The extension works with Google-served ads as well as with ads from companies that have signed up with AboutAds.info.


Other Initiatives


Last month, Microsoft announced that IE 9 will include a way for users to create lists of sites or companies that are blocked from tracking their data. This is significant because of reports that Microsoft previously removed similar features from Internet Explorer 8 at the behest of online advertisers.

The features and plugins proposed by Google, Mozilla, Microsoft and others are a good start in making it easier for users to opt-out of online behavioral ads; however, these solutions will only work if advertisers and browser makers can work together in a cohesive way.

Photo courtesy of swanksalot

More About: advertising, Browsers, chrome, do not track list, Firefox, FTC, Google, IE9, microsoft, mozilla, privacy, trending




The Social Analyst is a column by Mashable Co-Editor Ben Parr, where he digs into social media trends and how they are affecting companies in the space.

Google is preparing for war with Apple and Microsoft over the future of web video, and the rest of will be caught in the crossfire.

Earlier this week, Google quietly announced that it would be phasing out Chrome support for H.264, the video codec and standard supported by Adobe Flash, Blu-ray, Internet Explorer, Safari and others. Instead, it will be supporting WebM and Ogg Theora, which are supported by Mozilla and Opera.

What Google hoped would be a small footnote turned into a tidal wave of criticism. Google was chastised for turning its back on “open innovation” by dropping a more widely used codec for a lesser-used one. Compounded by the fact that Google is a strong supporter of Adobe and Flash, and it’s easy to see why the firestorm started in the first place.


Why Is Google Against H.264?


After several days of being slammed in the media, Google finally responded and wrote the post it should have written in the first place.

First, Google’s Mike Jazayeri clarified that Google Chrome would only stop supporting H.264 in HTML5, not in Flash or other forms of media. Then he dove into the problem surrounding the HTML5 <video> tag:

“As it stands, the organizations involved in defining the HTML video standard are at an impasse. There is no agreement on which video codec should be the baseline standard. Firefox and Opera support the open WebM and Ogg Theora codecs and will not support H.264 due to its licensing requirements; Safari and IE9 support H.264. With this status quo, all publishers and developers using the <video> tag will be forced to support multiple formats.”

Google has come to the conclusion that there will never be agreement on H.264, since it is proprietary technology owned by MPEG LA, a firm that forms and licenses patent pools. Thus the search giant decided to draw a line in the sand and double down on the WebM. WebM, for those of you who may not remember, is the open codec/standard for web video created by Google.

Unlike H.264, WebM/VP8′s patents have been released royalty-free. Apple and Microsoft are part of H.264′s patent pool, as are companies like Sony, Sharp, Cisco, LG Electronics, Hp, Toshiba and Dolby. Absent from the list: Mozilla and Google.

The tech titan also addressed the criticism that it should have selected H.264 as its baseline codec because of its wider adoption:

“To use and distribute H.264, browser and OS vendors, hardware manufacturers, and publishers who charge for content must pay significant royalties—with no guarantee the fees won’t increase in the future. To companies like Google, the license fees may not be material, but to the next great video startup and those in emerging markets these fees stifle innovation.”

Google also argued in its response that a community development process is superior to one where multiple parties have incentives to collect patent royalties.


Neither Side Will Budge


While Google may not have intended to start a war, it has essentially drawn the battle lines and made it clear that there will be no compromise. On the one hand, you have Google, Opera, Mozilla and and its WebM allies, which include WinAmp, Skype, AMD, Broadcom, Qualcomm, Logitech and Nvidia. On the other hand, you have the participants of the H.264 patent pool. There isn’t a single company that is part of both WebM and H.264.

The final paragraph of Google’s response may be the most telling thing in this whole affair, though:

“Bottom line, we are at an impasse in the evolution of HTML video. Having no baseline codec in the HTML specification is far from ideal. This is why we’re joining others in the community to invest in WebM and encouraging every browser vendor to adopt it for the emerging HTML video platform (the WebM Project team will soon release plugins that enable WebM support in Safari and IE9 via the HTML standard <video> tag). Our choice was to make a decision today and invest in open technology to move the platform forward, or to accept the status quo of a fragmented platform where the pace of innovation may be clouded by the interests of those collecting royalties. Seen in this light, we are choosing to bet on the open web and are confident this decision will spur innovation that benefits users and the industry.”

Google says that it hopes that the other browsers will adopt WebM, but it’s clear they already know that won’t happen. Why else would Google build Safari and IE 9 plugins to add WebM support into those browsers?

The inability for both sides to compromise will almost certainly stifle the growth of innovation surrounding HTML5 video. Why would anybody invest time and money into a technology that will only work in some browsers, when Flash is guaranteed to work in all browsers (except Mobile Safari)?

Unless both sides find a way to compromise, the future of web video will continue to be in Adobe’s hands. We doubt either side is going to budge anytime soon. The citizens of the web will end up being the losers of this affair.

More About: chrome, Google, google chrome, h.264, HTML 5, HTML5, Opinion, trending, video, webm, youtube




Modern commercials are a funny thing. Ad agencies are trying to figure out how to give their commercials viral appeal while balancing that against providing company or product information.

The Old Spice guy campaign is more or less universally lauded as an example of how to reboot a company’s image and turn a commercial idea into a viral phenomenon. However, the videos, and the campaign’s subsequent expansion onto Twitter, focused on entertainment and branding: We knew what kind of brand Old Spice wanted to be and we laughed at (most of) Isaiah Mustafa’s quips and sound bites. Old Spice prioritized those elements instead of explaining how its products work or even what they smell like.

That worked for Old Spice, because most people can intuit how deodorant works; the company wasn’t reinventing the wheel, it was reinventing its brand. Old Spice didn’t need to provide detailed product information to make its campaign successful. Other companies, like Google, don’t have the same luxury.

Google’s products are often less obvious to an everyday audience. Google Goggles? Do you wear them? Are they on your phone? Is that some Mountain View euphemism? Google has the task of both explaining its new tech and providing a viral kick with its ads — that’s not easy.

Despite the odds, the technology giant has done a great job with its latest round of marketing campaigns by applying some basic principles in some very creative ways. Read on for a look at what Google did right and what it could do better.


Make Your Audience Feel Smart


Google was not a company known for its commercials until last year’s Super Bowl “search” ad that simultaneously showed the narrative arc of a budding romance and the vast possibilities of Google’s core product all via search terms.

The next batch of commercials focused on the speed of its web browser, Chrome, by testing how quickly Chrome could load a page compared to a (usually silly or messy) science experiment, such as lightning hitting a small boat or shooting a potato gun.

The real gem is Google’s latest commercial focusing on Google Instant. While Google still handily owns the market for searches, it’s had real competition from Microsoft’s Bing. Google Instant was supposed to tip the balance by letting you see search results as you were typing — no need to hit enter. The idea is technologically complicated but not especially awe inspiring. To make the technology more compelling, Google put together a commercial that emphasized how much we take that instant gratification for granted while at the same time making the viewer feel smart (the video is embedded above).

People enjoy repeating patterns, and people especially like being able to recognize those patterns and start to solve them. “Humpty… ” “Fee Fi Fo… ” Both phrases are relatively simple to complete, but there is a certain satisfaction in being able to figure it out, as simple as it may be. That’s the concept behind the commercial: display a list of recognizable phrases that the viewer can piece quickly together as they watch. It’s an incredibly smart way of congratulating the user and making the case for Google Instant at the same time.

As smart as the commercial may be, it doesn’t have the same entertainment value or viral capacity as some of Google’s sillier campaigns. With just 22,000 views, the ads seem to be geared toward selling their product rather than making it onto sites like Reddit.


Make Your Audience Care


It is difficult to get people to care about things on the Internet. The Internet has a lot of things on it, and the average attention span is depressingly low. Still, there is one way to guarantee that people sit up and take notice: Put them in it. People are more likely to care about a product if they had some hand in its development.

The Google Demo Slam challenges users to come up with the funniest or most creative videos showcasing new Google technologies, like the aforementioned Google Goggles. It’s a total win for Google, combining the viral ridiculousness of dressing up like Mount Rushmore to test Google Maps, for example, while getting real people to demonstrate how the technology actually works.

There are some celebrity spots (like Maria Sharapova pelting helmeted reps with tennis balls), but many of the submissions come from independent groups (like Sony) or everyday people trying to out-do their competition. Submissions are screened by Google and then randomly paired against one another where users can vote for their favorites.

The interactive, gaming element helps draw the viewer into the competition (champions will be eventually named). The format of the Slam also necessitates that people spend some time on the site, creating stickiness and enticing people to watch more and more videos (each of which essentially is a commercial for a Google product). Google has started using the some of the videos as stand alone commercials, but they don’t necessarily play as well without the competitive aspect of the site.

By putting its faith into users to come up with interesting and genuinely funny videos, however, Google is potentially assuming some risk. It’s also hard to believe that all the videos submitted to the campaign are purely user-generated given their professional look and feel.


Offer Value in More Creative Ways


chrome experiment image

It may not take a lot of creativity to make your RSS feed into a mobile app; it does take some thinking to quietly sponsor an experimental music video that, by virtue of its complexity, showcases the power of Chrome. “The Wilderness Downtown” is a Chrome experiment directed by Chris Milk. It uses Arcade Fire’s song “We Used to Wait” and interactive HTML5 elements, synchronized pop-up windows and a spinning panorama using Google Maps. Not only was it an inventive and touching video, but it was an excellent showcase of Google products.

Of course, the video can be run on other browsers, and if you don’t like Arcade Fire or the chosen song, the video experiment won’t have much of an impact. So clearly not every view was a win for Google.

Google is far from a perfect company, but its marketing campaigns are often an intelligent blend of viral goodness and thoughtful product showcases. What do you think? What commercials or campaigns did you find memorable or smart? What were some of Googles major flops or missteps? Let us know below.


More Marketing Resources from Mashable:


HOW TO: Create a World-Class Online Community for Your Business
How Social Data & Mobile Tech Can Improve the Retail Experience
Making Data Relevant: The New Metrics for Social Marketing
Why Marketing Threatens the True Promise of Social Media
Why Every Brand Needs an Open API for Developers

More About: ad, business, case study, chrome, Commercial, demo slam, Google, MARKETING