What’s new in Chrome 17

For a couple of weeks after updating from Chrome 16, there weren’t that many changes to the Canary Channel, except for the usual under-the-hood tweaks. Only later did new features started to roll in, so here are some of the more visible:

A redesigned ‘History’ page

Available for older versions of the browser by manually accessing the address chrome://history2/, the new ‘History’ page has become the default beginning with Chrome 17. Beside the visual changes, including a new style for check-boxes and more emphasis on the domain you visited, the new version makes it easier to remove pages from the browsing history. You can either hover over a result to show the check-boxes and select multiple items by holding down Shift or use the small context menu on the right to remove the single entry or search more visited links from that site.

New webUI task manager

The task manager also got a face lift and is being implemented as a web page, like many other dialogs in Chrome. Feature-wise it added a small context menu for each process, allowing users – and more importantly web developers – to launch the web inspector directly or to jump to the web page, making it a bit more like the original Windows version. As before – something I haven’t noticed until now – you can choose which columns to display in addition to the default ones, including the user profile, image, script and CSS cache, JavaScript memory and, as an Easter egg in the dev versions, the number of ‘Goats Teleported’ by the page. Curiously, this was the first change to appear in Chrome 17, but it has been rolled back in the mean time and the new task manager is only available as an internal page under chrome://tasks/.

Spelling suggestions from Google

In an effort to improve the spelling suggestions, Chrome now integrates an online service to offer suggestions, similar to the ones offered for search queries. Personally I found it underwhelming, I rarely get a good suggestion out of it and it adds a noticeable lag when I right-click a misspelled word.  Curiously, in the final Canary builds the feature is on by default and cannot be turned off from the settings, even though that was possible previously… Given the privacy implications, that doesn’t sound like a good idea.

Other tweaks and experimental features

• Page zooming has been improved: the range of supported zoom values has been widened to 20% – 500% and the zoom levels are now preset and more consistent with other browsers.
• The crash page has a new handy ‘Reload’ button dead center. Guess some people complained because they didn’t knew what to do next… Unfortunately the fact that I noticed this also means the browser was less stable than before, crashing constantly during routine operations, like pasting in Gmail or Google docs.
• If you enabled the ‘Print Preview’ experiment, the preview isn’t shown in a separate tab anymore, but in a modal dialog in the current tab. Which seems counter-intuitive given the description of the experiment and goes against the general tendency to make most dialog boxes available in new tabs, something I have always liked. Also, ‘Print with Google Cloud Print’ has been added as a new option in the ‘Destination’ drop-down.
• When you create a fresh Chrome profile, the browser automatically installs three apps, YouTube, Gmail and Search. Even if those are simple links to the websites and the idea is to help people discover apps, it feels a bit wrong to force them on new users, all of them being Google products…
• HTTP pipelining has been implemented – something that Firefox had for as long as I remember – again behind a flag that can be turned on from chrome://flags/.
• Another useful experiment is ‘Enable extension alerts’; once turned on it will show desktop notifications about extensions, for example when they are updated or crash.
• The internal page used for network diagnostics, chrome://net-internals/, has evolved quite a bit since I wrote about it; in Chrome 17 it gained a ‘Timeline’, a graphical representation of the network activity, including DNS requests and the amount of data sent and received.
• Speaking of tools for developers, another new internal page, chrome://profiler/, provides a profiling and debugging infrastructure for tasks; more technical details here.
• As reported on the blog Google Operating system, you can now change the user agent easily through the web inspector, instead of messing around with command-line flags.
• The support for more interactive online games is being constantly improved. Chrome now supports Fullscreen API for websites – just like recent Firefox builds – mouse pointer lock and the GamePad API, the latter two as command line flags. The full screen support is already active on YouTube with the HTML5 player. The browser also added a couple of new content settings in order to provide better control over the sites trying to access these new features.
• If you enabled the flag ‘syncing open tabs’ you can take advantage of another internal page, chrome://sessions/, to see the list of tabs opened on other computers and to reopen them in the current window. Another one, chrome://sync-internals, offers a wealth of data and debugging information about the sync process. On the other hand, the option to sync search engines, added in version 15, has been dropped, which causes errors if you try to sync different browsers versions, like I am doing.

P.S. Did you know that the code name for the Chrome sync engine is Nigori, most likely after a sake variety?

Update: And here is the official announcement about this release on the stable channel:

Just launched: today’s Stable release brings new speed & security features to #Chrome goo.gl/nOxQQ

— Google Chrome (@googlechrome) February 8, 2012

Another small update that has been added to Chrome 17 is the ability to make Gmail handle mailto: links natively, without any extension with the tip described here.