Google enhances Gmail with IMAP support

Search giant Google has started to add internet message access protocol(IMAP) support to Gmail, allowing users to access their email from client applications such as Outlook and Thunderbird while simultaneously updating the status of messages in Gmail's main web interface.

Google has long supported client access to Gmail via the POP protocol, but this method has significant drawbacks.

Messages that have been read using client software are still marked as unread when email is subsequently accessed using a web browser. Similarly any replies or messages forwarded using the client do not appear in the sent mailbox on the web. IMAP support will resolve these issues.

Google has published a list of supported clients with instructions for enabling IMAP in each case. The list includes the iPhone's email application but mentions no other mobile devices and explicitly rules out Palm handhelds.

AdSense improvements

Google has clearly been busy of late. In addition to the Gmail changes, the company has added new features for altering the appearance of AdSense advertisements and adopted a new engine for its Google Translate service.

Adsense users, whose websites display Google ads, will soon be able to manage their ads from within their AdSense account. This means that ad unit settings such as colours and channels will be saved in the AdSense account each time code is generated. So if a user needs to change any of these settings in the future, all they do is make the update within the account, meaning it is no longer necessary to manually replace the ad code on all web pages that carry ads.

According to Google, this new feature will be introduced over the next few weeks. Google has more information in its Inside AdSense blog.

Google Translate

Alongside its other improvements Google has also announced that Google Translate is now using the company's own translation engine, which it has been testing since April.

The new engine employs a completely new technique, according to Google research scientist Franz Och.

"Most state-of-the-art commercial machine translation systems in use today have been developed using a rules-based approach and require a lot of work by linguists to define vocabularies and grammars,'he said.

"Several research systems, including ours, take a different approach: we feed the computer with billions of words of text, both monolingual text in the target language, and aligned text consisting of examples of human translations between the languages. We then apply statistical learning techniques to build a translation model."