Google decided to concentrate on this aspect of the Web. They aimed to improve the speed of downloading images and reduce the size of image files without significantly affecting the image quality.

It has been estimated that photos and images take up about 65% of the space on web pages, thus slowing down the transfer of information considerably, especially when bandwidth is limited, as in the case of mobile networks.

Engineers at Google worked on ways to compress the JPEG or other lossy images, as these constitute the vast majority of images. The product of their research was introduced this week, in the form of an all new format for images – WebP

Google recently released a developer preview of the new WebP image format, which has been found to reduce the byte size occupied by images and also helps to load them faster.

WebP is a method of compressing photo images. During a study of 900,000 images, WebP was found to compress the images by 39.8%, compared to 9.7-14.6% for JPEG images.

The WebP file consists of VP8 image data and a container based on RIFF. WebP works on the principle of predictive coding to encode a message. Predictive coding only encodes the difference (residual) between the actual values and the predicted values of a block, thus compressing the image more effectively. Google expects that once the developer community starts working with WebP they will be able to produce even better results and further compress the images.

At present the WebP format is not supported on any image viewer or browser. The success of this initiative depends heavily on whether or not browsers will support the format and web designers will be happy to use this new format in upcoming web pages.

Now, when a user clicks on an image, the top bar, which used to look very similar to the Google Image Search preview bar, will instead show a sleek interface on a grey background.

The new preview bar provides a much larger preview of the original image, in addition to smaller thumbnails of 7 other image search results and 5 thumbnails linking to other popular related searches, as shown below.

The old image search page only provided a thumbnail-sized image, accompanied by the name and details regarding the size of the image, in the preview pane.

A search box with search assistance is included in the new preview page, to enable users to quickly search for more images without having to go back to the search results page.

While the addition of the search box is the same as on Google’s updated image search, the image type and colour recognition functionality now provided by Google is not present on Yahoo! Image Search. The latter, however, has a nifty “Recent Images” feature that shows up on certain search terms, and does not include ads with image search results.

These changes certainly make the preview page look good and will likely prompt more searches via the presence of the search box on the preview page, it might, once again, not be quite enough for Yahoo! to score over Google in anything other than SERP aesthetics.

Where it comes to quality of image search results, when tested over a numbers of random, widely varying search queries, Yahoo! does seem to be slightly better than Google at bringing up relevant results. Though Google finds more in terms of quantity, many of Google’s image results are not relevant.

A software engineer at Google Image Search, Sean O’Malley, says “searching for the perfect image can be challenging if the search results match the meaning of your query, but aren’t in the style that’s useful to you.”

To begin with, they launched “face search”, which as the name implies allows one to find only those images related to the search query, which have faces in them. There are two options available; search with faces, and search without faces.

The next innovation was the introduction of photo search. If a user selects this option, the results will automatically delete all images containing cartoons, drawings etc. and focus only on images with photographic content. The reverse option is also available. i.e. users can opt to delete photographs, and select only cartoons and drawings instead.

The latest enhancement which Google has added, is the choice between clip art and line drawings. which allows users to decide beforehand, which category of results they wish to see.

To select any of these options, the user can click on the “Any Content” option in the blue title bar, of the search results page, or he can select “Content Types” on the Advanced Image Search Page.

These new facilities are very simple and practical ways to help users find what they are looking for without wasting their precious time and are sure to become very popular in a short while.

Digg Images

To keep in tandem with this new section, the category system of Digg has also been reconfigured. Now users can submit images, video and news into any category. However, it would be better to get back the subcategories bar instead of the click-drop-down-menu that we get to see now.

The new image section does look cool but the absence of an accompanying description might become an irritation in the future when users get misled by sensationalist headlines and taglines. Alternatively Digg could expand the image or pop up an overlay of the image when a visitor hovers over the thumbnail images. At the end of the day, though, a picture still remains worth a thousand words!

A search for Spiderman on Yahoo! shows regular image search results, followed by a differently coloured box on the bottom of the search results page, that says “Others who searched for spiderman also searched for:”. This can be seen in the screenshot below:

Just like Google, Yahoo! also stores user search query data and mines it to improve search results quality for their users. Personal search information on Yahoo! can be accessed via signing into a My Yahoo! account, where users can view their list of favourite searches.

Media attention has focussed mainly on Google and their potential exploitation of users’ private information for personalisation of search results. Features such as the suggested search explorer bar could draw some scrutiny towards Yahoo! as well. However, the sheer number of searches on Google outweighs that on Yahoo! enough to allow Yahoo! to take many more liberties with how they use a personals private search data.

Explorer bar is currently enabled for entertainment and location-related searches in the USA. Yahoo! plan to expand availability of this feature to further categories in the near future. This improvement was launched alongside the integration of Flickr results into regular Yahoo! Image Search results.

The article provides a checklist for Wikipedia image SEO, which is replicated here:

1. Do a survey to identify specific articles that lack quality images. Your aim should be to provide unique content where it did not exist before. Create a list of images that would each enhance at least one article.
2. Arrange the appropriate licensing and announce it on your client’s web site.
3. Upload images to Wikimedia Commons. Include a statement about licensing status and an outgoing link as verification. Categorize the images.
4. Over at Wikipedia, notify the appropriate WikiProject with a list that correlates available images to suggested articles and request assistance from volunteers. This indirect approach is diplomatic for conflict of interest situations.
5. Be low key and tasteful. Wikipedia is an encyclopedia.

According to Cary Bass of the Wikimedia Foundation, “The Wikimedia Foundation appreciates good quality, freely licensed images, and credit where credit is due is never an issue. Many of our images fall under an attribution license, which often takes the form of a web link back to the releaser’s web page.”