New Google Search Features

Google has introduced some new search features as experimental through Google Labs. They've added

• right- and left-handed search navigation
• keyboard shortcuts for search results
• keyword suggestions
• alternate views for search results



I like the left-handed search navigation and was playing with the layout and widgets a bit myself last summer when playing with Google's Web Toolkit. eBay has been doing some very similar UI work in its eBay Playground site that I've enjoyed. Amazon tends to overwhelm me at times with JSON this and AJAX that and they can't seem to resist the urge to package the search results and product descriptions to the extreme. I guess this shouldn't surprise me so much as they are a self-billed department store. Netflix, on the other hand, strikes the right balance with me through their consistent and concise detail drill-down through the AJAX essentials, XMLHttpRequest object and javascript onmouseover() event.

I don't particularly care about keyboard shortcuts and search results. This is a personal inconsistency however as I don't use keyboard shortcuts in Gmail either but almost always use the keyboard to navigate between applications, tabs and the OS in general. Maybe this is my unverbalized position that I just don't like the way Google implemented keyboard shortcuts. Maybe I'm just inconsistent after all.

The keyword suggestions have been available as a Google Labs offering called Google Suggest for a while and the search bar in Firefox provides JSON-enabled search term suggestions.

The alternative search results are a great move forward with regard to search result presentation, specifically addressing the need for better contextual-based and grouped/ordered search results. I've written about this previously and was eager for new search primitives to address this perceived shortcoming or at minimum search options that accomplished the same thing.

At least I'm not alone in liking the latest search presentation options. Ars Technica described it simply as "awesome".

Google Trends - Part 1

Huh...loads of news regarding Google Trends' new functionality: regularly updated info listing the most popular search trends/terms. In one manner, this is obvious - Google has the search data available to them at any given point in time through, presumably, numerous search entry points or via a search funnel. It also makes perfect sense that this functionality, the ability to see what your peers or organization or group is searching for, would be pushed down to the Google Search Appliance at some point.

That's really the interesting addition in my opinion: by localizing search trends, through the GSA or adding a new search primitive, such as fromDomain:, fromIpAddress: or fromNetBlock: (in the same way that the site: or url: search primitives are available now), search/trend consumers would have the ability, from a social networking perspective, to learn what is being searched for, relative to themselves in their own localized search bubble.

Let's consider an example:

Say you're a faculty member at a university and you're vaguely aware of a new collective bargaining vote that's upcoming and you want to know more. You search Google and find press releases and maybe a few local area newspaper items that have been indexed. This is the way in which we normally search or engage in absolute searches, modulo intentional deletions or censoring of search results, etc. Say you're a Google Master and break-out your Google-fu and attempt a more granular search through the selective use of one or two Google search primitives, for example, site:GoogleU.edu. Great! You weren't interested in what was being reported through channels on the other side of the country anyway.

Is this enough? Are the results sufficient in order for a search consumer to best utilize the data provided? Maybe, but it could more than likely be better by taking into account the context of the search results.

Now, say that advanced, localized trending were available to the faculty member. S/he would be able to search, relative to the university community or a specific department or a specific demographic, and determine what his/her peers are searching for, in the hope of learning about key issues being addressed at the collective bargaining vote. Add in the capabilities for search trends and results to be presented in such a way that they can be ordered or grouped and the faculty member may find him/herself better in the know and ready to cast an educated vote.

So...search, trends and localization step up as the latest social networking platform.

I wonder what I spend most of my time searching for and if anyone else cares?