Imagine surfing the Web, and being able to look at what other Web sites or other visitors wrote about the site you’re visiting.
For example, someone might be viewing a manufacturer’s web page relating to a product they are interested in purchasing.
A past effort at Web annotation was the Third Voice browser plug-in, which let people post public notes about a site that could be seen by other Third voice viewers. Many of those ended up being spammy and/or inappropriate. Click on the image below for a larger version of Google’s potential approach.
This invention would let people receive summaries of blog posts linking to the Web site being visited. Those people could also perform a Web search or blog search through a search engine requesting documents relevant to the site.
A thoughful and intelligent article from Shannon Watters at Digital Web this week, How to Choose an eCommerce Package, offers some great suggestions on what to look for when choosing software for an online shop offering goods or services or both.
Shannon writes about what she calls the “top eleven things to consider when choosing an eCommerce package,” and it’s difficult to argue with her selections, but I was hoping for an even dozen things to consider – with an addition of how the ecommerce software might interact with search engines.
Of course, an ecommerce system should be easy to use for both shopper and site owner. Updating the software, and adding functionality from third party toolmakers should be a breeze. The software should be able to scale with growth, and it should be easy to use with an analytics package, so that you can measure your traffic and see how visitors use the site.
Shannon’s suggestions regarding promotions and discounts and the ability to offer customer service are spot on. Security is essential, and an intuitive checkout process will be a major determinent as to whether visitors become customers. Her opinions on open source options, and on the community and company behind a software are filled with thoughtful suggestions.
The official Google Blog announced today that Google had acquired Zenter, Inc., manufacturer of software which provides online presentation tools.
The circumstances around the development of the presentation software is described in My Innoview with Wayne Crosby.
The company appears to have started out under the name Click ‘N Slide, Inc., and may have changed names after Evan Williams spoke at a presentation held by Zenter funder Y Combinator and offered them use of the domain name – see: Zenter -> Goog
There appear to be at least two provisional patent applications applied for by the company, including one on WebFonts which allows presentation authors to create, edit, and rotate text in an intuitive manner. A search through the US Patent and Trademark office doesn’t reveal these patents, which probably aren’t published yet.
Not every algorithm or process that a search engine comes up with is aimed at providing more relevant search results, or advertisements to go with those results. Some just focus upon getting results back to searchers quickly.
A nice paper from Yahoo research looks a some of the technical aspects behind how a search engine works, and some of the implications of those approaches: The Impact of Caching on Search Engines
Google’s sitelink feature provides searchers with shortcuts to in-domain pages “based on user behavior and possibly site structure” to bring searchers one step closer to a destination that they may be aiming at in a search that they had performed.
A study and paper from Microsoft explores a similar type of teleportation, but to a destination that may be a few steps further along a search query path, based upon “exploiting a combination of past searching and browsing user behavior” using information gathered through browser plugins and proxy server logs.
The paper is one being presented at SIGIR’07 in July at Amsterdam, Studying the Use of Popular Destinations to Enhance Web Search Interaction (pdf). There, they tell us:
In this paper, we present a user study of a technique that exploits the searching and browsing behavior of many users to suggest popular Web pages, referred to as destinations henceforth, in addition to the regular search results.
Advertisers and searchers can benefit when a search engine collects geographic information from the Web and indexes it by associating that geographic information with a system of overlapping and adjacent geographic boundaries for the locations.
It can mean using considerably less geographically related keywords to bid upon, and on smarter geographically related search results.
In a post from May, Geo Targeted Advertising for Google Maps and Google Earth, I described how Google may use, or anticipate using, an approach like this. Looks like Yahoo is considering it, too.
A trio of patent applications from Yahoo were published last week which describe some of the strategies and algorithms that Yahoo may use to gather and organize this kind of data, and use it in search results and advertising.
Ask.com’s patent application published this week, Methods and systems for generating query and result-based relevance indexes, notes that information from users’ queries and interactions with search results is limited in use to provide more relevant results to searchers.
That kind of information includes time of searches, the geographic sources of queries, and demographic variables about searchers. This information might be useful to businesses willing to pay for it.
Should a search engine offer that kind of information to businesses?
How might a search engine approve or reject ads automatically, without human review, on the basis that the ads are annoying or displeasing in some way?
Without considering the very large volume of ads that get presented to Google everyday, you might think that they would manually review every ad that advertisers present for publication, which would take a lot of people. While ads should attract some attention, they shouldn’t be annoying or offensive. There are a number of standards set from Google for image ads, video ads, and for text ads.
A patent application from Google goes into a good amount of depth on how it might take a programmatic approach to identifying ads, and Web pages that are “annoying.”
The patent filing describes some of the methods used when reviewing images and text and audio, with tools like Optical Character Recognition and pattern matching against large databases of images and sounds. It also details how Flash and animated images might be reviewed, but is silent on what it is looking at when it refers to things like a “Trust Score.”