joeandrieu.com

Evan Schuman writes about a new standard from the Association for Retail Technical Standards (ARTS) that will help online retailers interoperate on the semantic web, making it much easier for Search engines to return results that include accurate and timely information on current retail offerings. ARTS is a part of the National Retail Federation (NRF).

Most of the big players had some role, from Yahoo!, MSN, and AOL to Target, Penney’s, and REI.

It’s an early step, but definitely one in the right direction. Schuman mentions that Circuit City and AOL have been in trial for three weeks with the data formats, which use XML to communicate with Search engines. For example, one format allows retailers to submit updates about product information including product name, price, URL, image, description, color, size, in-stock status, and shipping fees.

When the bugs are worked out, this (or an alternative) will prove a critical component in any Vendor Relationship Management system.

Good work. I’m looking forward to seeing how it evolves.

It’s already being discussed on the microformats list, which could extend the reach of the standard beyond big retailers with well-honed IT departments, allowing smaller operators to use standard semantic XHTML to present the same data. Expect that process to take some time, but it should definitely be helped by the prior work done by ARTS.

Apparently, my problem with William Gibson and Prof. William Gibson at Amazon isn’t so much that they have the wrong William Gibson, but rather that Amazon has no mechanism for disambiguating authors.

When you click on an author’s link to find more books by that author, it simply takes you to a search based on the author’s name. Other authors with similar names also show up. So not only do other William Gibsons show up, so does “Henry William Gibson” and “James William Gibson” and “William Oliver Gibson” and so on.

Contrast this with Wikipedia’s results for William Gibson. Not only do they focus on the specific name, they let you choose which William Gibson you are looking for.

I do give them credit for prompt responses to my emails to customer service, complete with nice RSVP links that made it easy for me to get closure.

But with regards to the authors, couldn’t we do a little better, Amazon?

I’ve panned the AttentionTrust quite a bit in my first few posts, but I must say it is really because they are so close to spot on. Powerfully close. Now if we can just jump from there to something we can actually use…

Squirrel Tao writes a bit about attention from a creativity standpoint. That’s attention with a little “a”. It’s nice, though, how the thinking applies equally well to Attention, with a big “A”.

The post finishes with:

Willian James wrote, “If we wish to keep our attention upon one and the same object, we must seek constantly to find out something new about it.”

An absolute brilliant statement that subtly points out one of the key flaws of the GestureBank. What is the object of Attention in the GestureBank? Clickstream logs like the AttentionTrust Extension capture all activity and mish mosh it into a goulash, in the hope that after the fact, one can extract or identify the object of attention. The GestureBank then takes that data and makes an even bigger goulash. But as Chris Anderson writes in The Long Tail, top ten lists are useless without context. It is only in the niches that we can get value out of knowing the most common similar results. Clustering is one way to mathematically generate niches, but its use will prove limited to spaces where clean mathematical separation exists. So why does the GestureBank systematically strip the user context from the already context-free Attention log? That makes it pretty hard to discover the object of the user’s Attention.

Why not just let the user tell us?

Then we can avoid the whole post-activity reconstruction/clustering/meta-modelling thing.

Amazon surprised me with a recommendation today:

The Original And Institution of Civil Government, Discuss’d (Ams Studies in the Eighteenth Century) (Hardcover)
by Benjamin Hoadly (Author), William Gibson (Author)

This is probably because I’m a William Gibson fan.

However, I’m pretty sure William Gibson of Neuromancer and Idoru is NOT the William Gibson who co-wrote this book.

I sent an email to customer service. We’ll see how they respond. Thanks also to Google for finding professor William Gibson’s home page.

I discovered an interesting innovation in economic theory today, called Nomad Economics.

I’ve always felt that the market acting as an aggregate of individuals–the “invisible hand” setting prices–is pretty amazing. It can be a hyper-efficient tool for simultaneously setting value and assigning resources for maximal economic utility. But there are also some huge flaws. Not only are there several well-known market failures–such as monopolies and the management of public goods–but even the underlying assumption of “rational behavior” is clearly false. We are all guilty of acting irrationally from time to time when it comes to money.

One of the basic ideas behind Nomad Economics is that between the “invisible hand” of neoclassical economics and the “large entity dynamics” of socialism and marxism lies a range of “renegade” economic actors that explain a lot of relevant interactions.

Abe Burmeister writes that

Renegade economics objects are a class of entities that lies between these two poles. They are larger than individuals, yet smaller than governments (or at least the larger ones) and far more concrete than “society”. Corporations are renegade economic objects. So are social networks. Markets themselves (but not “the market”) are renegade economic objects. So is money itself. Non profit corporations are renegade economic objects too, as are open source software projects. Terrorists groups are often renegade economic objects, and perhaps the “military industrial complex” is too. Closer to home families are renegade economic objects and cities, neighborhoods and other urban concentrations are too. There is a whole world of these things, and it has just begun to be explored.

Fascinating stuff. In his book, Abe goes into much more detail about nomad economics –it is much more than just about renegade objects–and how it fits into/works with neo-classical market theory.

The fascinating thing is how Abe’s work could be so hyper relevant to figuring out the next phase of Search.

The net is not only accelerating and expanding the scope of classic market dynamics–making the invisible hand that much more adept–it is also giving rise to activities that don’t fit into classic economic theory, like the “gift economy” of open source software. Pay-Per-Click advertising has radically transformed most online–and many offline–business models, fundamentally shifting how we think about marketing and customer acquisition. It not only sped up our ability to perceive the Attention economy, it tells us immediately how Attention converts into sales and profit when we track our PPC into our shopping carts.

Yet, when Bill Gross first launched PPC with GoTo.com, it was practically dismissed out of hand. It was crass to let money decide the best Search result. The purity of Google’s page rank seemed most holy and just in comparison. But the genius of PPC is that it is a real-time market. It is the invisible hand, using market dynamics and the magic of market pricing to identify the most likely best result for users’ Attention. It’s brilliance was pure Adam Smith.

And yet, there is still something to be said for those PageRank results. Anecdotally, I find myself using the “organic” results from Google far more often than I do the paid ones. Sure, when my intent is to purchase, I’ll often use the sponsored links. I know those folks are offering me products or services. But much of the time, my attention wants to be spent on non-market transactions. I’m exploring or learning or just hoping to be entertained. In those cases, my highest estimated Return-On-Attention will be from websites that have nothing to sell in the classic sense and hence will never be part of the PPC marketplace.

That’s pretty interesting. While we have the worlds largest, fastest real-time market for converting attention into revenue, a huge portion–perhaps the majority–of that attention seeks out non-market resolution. Seems to me that there’s more going on here than the invisible hand.

Check out Abe’s work. If he can give us a few new ways to think about how to make the most of our attention, perhaps we might just be able to think up some new ways to improve Search.

I’ll have more to say once I’ve digested his book.

Brian Smith kicks off a series over at Search Engine Watch looking at how Ask, Yahoo!, Google, and Microsoft are providing bonus, focused, “added-value” search results.

First up: Ask’s Smart Answers. Basically, Ask uses a handful of trigger words and specific terms to provide hand-crafted Added-Value results in addition to the organic, “normal” results we are used to.

For instance a query at Ask for president of tunisia returns with the following listed at the top of the page:

The Chief of State of Tunisia is President Zine El Abidine Ben Ali, and the Head of State is Prime Minister Mohamed Ghannouchi

World Factbook | Encyclopedia | BBC Profile | US Government Travel Info|
Maps

Smart.

Rather than rely on the underlying search algorithm to simply find the right URLs and display a summary of each, Ask is using human engineering to provide likely answers to the implied questions.

According to Ask.com’s Director of Online Information Resources Gary Price,

“We look at how people are searching—the keywords, the click through pattern—to figure out good Smart Answers. People out on the road suggest them all the time. Smart Answers is run by the product team. They make the necessary content deals to get things up. Plus an editorial team gets involved when necessary for special events like hurricanes or the San Francisco earthquake anniversary. There’s an entire team of engineers who decide when a Smart Answer will be triggered and how it will be triggered—what shows up and how it shows up.”

This blurs the line between “search” and “editorial” content, but then About.com has been doing that for years. The architectural innovation–for me–is that while Ask continues to serve the long tail of search organically, they are investing human resources to handcrafting results to the most common and most obvious added value search queries. This doesn’t necessarily scale well, but it does cherry pick the high value opportunities, and as long as the profit margins on the click-through for these bonus sections is enough to pay for the team of engineers, it should work.

Because the Smart Answers are sometimes defined by trigger words independent of the subject terms, such as “map” or “definition” or “market cap”, Ask is providing vertical specialization seamlessly integrated with organic results. Of course, Google has been doing this for a while, with “define:” and “recipe”, so I’m looking forward to Brian’s run down on the rest of the Search Engines’ approaches to this type of added-value results.

This whole phenomenon demonstrates how far we still have to go before we seamlessly transform user intent into simple, complete answers.

I recommend reading the full article.