Windows Phone Resources
At last night’s meeting of the Learn Silverlight group I’m a part of, the topic of search visibility of Silverlight pages was brought up. I had heard a few times that one of the advantages of XAML being made up of text, and of the Silverlight .xap container being a standard zip was that it was supposed to make things easier to parse and search, but I hadn’t seen an actual example of this. There was quite a bit of interest in figuring out how to index a site that was made up primarily of Silverlight with the content accessed through the container, so we began poking around.
As with many of the questions that are asked in our group, we started looking for answers by seeing how Vertigo did it. I remember that you could share links to specific pieces of memorabilia on the Hard Rock site, so we went there. I grabbed a random piece of memorabilia and grabbed it’s permalink.
From the link, we could see that they were just passing in an argument to the base page. Loading this page, we could see that the title and meta tags had changed, but it was basically the same page. It made sense, but how were they exposing this to the search engines. I navigated over to their robots.txt file, and there was the answer: they had defined a site map, which contained links to every item on the page. Each item was just an argument to the same page, but the search engine was seeing them each as a unique page. It didn’t need to read the Silverlight control at all. The Silverlight control was able to show the item based off of the page argument, and the page prettied itself up with a title and some meta content to make itself relevant to the search engines. The same thing could be done inside the Silverlight page, but they were removing the necessity to do so. They were basically taking the same approach as they would for any site with dynamic content. Well played, Vertigo.
If you head over to Bing.com, you’ll see a landing page with a coming soon message. If you click on the “Find out more” link, and you should, you can watch a short video on the features that make me excited about the new search engine.
Some of these features have been available through separate tools, but this looks like it’s a lot better integrated. They’re calling it a decision engine, rather than a search engine. By categorizing their results, they give you the ability to find what you’re looking for. I could have used something like this when I was searching for books on technical evangelism, and kept getting information on religious evangelism on the web.
They broke down the key features of the new decision engine into four groups: local search, travel planning, health search, and shopping.
Shopping integrates their Cashback features with some great looking product reviews, ratings, and discovery features.
Health results come from some top medical sites, like the Mayo Clinic, giving you confidence in the results. It looks like a lot of effort was done in identifying common topics, and giving some customized search features for those topics.
Travel brings in the features from Farecast, which I use whenever I travel, although I’m still waiting for them to implement fare prediction between San Francisco and Narita, which is the most expensive trip I take.
Local search gives you some qualifiers you can add in to filter your results based on things like parking, price, reservations, and atmosphere. This sounds great for finding a place I can take my kids, or finding a place where we can go without them. They also give you reviews, hours, and contact information.
I’m not completely sold on the name, but with sites like twitter and digg, I guess it’s really the product that defines the name. I’m also not sure what we’re planning on doing with the Live Search page, or whether this is going to end up being something like Microsoft Live Bing. I’ll just have to wait and see what comes next.