Google has released a number of open source code projects developed by their staff: Google Code. A lot of it is pretty esoteric. One that caught my eye was PyGoogle, a python module that can be used to call the Google search API. We use a search engine at work that uses Python, so in theory we could use the PyGoogle library to incorporate google search results with our own. Nifty.
Google just launched another search appliance: the Google Mini. It costs $5,000 which seems to be a good price point for a lot of non-profit organizations to make an investment for their site or intranet. It will index up to 50,000 documents, which should be plenty for most needs in this market as well.
Lou Rosenfeld’s recent post on where to position search and taxonomy management within the organization was a nice validation of how we have it set up at our office. According to Lou:
To rant a bit, it really drives me nuts to hear people talk of “search and IA” (which they often understand as browsable taxonomies). This is an absolutely false distinction, and leads to poor search design, poor taxonomy design, and perhaps worst of all, missed opportunities to better integrate the two to support finding, IA’s ultimate goal. For example, search often is greatly improved when it leverages metadata tags. Metadata therefore should be designed with search in mind. So why separate teams? I don’t see any good reason, just a lot of bad ones.
At ASHA, we have two teams in the Web Cluster (our label for a division): the Content Management Team (CMT) and the Knowledge and Community Management Team (KCMT). CMT has responsibility for IA, visual design, general content development and managing the stream of content that comes from our 40+ content contributors. KCMT is responsible for managing our search engine, the ASHA intranet, the member community, online events and the ASHA thesaurus of terms. Both teams sit next to each other in our office and have easy access to one another. We also have a full staff meeting every two weeks where the topic of discussion is often on how we can improve the overall findability of content and services on our site by tweaking our search, metadata, etc.
While they are technically two separate teams, they operate as one in effect. I’m very happy with how well this arrangement has been working for us.
Came across this link on RC3 that translates Yahoo’s new paid search listings: Yahoo Paid Search, Translated. I really don’t like that these paid listings won’t be identified as such in the search results. Ultimately this will hurt them if it degrades confidence in their search results due to not-so-valuable paid listings crowding out other content.
Martin Belam posted a pointer to a new advanced search “tips & tricks” page from BBCi Search developed by the team he is a member of. It is a succinct and easy to read list of tips on how to narrow down your search down on the bbci site. See CNN’s search help page for a comparison.
I still find it extraordinary that a media site has a full team of people dedicated to its search function.
Today I was looking for a list of variables used by search engines in their query strings and found this excellent page.
Analysing search logs allows us to predict what those queries will be, and work out how best to answer these queries to fulfill the users requirements without the need for them to understand the naming conventions, departmental demarcation or strategic aims. Instead we can use the interpretation of their actions to shape a better interaction with the site for future visitors, and we can use that interpretation to influence the content and navigation of a site to better suit all visitors, not just those who have resorted to search.
Another great article from Martin Belam that is based on his experience on the BBCi search team.
The previous iteration of BBCi Search had the best links offset in a box above the tabbed interface, and we found that they were effectively in a blind spot, which is why they were moved into the run of the results.
VeriTest was commissioned by Inktomi to conduct the test. It found that in raw scoring (where URL position wasn’t taken into account), Inktomi came out tops — but just barely. Inktomi earned 1630 points, with Google just behind at 1597. That’s so close that I’d essentially consider the services tied. Behind the leaders came, surprisingly to me, WiseNut at 1277, followed by Teoma at 1275, AltaVista 1222 and AllTheWeb at 1173, another big surprise for coming in last.
Isn’t Inktomi now UltraSeek, which is now owned by Verity?
This piece also has some interesting things to say about bias in commissioned tests.