SharePoint Search: Best in the Enterprise: NOT (Part 2)

SharePoint search technology is so great that Microsoft just spent 1.2 billion dollars to buy an enterprise search company from Norway called “Fast Search and Transfer”.  I wonder if they could have had “Merely Decent Search and Transfer for only 500 million?

SharePoint Search: Best in the Enterprise: NOT

Victor thinks SharePoint Portal Server Search is the best in the enterprise. Here’s a glowing tribute from MSMVPs:

I have upload a Excel document wih the name – printerconfig.xls in a document library.If i type in at MOSS 2007 search – printer, he didn’t find the file. If i type in config, he also didn’t find the file. If i type in .xls, he find the file. I have tried Wildcards for example printer*, but he also didn’t find the file!

Google is quaking in their boots right now about SharePoint’s über file-extension search technology.

Blogged with Flock

Tags: SharePoint, SPS, SharePoint Portal Server, Search

TechCrunch Search Sucks Eggs

Call me a stickler for good search results. I’d previously blogged about how terrible MSN search is, and how Safari Bookshelf was only good for leather-masked masochists, and even how WordPress search can’t find its butt with both hands, but I expected better from the 2^nd biggest celebrity on the tubes.

Here’s me entering the search term “lane hartwell” into the TechCrunch search box:

Here’s me getting only two results, neither of which have nothing to do with Lane Hartwell. At least the second result actually contains the string “lane hartwell,” although it’s a linkie to one of the ACTUAL POSTS that should be coming up for this search.

No “Best Search Results of 2007″ Crunchie for you, Mike Arrington! You can’t use a search engine that combines a dartboard and a blindfold and expect to win!

UPDATE: Is this why Edgeio folded? zing!

Google Blog Search Broken?

I’m very new to search analytics, so I’m trying lots of different things to get a fuller understanding.  One thing I noticed today is that Google’s Blog Search is not pulling up results for my blog as well as the main Google Web Search is.  This differs from my experience a week or two ago, when Blog Search seemed to be picking up items within minutes.

Searching for “Spire Viro” in Google Web Search brings back two results from Xidey in the the top three results, including a recent result from today at 11:28:57 GMT (about 18 hours ago).   That cached page contains a post whose title begins “Spire Viro” and whose sole tag is “Spire Viro”.

Google Blog Search for the same search term only brings back a much older post from Xidey, from about six weeks ago.  When I narrow down results to Last Day or Past Week, no results are returned.

Odd.  Something is going on.

For what it’s worth, Technorati Blog Search sucks goose eggs for the same search term, and WordPress.com doesn’t have a built-in search function to compare against.

Thoughts on Technorati Percolator

Technorati, the UI you Love to Hate™, just relaunched with a new user experience, some new features, a rediscovered emphasis on What’s Happening on the Web Right Now. Fred Wilson is hopeful. I’m not that impressed. How is this any different than the huge relaunch 6 months ago?

Sound familiar?

Whereas folks using Technorati a couple of years ago were predominantly coming to us to search the blogosphere to surface the conversations that were most interesting to them, today they are increasingly coming to our site to get the 360 degree context of the Live Web – blogs of course, but also user-generated video, photos, podcasts, music, games and more. They want all the good stuff out there, all in real-time, and we’re using the power of 80 million bloggers to help organize it and make it fun to browse; using the wisdom of crowds as a mirror on ourselves.

That was from May 2007.

First, an aside — are they searching, discovering, or percolating? Or all three? My Marketing Buzzmeter flashes uncontrollably when I read some of this. It sounds focus-grouped. It sounds like Mindheads. It sounds smelly.

I jest (a little bit); their theory seems sound enough. Blog posts are

primarily ranked by the age of the item, the authority of its source, the authority of the referring blogs and the density of recent links to it [...]

I did a little test: I had a lengthy post last night on performance reviews titled Performance Reviews Are Asinine. I searched for “Performance Reviews” on Technorati.com:

The top two results (above the fold) are from a music review blog and an Oscar-awards blog. Then we have something ACTUALLY about performance reviews from the Frugal Duchess, then crap: Manga, Cell Phones, London Theater, Erotic Art. I mean, come ON. Only one of those seven posts includes the exact term “Performance Reviews”, and only one (the same one) concerns the topic at hand. A few of those blogs have barely any more authority than does Xidey. The common factor seems to be timeliness — the oldest post in this list is only 6 hours ago.

Just for kicks, I searched for the exact title of my post — Performance Reviews are Asinine — and (predictably) didn’t get any better luck.

Note the age of these posts: 25 to 98 days ago.

My verdict on Technorati search, whether or not it’s called “Search”, “Discovery”, “Percolating”, or whatever — is that it needs some fundamental changes. If you can’t find posts with the exact title you’re looking for, it’s not really search at all. The good news? They are probably better than Safari Bookshelf.

Search: Google Impresses, O’Reilly Depresses.

Check out these two search results for an example of the right way and wrong way to do search. In the first case, I search on O’Reilly’s Safari Bookshelf for a book I had been reading just yesterday called Programming Collective Intelligence. The search term is the exact string “Programming Collective Intelligence” — it gets a little cut off by the search box (itself a mini-lesson).

The second screenie shows the results of the exact same search in Google:

On their own website, and with the information that I had spent probably an hour yesterday reading this very same title, O’Reilly couldn’t even get this recent book in the first page of search results! Because I’m a sucker for punishment, I paged through 79 pages of results and finally found the book whose title matches the exact search string I typed in at #799:

Can the Safari Bookshelf search from O’Reilly get any worse?

Mozilla auto-search feature is impressive

I’ve just found a new feature in Flock that actually is present in Firefox as well, but who knew? It speeds up “Quick Find” (in-page search).

Go to to Tools | Options | Advanced and click on the General Tab:

Check the “Search for text when I start typing” checkbox, and save.

When you visit a web page, just start typing and the normal Mozilla”Quick Find” comes up:

As items are found, the first result is highlighted in the page just like normal:

This little tip will definitely make your web-browsing experience better. Give it a try!

How important is metadata for search?

Tim Anderson’s ITWriting – Tech writing blog » Making search better: smarter algorithms, or richer metadata?

Metadata has its place, it will always be there. But in the end I don’t think the clock will turn back; I think plain old full-text search combined with smart ranking algorithms will always be more important [...]

Couldn’t agree more.