Hacker News Reader

Comment by bretthopper | Hacker News Reader

Most interesting part to me:

"To track our progress, we conducted tests that removed any trace of Google and Bing branding. When we did this study in January of last year 34% people preferred Bing, while 38% preferred Google. The same unbranded study now shows that Bing Search results now have a much wider lead over Google’s. When shown unbranded search results 43% prefer Bing results while only 28% prefer Google results."

View on HN

I switched over to Duck Duck Go, which uses the Bing API, with every intention of using them full time and not going back to google. But the results were much, much worse and I had to go back to google.

That wasn't just a blind taste test, I was actively rooting for pepsi but I just couldn't stomach the taste and had to go back to Coca-Cola.

Switched to DDG as well, noticed that it was a little too slow for my liking at times, and often did produce less desirable results than Google,

but after a couple of months those dried up. It seemed to speed up a little (don't know if it actually did) and the results seemed to get better (don't know if they actually did or I just kicked the subconscious habit)

I still find the ddg results are lacking about half the time. Fortunately, its only two characters more to redirect the search if ddg came up empty handed.

I've done a quick study of this at my office, and I'm not saying the results were what you experienced, just that this was one sample. When people put the same simple search term in both, they found the result quicker in Bing. When the search term was more complex and used operators, they found the result quicker in Google. The difference I saw was that people were used to Google and how to format a search so Google finds it better. The same search operators didn't work as well in Bing.

Whether this is a problem with complex queries in Bing or a problem of the computer training the user, I can't say. This was just what my office found out. For the record, no one changed their search provider.

I have found Bing results unimpressive, too. However, I personally suspect I've just been trained to use Google effectively, giving Google the edge when I compare the two.

Blekko lets you do this kind of experiment any time you put '/monte' after any query. It's pretty fun.

This strikes me as being similar to the Coca Cola / Pepsi rivalry. Supposedly, in blind tests, people prefer Pepsi, but does that correlate with them actually buying Pepsi more often? It's not clear that it does.

I wish I could track down the source for this, I think I heard it once at a conference. In regards to Coke vs. Pepsi, Pepsi is sweeter so when tasted in a small amount people tended to choose the sweeter drink. However when people have a full can the extra sweetness was a turn off.

Here is a good source: http://www.snopes.com/cokelore/newcoke.asp

Warren Buffet actually mentions this in interviews. He is a long time Coca Cola investor.

Yeah, but until we can see how that research was done, we do not know if those results are relevant.

Reply on news.ycombinator.com