Sigh. The Future of Search. Jeopardy
October 21, 2010
Think back to college. Remember freshman English. You know. You had to write essays every week or two. Most students got Cs. Now everyone is a great writer. I am not. I suck. My knack was finding information. I skipped fun and games in order to dig up facts for my crappy papers. I got by. Now everyone is a great researcher. Search, the browser, and the mouse has made it easy for anyone to find “everything” about a topic. Baloney.
I just read a remarkable summary about the future of search called “The Future of Search Looks a Lot Like the Present (on Steroids).” Even the title scares me. See I like Boolean queries. I don’t like algorithms, 20 somethings, or programmers who “know” what I want. Research means collecting information, synthesizing, and assembling. I don’t want research to be easy. I want to do the work. Pretty shocking.
Now you navigate to the article and read the future of search. Look at this item:
Keyword research may be narrowed down by all three search engines
What does this mean? The notions of precision and recall are gone. Today it is clicks. Lady Gaga type tricks. Easy answers. Hey, those C students wouldn’t know disinformation if it covered their Honda windshield, right?
Work? Thinking? Slogging through data? Too darned hard. Today search is information with training wheels. Get off the marketing bicycle and walk. C students will be walking, probably to the unemployment office. What about the top dogs? Working at marketing agencies, making millions from their blogs, or raking in the dough at a conference sponsored by third tier consulting firms. Who wants the dudes and dudettes in the Math Club making decisions? The C students. Sigh.
Gimme Boolean or gimme death.
Stephen E Arnold, October 21, 2010
Freebie for sure.
Comments
One Response to “Sigh. The Future of Search. Jeopardy”
-
- Subscribe to Beyond Search
Feature archive
News archive
-
> See I like Boolean queries. […] Gimme Boolean or gimme death.
People who prefer Boolean do so because they would like to stay in control,
and they see the alternatives to Boolean querying as a lack of control
However, while I sympathize with the attitude, and while there’s something to be said for different levels of “letting the user do the work” (manual influence), i.e. use different query languages for power users versus newbies, the fundamental problem of Boolean queries is that they are merely set descriptions that specify what is and what isn’t contained on the resulting page, and if you have a term such as “A or B or C” you will get max(#A, #B, #C) results/pages back. In the 1960s that worked because we were talking dozens of pages (can be sifted through manually) but nowadays where we have billions of hits that’s not feasible anymore. Also, most Boolean search systems don’t have a notion of ranking (there’s no obvious way to do that), so the most relevant result could be on page 293346285392.
So maybe what you really wanted to say is that you would like to have a mechanism to communicate your query intent to the system (being a power user) instead of having the algorithm do the guess work, regardless of the notation (which may be Boolean or not).