Yeah, I started noticing this a while back. Incredibly annoying and just plain stupid. As per the above link, Google has decided on our behalf to include “similar” search results that may not include the actual words being searched for. So the more terms you throw in, the more hits you get. With no way to turn it off.
Thanks a lot idjits.
Over and over, Google is just going in doing things the exact opposite of it’s original style.
Does it really matter, though? I mean, the results are sorted by relevance, and the most-relevant results will be ones that include all search terms. Are you actually going to find any hits that don’t in the first, say, ten pages? And who even actually digs that deep into a Google search?
If you don’t want merely similar things, though, you can put a plus sign in front of each search term.
More hits = more revenue*. Pretty clear where their motives lie.
I could have sworn you can turn off the behavior by prepending every search term with a “+” to mean “must be included,” but that doesn’t work anymore (or perhaps never really worked like I thought it did), at least not with your example. It does seem to cut down on other examples, though. Try zymurgy solipsism hyperfocal vs +zymurgy +solipsism +hyperfocal. So, I have no idea what is really going on there.
Google counts are notoriously unreliable; they aren’t fully accurate counts of the number of hits in their entire database, they’re just quick heuristic approximations. So they don’t always follow the rules you think they would.
They don’t even display consistency within themselves: type in a random string of words until you get the number of results down to something around, say, 600. Then click through to the end: the number of results will almost certainly change to be something like 80. The first rough approximation gets refined as you actually jump to the end of it.
Example (jumps from 398 on page 1 to 70 on page 10 (which turns into page 7))
(And if you choose to include omitted duplicate results, it goes from 381 (yes, it in fact starts lower) through 371 through 366 down to 249 at the end. The point being, the displayed count keeps changing. Don’t expect any consistency from Google search counts.)
A. Google uses their own method to decide which order to display them (page ranking). Whether the words actually appear on those pages is secondary. So the first hits can easily not contain all the words while later ones do. You have no idea if they do or not until you open the page and search.
B. As per the link earlier, the “+” sign omits “similar words”. E.g., in searching for “Tery”, Google will include “Terry” unless you include the “+” sign. (Supposedly.) It does not exclude pages that Google feels are “related” in some mysterious way.
So it matters. A lot. I know enough about search terms that I can choose ones that will limit the number of unrelated hits, or at least I use to. So if I include 5 specific terms, I expect only pages with those exact terms.
Google is becoming progressively less and less useful and harder to use.
Sometimes Google includes results where one of the filters (search terms) appear in links to a page. If you search on cats “cute puppies”, the first result you get (at least at the time I did it) is to The Daily Cute | Cute Puppies, Cute Kittens, Cute Animals!. (Og have mercy. Where’s my insulin?) If you click on the Cached hyper-link for this entry you’ll see in the gray box, “These search terms are highlighted: cute puppies These terms only appear in links pointing to this page: cats”
And Diablo!
I’m pretty sure SeaCanary’s reason is why things are related, even if they don’t contain the term.
The other is them expanding abbreviations and synonyms, but I believe that is handled by the + operator.
If this made any sense, then you’d figure that Boolean searches would never have been invented in the first place.
I dunno, people using Google for research, as part of their jobs?
If you’re getting ten pages deep and still not finding what you’re looking for, then you probably need better search terms. I’d expect that people who use Google as part of their jobs would develop the skills to refine their searches better.
:dubious: “Better search terms,” as in, say, Boolean operators?
Are you aware of the premise of this thread?
I … really don’t know what to say … here.
We’ve tried, Chronos, we’ve tried to explain this to you. Try reading all this again.
Your view was correct 5 years ago. It’s not remotely in line with Google’s behavior now.
So, you’re saying that there is information which can be found on Google, but only after ten or more pages of irrelevant results, and for which a better search wouldn’t bring it up quicker?
Yup. I don’t know what you mean by “better search”. But I would mean “Like Google used to work.” That is, exactly the words I type in appear on each page.
So use the plus signs. And I don’t know what you mean by “like Google used to work”: They’ve always included words that link to a page in addition to words on the page. That’s what made them so much better than all the rest in the first place.
Wait a minute. By “exactly those words” you mean those specific words, and no others?
And hold on. By “type in,” you mean input alphanumeric characters via a keyboard interface?
And by “page” you mean each successive numbered screen of search results?
:dubious: These things need to be clarified, you know. Otherwise, I’m just not seeing it. Why do you people seek out so many things to complain about, anyway?
Erm.
Google’s big success was that it was better and quicker at delivering relevant results, rather than simply being a search engine like, say, Lexis, which finds exactly what you ask and nothing else (which is why Lexis recently introduced a Google-style natural language search that tries to find relevant documents). In other words, you have it exactly backwards – if boolean searches were so great, Google never would have been invented or become the best search engine virtually overnight.
Also, I feel compelled to point out that if you are doing serious research using Google, you probably aren’t doing serious research.
Why?