Yes. Not just over the past 2-4 months, but over the past five years or so.
It's become so bad that Google is no longer the most useful search engine for me.
He said they haven't noticed any regressions. I said I figured that would be the case but I can definitely feel the difference as a daily user.
I tend to believe that if user complaints about new problems or regressions increase over statistical noise - there is a problem.
Well said. This is a big problem. We see a similar problem with the use of telemetry data as well.
I had no idea about Hummingbird though.
I had assumed that Google search had gone downhill because it started trying to "personalize" my search results. That wasn't a great explanation though, as I don't use a Google account.
Hummingbird seems a much more likely explanation.
I do a full clear on my web browser (cookies / offline storage / history, everything) and then open YouTube in a private browsing window and it asks me which of my two Gmail accounts I want to log in with. I'd guess it's just a combo of external IP and browser fingerprint, but it's creepy.
I understand this being default behavior, but there really needs to be a way to disable it.
On the other hand the non-exact hits that it returns push me from time to time in the right direction.
Having said this, I don't know of course if A) I'm too old (40) and the mindset of the younger search-people has now changed and/or B) Google just doesn't index tech forums as much as it used to and/or C) there are just fewer forum-posts and/or D) my problems became more complex (don't think so) and/or etc... .
I tried (and still try from time to time) to use DDG and Bing but without success.
Does anybody else have the same impression?
So ... although I feel like we might be having the same issue, I'm not sure I'm using DDG correctly enough to say it's a problem.
Any search term not wrapped in quotes can be randomly ignored today. It can inject keywords it thinks you want (but really don't). Google is great for searching modern sites like Stack Overflow, but it seems to have lost interest in servicing power users.
I agree.
Could something as innocent as training a new neural net or testing a buggy version of the algorithm on subsets of users. But it could also be as sinister as driving traffic to those in bed with Google, silencing opposition, or effectively whitewashing the entire internet...
They are maximising ad revenue, not the search relevance/usefulness.
Stopped using "google" as a verb a long, long time ago, in favor of just saying "search". I don't think that's ever confused anyone.
It's my opinion that a large portion of the websites on the front page of any search (quora and pinboard anyone) are completely bought and paid for.
I frequently suspect they're starting to optimize more for $ than they were before, and ML just gives them more ways to make that number go up another % or so... but it often comes with impossible-to-predict and wildly inhuman edge cases. It's a pretty common trend when companies start focusing on small number increases - each A/B test shows improvement, but the product as a whole worsens and it drives people away in time.
Maybe some similar change is coming to Search.
To me they started spiraling down when they started to give too much power to designers. Form over content is a terrible idea for a search engine ...
Aspiring science fiction authors, or Neal Stephenson, should write a novel about a world where ML tuned models optimize everything to be just good enough not to churn customers while maximizing margins. (Also applicable to non-profit items like politicians and universities)
So then I do the quotes thing, especially quoting phrases that 100% for sure must exist on some web pages, along with all my other keywords and pretty soon I'm at "no pages found". Pull back just a little, and it's page after page of entirely unrelated-to-what-I-want blogspam.
Looks like only page 6 is indexed for some reason. The site owner would be able to check the webmaster tools on Google to see why.
the end user searching for the content
the webmaster or author of the content
the search provider
If I'm searching for something that I know exists and I cant find it there is no excuse. The search provider failed to do its job.
There is not but the webmaster should have done this and that. He was hit by a bus 10 years ago and we should be happy the content is still available.
A good search provider would link a vanished website to archive org if the content is exactly what the customer wanted.
Long long ago when posting interesting links in comments didn't trigger commercial hysteria people would cite bits of texts and have a link to the full text. Later this became simply citing a chunk of text. I use to drop a few lines from the citation into the search engine and find the original work.
Just look!
https://www.google.com/search?q=Looks+like+only+page+6+is+in....
As i'm writing this there are exactly 45 search results above the one that should have been displayed.
There is no excuse like HN not ranking enough, they didn't not index the page, the other results didn't match the query better.
If we do this with 4 exact lines from a less popular site it will end up some place on page 20 of the search results.
Another example, I really don't care for indexing but here is an article that I always (jokingly) refer to as my greatest work.
The exact title:
https://www.google.com/search?q=%22the+wrath+of+the+book%22
A really weird result. Safe to say nothing matching is there.
The first many words from the text:
https://www.google.com/search?q=I+think+someone+%28you%29+sh...
It doesn't find it.
Then we check if it is even indexed...
https://www.google.com/search?q=http%3A%2F%2Fblog.go-here.nl...
And there it is! Why does it even crawl the page?
It also lists websites that have the number 8616 on them and ones with both the word "blog" and "here" in the text.
I'm not suppose to laugh?
you can force the site with "site:... ": https://www.google.com/search?q="metallica+only+played+2+son...
it's doesn't find page 5 with these terms but find page 6.
There is probably an issue within the page 5 itself.
How is Google supposed to find that out?!
Which goes against the original mission of Google to "organize the world's information and make it universally accessible".
A "bug" could be an option, but I don't expect that to be the reason. It's too easy to find examples of forgotten content. And I don't think a bug of that magnitude in Googles core business would go unnoticed.
Which core business are you referring to?
It forced my to solve a bunch of CAPTCHAS too.
Also on the "Million Short" search engine mentioned by kickscondor:
https://millionshort.com/search?keywords=%22Metallica%20only...
I never saw that one. Do they have their own crawler?
http://www.gnoosic.com/discussion/metallica__5.html
No matter how you search for the content on Google, nothing comes up:
https://www.google.com/search?q="Metallica+only+played+2+son...
DuckDuckGo has it:
https://duckduckgo.com/?q="Metallica+only+played+2+songs+fro...
I checked the wayback machine and the content has constantly been on that url for over 10 years.
This is the first example of an old forum page I tried after reading the article. So I tend to think it's true. Google is discarding the "classic" web.