Ok, but accurate-and-shallow is still shallow. What's needed is a substantive accurate statement. It's particularly the shallow dismissals which harm discussion—hence that guideline.
uhmm, weirdly enough it might explain why some (myself included) have found that Google's results were shit /sub-optimal. Turns out I might have been using Google wrong for the last few years. I am used to using specific boolean search parameters while it seems that Google have been optimizing for natural language. An example from the announcement may clarify - "Here’s a search for “2019 brazil traveler to usa need a visa.” The word “to” and its relationship to the other words in the query are particularly important to understanding the meaning. It’s about a Brazilian traveling to the U.S., and not the other way around. Previously, our algorithms wouldn't understand the importance of this connection, and we returned results about U.S. citizens traveling to Brazil. With BERT, Search is able to grasp this nuance and know that the very common word “to” actually matters a lot here, and we can provide a much more relevant result for this query." Personally I would never have formatted a search query in natural language. Perhaps I should have been.
Yes, this is the exact problem I'm running into, I hammer out my search queries like its a wildcard CONTAINS SQL statement, because a simple inclusive search should bring back predictable results.
If you can remember any of the queries that failed, I'd be happy to pass them along to debug. If you have it turned on you can look in your search history here: https://myactivity.google.com/myactivity?product=19
I was after a specific legacy driver last night which the vendor no longer has available. Google returns zero results for this, bing returned a few relevent results (but sadly still didn't help me get what I needed) in the end I went mooching through the way back machine.
It's this class of search that bugs me the most, I know for a fact something with that exact filename is out there on the web, I just can't find where easily.
If I run into more issues or can recall anything else, I'll forward it on.
The improvements are likely for non-power users. As they make up the majority of users, it's understandable that Google optimizes for them. Still, it gives power users worse results because they know what they are looking for and they (often) use certain words for a reason. Google's default assumption ("the users have no clue what they are doing") leads them to a paternal "I think I know what you're going to ask ..." approach which often fails and leads to shit results.
The method to combat that is to start wrapping every word in quotes which is annoying. I'm sure intermediate users will catch on at some point, start doing it too and Google will drop the quote modifier. Let's keep it a secret so that happens later rather than sooner.
Natural language is a lot more expressive than boolean expressions of keywords. Perhaps experienced searchers who use the same approach to query formulation that worked well two decades ago are no longer the true power users. Maybe they are just dinosaurs.
This assumes a simplistic view of how its core search algorithm works, though. It's obvious through even some basic querying that it uses different strategies for recognizing intent based on the nature of the input: single words, addresses, arithmetic, business names, full text queries, etc.
There's no reason to think that they will enact this to the detriment of other queries. It _could_ happen, but I am skeptical - optimistic even. As they mention, this improves ~10% of queries. The other 90% likely represent different forms of query input and I would hope remain unaffected.
I think most of us technologically inclined users are in the habit of using specific words to get search results as we have been doing that for years. But I think a large portion of newer users actually ask full questions so google is optimizing for it. For better search results we might need to actually start asking exactly what we are searching for instead.
Google works well when you are searching for things you don't know much about, your input is imprecise and it generally points you in the right direction.
However for the opposite case, when you are trying to find something highly specific, even down to an exact substring match I find the results to be very poor.
Great, so the search results are going to get even worse?