You do get different (and better) results if you search "What is Σ" rather than "why is Σ" or just "Σ". Like the other commenter said, it's better when you'd rather find a forum or SO post matching that title, rather than getting back Google's own answer.
I dont think you and the other response understand.
Google used to conduct a relatively straightforward string matching search. That means if you used words like "what" you could end up with unrelated matches, or lower quality matches, because they would match with the word "what".
Now, bear with me, these days google seems to use some kind if machine learning to suggest results to you based on what others have searched and chosen.
What does this mean, practically? At least two things:
1.Laymen have to think less critically, less technically when searching. Considering google is, for the average person, basically the window to knowledge, I think this is ultimately a disservice to society.
2. Search quality for technical information seems to be declining, now that laymen and non-laymen alike are searching using the same " extra" words and, I may just be projecting a poor understanding of neural nets here, but our technical results end up getting sort of clustered with everything else.
Am I the only one who has this problem with the decline of google search's technical relevance? Maybe I'm doing something wrong...I still think what google has done to make the internet more accessible may be a net harm to society.
Nah, I understand. I had to attend a 1 day class in my university on how to properly use Google and Google Scholar. I also miss some of those operators and tricks.
But nowadays, it's not as simple as just machine learning to understand your query, but the way the information is stored is less and less like a text storage. If I search for something involving "car' it's very likely I'm okay with results containing "sedan", "vehicle", or various brand/model names of cars. I'm also probably fine with "vehcile" or "vehiclle" or "carr" but not "cat". And Google is well aware of this and considers this when building their model.
Furthermore, the way you ask SHOULD change answers, especially because it attempts to automatically answer you at the top of the page. A search for Sigma and What generally means I want to know what it means, how to use it. But a search for "why" should bring up the historical reasoning Sigma was chosen over another letter, or another culture's alphabet entirely. That's too much information to fit into any result for simply "Sigma", and the query is too vague to help pare it down.
I do agree with you though that it is declining for technical information. But I think overall it has gotten much better at becoming a general "window to knowledge" and I think it's worth the tradeoffs.
