That verbatim option looks like it might be a way to largely bring back the old Google which tended to work quite well for me (except that modern SEO spam means we can probably never have the old Google again).
I’ve been pissed off with Google for (10+?) years now, ever since it became apparent to me that they changed their search from returning what you queried to some kind of fuzzy guess at what they think you’re looking for (probably based on what average, ie, non technical, people look for).
Possibly that verbatim option puts it back to what used to work pretty well for me, but because they changed it without really saying and then made the option for the old behaviour hidden away with an obtuse name, I had no idea about it until reading this article and had largely written them off as being more interested in showing adverts than what I was actually looking for.
Once I discovered that ChatGPT and similar are almost always way better at finding what I was actually looking for than Google ever was, Google is now very far from my first choice of search service.
It also helps that LLMs tend to be very good at helping you find the correct term, project, technology name, etc, for something when you’re not sure up front of what term to search for exactly.
lots of useful Google search tricks and syntax all in one place. I already knew many of these. But verbatim mode is new to me and addresses a major complaint I’ve had about increasingly fuzzy semantic search.
FWIW, this article isn't your usual substack slop. There are some Googling tricks and techniques here that I've never seen documented elsewhere, such as AROUND(n).
Is the AROUND(n) one real? I've never seen it before, and trying "climate AROUND(3) policy" as mentioned in the article just gives me results where "Around 3" is in the body:
European Central Bank
Climate, Nature and Monetary Policy
1 day ago — ECB research has found that four years after a drought or flood, regional output remains depressed by around 3 percentage points on average
> There used to be a professional layer between most people and raw information. Librarians, researchers, editors, fact-checkers: people whose entire job was to understand how information was organized, who produced it, what motivated them, and where the gaps were in any given source. You didn’t need to think much about any of that, because someone else already had.
> That layer has largely dissolved. Search engines replaced the card catalog, algorithms replaced the reference interview, and AI summaries are now stepping in where a librarian’s judgment about source quality used to sit. What’s been left in place of all that professional mediation is a search bar and the assumption that you’ll figure it out. - https://cardcatalogforlife.substack.com/i/188856309/the-skil...
An unfortunate conclusion that smuggles in unwarranted good-old-days nostalgia to an otherwise excellent overview. The previous system that they're describing had serious problems, limited access to raw data compared to now, and could not have scaled up to the level of access to information that the internet provides.
The information environment prior to the early 2000s was quite terrible. We shouldn't pretend this was a golden age of truth-oriented gatekeeping, although there were certainly gatekeepers. There were a lot of misconceptions, errors, and unchecked biases on dead trees and in late 90s databases. The idea that those librarians, researchers, editors, fact-checkers were aligned with anything you might care about is also unsupported and dubious.
I distinctly remember being an 8 year old in primary school and not being believed by a teacher that tungsten existed. I was told I must be wrong about the density of this metal being higher than lead and unless I could find a book to prove it I should shut up about it. In reality I'd been to a museum and learnt all about wonderful wulfram and probably just must have been insufferable.
I’ve been pissed off with Google for (10+?) years now, ever since it became apparent to me that they changed their search from returning what you queried to some kind of fuzzy guess at what they think you’re looking for (probably based on what average, ie, non technical, people look for).
Possibly that verbatim option puts it back to what used to work pretty well for me, but because they changed it without really saying and then made the option for the old behaviour hidden away with an obtuse name, I had no idea about it until reading this article and had largely written them off as being more interested in showing adverts than what I was actually looking for.
Once I discovered that ChatGPT and similar are almost always way better at finding what I was actually looking for than Google ever was, Google is now very far from my first choice of search service.
It also helps that LLMs tend to be very good at helping you find the correct term, project, technology name, etc, for something when you’re not sure up front of what term to search for exactly.
This hasn't been the case for at least five years surely?
European Central Bank Climate, Nature and Monetary Policy 1 day ago — ECB research has found that four years after a drought or flood, regional output remains depressed by around 3 percentage points on average
(compared to e.g. https://www.google.com/search?q=climate+policy+ecb) which has the same result but does not show the "around 3 percentage points" snippet
> That layer has largely dissolved. Search engines replaced the card catalog, algorithms replaced the reference interview, and AI summaries are now stepping in where a librarian’s judgment about source quality used to sit. What’s been left in place of all that professional mediation is a search bar and the assumption that you’ll figure it out. - https://cardcatalogforlife.substack.com/i/188856309/the-skil...
An unfortunate conclusion that smuggles in unwarranted good-old-days nostalgia to an otherwise excellent overview. The previous system that they're describing had serious problems, limited access to raw data compared to now, and could not have scaled up to the level of access to information that the internet provides.
The information environment prior to the early 2000s was quite terrible. We shouldn't pretend this was a golden age of truth-oriented gatekeeping, although there were certainly gatekeepers. There were a lot of misconceptions, errors, and unchecked biases on dead trees and in late 90s databases. The idea that those librarians, researchers, editors, fact-checkers were aligned with anything you might care about is also unsupported and dubious.
Wikipedia is a godsend.