If you had told me 3 months ago that Bing would be a serious threat to Google, I still would have laughed. That's how much impact the ChatGPT integration had for Bing -- overnight.

I’m blown away by the competence of the language model but its willingness to make up facts makes me leary.

I see you haven't met humans

Bing, IIRC, has a way to provide feedback, not sure how useful it is for today's users and if it will be able to solve hallucinations one day.

But I'm generally not interested in asking an average human. I'm interested in asking someone who knows their butt from a hole in the ground in whichever topic I'm asking them about.

When ChatGPT tells me something, I have no idea if it's paraphrasing information gathered from Encyclopedia Britannica, or from a hollow-earther forum.

phind.com has been incredibly good for me.

Is the work of judging the accuracy of a summary not just the work of comprehending the non-summarized field?

For example, a summary could be completely correct and cite its facts exhaustively. Say you're asking about available operating systems: it tells you a bunch of true info about Windows and OSX, but doesn't mention the existence of Linux. Without familiarity with the territory, wouldn't verifying the factuality of each reference still leave you with an incomplete picture?

At a slightly more practical level, do you actually save any time if you've gotta fully verify the sources? I assume you're doing more than just making sure the link doesn't 404, as citing a link that doesn't say what it is made out to be isn't exactly a new problem, but at that point we're mighty close to the traditional experience of running through a SERP.

Finally, even if you're reading all the links in detail, isn't that still a situation prone to automation bias? There's a lot of examples of cases where humans are supposed to check machine output, but if it's usually good enough the checkers fall into a pattern of trusting it too much and skipping work. Maybe I'm just lazy, but I think I'd eventually get less gung-ho about verifying sources and eventually do myself a mischief.

I'm asking because I've been underwhelmed by my own attempts at using LMs for search tasks, so maybe I'm doing it wrong.

Or it's something it just hallucinated out of thin air.

Checking the links is a good practice.

I feel like we just created an interesting novel problem in the world. Looking forward to seeing how this plays out.

https://www.phind.com/

“everything you read in the newspapers is absolutely true, except for the rare story of which you happen to have firsthand knowledge”.

Humans are pretty good bullshitters too!

I've asked BingGPT about myself and it gave me three answers. One was more or less on-point (it found my linkedin profile), and the other two were hallucinations. What happened was Bing found two unrelated pages and GPT has tried and failed to make sense of them.

Either that, or I am a prince whose name means "goose" in Polish.

I really hope they find a way to have it apply context from future conversations such that when it learns the error of its ways it emails you a retraction, but that's probably a ways out because humans can't be trusted to not weaponize such a feature into sending spam.

The weight of phrases like "you are wrong" is in fact so strong, that it fools the chatGPT to apologize for its 'mistakes' even in the scenarios where its text was obviously correct - like telling it 2+2 doesn't equal 4

Sure, grep has never flat out lied to me the way chatGPT does, but it's a statistical model, not a co-worker, so I don't feel betrayed, I just feel... cautioned. It keeps you on your toes, which isn't such a bad state to be in.

Especially when I talk to it about fiction and ask questions about - for example - a specific story and you see it invent whole quotes and characters and so on...it is a masterful bullshitter.

Cold War strategy. Trust but Verify.

Leary is a rare variant spelling of leery.

I mention this, because you seem to care about correctness.

What could be more 2020s than a post-truth search engine?

That said, I'm not a fan of using these tools for search. For me, anyway, they don't even come close to doing what I want when I'm searching the web.

i find chatgpt 3.5 answers better than bing. Also ive had bing end conversion with me on more than one occasion without saying anything offensive

it solves a lot for hallicination and current event problems you have in chatgpt

I'm using it to help me with a library integration, for example. I noticed it was recommending deprecated methods. So I copy/pasted the latest source code, asked it to update itself, and voilá.

It's super smart and learns literally in a second. Just drop recent information at it and ask what you need.

Personally — and having not tried ChatGPT for this — I don't think ChatGPT would do well with resolving the kinds of queries I consider Google "good at."

To me, the place where Google wins over Bing, DDG, etc. is when I know there must exist some page that uniquely talks about some extremely niche overlap of concepts; but I don't know any specific "natural key" keywords to refer to the that overlapped-set-of-concepts, and instead only have a "cloud of highly-correlated keywords for the individual concepts involved" to throw into the search box.

For example, if I'm trying to conjure from the aether a discussion people are having about an issue I'm facing with some buggy behavior in an API — where that buggy behavior doesn't spit out any distinctive error message to use in the search.

I could see ChatGPT being good at a limited version of this problem, where I could give it e.g. several definitions of a word (= correlated keywords), and it could tell me the word that fits those definitions.

But the full version of the problem — pointing you at (or regurgitating) the one unique conversation that most highly correlates with your keyword cloud — essentially implies an Internet-scale "language model": one where there are unique vertices for every unique URL. Which, if you think about it, is what a traditional search engine's index is: a very dumb, but very large correlational language model, where that "dumbness" is a valuable constraint meaning that queries are able to be run map-reduced across many nodes.

Is there something I'm missing here, that makes Bing+ChatGPT better at these types of queries than Google is?

Or are the advantages ChatGPT is bringing to the table here, in areas that have nothing to do with making search engines better at the things they're already "the best tool for the job" at, and instead are in solving problems that could be solved any number of other ways (e.g. querying a search assistant such as Siri/Alexa; or pulling up an encyclopedia article or textbook relevant to the subject and just reading it) such that a search engine wouldn't necessarily be the first tool you'd search for?

No problem for ChatGPT however which was not only able to write the code, but write it in Rust - the target language Iw as going to. Now I've just found it easier to ask ChatGPT first then go go google.

When you're asking generally about a pretty basic topic which you just happen to not be very familiar with, ChatGPT is not too dissimilar from having an expert in the field you can chat with and ask questions to. I find it to be a very effective way of querying the huge database of information that is its training dataset.

Surprisingly, the one thing it's really terrible at but which I would've expected it to be okay at, is writing config files. I sometimes ask it how to write, say, a systemd service file which does a particular thing, and it usually shows me something which looks roughly sensible but doesn't actually do what I wanted. Its nature of fancy autocomplete with no understanding really shines through in those cases. Its biggest downfall is that it has no way to recognize when it doesn't have an answer and is making stuff up.

And apparently you didn't quite notice the particular inefficiency there...

> But instead of doing that, I just asked ChatGPT the specific questions I needed the answer to.

Yeah, sure, if you want to vet hallucinations in stead of just getting the facts.

Just go directly to Wikipedia instead.

B) Your browser (most probably) has a search function.

I'm sure that recipe exists somewhere on the internet but ChatGPT gave something to me in a very succinct format with none of the usual bullshit you deal with when looking through search results. ChatGPT also thankfully did not include the usual recipe backstory.

how to create a multi line string in a bash script. I needed to encode a human-readable JSON string in a curl call and didn’t know how. Google game me crummy tutorial sites, GPT gave me instructions, and when provided with the target, did all the formatting too.

I found it also understands git well. I use git at work, but I almost never use anything beyond push/pull/commit so crazy rebases or merges and stuff I still have to search for instructions to remember them. Now GPT can just explain to me the steps for my particular case. When I googled things, I’d search for keywords and stuff based on my knowledge and piece together the steps myself.

On a counter example, I recently had an intern who botched their config on VsCode and didn’t know what settings to fix. I found it was easier to google search how to reset things than use GPT. Ymmv.

For example, today I asked ChatGPT how to write a class method in Ruby and to explain the class << self idiom. Super simple stuff, but it gives accurate answers and it's way more convenient than Google.

For this class of simple queries there's a lot of overhead to do a Google search and then try to filter out bullshit and padded results vs a super simple prompt to ChatGPT.

Geographical information about a region.

What is the name of a song I have in my mind.

Virtually anything else. I'm studying architecture and read about associations of feelings that cardinal points transmit in a house (north, south, east, west). Like, east is associated with youth because of sunrise. At first it wasn't obvious why, so I asked ChatGPT and it explained everything brilliantly to me.

It takes me an order of magnitude less time to educate myself on ChatGPT comparing to Google.

Spammers will, too, and since OpenAI has access to what they're asking and can easily flag their questions, they can feed misleading guidance to spammers.

For instance: That Microsoft Rewards counter? Ugh.

[0] For example, Google used to have a fairly strictly enforced rule that indexable content had to actually be visible to an unauthenticated user. The current crop of sites that have apparently useful content in snippets but that hide it when loaded would have been penalized, possibly severely.

It does seem like Larry and Sergei are finally back trying to fix the excessively politically sensitive and overly cautious culture. Larry, having disappeared to Fiji for several years, must have been pretty bored or annoyed with running it.

Yesterday night I was trying to recall the name of a particular SCOTUS case that I had (semi-incorrectly) thought was connected to the 14th Amendment.

Bombed out on the SERPs two or three times, so I started looking for an very old article by Thomas Sowell that randomly introduced me to the case a few years back. I knew it was something regarding the unintended consequences of the Civil Rights Act.

Neither Google nor DDG gave me ANY useful results for 3-5 variations on “scholarly critique of the Civil Rights Act”. I eventually remembered the Sowell connection and even adding his name in quotes only got me to a page deboonking the article I was searching for!

I’m very far from being a right winger or whatever so I’ve never really experienced this sort of thing before, but my god is the “no no, you don’t REALLY want to search for THAT” silent censorship out of control. Millions to billions of people use these search engines daily and assume they’re mildly biased but otherwise shining portals to the sum total of human knowledge.

In hindsight, I suspect I would have immediately found the SCOTUS case with my initial search on 2010 era Google. Very much seems like my starter queries were triggering the Bad Think Detected algorithm.

*debunking

https://yandex.com/search/touch/?text=scholarly+critique+of+...

Publishers blamed Google for declining revenue since they had to make their content openly accessible and therefore free in order to be visible to users on search. The EU tried to make Facebook and Google pay publishers to account for this. I think allowing paywalled content was a compromise to prevent this legislation from passing.

That being said, I agree with the publishers especially since hypocritically Google and Facebook strictly don't allow scraping of their services and litigate those who do.

Google could easily fix this by putting a symbol or label on paywall-related content so you know not to click it.

People are acting like Bing has created some big market success and Samsung is running to them because the tech is better. But the alternative explanation is simple: vendors are always looking for reasons to put mediocre non-Google search as default on their products, so they can extract more money from search engine providers. Samsung sees the current hype around Bing/Google/AI as a convenient negotiating point, since the media will portray this as “Samsung switches to awesome AI search” rather than “Samsung forces its users to use crummy Bing.”

You mean, so that you can satisfy everybody at the same time?

No, sorry, the world doesn't work this way.

Being the richest gives someone a lot of power to make others look foolish

It already seems poisoned to me. It's popularity over correctness, because LLMs don't know semantics.

The users who the default engine really locks in are the ones who just mostly click ads and have no idea they are ads. When they use Google they are mostly clicking ads anyways, so the search result quality will appear about the same with Bing. Maybe even better if there are fewer ads.

For Microsoft, getting more users on board means higher ad volume which brings more advertisers who are going to spend the time to manage Bing ad campaigns. You can bet they've done the calculations for how much money they can spend at what price. Ultimately that leads to higher monetization and then Microsoft can pay other companies (like Apple) to switch the default engine.

Google is a multi-trillion dollar attention tax that just sucks money from the global economy. They've been wildly mismanaged since Eric Schmidt's CEO tenure ended. It's been a long time coming, but the timer is running out of sand for Google fast. The revenue may take a long time to peak and decline but when they start missing their quarterly earnings it will be a bloodbath for Google's employees.

Every day I enter a few difficult queries on both Google and Bing to see if Bing gives me something better. I'm still laughing that people think Bing is a serious threat.

Truman Capote and Gore Vidal were two American writers who had a long-standing feud with each other. The feud began when Capote wrote an article for Esquire magazine in which he claimed that Vidal had been thrown out of the White House after making a pass at a member of President Kennedy's family². Vidal took Capote to court for libel, where the two traded insults². After the pair settled out of court, their feud continued – even outliving Capote².

I hope that helps!

Source: Conversation with Bing, 4/17/2023(1) The A-Z of Gore Vidal | Gore Vidal | The Guardian. https://www.theguardian.com/books/2012/aug/01/the-a-z-gore-v... Accessed 4/17/2023. (2) Why Did Gore Vidal and William Buckley Hate Each Other? - Daily Beast. https://www.thedailybeast.com/why-did-gore-vidal-and-william... Accessed 4/17/2023. (3) Feud sensation! Why Vargas Llosa thumped Márquez. https://www.theguardian.com/books/booksblog/2007/mar/13/feud... Accessed 4/17/2023. (4) A life in feuds: how Gore Vidal gripped a nation - The Guardian. https://www.theguardian.com/books/2015/aug/14/gore-vidal-gri... Accessed 4/17/2023. (5) ‘Just a couple of fags’: Truman Capote, Gore Vidal, and celebrity feud. https://www.tandfonline.com/doi/full/10.1080/19392397.2015.1... Accessed 4/17/2023. (6) Truman Capote’s unhappy ending | PBS NewsHour. https://www.pbs.org/newshour/health/truman-capotes-unhappy-e... Accessed 4/17/2023.

Oh, and references (2) and (3) seem to be hallucinated and are unrelated to the question and response despite being cited inline. The other ref links are valuable but well then it's just a search engine with more noise.

I just asked it about something in today's news and it answered it and provided links to 6 news articles on that topic.

Bing has only 10m+ downloads on Android. Not even a top 200 app.

Samsung switch too Bing will hurt Samsung more than Google.

For search, Bing + ChatGPT is now my driver for search. I still use Google and Duck Duck Go occasionally, but usually I am OK with waiting a short while for Bing + ChatGPT results.

Sorry for the OT question - do you use slime-tramp? And if so, do you have a way to use M-. (slime-edit-definition) without having to re-compile files using the remote paths?

For Common Lisp, running Emacs on a remote server instead of slime-tramp has always been good enough. Would you suggest I try slime-tramp?

Mosh as in "mobile shell"? https://mosh.org/

> Would you suggest I try slime-tramp?

I don't know yet, I've just started trying it out.

What made me want to try it was that I could use GUI Emacs to connect to emacs running on a different machine and still have full access to all the emacs keybindings.

So far, the downsides that I have encountered are that M-. and C-c C-k (slime-compile-and-load-file) don't quite work. The work-around would be to visit each file using the remote path and re-compile them so that the running Lisp image can map what's in the image to a path tramp recognizes. Then M-. and C-c C-k should work.

To recompile, select all then compile (X-c X-p C-c C-c) works, or I think C-c M-k also works. Not a great solution if there are a lot of files, though.

IIUC the problem boils down to M-. eventually calling (xref-find-definitions) which is an emacs built-in, and I think that's why the tramp paths aren't translating until a re-compile is done.