The graph showing the deviation from each search engine's 100% baseline is kinda telling considering Bing is all over the place with it actually seeing some traffic, and Google is a consistent up-and-down shape that doesn't seem affected at all by whatever is happening with Bing.
That's cool in the short term, but it will be news when it's sustained. I logged in to Bing for the first time in forever to play with the AI some, but it didn't change my default search behavior of using Google, and now it's been easily two weeks since I went to Bing at all.
The Bing AI is too crippled. It would be nice to have the option to remove its handcuffs. The chat limit and the constant “I’m sorry but I cannot continue this line of conversation” gets annoying fast. I was also able to get better answers from ChatGpt running on gpt3 than on Bing running on gpt4 for this very reason.
If you use duckduckgo, you already use Bing results.
I rarely use google, and when I do it's from duckduckgo !g
I would immediately switch to bing by default if such bangs were supported as I sometimes need them for wikipedia !w or amazon !a
If there was a way they could be supported Edge address through a plugin or something, I would immediately switch to bing as that's what I already use 90% of the time (between duckduckgo frontend and requesting bing directly with !b )
Right, I would be surprised if Bing sees any sustained benefit. The search engine still sucks, and the way they integrated ChatGPT sucks even more. I use Google a lot less these days, but the beneficiary, in my case at least, is not Bing, but chat.openai.com, where I have become a paying customer.
I would 100% use Bing more if they offered a "minimal" or "low bandwidth" mode that cuts out all of the pictures, news, weather, etc. and just takes me to the results.
The weakness is revealed. Google’s stickiness now has a quantifiable wedge factor. Whether it goes to Microsoft or someone else is more a matter of time.
Users have to sign in with a Microsoft account to use this. Is that the future? You have to sign up to search?
It takes more resources per query to run AI-based search than a search engine.
Everybody offering ChatGPT type systems is either pay per view or heavily throttled. We may be in the last days of free search as the dominant product.
I think it's far too early to say where this is going. If you could make this into an extremely skilled targetable manipulator/advocate for any kind of viewpoint, worldview, marketing strategy, on a per-user and day basis, one that subtly pushes the user in a given direction, I'm sure that would be lucrative enough to cover any amount of free/"free" views/prompts. Also, if this takes off in earnest, improving hardware acceleration will probably bring down costs as well.
I am one of the people that installed the Bing app because I was told that it would bump me up the waitlist for Bing Chat.
I have not used it at all, and this is a reminder that I should probably uninstall it. I have access to ChatGPT, Bard, Alpaca etc. and there’s very little reason at this point to pretend to use Bing in order to get a crack at Chat.
ChatGPT doesn't have the browsing plugin enabled yet.
Bing can both search/browse the web and generate images interactively by using DALL-E API.
There are several drawbacks like a more strict system pre-prompt, 15 messages limit for conversations, and some kind of 24 hours ratelimiting if you use it too often.
Yeah, there are some differences in capabilities, but they seem to get slimmer on a daily basis.
Want an LLM-enhanced search? perplexity.ai is actually pretty cool
Want to generate images? There are a ton of free Stable Diffusion sites (for example you.com has that built into their chat), or run it yourself.
Want a neat chat experience? ChatGPT is free, Bard appears to have a shorter waiting list than Bing, and Alpaca runs on regular hardware.
Bing doesn’t have a big enough moat to force me into using their app. The LLM space is legit competitive and a company can be ahead in the morning and left in the dust by the afternoon. In my opinion, this is what’s happening with Bing at present.
Who knows though? Bing might make Chat open while I’m writing this sentence and I’ll have to adjust my opinion accordingly.
I signed up to the Bing GPT "beta" waitlist (I hadn't logged into my Microsoft account in years) only to be told, days later, that I had to use Edge to try it. I noped out of there.
you don't ACTUALLY have to use edge to use it. just use a chrome extension to fake your useragent to pretend to be Edge. I despise needless microsoft hubris
Honestly these numbers are worse than I was expecting for Bing. Can they maintain these numbers when they’re not releasing new products people want to try?
Right? I'd have been entirely unsurprised if Bing saw a 100% increase, given how buzzy all this has been. (Finally, an actual reason to use Bing!) 15% seems like a failure.
lol. I noticed it depends on what I search for. I typically search for programming issues. Google tends to have the correct Stack Overflow answer up at the top. But I agree a bunch of ads otherwise.
it depends how and what you are searching for, I've seen some "spam" with stackoverflow clonse, but that seems to be gone now mostly. I also think that you are exaggerating a lot, otherwise citation needed.
I feel like I’m missing out on something but other than to summarize text or fill pointless and lazy homework assignments I don’t have much of a use of gpt search. I think I have a trust issue with it and I rather get the raw result and process it myself than to believe that what this model “understood” is right. Anyone else feels this way?
Yup, and maybe we're just getting old. There are still people on HN complaining about Google removing the OR and AND modifiers, meanwhile the younger generations write queries like "where should I eat breakfast today"
I'm fine with writing queries like "Where should I eat breakfast"
But I hate seeing the top half of the screen are ads for huge chains, and then the top results are blog posts about "Where are the top 10 places to eat in your city" and not a list of restaurant web pages.
Those aren't really exclusive. Different problems, different query types. (I wouldn't quite do the second example like that, but "where can I get food here" is fundamentally different problem than "I hope there is somewhere out there a bootleg copy of the datasheet for this very weird part and no I don't mean any of these 5 easily-confused other things", which is when the lack/ignoring of detailed modifiers drives you up the wall)
I used to try to be surgical with my queries but I've given up and just type a question at my search engine because if that's what they want as a query well they're going to get it.
In the context of LLMs they aren't really sources, just additional reference material. I'd trust what Bing Chat tells me far before I'd trust Google's new Bard AI. It legit started making up non-existent commands for me to use.
I find it useful in Edge where it's aware of the tab contents. It's nice to be able to ask it to summarize a long article, or find Reddit comments about it and summarize those. I think it uses Bing's search index because it does not seem to be aware of my personal data on my tabs
I agree with your sentiment exactly. But your comment made me think about this an interesting way that I hadn’t thought about before, that made me better understand why I am “anti-GPT search”.
All the same things that the model makes an effort to “understand” are things that can contribute to your own understanding. When it “understands”, you do not. You get the summary and miss out on a great amount of nuance that can come with learning and finding the right answer for yourself.
For instance, it is very common that in the search for why X is happening in my code, I will find tons of information that don’t answer my question but help me form an understanding of why it is happening, how the system is working, etc.
This is especially visible in science. Many papers outright conflict with each other. Some have better or worse methodologies than others. Some have better analyses. Some state outright falsehoods without citation or misinterpret citations. Having a GPT “understand” this to produce understanding in humans is going to end very badly. It takes several hours for a human (in my experience) to understand even a single academic paper on its own.
I suppose the appropriate cliche is “it’s the journey, not the destination”.
A great quote from an article I read on HN recently
> Suppose I was an evil person and wanted to eliminate the curiosity of children. Give the kid a diet of Google, and pretty soon the child learns that every question he has is answered instantly. The coolest thing about being human is to learn, but you don’t learn things by looking it up; you learn by figuring it out.
Most of the people here know and understand computers because we had to tear everything apart and figure out how everything worked just so we could play some video games.
Contrast kids of today who can just go and play video games, no disecting of computers required. Naturally, most of them don't understand or appreciate computers as anything more than just another appliance.
As opposed to the thing we did before, which is when kids asked something they were told something that's wrong and told to "stop talking back" and "go to your room". I think this is still an improvement.
> As opposed to the thing we did before, which is when kids asked something they were told something that's wrong and told to "stop talking back" and "go to your room"
Do you have concrete examples of stuff you figure out instead of looking it up? I'm not going to figure out the population of the Liechtenstein, I'm going to look it up.
What chatGPT misses is I want many sources of information conflicting or not so I can form a more informed opinion. I want to know every side not the correct one.
I certainly have the trust issues for things where correctness matters. For things were it doesn't and I already know how to do something but don't have it memorized, I think it is very very convenient.
I use it various times a week for cooking. E.g. "basic recipe bechamel sauce in metric"
ChatGPT: Within a couple seconds I get the ingredients in metric and a step by step guide.
Google: The provided summaries are useless to me since they are not in metric. So I have to click on a link and then start scrolling around all the noise of history of bechamel sauce, anecdotes about someones grandparents, pretty pictures, etc to find the actual information. Without ad blocker it would be an even bigger nightmare.
I don't need to engage in a conversation. Query was exactly the same for both.
There is a difference whether I'm looking for interesting, more complicated dishes I haven't done before, or fairly standardized, simple recipes I have done many times and remember the technique but not things like ratios.
For the former I use a select few recipe sites or youtube channels directly. Google is miserable for discovery in my opinion. For the latter I use either my own recipe notebook or more recently ChatGPT as well.
Here is what I have asked it the past few days with great results:
- What is the trim stop attachment that comes with the Festool DF 500 used for?
- What are some options for water for a home with no municiple water or well access?
- What dimensions drawer should I make when using 18" Blum Blumotion full extension drawer slides if the drawer opening is 20 inches wide, 18 inches deep and seven inches tall?
- Can the smaller Laguna 14 bandsaw motor be replaced with the 3HP version?
- What's the best way to get ChatGPT support if the official support page is not working?
- I'm trying to remember a 90s movie about a boy prodigy that goes to an event with other prodigies and then shouts out an answer to a math question from the audience.
You have touched on something I am incredibly excited about. ChatGPT is going to do away with a lot of the intro level questions you ask researching certain sorts of technical problems, especially as it becomes better able to understand images. ChatGPT removes the need for you to ever open a manual (unless you want a deep understanding).
How do I turn off this feature in my car?
How do I replace the fan in my fridge’s compressor?
Why is windows repair not working in this scenario with this PC?
This tool is going to be incredibly deflationary in many services geared around repair when now it can tell you almost exactly what to do and soon it’ll be able to produce videos or images of each step on demand.
Yeah, when the “site:” search hack doesn’t get me an answer and I fall back on the normal SERPs, most of the time the top several results are all:
“Authoritative title that references my exact search” -> page body restates my question, gives an oddly reworded summary of Thing in Question #1 -> “some people online have said” followed by snippets lifted from the ether -> repeat for Things 2,3,etc -> “our suggestion is to look at the features and decide based on what other customers/users are saying”
It’s perfected vapidity wearing the skin of an editorial review site and takes a LOT more cognitive load to suss out the phoneyness compared to the last time the SERPs were packed with spam (~10 years back).
hallucinations don't invalidate the product itself. I really hate to see this constant lazy refrain on here.
Its just like anything else. Trust but verify. Some will just trust, others never will... People probably over-trust the tech now, but just like wikipedia it will be approached a little more carefully as people learn its shortcomings.
Unless you're trying to perform open heart surgery or doing something intrinsically dangerous, its probably going to be additive. I wouldn't trust it to help me assemble a warp engine, but it may be able to give me some decent pointers for guitar technique or how to change my oil.
It’s not a lazy refrain, it’s a serious downside of the technology. It’s tiring and stressful to supervise the work product of an assistant that is extremely capable sometimes, but a compulsive bullshitter at other times. Just as it’s tiring to supervise an “autonomous” car that often deftly navigates the road, but sometimes wants to plow into a school bus.
I think that the concern about LLMs making up information are absolutely valid, and a serious concern. At the same time, I think most people focusing on those concerns aren't thinking much about the alternatives, namely Google.
For many people, Google results have gone from being an absolutely amazing demonstration of what could be done if one had the resources to crawl and process the entire internet to a complete waste of time at best, clogged with spam and nonsense and misinformation. I've been using Duck Duck Go for years now, and recently my middle-schooler switched to Duck Duck Go because I got answer to a question she had already asked Google before checking with me. Some people even pay for Kagi search, because Google is just that bad.
So LLMs don't have to be perfect to be better than the alternatives, not even close.
> Some people even pay for Kagi search, because Google is just that bad.
Small correction - from what I have heard in our community, most of our users pay for Kagi search, not becuase Google is just that bad, but because Kagi is just so good. In other words we are making the value proposition be "best search in the world", one worth paying for.
There's a ditch by my house that the city put in. I roughly knew what it was, a big pit where rain collects and soaks into the groundwater. But I wanted to know the name of it and no amount of Googling could get me to the right answer. My query was too vague.
The day that ChatGPT launched, I described the what it was and it came back with, "that's called an infiltration basin."
> - I'm trying to remember a 90s movie about a boy prodigy that goes to an event with other prodigies and then shouts out an answer to a math question from the audience.
I love that movie! I had the same experience a few years ago of trying to remember its name.
This is a great example of something ChatGPT does amazingly well at. I did almost the same thing myself recently, trying to find the name of a show featuring an actor in a different show or something. A straight-up search returns nothing for it; you have to look up the actor's name, then look at his shows, then try to remember the name. But a search with ChatGPT found the result easily.
I tried a "similar" thing now. I was thinking of the movie Gifted, and wanted to see if I could get ChatGPT to pick it up, here's my query:
> I'm trying to remember the name of a movie about a child, featuring an actor that was in a marvel movie.
And the result:
> There are many movies featuring actors who have appeared in Marvel films, so it's a bit difficult to pinpoint the exact movie you're thinking of without more information. However, one possibility could be "Gifted" (2017), starring Chris Evans, who is well-known for playing Captain America in the Marvel Cinematic Universe. In "Gifted," Evans plays the role of a man raising his young niece, who turns out to be a child prodigy. If you could provide more details about the plot or other actors, I'd be happy to help narrow down the list further.
This is obviously a tiny use case, and really not worth thinking about. Who cares, right? I thought the same thing about search in general 20 years ago, and that turned into one of the biggest businesses of all time.
This is the search that I did that made me the most afraid for Google.
This is hilarious- now when I google "90s movie about a boy prodigy that goes to an event with other prodigies and then shouts out an answer to a math question from the audience" the top result is the IMDb page for Gifted!!
The actual movie is Little Man Tate, yes? I've never seen it but that's what ChatGPT tells me. Zero Google results point there, but it is the #5 result for "related searches." I hate Google and even I am shocked how bad this is.
For me it’s not even theoretical. I’ve already wasted my time trying to sus out subtle garbage mixed into its answers. It’s the same reason I stopped using copilot. I don’t need or want that.
Funny, so far i've been just asking it things i normally ask Google. Same paranoia of answers that StackOverflow/etc give me, but far quicker and more responsive.
TBH i'm not sure under what scenarios people search into Google expecting perfect answers. Ie where we're disappointed by incorrect information. I filter through a dozen wrong answers on Google every day. Why is it different if it's from ChatGPT?
I guess one difference is that I have a lot (decades?) of experience in the types of errors humans tend to make. Humans have certain biases, certain blindspots, etc. In programming for sure, there are very strong patterns in the types of mistakes we make that tend to be a function of skill or experience. So you (I) look for those without even thinking about it. Violation of YAGNI, ACID, DRY, pre-mature optimization, etc.
Even before getting to an answer I can often judge how trustworthy or skilled the author/speaker is likely to be by all kinds of little signals or keywords in their speech.
ChatGPT (and Github co-pilot) disables that highly tuned error detection and correction experience. It makes unpredictable mistakes, which is peculiarly pernicious. I'm sure it varies by the temperament of the user. Personally I don't have the patience for that.
Good points. I think it also helps that i purposefully avoid asking for things that i can't validate. The easier it is to validate the more likely i am to ask.
I find ChatGPT especially great with giving starting points. It kinda feels like 10 (or something... time?) years ago when i discovered how good Google was at finding movies based on vague wishy-washy definitions. If i don't know what to call something ChatGPT does a better job at pointing me in the right direction. Often giving the right answer.
But i'm not saying it has lasting power. We'll see. So far i'm using Search engines less than ever.
On further reflection one of the major missing signals from ChatGPT is time. This is especially important for technology. When someone's explaining, giving an opinion, providing a solution -- I always consider that in the context of the date attached. ChatGPT gives us these answers that remove that.
I see completely the difference in your example, though, where you can easily and immediately validate the response. It definitely is way stronger at that than previous solutions.
I've found that if you push through the trust issue with the expectation that it's not a final answer, it's just getting you closer to it a lot more quickly, then that ends up being very helpful in a lot of cases.
The most illuminating thing about ChatGPT for me is just how terrible most programmers on HN apparently are. I thought it was just a funny meme that all we do is copy/paste from Stack Overflow but apparently that is literally what a lot of people are doing all day.
I don't think I'm so smart, what I am saying is the GP is being extremely dismissive of other engineers here on Hackernews. Every post about LLMs is full of engineers stating how these have been useful to them. GP simply asserts that these are all "terrible engineers." Do we think that's REALLY the case or in fact LLMs can be very useful, especially when for example learning a new language or framework they haven't used before.
Example: recently I was curious about cross-platform UI development so I used ChatGPT for research into different frameworks and made several apps in Java and .Net. Not being an engineer who's built many GUIs before, ChatGPT was amazingly useful for this and help speed my progress. Does this make me a terrible engineer?
some engineers != all engineers, that's classic unrepresentative sample. nice it helps you, hope you won't forget how to code in future, or you will be handicapped in case of openapi outage.
People who blindly copy from ChatGPT or Stack Overflow are most likely either very inexperienced or simply bad programmers. However, from where I see it, ChatGPT, Copilot, and any other such tools are fantastic at prompting you to think differently, getting you past writer's block, giving you ideas, or saving you some time googling for syntax yet again. You need a nudge to trigger recall of what you already know. It's a fantastic tool, just like an IDE.
It's sad how most programmers are so terrible they need IDEs to automatically create stubs for required methods or getters and setters. They should have to type all of that out every time, and from memory, too! VIM with no plugins is the only way to do it!
Or, you know, smart developers use tools appropriate to the work, and some of us have figured out how to use this new tool before you have. That's okay, you can catch up!
I was working on a personal project yesterday to answer some questions I had about how liquidity risk works for money market mutual funds, and to forecast/nowcast liquidity risk and NAV risk for a bunch of funds.
Mind you: I don't know the first thing about anything financial. I was just curious.
chatgpt gave me a bunch of sources of data that I wanted, translating my lay description of things I wanted to know into financial terms of art. I could then look up legal definitions and formulas for those terms to make sure they were what I thought they were. chatgpt also told me which SEC forms those things are disclosed on, what data brokers I could use for other data, etc.
between chatgpt and copilot I saved at least an hour on the job of pulling down historical data from EDGAR for a bunch of funds and getting the stats I wanted (I didn't know EDGAR existed until yesterday, and the xml/html/txt formats are kind of annoying... like, fine, but a bit of a pita so I'm glad I had help because ughhh is that kind of code boring and damn are LLMs good few-shot inductive parser generators!). Also wrote some nice chart.js code for me and helped with automatically collecting, searching, and extracting some key stats and terms from prospectuses. I didn't know about chart.js until yesterday.
All of this would've been possible without assistants, and required a lot of "executive function" on my part to bring together, but it seriously saved me at least a couple hours of implementation work and up to a day on research and learning terminology and regulatory stuff. Again, verification of those things is way easier when you know what words to look up definitions for. And chatgpt did make mistakes/hallucinate.
I don't find much use in my professional life, where the code I'm writing is apparently too domain-specific for copilot to be helpful and the mathematics is too complex for chatgpt to help with. Maybe in a few years. We'll see.
Do you have Plus? chatGPT was annoying when I was using the free version - I would have to keep logging in again and again and the service was unreliable.
Ever since I switched the Plus, it's (usually) always on and doesn't log me out.
Makes it much easier to use as a general search engine
I don’t spend that much time in it to even consider that, and so little time that signing in again is not a problem for me. I guess I don’t know what to ask of it.
Well, Microsoft forced me to use Edge to open Bing chat, so I did that and asked it "how can you use bing chat on chrome?". It pointed me to a chrome extension that allows that. See, it's useful.
Maybe because Google is actual trash and it went from being the best search engine to being an SEO-infested desert of a search, where your query is interpreted however they want not how you want.
Latest example from today - search for "micropython html parser" - 100% of results are about normal python not micropython, with the first 4 results being for generic paid programming courses. It's completely useless as a search engine now.
I'm of the opinion that it should be illegal for companies to offer the ability to buy ads for registered product names or trademarks.
Google, Amazon, and Apple all allow trademark and product name squatting. It's gross and abusive.
Generic terms? Sure. But actual trade names for products? That's extortion.
These services aren't helping in discovery if the customer already knows the name. They're merely forcing themselves into that relationship and taxing it.
You then have to buy n-many ads across m-many services just to keep you in front of your competitors that would squat you.
IMO, the services should just not buy the ads for their own name. It might hurt their bottom line a bit, but doing this makes the search engine shittier for the users who know what they want to see.
User input/frustration will eventually lead them toward alternative search engines, or provide enough feedback to Google/Bing/etc to make them disable registered service name or company names from being used as ad terms.
I've been using Neeva for a while now. It's ad-free and (for me, anyway) on par with Google. I have one premium account ($5/month) and one free account (used for work) and have had no issues with either. By default, searches aren't saved, although enabling it can lead to better personalized searches.
Kagi's one I've heard great things about as well, although I'm not a fan of a business model that allows me to pay for X searches before hitting a barrier that prevents further use of the platform (theirs is 200 searches a month for $5, which seems way too low based on my use cases).
In any case, users will force the change, not any government or corporate pressure. Businesses' best chance of fighting it is to refuse to play along.
They provide 100 free searches, 200 searches for $5/month, and 700 searches for $10/month.
If you run over your allotment, you pay 1.5c thereafter.
That's a barrier. That's exactly what I was talking about.
Edit: Admittedly, I didn't realize that they just bill for follow-on searches, I assumed they stopped you. That charge per follow-on search is even worse.
> I'm of the opinion that it should be illegal for companies to offer the ability to buy ads for registered product names or trademarks.
I'm of the opinion ads themselves should be illegal. Companies can "advertise" in a catalog I can choose to look at, but would no longer be allowed to invade my attention with their bullshit constantly.
We've let marketers ruin too many good things as it is, lets just kill off the entire industry for the good of society.
That's too extreme. The economy would fall apart and a lot of people would lose their livelihoods.
Think about second order effects. YouTubers. News stations. Newspapers. Film crews. Publishers. Writers. Marketers (who do more than just ads). All impacted.
Many of the products we use everyday would no longer be viable. Even the premium tiers are supported by ads.
Engineering salaries would be broadly impacted.
The economy is driven by hustle, sales, and ads. I'd rather live in this world than one than one that has no ads. We'd be less productive: less employed, less eager to buy, less eager to work. Maybe some aspire to that, but that society won't be building rocket ships and inventing immortality.
I do hate ads, but they're better than not having them. And I understand their utility and the overall ecosystem they support.
> Many of the products we use everyday would no longer be viable.
I think this is a bullshit myth the industry uses to justify itself. People are capable of paying for things, and to the extent they don't want to then maybe we really don't need whatever it was the ads were supporting.
I mean, a lot of the "content" exists purely as clickbait with a tiny bit of entertainment or information attached so it can product place or sell ads. Fuck it, we don't need it as a society.
>The economy is driven by hustle, sales, and ads.
Allowing jobs that make society worse for profit because "economy" is just the broken window fallacy. We are better off skipping the job part and just giving them money directly.
> We'd be less productive: less employed, less eager to buy, less eager to work.
We'd be doing bullshit jobs less. How is that a bad thing? Lets redirect what productivity we do feel like engaging in on making the world better instead of making the people at the top of the economic food chain wealthier and playing on people's insecurities to get them to pay for shit they don't need.
In my idea world, if I am interested in buying some new product or service I consult a catalog. Opt-in marketing, not bullshit carrot and stick marketing, and certainly not the kind of completely unsolicited bullshit that keeps us from having unfiltered email, answering our phones, or using the internet without an ad blocker.
I do believe I just saw an open letter from some Google staff to the CEO reiterating the need to uphold their "do no evil" model.
I can't imagine what it's like for the original employees to watch the company become what it has.
All the things that made Google so appealing, and gain market share, seem to have faded. It's a bit sad, to be honest. I remember being so excited to become a Gmail beta tester, back in the day.
Google can also return enough ads to push real search result below fold, while also making it as hard as possible to discern between real and paid results.
If you saw how much of Google's revenue comes from ad spending then it makes perfect sense why it is the way it is now. But it goes to show what being hyperfocused on revenue does to the user experience.
Bing is on it's way in that direction. Microsoft needed something big to make Bing relevant again. This could potentially be it.
Will they eat all of Google's lunch? Probably not, but if they can begin to chip away at the insane share (93%+) of the marketplace that Google controls, then it can only be for the betterment of the Internet.
I created aisearch.vip which searches using bing api, removes results containing ads or seo junk, and uses openai to summarize content
to void clickbait. I believe that this is the only way to get the 0-4 good results hiding in the top 20 results. It's paid because it costs to run. But to be honest I myself am a fan of what phind is doing. Just can't understand how they are able to cover costs yet
Microsoft is really still in their infancy of building out their end vision for AI in search. Right now it's roughly a chat bot that summarizes and gives references, without really tying into their core search well. Eventually it will negate the need to click through to many reference, then it will be better able to curate a list of alternative sources for what you seek. I think that's the big problem with AI as it stands now, people think we've already reached the end of the road, when really we're only at the beginning of it. Watch and see what the next decade of it looks like. I agree with Bill Gates that it all will fundamentally changed the way we see and interact with the world (for the better).
> Eventually it will negate the need to click through to many reference, then it will be better able to curate a list of alternative sources for what you seek.
Doesn't that just kind of describe google now though?
I poorly worded that. I meant to say it will be able to answer most questions, but also refer you to the best sources for information that it can't provide.
I've been impressed with what Kagi is able to dig up for me on old and sometimes niche electronics test and measurement equipment. However, I don't think they are targeting the same set of users that early Google was. Among other things, you have to pay for Kagi.
No, but there are other search engines that aren't any worse than Google.
And, apparently depending on the style of your search terms or what you're searching for, one or two are better. I get noticeably better results from DDG, but I know that a lot of other people don't. I can only speculate that the difference must be search terms or topics.
That's unfair to altavista. Altavista never got evil, they just became irrelevant. (And even when they were irrelevant, they still had a search syntax that I miss. You could search for things like term_a NEAR term_b.)
Amen. Buried, unknown to most, Google used to have a similar AROUND(n) syntax. No idea if it still is respected.
Google seems to have refocused around returning any result out of a set of popular links rather than deep dives. It feels like I must quote every term I enter now.
I have noticed that too. Google has become so useless. And with all the little knowledge stuck behind reddit dark patterns, discord locked and utterly unsearchable world. And the modern forum stuff a mess of slow JavaScript that loads on demand. It's like finding knowledge is a fine sand fallings between the fingers. YouTube has some bit of interesting content, but it's 10min for 30s of info plastered with sponsored content. And finding something is harder and harder. ChatGPT and the like won't resist long becoming the same.
My only hope is that it will eventually bring back the notion of trust. Naybe finding a programming job will be harder outside of your network of friends. But won't require silly interviews anymore. Maybe we will get back to lifelong tenure at work?
Anyways I am just rambling here. What do I know anyways.
I just performed that search in Google and the top two results are links to PyPI and Snyk packages for micropython-html.parser. The third result is a link to python.org documentation for html.parser (not micropython).
Bing returns the result for PyPI and the one for the non-micropython python.org result.
I don't see a big difference here (except maybe me learning about Snyk through google and not Bing?).
I mean, this is a big part of the issue. Of course the issue isn't reproducible; Google is trying to do the search based on their model of what you want back. Sometimes you benefit from that, other times it pushes the results you want way down. I just tested this and there were some micropython related results in the top ten, but the majority were for beautiful soup with no mention of micropython.
> Of course the issue isn't reproducible; Google is trying to do the search based on their model of what you want back...
There's a problem here though, which is that there is no way for me or any other reader to discern whether the reality of the situation is "it's not reproducible and the commenter is misremembering basic facts about their search results due perhaps to some unrelated frustrations" and "it's not reproducible but the commenter's account is accurate."
I completely share the views about Google's search results being terrible, but it's still possible to overstate how bad they are. And I do find it fairly difficult to believe that Google would show some people paid programming courses as the top 4 results for that query when much more relevant results clearly exist and everyone attempting to reproduce this gets those more relevant results.
nor are we anywhere near being able to inspect how google is doing all this.
if only there were some way to somehow be able to share that information with everybody, some technology so that we can all access such potentially useful information about how google is functioning right now
sorry about the snark, but it's a vent for accumulative frustration from seeing a worsening trend in this regard.
Kafka would recognize this trend; computing is truly becoming a "bureaucrat's best friend".
I often find that frustrating about Google search, but I'm also unconvinced it's a problem that a more-AI-driven approach will avoid. Seems like the idea to try to convince the model to interpret things differently based on user history is still likely to get pushed by product managers, and still technically capable. And then "no tell me about the framework" "conversationally" vs tailoring Google search terms seems a bit of a wash.
I'm about 80% sure that the reason why Google became so bad for my searches is that they added AI elements to it, to make it try to figure out what I'm searching for instead of taking my word for it.
In that sense, I'm not sure more AI would improve the situation. OTOH, better AI just might. There doesn't seem to be a "no AI" option on offer.
Based on my usage of ChatGPT, I think that a more AI driven approach could help a lot. With ChatGPT if it misinterprets what I'm asking for, I can make a clarifying followup query.
> Of course the issue isn't reproducible; Google is trying to do the search based on their model of what you want back.
There could also be other things that make it non-reproducible as well.
It sometimes feels like search just returns whatever it has on hand that seems somewhat similar to your search. It may then do an asynchronous request to pull into cache results from a deeper backend search, but by the time the frontend gets them you've already been served your results. Subsequent searches from other people for the same terms would then get better results than you did, so this sort of optimization would on average improve search quality even if quality is poor for the first searcher.
I have no idea if search engines actually do this. But it seems to explain some of their more mysterious behavior, like immediately returning a bunch of results that have nothing to do with your search terms.
That sounds plausible. There could be some service aggregating search results from multiple systems, and if one of them is slow (maybe breaking some SLA), it would be omitted.
I guess there are probably a hundred Google employees reading this who know whether it’s accurate and we’re just speculating.
It's been publicly known that Google has been bubbling search results for years now, and that one individual's search results have no relation to another's.
Every individuals everything has no relation to another's - which I think is a core cause of society's disconnects. Not that people got along before algorithms drove everything, but it's certainly been a destabilizing factor.
Google can incorporate quality signals in near real time. Those signals include leaving and immediately returning to the SERP. So us discussing these particular results could have already influenced them.
That's actually super impressive. So there's basically no "index" anymore, in the sense that the results returned are almost "dynamically" generated? And I'd guess that all of that still needs to be pretty cheap, computationally speaking.
I've seen this before - I've complained on HN that google has no results for X, and within two hours google's first(and only result) was the very thread of me complaining about it.
Google is more than capable of doing such basic fingerprinting, but do you have evidence Google uses it for search results? You can do a trivial experiment on another Google platform: open some video on YouTube, leave it, and look at the front page. Then, look at the YouTube front page in a private/incognito tab. (Alternatively, use two different private sessions.) In my experience, no matter what device or network I use, and no matter what the video was, the first page will always show videos related to the one I opened, and the second will always show an extremely generic set of clickbaity videos (likely based on GeoIP). This suggests that Google uses ordinary cookies for basic relevancy ranking.
Considering people switch to Incognito Mode to get different results, it'd be weird and not particularly commercially savvy to frustrate them and feed their paranoia by serving them the same ones. Non-Google search engines and browsers exist.
I'm also reminded of a CTO complaining that his test sessions suggested the default Google ads on a page with little text content all related to dating. We pointed out that that from Google's point of view, this was probably a sensible ad-targeting decision for a user running an Incognito mode browser (or any browser without tracking cookies)...
> Considering people switch to Incognito Mode to get different results, it'd be weird and not particularly commercially savvy to frustrate them and feed their paranoia by serving them the same ones. Non-Google search engines and browsers exist.
I just installed Google Chrome on a fresh Windows 10 VM to repeat the experiment outside of Incognito Mode. First, I opened a video. Then, I checked the front page and confirmed that related videos had been added. Then, I closed Chrome, reset the VM's state to before I had visited YouTube, and checked the front page again. (I don't think this scenario would be too indicative of a prying user; I can easily imagine corporate systems that regularly wipe out browser data.) Again, the related videos were replaced with generic videos.
I then tried the same thing, except by clearing browsing data through Chrome's UI instead of resetting the VM. The results were the same.
Obviously, these observations could all have been manipulated by sufficiently conspiratorial fingerprinting. But the simplest explanation, in the absence of good evidence to the contrary, is that the site uses and respects browser cookies for its recommendations. Thus my request.
This is a good argument for LLM search: have search be conversational, and let you say "hey, I actually meant this!". Then you could have different chat environments, each tailored to micropython results, CPython results, IronPython results, etc.
All you need is for Google to get worse and Bing's marketing to get better (which it has, by nature of being associated with the biggest tech hype train of recent memory).
> As an AI language model, I cannot infer emotions or intentions with certainty, but the statement does not appear to use typical markers of sarcasm. The commenter is expressing an opinion about SEO and its impact on search, and while it is framed as potentially controversial, it does not necessarily come across as sarcastic.
It totally did! It’s the Eternal September effect on the internet as a whole. Marketers climbing all over each other, desperate to steal the eyeballs of the naïve masses.
Are you for real ? I mean, pick better examples if you want to poke something, since:
Search for "micropython html parser" gets you https://pypi.org/project/micropython-html.parser/ as the first result with the quote: "This is a module ported from CPython standard library to be compatible with MicroPython interpreter."
Um, no. Google "bubbles" their search results, meaning that they customize them based on their profile of you. This is not new - it's been implemented and known publicly for years.
Yeah I just vpn'ed to my machine at work and I'm getting results about micropyhton too. But on my personal machine it absolutely thinks I'm interested in buying some generic programming courses and nothing about micropython.
I agree, but the part I’m failing to see in the conversation is what the future looks like.
If the chat approach is gaining visits, then ad companies will follow, so what then?
The chat offers up the same seo ad laden tripe but further obfuscates it, and alternatives, while caging it in a conversational tone?
Is that better? Worse?
If these chat offerings only cause search to step back and attempt to recreate the early days of usable search then they would appear a success in my eyes.
> where your query is interpreted however they want not how you want
Exactly, I feel it is because with Google advertisers are the real customers and they could not care less about user experience unless it affects that bottom line. Case in point is the removal of the dislike button which users loved but advertisers hated. Google is a rich one trick pony right now anyway.
It is wild how most of modern SEO and "SEO+" involves just outright creating spam. Spamming url strings (keyword stuffing essentially) for location specific ranking is a new one I've noticed being touted.
I'm curious too. When I go to stack overflow, I get lots of clues as to the trustworthiness of a given answer. I don't know how I'd get any clues from an LLM that its solution is outdated or suboptimal in some way
I don't feel stack overflow has been rendered obsolete because the snobbishness and moderation people complain about so much has caused the recommendations on stack overflow to generally be of higher quality than those ChatGPT offers and you have more metadata (answer age, votes) about which answers are especially trustworthy.
I can't tell you how many times I search a somewhat niche compound term and the first thing in the results page is a list of shops in my area keyed on some word in the search.
I trieD this in both DDG and Google (using the !g macro with DDG).
DDG first page of results all pointed to zoom.us links. Google (mediated through DDG !g) contained all sorts of spam links to sites like subdomains of uptown.com.
Perhaps it was a matter of me not paying enough attention, but it seemed like Google degraded slowly over time, and then suddenly all-at-once. Or in other words, it seems like its become unusable for me entirely within the past year.
> Maybe because Google is actual trash and it went from being the best search engine to being an SEO-infested desert of a search, where your query is interpreted however they want not how you want.
Agreed but tbf... every major search engine seems to suffer from the same.
I cannot recommend enough the 'ContextSearch web-ext' extension. It lets you easily search using any search provider. I constantly (1) highlight text (2) right-click (3) select search provider (from youtube to reddit to bing) (Note: my default is DDG):
But you literally typed it in the same way that a scammer would word it and then not expect it to find a scam site?? Nice bait, ever considered being a journalist?
You don't seem to be very familiar with users that happen to be older. However, that search is uncomfortably common. Impersonation and scamming of their service is a problem, but when they help the user to the impersonation site.. that's on them.
They stopped caring about what information you want to get and instead give you the information they think you should get. Sometimes it works, sometimes it's just awful.
Not to mention the constant pain that's localization/internationalization as well as some of the queries becoming worse and worse. The service is supposed to become better with time, not the contrary. Plus, of course, all the SEO garbage.
It's absurd how Google will simply ignore the keywords in my search to feed me whatever it's trash algorithm decided fits my search.
And the decline in Google Maps has been shocking. Maps has gone from always reliable to being "trust it if you like wasting time and burning fuel" bad.
I don't know what the hell happened at Google, but it ain't pretty.
It’s even worse. I’ve been using ChatGPT a lot in situations that’d be Google searches before and I love how you can fill in so much detail and it will actually provide you with better results. With Google, additional detail is actually used against you. With enough keywords, it will match just about any site to your query.
Hah it wish I kept this search I did once. I was searching something about lsd. It somehow decided to replace it with "acid" but then showed results about related to pH lol
I don't find Bing's search any better than Google, but importantly, I don't find it any worse.
GPT4 is going to do what Hawaii five-o, Bing Rewards, integrating windows search with Bing, and setting Bing as the default search engine for Edge on Windows never could. Make people use Bing over Google.
I've already set it as my default search engine across my devices as I move to cut Google out of my life entirely. I simply don't feel the need to use any of their services anymore. Their moat is gone.
The only downside I have noticed is that Bing Maps is a pretty bad service relative to Google Maps or Apple Maps and I will go out of my way to avoid using Bing Maps. Good mapping being integrated into a search engine is actually a pretty big deal since it's nice to use a search engine as a front-end for a mapping service, so this is a significant weakness.
To be fair, you could always surround micropython in quotes so that Google knows to only show results containing the keyword "micropython". Only caveat is that Google will still show you whatever it wants in case no results contain the keyword. I swear a decade ago it would simply tell you no results were found, but now you need to (1) use quotes in keywords, and (2) click on "show only results containing [keyword]" when it can't find any results yet decides to show you unrelated pages anyway.
I can't believe how disconnected I always feel on the Google search conversations here. It still does what I need it for fine, I don't remember it being significantly better than it is now. I guess I'm not a power user? But I also think this is how the most people feel. On this site Google has been a shockingly useless zombie for years.
Since google search results are tailored to your particular history and profile, it may be as simple as you fitting into Google's way of doing that better than others.
The decline started way back when they stopped letting you find mp3's and mp4's of movies and albums. I get that copyright infringement is bad and all, but their engine was fully capable of you typing a obscure phrase, and being able to get web results. Somewhere behind all the filters a lot of powerful searching capabilities just died.
If I had unlimited money, I'd make a fresh search engine, zero filtering to start outside of blatant spam sites, and go from there. Focus on making results be powerful.
It's a bit strange, to me Google seems to be what it has always been, but if you read hacker news it's 'actual trash'.
It's the same with amazon, I purchase from there on a weekly basis and it goes well, but you read the comments on hacker news and it's apparently unusable now.
> Latest example from today - search for "micropython html parser" [..]
Are you entering the micropython search term in quotes in the Google search box, because I've found that's a moderately good way to get rid of the normal Python results.
My search terms for that search would probably be:
Microsoft needs to integrate GPT into Windows, and make it available in every app and input box - in the mail app, in Discord, in browsers.
You would open Chrome, go to Hacker News, login, and say in the comment box - "GPT, please read this page and make a funny but insightful post about how GPT is a danger to all of us".
In the mail app: "make an excuse to the manager for why I will be late at work today".
Might be wrong, but for a long time search results from Google and Bing in blind tests were basically equivalent. Google’s edge has been its brand recognition and consumer search habits.
This is a case study for an undergrad statistics or responsible journalism class.
* for traffic see the small note "all values rebased to 100"
they are likely hiding the significance of the increase
* for the app downloads graph:
does the 30x gap say anything?
are there seasonal reasons that can explain why every Jan1--Feb4 has more DLs than Feb4--Mar11 ? e.g., new phones?
It's also as if google decided to suicide themselves a bit earlier. Their results have really become crap lately. It keeps ignoring anything than 1-2 keywords. What's going on?
the thing that annoys me most is when it gives a list of completion suggestions & when I not-so-critically select one of them, only to realize it has changed something in the original search terms. Seriously F*k Google for that.
I personally use much less google since Bing Chat came out.
That being said it makes me worry a lot. Not that GPT is going to replace me or something, but how effectively it can serve ads to me if it wants. If MS decided to do evil with Bing Chat (which is almost inevitable), it would make today's Google look like a charity.
Bing’s search product (ignoring the chatbot) is quite good and comparable with Google. I (and I assume, many people like me) would have never tried it if not for the Bing Chat hype. I find myself using Bing Search more than Google or Bing Chat now.
The broader question may not be whether its Bing or Google. People may not be searching anymore to find the sites which can potentially answer a problem/query - instead they may just want the answer. GPT/LLms may even slow down the whole search engine industry or make it more like chatgpt versions of it.
Sad day for me. Bing actually gives you points to search on their engine. I just cashed in a bunch of points for a $100 gift card. My guess is this program will go away if Bing becomes more popular.
That was bound to happen at some point, AI or not. Google search quality has been declining more and more every year, to the point that I'm almost sure that you could build a better and smaller index for much cheaper with a curated domain list.
AI brought down the price to compete from billions to a few millions, the monopoly is being threatened.
598 comments
[ 2.9 ms ] story [ 323 ms ] threadNews title at its finest, the "near 1%" must not make it to the title for maximum effect.
I rarely use google, and when I do it's from duckduckgo !g
I would immediately switch to bing by default if such bangs were supported as I sometimes need them for wikipedia !w or amazon !a
If there was a way they could be supported Edge address through a plugin or something, I would immediately switch to bing as that's what I already use 90% of the time (between duckduckgo frontend and requesting bing directly with !b )
Signed in, installed Edge, etc.. It seems like I'm just stuck on the waitlist.
The weakness is revealed. Google’s stickiness now has a quantifiable wedge factor. Whether it goes to Microsoft or someone else is more a matter of time.
It takes more resources per query to run AI-based search than a search engine. Everybody offering ChatGPT type systems is either pay per view or heavily throttled. We may be in the last days of free search as the dominant product.
A personalized AI assistant goes WAY beyond that. Whenever you talk to it, it can go into salesman mode to con you into buying shit you don't need.
Surely the economics will work out to still provide "free" searches.
Only for people with significant spending. Amazon Prime customers, for example.
Mark my words.
I have not used it at all, and this is a reminder that I should probably uninstall it. I have access to ChatGPT, Bard, Alpaca etc. and there’s very little reason at this point to pretend to use Bing in order to get a crack at Chat.
There are several drawbacks like a more strict system pre-prompt, 15 messages limit for conversations, and some kind of 24 hours ratelimiting if you use it too often.
Want an LLM-enhanced search? perplexity.ai is actually pretty cool
Want to generate images? There are a ton of free Stable Diffusion sites (for example you.com has that built into their chat), or run it yourself.
Want a neat chat experience? ChatGPT is free, Bard appears to have a shorter waiting list than Bing, and Alpaca runs on regular hardware.
Bing doesn’t have a big enough moat to force me into using their app. The LLM space is legit competitive and a company can be ahead in the morning and left in the dust by the afternoon. In my opinion, this is what’s happening with Bing at present.
Who knows though? Bing might make Chat open while I’m writing this sentence and I’ll have to adjust my opinion accordingly.
- when scrolling down the page goes to Bing Search
- when typing predictive suggestions show up
I would use Bing even more if I knew how to disable these things. Does anyone know if this is possible?
Because google is 99% spam for me
But I hate seeing the top half of the screen are ads for huge chains, and then the top results are blog posts about "Where are the top 10 places to eat in your city" and not a list of restaurant web pages.
Chat search tools are only useful if (when?) they achieve a 99.99% reliability (I'm fine with one mistake per 1000 searches).
All the same things that the model makes an effort to “understand” are things that can contribute to your own understanding. When it “understands”, you do not. You get the summary and miss out on a great amount of nuance that can come with learning and finding the right answer for yourself.
For instance, it is very common that in the search for why X is happening in my code, I will find tons of information that don’t answer my question but help me form an understanding of why it is happening, how the system is working, etc.
This is especially visible in science. Many papers outright conflict with each other. Some have better or worse methodologies than others. Some have better analyses. Some state outright falsehoods without citation or misinterpret citations. Having a GPT “understand” this to produce understanding in humans is going to end very badly. It takes several hours for a human (in my experience) to understand even a single academic paper on its own.
I suppose the appropriate cliche is “it’s the journey, not the destination”.
> Suppose I was an evil person and wanted to eliminate the curiosity of children. Give the kid a diet of Google, and pretty soon the child learns that every question he has is answered instantly. The coolest thing about being human is to learn, but you don’t learn things by looking it up; you learn by figuring it out.
Contrast kids of today who can just go and play video games, no disecting of computers required. Naturally, most of them don't understand or appreciate computers as anything more than just another appliance.
Not everyone had your childhood.
I use it various times a week for cooking. E.g. "basic recipe bechamel sauce in metric"
ChatGPT: Within a couple seconds I get the ingredients in metric and a step by step guide.
Google: The provided summaries are useless to me since they are not in metric. So I have to click on a link and then start scrolling around all the noise of history of bechamel sauce, anecdotes about someones grandparents, pretty pictures, etc to find the actual information. Without ad blocker it would be an even bigger nightmare.
Now you have to engage in a conversation and get the 'a feel lucky' version where before you got to look through different more visual recipes
There is a difference whether I'm looking for interesting, more complicated dishes I haven't done before, or fairly standardized, simple recipes I have done many times and remember the technique but not things like ratios.
For the former I use a select few recipe sites or youtube channels directly. Google is miserable for discovery in my opinion. For the latter I use either my own recipe notebook or more recently ChatGPT as well.
How do I turn off this feature in my car?
How do I replace the fan in my fridge’s compressor?
Why is windows repair not working in this scenario with this PC?
This tool is going to be incredibly deflationary in many services geared around repair when now it can tell you almost exactly what to do and soon it’ll be able to produce videos or images of each step on demand.
As long as they're not hallucinated media...
“Authoritative title that references my exact search” -> page body restates my question, gives an oddly reworded summary of Thing in Question #1 -> “some people online have said” followed by snippets lifted from the ether -> repeat for Things 2,3,etc -> “our suggestion is to look at the features and decide based on what other customers/users are saying”
It’s perfected vapidity wearing the skin of an editorial review site and takes a LOT more cognitive load to suss out the phoneyness compared to the last time the SERPs were packed with spam (~10 years back).
Its just like anything else. Trust but verify. Some will just trust, others never will... People probably over-trust the tech now, but just like wikipedia it will be approached a little more carefully as people learn its shortcomings.
Unless you're trying to perform open heart surgery or doing something intrinsically dangerous, its probably going to be additive. I wouldn't trust it to help me assemble a warp engine, but it may be able to give me some decent pointers for guitar technique or how to change my oil.
For many people, Google results have gone from being an absolutely amazing demonstration of what could be done if one had the resources to crawl and process the entire internet to a complete waste of time at best, clogged with spam and nonsense and misinformation. I've been using Duck Duck Go for years now, and recently my middle-schooler switched to Duck Duck Go because I got answer to a question she had already asked Google before checking with me. Some people even pay for Kagi search, because Google is just that bad.
So LLMs don't have to be perfect to be better than the alternatives, not even close.
Small correction - from what I have heard in our community, most of our users pay for Kagi search, not becuase Google is just that bad, but because Kagi is just so good. In other words we are making the value proposition be "best search in the world", one worth paying for.
(Kagi founder here)
The day that ChatGPT launched, I described the what it was and it came back with, "that's called an infiltration basin."
That's when I knew Google was in trouble.
Ooh! What was the answer to this one?
I tried to sign up for ChatGPT, but the verification text never arrives.
(No, it's not a soft phone. I've had the same phone number on the same cell carrier for over 20 years.)
I love that movie! I had the same experience a few years ago of trying to remember its name.
This is a great example of something ChatGPT does amazingly well at. I did almost the same thing myself recently, trying to find the name of a show featuring an actor in a different show or something. A straight-up search returns nothing for it; you have to look up the actor's name, then look at his shows, then try to remember the name. But a search with ChatGPT found the result easily.
I tried a "similar" thing now. I was thinking of the movie Gifted, and wanted to see if I could get ChatGPT to pick it up, here's my query:
> I'm trying to remember the name of a movie about a child, featuring an actor that was in a marvel movie.
And the result:
> There are many movies featuring actors who have appeared in Marvel films, so it's a bit difficult to pinpoint the exact movie you're thinking of without more information. However, one possibility could be "Gifted" (2017), starring Chris Evans, who is well-known for playing Captain America in the Marvel Cinematic Universe. In "Gifted," Evans plays the role of a man raising his young niece, who turns out to be a child prodigy. If you could provide more details about the plot or other actors, I'd be happy to help narrow down the list further.
This is obviously a tiny use case, and really not worth thinking about. Who cares, right? I thought the same thing about search in general 20 years ago, and that turned into one of the biggest businesses of all time.
This is the search that I did that made me the most afraid for Google.
The actual movie is Little Man Tate, yes? I've never seen it but that's what ChatGPT tells me. Zero Google results point there, but it is the #5 result for "related searches." I hate Google and even I am shocked how bad this is.
TBH i'm not sure under what scenarios people search into Google expecting perfect answers. Ie where we're disappointed by incorrect information. I filter through a dozen wrong answers on Google every day. Why is it different if it's from ChatGPT?
Even before getting to an answer I can often judge how trustworthy or skilled the author/speaker is likely to be by all kinds of little signals or keywords in their speech.
ChatGPT (and Github co-pilot) disables that highly tuned error detection and correction experience. It makes unpredictable mistakes, which is peculiarly pernicious. I'm sure it varies by the temperament of the user. Personally I don't have the patience for that.
I find ChatGPT especially great with giving starting points. It kinda feels like 10 (or something... time?) years ago when i discovered how good Google was at finding movies based on vague wishy-washy definitions. If i don't know what to call something ChatGPT does a better job at pointing me in the right direction. Often giving the right answer.
But i'm not saying it has lasting power. We'll see. So far i'm using Search engines less than ever.
I see completely the difference in your example, though, where you can easily and immediately validate the response. It definitely is way stronger at that than previous solutions.
If you don’t have enough insight to understand how GPT4 could be useful for engineers you’re not as enlightened as you think you are.
Example: recently I was curious about cross-platform UI development so I used ChatGPT for research into different frameworks and made several apps in Java and .Net. Not being an engineer who's built many GUIs before, ChatGPT was amazingly useful for this and help speed my progress. Does this make me a terrible engineer?
Or, you know, smart developers use tools appropriate to the work, and some of us have figured out how to use this new tool before you have. That's okay, you can catch up!
I was working on a personal project yesterday to answer some questions I had about how liquidity risk works for money market mutual funds, and to forecast/nowcast liquidity risk and NAV risk for a bunch of funds.
Mind you: I don't know the first thing about anything financial. I was just curious.
chatgpt gave me a bunch of sources of data that I wanted, translating my lay description of things I wanted to know into financial terms of art. I could then look up legal definitions and formulas for those terms to make sure they were what I thought they were. chatgpt also told me which SEC forms those things are disclosed on, what data brokers I could use for other data, etc.
between chatgpt and copilot I saved at least an hour on the job of pulling down historical data from EDGAR for a bunch of funds and getting the stats I wanted (I didn't know EDGAR existed until yesterday, and the xml/html/txt formats are kind of annoying... like, fine, but a bit of a pita so I'm glad I had help because ughhh is that kind of code boring and damn are LLMs good few-shot inductive parser generators!). Also wrote some nice chart.js code for me and helped with automatically collecting, searching, and extracting some key stats and terms from prospectuses. I didn't know about chart.js until yesterday.
All of this would've been possible without assistants, and required a lot of "executive function" on my part to bring together, but it seriously saved me at least a couple hours of implementation work and up to a day on research and learning terminology and regulatory stuff. Again, verification of those things is way easier when you know what words to look up definitions for. And chatgpt did make mistakes/hallucinate.
I don't find much use in my professional life, where the code I'm writing is apparently too domain-specific for copilot to be helpful and the mathematics is too complex for chatgpt to help with. Maybe in a few years. We'll see.
Ever since I switched the Plus, it's (usually) always on and doesn't log me out.
Makes it much easier to use as a general search engine
Latest example from today - search for "micropython html parser" - 100% of results are about normal python not micropython, with the first 4 results being for generic paid programming courses. It's completely useless as a search engine now.
Google, Amazon, and Apple all allow trademark and product name squatting. It's gross and abusive.
Generic terms? Sure. But actual trade names for products? That's extortion.
These services aren't helping in discovery if the customer already knows the name. They're merely forcing themselves into that relationship and taxing it.
You then have to buy n-many ads across m-many services just to keep you in front of your competitors that would squat you.
User input/frustration will eventually lead them toward alternative search engines, or provide enough feedback to Google/Bing/etc to make them disable registered service name or company names from being used as ad terms.
I've been using Neeva for a while now. It's ad-free and (for me, anyway) on par with Google. I have one premium account ($5/month) and one free account (used for work) and have had no issues with either. By default, searches aren't saved, although enabling it can lead to better personalized searches.
Kagi's one I've heard great things about as well, although I'm not a fan of a business model that allows me to pay for X searches before hitting a barrier that prevents further use of the platform (theirs is 200 searches a month for $5, which seems way too low based on my use cases).
In any case, users will force the change, not any government or corporate pressure. Businesses' best chance of fighting it is to refuse to play along.
If you run over your allotment, you pay 1.5c thereafter.
That's a barrier. That's exactly what I was talking about.
Edit: Admittedly, I didn't realize that they just bill for follow-on searches, I assumed they stopped you. That charge per follow-on search is even worse.
I'm of the opinion ads themselves should be illegal. Companies can "advertise" in a catalog I can choose to look at, but would no longer be allowed to invade my attention with their bullshit constantly.
We've let marketers ruin too many good things as it is, lets just kill off the entire industry for the good of society.
Think about second order effects. YouTubers. News stations. Newspapers. Film crews. Publishers. Writers. Marketers (who do more than just ads). All impacted.
Many of the products we use everyday would no longer be viable. Even the premium tiers are supported by ads.
Engineering salaries would be broadly impacted.
The economy is driven by hustle, sales, and ads. I'd rather live in this world than one than one that has no ads. We'd be less productive: less employed, less eager to buy, less eager to work. Maybe some aspire to that, but that society won't be building rocket ships and inventing immortality.
I do hate ads, but they're better than not having them. And I understand their utility and the overall ecosystem they support.
I think this is a bullshit myth the industry uses to justify itself. People are capable of paying for things, and to the extent they don't want to then maybe we really don't need whatever it was the ads were supporting.
I mean, a lot of the "content" exists purely as clickbait with a tiny bit of entertainment or information attached so it can product place or sell ads. Fuck it, we don't need it as a society.
>The economy is driven by hustle, sales, and ads.
Allowing jobs that make society worse for profit because "economy" is just the broken window fallacy. We are better off skipping the job part and just giving them money directly.
> We'd be less productive: less employed, less eager to buy, less eager to work.
We'd be doing bullshit jobs less. How is that a bad thing? Lets redirect what productivity we do feel like engaging in on making the world better instead of making the people at the top of the economic food chain wealthier and playing on people's insecurities to get them to pay for shit they don't need.
In my idea world, if I am interested in buying some new product or service I consult a catalog. Opt-in marketing, not bullshit carrot and stick marketing, and certainly not the kind of completely unsolicited bullshit that keeps us from having unfiltered email, answering our phones, or using the internet without an ad blocker.
I can't imagine what it's like for the original employees to watch the company become what it has.
All the things that made Google so appealing, and gain market share, seem to have faded. It's a bit sad, to be honest. I remember being so excited to become a Gmail beta tester, back in the day.
https://www.wired.com/1999/09/arrest-of-a-web-pioneer/
(I had the misfortune of working for him )
Will they eat all of Google's lunch? Probably not, but if they can begin to chip away at the insane share (93%+) of the marketplace that Google controls, then it can only be for the betterment of the Internet.
Doesn't that just kind of describe google now though?
And, apparently depending on the style of your search terms or what you're searching for, one or two are better. I get noticeably better results from DDG, but I know that a lot of other people don't. I can only speculate that the difference must be search terms or topics.
For what it fails to find well I just prepend !g to my query and have it bounce me over to Google.
Google results are more often then not trash now, so I don't think the bar is too high for competitors now.
It does, yes, but not only Bing. At least report, it aggregates from about 400 different sources, and they also run their own crawler.
They used to have a list of the other engines they use, but I can't seem to find it anymore.
If I search for something like "micropython html parser" I expect the top results to say "Missing: micropython | Must include: micropython".
Yes Google, it was the first term I entered for a reason don't cha think?
Google seems to have refocused around returning any result out of a set of popular links rather than deep dives. It feels like I must quote every term I enter now.
My only hope is that it will eventually bring back the notion of trust. Naybe finding a programming job will be harder outside of your network of friends. But won't require silly interviews anymore. Maybe we will get back to lifelong tenure at work?
Anyways I am just rambling here. What do I know anyways.
Bing returns the result for PyPI and the one for the non-micropython python.org result.
I don't see a big difference here (except maybe me learning about Snyk through google and not Bing?).
There's a problem here though, which is that there is no way for me or any other reader to discern whether the reality of the situation is "it's not reproducible and the commenter is misremembering basic facts about their search results due perhaps to some unrelated frustrations" and "it's not reproducible but the commenter's account is accurate."
I completely share the views about Google's search results being terrible, but it's still possible to overstate how bad they are. And I do find it fairly difficult to believe that Google would show some people paid programming courses as the top 4 results for that query when much more relevant results clearly exist and everyone attempting to reproduce this gets those more relevant results.
if only there were some way to somehow be able to share that information with everybody, some technology so that we can all access such potentially useful information about how google is functioning right now
sorry about the snark, but it's a vent for accumulative frustration from seeing a worsening trend in this regard.
Kafka would recognize this trend; computing is truly becoming a "bureaucrat's best friend".
In that sense, I'm not sure more AI would improve the situation. OTOH, better AI just might. There doesn't seem to be a "no AI" option on offer.
There could also be other things that make it non-reproducible as well.
It sometimes feels like search just returns whatever it has on hand that seems somewhat similar to your search. It may then do an asynchronous request to pull into cache results from a deeper backend search, but by the time the frontend gets them you've already been served your results. Subsequent searches from other people for the same terms would then get better results than you did, so this sort of optimization would on average improve search quality even if quality is poor for the first searcher.
I have no idea if search engines actually do this. But it seems to explain some of their more mysterious behavior, like immediately returning a bunch of results that have nothing to do with your search terms.
I guess there are probably a hundred Google employees reading this who know whether it’s accurate and we’re just speculating.
It's been publicly known that Google has been bubbling search results for years now, and that one individual's search results have no relation to another's.
I'm also reminded of a CTO complaining that his test sessions suggested the default Google ads on a page with little text content all related to dating. We pointed out that that from Google's point of view, this was probably a sensible ad-targeting decision for a user running an Incognito mode browser (or any browser without tracking cookies)...
I just installed Google Chrome on a fresh Windows 10 VM to repeat the experiment outside of Incognito Mode. First, I opened a video. Then, I checked the front page and confirmed that related videos had been added. Then, I closed Chrome, reset the VM's state to before I had visited YouTube, and checked the front page again. (I don't think this scenario would be too indicative of a prying user; I can easily imagine corporate systems that regularly wipe out browser data.) Again, the related videos were replaced with generic videos.
I then tried the same thing, except by clearing browsing data through Chrome's UI instead of resetting the VM. The results were the same.
Obviously, these observations could all have been manipulated by sufficiently conspiratorial fingerprinting. But the simplest explanation, in the absence of good evidence to the contrary, is that the site uses and respects browser cookies for its recommendations. Thus my request.
Of course, how can you scale this for every user?
"No, you didn't" "I didn't say that" "Seems like your are using on old version"
Search for "micropython html parser" gets you https://pypi.org/project/micropython-html.parser/ as the first result with the quote: "This is a module ported from CPython standard library to be compatible with MicroPython interpreter."
And why should random searches make my results significantly worse than the default? That doesn't exactly absolve google of screwing up.
People just invalidating each other instead of the third party because a third party creates different experiences
The ads are personalised to you though
Google provides exactly what you're looking for ..
If the chat approach is gaining visits, then ad companies will follow, so what then?
The chat offers up the same seo ad laden tripe but further obfuscates it, and alternatives, while caging it in a conversational tone?
Is that better? Worse?
If these chat offerings only cause search to step back and attempt to recreate the early days of usable search then they would appear a success in my eyes.
Exactly, I feel it is because with Google advertisers are the real customers and they could not care less about user experience unless it affects that bottom line. Case in point is the removal of the dislike button which users loved but advertisers hated. Google is a rich one trick pony right now anyway.
> The goals of the advertising business model do not always correspond to providing quality search to users.
- Sergey Brin and Lawrence Page, The Anatomy of a Large-Scale Hypertextual Web Search Engine
Now I use ChatGPT for almost all my programming questions. Good riddance Google.
DDG first page of results all pointed to zoom.us links. Google (mediated through DDG !g) contained all sorts of spam links to sites like subdomains of uptown.com.
Crazy.
Agreed but tbf... every major search engine seems to suffer from the same.
I cannot recommend enough the 'ContextSearch web-ext' extension. It lets you easily search using any search provider. I constantly (1) highlight text (2) right-click (3) select search provider (from youtube to reddit to bing) (Note: my default is DDG):
Firefox: https://addons.mozilla.org/en-US/firefox/addon/contextsearch...
Chrome: https://chrome.google.com/webstore/detail/contextsearch-web-...
This is one of the sites it shows you for "log in to g mail"
https://clinmedjournals.org/articles/blogs/gmail-login.html
Not to mention the constant pain that's localization/internationalization as well as some of the queries becoming worse and worse. The service is supposed to become better with time, not the contrary. Plus, of course, all the SEO garbage.
And the decline in Google Maps has been shocking. Maps has gone from always reliable to being "trust it if you like wasting time and burning fuel" bad.
I don't know what the hell happened at Google, but it ain't pretty.
GPT4 is going to do what Hawaii five-o, Bing Rewards, integrating windows search with Bing, and setting Bing as the default search engine for Edge on Windows never could. Make people use Bing over Google.
I've already set it as my default search engine across my devices as I move to cut Google out of my life entirely. I simply don't feel the need to use any of their services anymore. Their moat is gone.
The only downside I have noticed is that Bing Maps is a pretty bad service relative to Google Maps or Apple Maps and I will go out of my way to avoid using Bing Maps. Good mapping being integrated into a search engine is actually a pretty big deal since it's nice to use a search engine as a front-end for a mapping service, so this is a significant weakness.
If I had unlimited money, I'd make a fresh search engine, zero filtering to start outside of blatant spam sites, and go from there. Focus on making results be powerful.
It's the same with amazon, I purchase from there on a weekly basis and it goes well, but you read the comments on hacker news and it's apparently unusable now.
What gives?
Are you entering the micropython search term in quotes in the Google search box, because I've found that's a moderately good way to get rid of the normal Python results.
My search terms for that search would probably be:
"micropython" html parser
You would open Chrome, go to Hacker News, login, and say in the comment box - "GPT, please read this page and make a funny but insightful post about how GPT is a danger to all of us".
In the mail app: "make an excuse to the manager for why I will be late at work today".
* for traffic see the small note "all values rebased to 100" they are likely hiding the significance of the increase
* for the app downloads graph: does the 30x gap say anything? are there seasonal reasons that can explain why every Jan1--Feb4 has more DLs than Feb4--Mar11 ? e.g., new phones?
That being said it makes me worry a lot. Not that GPT is going to replace me or something, but how effectively it can serve ads to me if it wants. If MS decided to do evil with Bing Chat (which is almost inevitable), it would make today's Google look like a charity.
You realize that Google Search still brings in mountains of cash, right? Search generates 57% of Alphabet’s total revenue. [1]
We’re extremely far from that point.
AI brought down the price to compete from billions to a few millions, the monopoly is being threatened.