Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: What would it take to unseat Google?
19 points by janalsncm 4 months ago | hide | past | favorite | 30 comments
There have been a ton of articles about the decline of Google's search quality. Just the other day Planet Money released an episode that talked about a longitudinal study of Google search quality: https://www.npr.org/2024/06/13/1197965227/google-search-quality-algorithm-documents-leak

Here are some other examples:

https://future.a16z.com/the-future-of-search-is-boutique/

https://dkb.io/post/google-search-is-dying

https://nypost.com/2022/05/01/google-critics-say-ads-spam-sites-are-killing-search/

https://www.theatlantic.com/ideas/archive/2022/06/google-search-algorithm-internet/661325/

https://freakonomics.com/podcast/is-google-getting-worse/

This is certainly a popular topic in the media, and maybe we can even say the reports of Google's death are greatly exaggerated. An important consideration is the fact that SEO has also gotten a lot more sophisticated, so the number of high quality articles on any particular topic might not be that high.

But my question is, if we accept the premise that 1) Google search is getting worse and 2) Google is responsible for this, what would it take to unseat them?




I think the most likely way Google dies is if web search itself dies.

Maybe because something else replaces the "web" and Google failed to catch up. Much like how Microsoft monopolised desktop, but failed to capture the rising (and now dominant) mobile market.

Maybe AI agents become reliable enough to a point where a vast majority of people looking for information will reach out to ChatGPT instead of searching on the web.

Maybe all of the things we used to do on the web, we instead do in a handful of apps that Google has no access to.

To be honest I'm not going to bet on Google dying off anytime within my lifetime.


Let me give you some thoughts from a shifted point of view:

Techy people here on HN can easily circumvent Google Search result issues, the real harm is for the average user.

Ways to deal with terrible Google Search results:

1. For complex computer problems, Google search is still has the biggest index, and less affected by SEO spam.

2. When using Google as substitute for a websites builting engine, google dorks get the job done.

3. Modern browsers allow you to switch enginees on the fly from the address bar (eg. Brave Browser has :w which instantly switches to Wikipedia) and there are similar keywords for opening bookmark.

4. If I know where I want to go, I go there directly using those shortcuts.

5. When looking for inspiration or quality stuff to read I search hn.algolia.com, use Brave Search Goggles or any other "custom" browser enginees that give results for specific topic.

6. Ask ChatGPT or some other Chatbot. (Results differ, but it works sometimes)

7. I have collected by own sites and bookmarks over time.

The only thing I don't know how to search are Product Reviews.

I think "general" search enginees are in decline, and (techy people) are gking back to pretty much manual indexing.


Blogs used to be good for product reviews, now most I see are infected with affiliate links. I think YouTube video reviews and Reddit reviews are the as close to being "trustworthy" as you can get online these days. I prefer video reviews because it's easier to spot any embellishing when it's right in front of you.


I have user rtings.com for product reviews. Limited items of course but quality reviews.


Google was unseated for me by Kagi. I find the results and overall user experience is much better. However, since it isn’t a free service, I’m probably in the minority of people who are willing to pay for search.

I’ve been telling pretty much anyone I think will be interested about them to try and get people to see there is life without Google.

That said, I still use YouTube, and also use Google Maps for looking up businesses… though I tend to use Apple Maps for actual turn by turn directions in the car. My Gmail account has been relegated to my junk email.


1) Do ads better

2) Kill YouTube (somehow)

3) Realize that if your product isn't free, it's not competing with Google.

The hard part is reconciling all of these things at once. Google wins because they know people like low barriers-of-entry. Type in a search query and get "instant" results mixed with "instant" advertising. Delve into a YouTube video before getting sucker-punched by an ad a minute or two later. Fulfilling and monetizing our knee-jerk desires is what makes and keeps Google successful.


YouTube has two things that I think make it hard to unseat:

1. It is very popular, people publishing there can pick up views for basically free. People go there to find videos. There are lots of eyeballs that you can pick up just by showing up.

2. It has ready-to-go monetization. Most major publishers will supplement YouTube ads with direct sponsors and merch sales, but having a baseline income available is a major benefit. Also people blame YouTube for the annoying ads, not you.

Both can be worked around, but you are fighting a major uphill that doesn't really have anything to do with the technical challenge.


The most important difference between YouTube and any other video entertainment service is that they do not pre-audit content creators and content. Anybody and everybody can publish there and start building their audience. You don't have to be from the right family, have the right political connections and right political opinions, or do unspeakable sexual acts for casting managers. You can be independent, and unlike Instagram, Facebook and etc, you can make money on your work.


Android is also big, if someone comes out with really great smart glasses before existing big tech can, then they might have a chance to be part of big tech

Another advantage Google had was timing, they are now so vertically integrated, such as custom asic for YouTube encoding tasks, that it becomes quite capital intensive to compete


Needn't do a thing. As you alluded, they're on a trajectory to irrelevance all on their own.

It started the moment Google stopped putting the user first.

That opened up an unparalleled opportunity for someone to do better, and they will. We've already seen flickers of this across their other products, with competitors besting them in cloud infrastructure and AI.


It would hardly accomplish the unseating, but influencing people to stop using it even if just for search, would be helpful.

Duckduckgo is the only thing that has come close to doing this, but insignificantly. Unfortunately, while DDG seems less sinister than Google, it is also less functional - I find it one of the worst engines for most things. I really miss Scroogle!

Google is a monster, and as much as many would love to see it go away, it's here to stay without some unprecedented paradigm change. Thankfully the work of some few provides some options around it. I try to support this, though am frequently perceived as odd for doing so.

If it helps any, Amazon is shitting its own formidable pile of waste upon the internet. I can't find an article on any consumable product without wading through dozens of blogurgitations exalting the magnificent cloaca of Lord Bezos - regardless of search engine!


Have you tried Kagi? I am very happy with it. I went Google -> DuckDuckGo -> Kagi in the last 6 years.


I was under the impression that kagi is a paid service. I don't mind what others do, but I'm personally opposed to paying what I consider extortion fees to avoid being smothered in toxic waste. As a bit of an idealist, I hold that some things should be free, notably access to freely extent digital data.

That said, even if I've mistaken, I frequently see positive reports of kagi.

Thanks for the recommendation.


You're not paying for access though. You're paying for curation. Access to any given site is free (ignoring the cost of internet access itself) assuming you know the URL. What is not free is someone taking your query and finding sites that might be what you want. You pay in currency or attention. Take your pick.

We all need to think about paying for the internet because literally none of it is free. Not even the hobby blog on a random desktop in someone's basement. It still costs them that bit more in electricity. If no one will pay then we all deserve to drown in ads because that's what we say we wanted with our non-dollars.


I do not deny that there's validity to your points. For me, it's the wrong interpretation. Again, I understand the points, but the way I gaze into the future reflects an ugly picture if we allow ourselves to go this direction. We're immersing ourselves in a service (no production) economy which could easily result in a nightmare situation where too many people are trying to survive by creating problems to solve and every individual's primary existential purpose is to be economically, perpetually gnawed at by fellow parasites. The entire concept of currency is problematic and I foresee such things severely exacerbating it.

I won't hold you back, but I will myself. And don't perceive my words as a challenge, insult or anything negative - it's just something that makes, perhaps, too much sense to me.


Are you saying that "the entire concept of currency is problematic", and therefore you want to get stuff for free, even though you know that those who provide this stuff do it in exchange of money?

Sounds like saying that in order to solve poverty, you steal everything... your reasoning makes absolutely no sense to me.


>What would it take to unseat Google?

Same old thing.

You've got to have more real capital than Google has at their disposal, and you've got to deploy it irrationally in their markets for longer than they can afford to remain solvent.


The question is: how can we organize the world's information better than Google?

I've designed a potential solution over the past 12 years, and just launched the beta this week.

It's called the World Wide Scroll: https://scroll.pub/wws.html

If you had all the world's best information on your local machine, you would need Google a lot less.


This won't work because you haven't (as far as I can tell) solved the chicken or egg problem. There's no reason to post content there without readers and no reason for readers to show up without content that isn't yours. Also, it seems entirely centralized.

If I have good content why would I post it on your system when I could create a website of my own and probably get similiar traffic but have complete control, or I could post it on exiting walled gardens and have a decent shot at way more traffic in exchange for similiar control? Your proposal is the worst of both worlds.

And don't say that you'll avoid controversy about giving up control by just letting people post whatever. Best case, you'll end up like Cloudflare and have controversy anyway. Worst case, you'll end up as a magnet for quite unsavory content which will then scare away other content.


>> as far as I can tell

The second derivative is the key metric.

>> it seems entirely centralized.

The `wws.scroll` file (our version of the Root Zone Database) is ultimately controlled by me for the time being (and the people I've given access to approve pull requests), but that just contains meta data on the folders. All of the _content_ is decentralized.

>> If I have good content why would I post it on your system

The WWS has no exclusives. We want you to post your publications to the web as well (and any other place you wish).

Our language and system is designed to help you make the best possible publications you can make.

Anything that makes publications worse we reject. Paywalls? No. Internet connection required? No. Ads? No. Copyright? No. Cookies? No. Closed source? No.

If you love your ideas and want to see them thrive, if you put them before yourself, then our platform is the place for you.

If you have a deep love for ideas--writing, math, music, art, film, data, hardware, software, science, poetry, et cetera--there is no better place for you than the World Wide Scroll.

I am perfectly okay if this rules out 99 out of 100 people.

We want people with deep conviction who truly love their craft and want to do their best work and see their ideas have the biggest positive impact on the world.


With a LLM. Case closed


WWS could be a great source of training data for LLMs.

Regardless, it will be a great place for meaning and connecting with other curious minds.


Apple's deal with ChatGPT could lead to a significant decrease in Google searches. Queries going to ChatGPT 4.o+ from Apple devices could directly lead to less usage of Google -- a result that could compound significantly over time.


I think a couple of generations, or never. IBM still exists, for example.


Already replaced by chatgpt for me. I use it for all search. Rarely use google.

And is google's AI solution going to work out? Momentum is not on their side.


Google = public infrastructure since its ubiquity. If its recognised as such - perhaps there will be a shift in the tide.


The same thing that it'd take to unseat all the other major tech companies at the same time.


Yahoo still has more traffic than Bing, but nearly everyone agrees that Yahoo is dead.


I can say that I'm fed up with Google's current performance, I'm becoming increasingly disappointed with Google, I can feel that Google has to take advertising revenue into account, this is an unchangeable fact. Therefore, I now use more AI-based search engine tools to find information. Although this is the case, I have to find multiple tools to verify the accuracy of the information, this may be the only reason why I can't completely trust AI.

I often use the following three AI search engines (Perplexity.ai / Felo.ai / you.com).

For research and academic purposes, I often use Perplexity and Felo. If it is for general information retrieval, I will use You. However, for searching information from other countries, I mainly use Felo, its performance in bilingual search is the best. Of course, Perplexity is also an important choice, this is the biggest change in the past year.


Build a better Google.




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: