Live data from Hacker News

Is Google’s 20-year search dominance about to end?

economist.com

281–290 of 550 posts

Re: Is Google’s 20-year search dominance about to end?

#281
post #221

Earlier quoted context omitted.

Similar statements have been made about most dominant technology when it was in its infancy. You are not wrong, but it can quickly go from novelty to necessity once folks start to rely on it.

Do you see yourself relying on it? Not when you know it’s KNOWN to give wrong answers. It’s also worth mentioning that Google PageRank was not the first search engine. No, it was good 4-5 years after the first usable search engine came to market. I think the same will be true for AI chat.

It excels at conversational style pedagogy when exploring new topics.

Re: Is Google’s 20-year search dominance about to end?

#282

It's been a while since they stopped innovating. Although Google, Maps, Youtube are of daily use they are monetized exclusively by advertising which is annoying and hated by many. It has been many years since Google has launched an innovative new product. I don't think ChatGPT will gain daily traction after this hype. Anyway we could say that MSFT and AMZN have demostrated more power to innovate with different busine…

Ads aside, the UI of Youtube and Maps is atrocious. There is so much room for improvement here, but it will never happen from Google. It's just not in their DNA. Unfortunately, they have such a huge moat here I doubt we'll see anything better for a long time.

What things about maps and YouTube’s UI annoy you?

I find YouTube quite easy to use, esp the shortcut keys.

Recommendation system is another story.

Re: Is Google’s 20-year search dominance about to end?

#283
post #8

While Google certainly has its faults and I am frustrated with its trajectory, I would bet money that this is an article pushing the nonsensical notion that AI language models will replace search engines. Edit: based on the subtitle, it seems like I am correct.

How is that nonsensical? Seems quite likely to me, especially if they can ground them better in a factual knowledge base. Most search engine queries are looking for information.

Most of the information people are looking for isn't really the sort of thing an LLM is set up to cover though. If I want information on anything since 2021, or to look into stuff near me, or access a price, a download, or information in visual form, or anything that's too niche for its corpus, ChatGPT is not good. When ChatGPT can offer an answer it's fast, but about as reliable as a search engine's spammiest result

And answers as text wrapped around a factual knowledge base just sounds like the sort of queries Google Home already handles in natural language form. That can certainly be improved upon, but it isn't obvious that the LLM engineers will beat the search engineers to it...

Re: Is Google’s 20-year search dominance about to end?

#284
post #95

Earlier quoted context omitted.

> What are people expecting out of Google that it's not delivering? Not optimize for "most documents indexed" but "highest quality of results". One of them encourages adding spam to their index, the other encourages removing spam from their index.

There are giant teams of people at Google whose only job is trying to define things like "spam" and "seo abuse" more precisely in more diverse contexts. There are equally large armies of people outside of it trying to outwit the first group and find more and more elaborate loopholes and workarounds. HN is constantly pushing this notion that "spam" is some well-defined, solvable problem, so obviously Google wants it.…

What's the reason for these giant teams of presumable well-paid experts and geniuses to not define "a clone of SO with some of the answers juggled around between questions" as spam?

> The notion that more click bait improves Google's bottom line is absurd

If you don't find what you're looking for on the first try, you'll need to try again, and see more ads. What else are you going to do, go elsewhere, visit a library, ask the town elders or give up on looking for things you want to know? You don't have a choice, you know it, they know it.

I find it equally plausible that Youtube's search sucks badly because they don't care what you're looking for, they want you to watch videos that they predict will lead to the maximum time spent on the site, again so you watch more ads. What other explanation is there that the world's leading search engine has the search of one of their flagship products run at 1999 quality? Presumably they have giant teams of people working on that too?

Re: Is Google’s 20-year search dominance about to end?

#285
post #271

Earlier quoted context omitted.

Google stopped being innovative the second their "20% Time" got gutted: > Employees were encouraged to spend up to twenty-percent of their paid work time pursuing personal projects. The objective of the program was to inspire innovation in participating employees and ultimately increase company potential. For Google's part, Gmail and AdSense both arose out of side projects. In 2013, Google discontinued 20 percent tim…

“What has Google released since 2013[0]? Stadia?” - GCP Suite (a full integrated and consistent suit, answering so many things) - Google Workspace (trust me when you get to know everything that is possible you will be amazed) - Tensorflow and other Machine Learning applications

Google Workspaces originally released in 2006 (inc. Sheets, Docs, Drive was 2012, Calendar was 2009, etc), it was rebranded "G Suite" in 2016 then rebranded "Google Workspaces" in 2020. Rebranding an existing product isn't innovation.

Google Cloud is a worse clone of AWS, is that innovative? Redmond-style photocopying?

I'll give you TensorFlow, but we're talking about that being their death knell, so congratulations about signing their own death-warrant?

Re: Is Google’s 20-year search dominance about to end?

#286
post #127
post #95

Earlier quoted context omitted.

There are giant teams of people at Google whose only job is trying to define things like "spam" and "seo abuse" more precisely in more diverse contexts. There are equally large armies of people outside of it trying to outwit the first group and find more and more elaborate loopholes and workarounds. HN is constantly pushing this notion that "spam" is some well-defined, solvable problem, so obviously Google wants it.…

This discussion is pretty sad and frustrating for me. I think your conclusion (“Google has no motive to encourage spam content that rises to the top using SEO tricks”) is about as convincing as it gets, but it’s completely drowned out by a very vague sense that google is a big company and therefor it’s making search worse on purpose. No amount of well-worded HN posts are gonna sway people, imo… Luckily HN posters don…

Have you looked at Youtube comments recently? It's a mess. There's no organization, half of the replies are censored but are still being counted, so you have plenty of "1 Reply" links but nothing shows up when you toggle. Half of the top level comments are completely useless comments that are only posted to help the creator "beat the algorithm", every successful creator now has a call to action to write a comment.

I see two options: a) Google can't do any better than that, b) Google has a reason to keep it in the current state (I'll put "Google doesn't know because nobody at Google has used Youtube in the last 5 years" and similar options under "a").

a) sounds ridiculous, b) sounds conspiratorial. What are the other options?

And again, I'm not saying they are making search worse on purpose (no "from now on our core mission is to make search suck"). I'm saying they aren't optimizing for SERP quality. They seem to care about index size (maybe it's an internal KPI? would certainly explain their aggressive guessing at additional URLs that you might have on their page but don't link to, don't add in sitemaps etc, and their stubbornness in keeping results from the index even if they've been 301ed or 410ed ages ago (they do get downranked after a while though)), but I assume that they mostly care about paid ad clicks, and if something increases ad clicks while the result quality decreases, they'll do it.

Re: Is Google’s 20-year search dominance about to end?

#287
> For generative ai to make real money, it will need to find uses in “navigational” searches (looking for a site’s internet address by its name) and especially “commercial” ones (“Best new ski boots this season”).

Serve users what they want, deals to hunt for: Apple Maps. Leverage Foursquare and Yelp lessons.

Re: Is Google’s 20-year search dominance about to end?

#288
post #221

Earlier quoted context omitted.

Do you see yourself relying on it? Not when you know it’s KNOWN to give wrong answers. It’s also worth mentioning that Google PageRank was not the first search engine. No, it was good 4-5 years after the first usable search engine came to market. I think the same will be true for AI chat.

It excels at conversational style pedagogy when exploring new topics.

How can you say that with a straight face? It’s a regurgitation of things it’s collected from the WWW. It is a step up from Siri/Alexa but completely lacks actual conversational skills.

Re: Is Google’s 20-year search dominance about to end?

#289
post #288

Earlier quoted context omitted.

It excels at conversational style pedagogy when exploring new topics.

How can you say that with a straight face? It’s a regurgitation of things it’s collected from the WWW. It is a step up from Siri/Alexa but completely lacks actual conversational skills.

You've just described most actual human conversation on the internet.

Re: Is Google’s 20-year search dominance about to end?

#290

Earlier quoted context omitted.

I'm actually expecting AI training data to turn into a snake eating its tail. I've been spending the last month not sleeping so I could try to get up to speed on deep learning, and my understanding is that AI trained on AI generated output becomes crap surprisingly quickly, and AI generated content is already starting to proliferate. I have no idea the extent to which this will hinder new models being generated, but…

> AI trained on AI generated output becomes crap surprisingly quickly Yet GANs work quite well

Sorry, I figured in the context of transformer based language models it was contextually clear I was talking about those
Post reply on HN