Hacker News Reader

Comment by crazygringo | Hacker News Reader

So many things about this don't make sense.

1) A proper, popular search engine is incredibly expensive to run. You can't do it for free. It's going to have to be ad-supported. But ads are the very opposite of Apple's brand.

2) And financially, they'd trade their incredibly lucrative deal with Google for such a high risk? It just doesn't seem like something a board would approve.

3) The timing of this feels very suspicious, though maybe that's just sudden journalist interest rather than leaks... but if you were Google, this is the best thing that could happen to you lawsuit-wise. Can't you imagine a conversation one or two years ago? "Hey Apple, OK we'll pay you the $8-12 billion, but you also have to spend a little chunk of it -- just $10 million, really -- to credibly claim you're 'developing' your own search engine. Don't ever deliver it, just always be 'working' on it. Cool? Awesome, thanks."

View on HN

1# It's not expensive as 2 decades ago, most of the data is behind gardens now, Reddit, FB, Twitter, Amazon, and less than 1000 other sites. Everything else can be crawled slowly.

2# Search is one of THE primary usage of the device, they have enough throwaway money to take this on. Just like Maps.

3# Sounds suspicious. But anti trust discussion started few years back and if you look through those optics EVERYTHING will look suspicious this year. There is no non-suspicious time for one of biggest companies in the world/human history to compete with each other.

The main cost of search isn't crawling.

The main cost of search is developing incredibly sophisticated ML algorithms for parsing search queries, for ranking results, and parsing web pages to provide factual results (like Google's snippets and knowledge base items).

And this is orders of magnitude more expensive than it was two decades ago. Because you've got to compete with Google which has a 20-year head start in throwing the best engineers in the world at these problems.

Nobody's going to use a search engine that isn't at least 90% as good as Google in these things. See: Bing.

How do you explain Apple building Maps then and forgoing a huge chunk of money they made there?

Also no reason they couldn’t sell ads so long as they don’t use personalization, people almost forget that invasion of privacy and ads need not be coupled.

Click tracking is MUCH more significant than ML/AI, Google results were very good even 10 years back when they were just based on click tracking. Whatever they are doing to parse/steal other websites data to show as snippets is only complicated because there is no schema. Look what Amazon did, they made people implement "skill" so it's all API now.

All that was solved for majority of use cases and leaped ahead by Amazon in under 5 years with Alexa, deployed as tiny hardware.

Remove click tracking and top 1000 sites from Google and it will be shit, no amount of ML/AI dance will solve that. Apple can just do top 1000 sites from each country with click tracking (to track which result users are preferring) and significantly cut traffic sent to Google. They already try to do this with Siri.

Bing was chicken(user) and egg problem (click tracking to optimize results).

I'm sorry, but nothing of what you're saying makes any sense to me.

All click tracking does is measure the quality of your results -- whether a user clicks on the first result or third or none at all. It doesn't help generate quality results in the first place. Google's main first innovation was PageRank, not click tracking. Click tracking is just a metric to judge the underlying algorithms.

(There is one exception: it can be used to fine-tune the order of the top 5-10 results for the most popular queries, but of course most of search is in the long tail where that fine tuning can't be done. And in any case, you still need excellent candidates for your first 10 items to begin with.)

Click tracking does nothing for parsing a query, nothing for ranking the top 10 items out of a million matches, and nothing for extracting information.

And why are you talking about the top 1,000 sites? My whole point is crawling the web is the easy part.

> 1) A proper, popular search engine is incredibly expensive to run. You can't do it for free. It's going to have to be ad-supported. But ads are the very opposite of Apple's brand.

They could also use ads, just in a more subtle and privacy friendly way like the search ads (https://searchads.apple.com/ or previously iAd: https://en.wikipedia.org/wiki/IAd) in the App store already.

>1) A proper, popular search engine is incredibly expensive to run.

Apple's recent testimony before the House revealed that they have already spent Billions of dollars developing Apple Maps.

https://www.reuters.com/article/us-usa-tech-antitrust/u-s-te...

Could be the other way around though.

What happens if the Justice Department finds Google has a monopoly, and somehow decides that Apple can't use Google's Search Engine anymore? I mean, it's a bit of a long shot and a weird idea, but it's not outside of the realm of possibility. Google could get themselves into all kinds of trouble in this investigation.

If you're Apple, investing a few tens of millions now so that you have a half-decent search engine ready in case Google goes away is a smart investment. And if nothing happens, you quietly shelve the project.

> decides that Apple can't use Google's Search Engine anymore?

I don't think it would happen like that. What might happen is that Apple users would be presented with the choice of what search engine they want to use.

It's like that on Android except Google Search and Chrome are already installed and the default, but you're presented with a choice of other search engines and browsers (might be a EU thing though).

> But ads are the very opposite of Apple's brand.

They have ads in the App Store already. Search ads are probably the one kind of ad they could make work, since you can sell keywords without tracking and profiling your users.

Keyword ads are significantly less likely to be clicked on than tracked advertising, because keyword ads are less targeted. This is fact confirmed by Google’s and Facebook’s success at selling targeted advertising. If Apple sells keyword ads their search product will never be able to generate as much revenue as Google because the ads will be less effective.

I thought it was the other way around? Keyword ads on search results are by far the most profitable since you have intent right there ("what is the best lawn mower"), whereas the vague stalker-ish amazon ads about buying another one of the thing you just bought (or just trying to guess "well homeowners in their 30's might be interested in a lawn mower") have way lower value.

Especially on Google Search results, the top ads for brand names are usually for the first search result anyway and people click on them because they're first, making google lots of "free" money as people have to protect the top spot for their own brand

It's more that some tracking is important to result quality. For example, if I search for "eclipse" on Google, I get 5 results about the IDE, but my wife gets local maps & youtube videos about eclipses.

Parent comment was about Apple not being able to do it because of massive expense, if apple can make any profit that's not true, they don't need to make Google profit. I wonder if they would be able to make enough to cover Googles payment to be the default, though.

There’s ~1 billion Apple products and ~10 billion annually in payment from Google to Apple. Apple would need to make $10/yr per device in order to replace the revenue from Google. That’s a fairly achievable goal.

Of course, there are more complications than that simple figure, but it does provide a order of magnitude estimation.

> "A proper, popular search engine is incredibly expensive to run. You can't do it for free. It's going to have to be ad-supported."

Not necessarily. Firstly, you might not want to search the whole internet anyway, given how much clickbait spam, link farms, blackhat SEO, and whatnot there is now, thanks to the damage the advertising-driven search model has done to the internet. And secondly, there are alternative search funding models, e.g. subscription fee or listing fee. FWIW, that's the premise behind my side project - search just the good stuff, with no direct adverts, and result pages which contain adverts clearly flagged and heavily downranked.

> listing fee

Can you elaborate on this?

A basic listing for free, but a paid-for listing with additional benefits, e.g. greater control over indexing such as the ability to trigger a reindex on demand.

I'm wondering if they are not planning to just sell it as (of course) privacy focused unique search experience for monthly fee or as part of package with most expensive device or other streaming service.

Yes, that is the way I would see it, and, I personally, would be willing to pay a reasonable (low) monthly fee for such a service.

Is a search engine that much more expensive than mapping?

  A proper, popular search engine is incredibly expensive to run

Is it? Google rakes in a lot of money, and so has the rope to publicly hang itself with silly, expensive initiatives.

That doesn't mean the Search product category is inherently expensive; it means Google has more money than it knows what to do with.

Reply on news.ycombinator.com