Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Perplexity got ads

twitter.com

Post reply on HN
Print view View original

281–290 of 369 posts

Re: Perplexity got ads

#281

Post by tomrod »

relativ575 wrote: post #116
tomrod wrote: post #56

Earlier quoted context omitted.

Ads are myopic and signal lack of ability by a company to be viable.

Really? Ad is the most popular business model for Internet businesses. Any you saying none of those companies is viable?

That's correct.

Turn off the ads (which is visual and aural pollution in 99.99999% of cases) and said company is literally not viable.

We accept ads because we're conditioned to, not because they represent a real business. The vast majority of advertising spend is inefficient and wasteful in terms of ROI. Most advertising in the space show substantially low statistical power, in effect being unable to identify an effect from random noise.[0] This means ROI measures are normally bunk.

Ad-supported businesses are subject wholly to the whims of unrelated companies and people outside their target client base.

We worry about AI slop, rightfully, yet do little to nothing to stop the memetic, exploitative mind virus advertising (and adtech especially) is.

Recommended reading: Existence, David Brin

[0] https://academic.oup.com/qje/article-abstract/130/4/1941/191...

Re: Perplexity got ads

#282

Post by TylerE »

wrasee wrote: post #272
xigoi wrote: post #200

Earlier quoted context omitted.

According to Kagi, they can’t make the price much lower and still be profitable.

If they get 100x the customers on a $2 tier than a $5 tier then yes, they can be profitable. The question is, could they get 100x? Or even 10x? I don’t know, and I assume they’ve done their homework and believe that they can’t. I just wonder, what if they’re wrong?

It if supporting each customer costs them, say, $2.50

Re: Perplexity got ads

#283

Post by currymj »

wrasee wrote: post #272
xigoi wrote: post #200

Earlier quoted context omitted.

According to Kagi, they can’t make the price much lower and still be profitable.

If they get 100x the customers on a $2 tier than a $5 tier then yes, they can be profitable. The question is, could they get 100x? Or even 10x? I don’t know, and I assume they’ve done their homework and believe that they can’t. I just wonder, what if they’re wrong?

for each Kagi search they hit APIs of other search providers etc., so they need to cover that cost.

Re: Perplexity got ads

#284

Post by tomrod »

lmm wrote: post #61
tomrod wrote: post #56

Earlier quoted context omitted.

Ads are myopic and signal lack of ability by a company to be viable.

Google is the world's fourth most valuable company.

I agree. Society is sick due to the exploitative nature of Google and other ad-tech companies in relation to ad-tech.

They could choose to be viable in other ways.

Re: Perplexity got ads

#285

Post by re5i5tor »

dewey wrote: post #259
stavros wrote: post #217

Earlier quoted context omitted.

I run out of the 300 searches/mo in around ten days. Everyone raves about Kagi but I've paid for it around three times now, and it's never been better enough than DDG to justify the $10/mo for me...

Very different experience for me, I’ve tried to switch to DDG many times and always switched back to Google after being disappointed. Using Kagi for a year now and only a handful of times I’ve checked if Google would give better results.

Same for me. Also like and regularly use Quick Answer feature (similar to Perplexity).

Re: Perplexity got ads

#286

Post by hoppyhoppy2 »

nickpsecurity wrote: post #265
i_am_proteus wrote: post #230

Earlier quoted context omitted.

The "right way"[0] to use Kagi is the unlimited search tier ($10/mo) and then aggressively rank domains in searches. Within a few weeks, the search quality improves tremendously, because I have filtered out the domains responsible for low-quality results, and up-ranked the domains I find to be most useful. [0]My opinion. Probably true.

There’s also the idea of vote with your wallet. What we spend can be a tool to change the landscape. Enough people paying $10/mo. for Kagi might keep a good, ad-free option in business. Similar to why many people pay for OpenOffice and Ubuntu. They want them to keep existing.

FYI OpenOffice is pretty much a dead project; nearly all development effort shifted years ago to LibreOffice.

Re: Perplexity got ads

#287

Post by fweimer »

scarface_74 wrote: post #276
fweimer wrote: post #221

Earlier quoted context omitted.

I think it's a bit of a cultural defect that we consider a product with 29K (supposedly satisfied) users a failure. More problematic (for Kagi) is that it's not a free-standing engine (“Our search results also include anonymized API calls to all major search result providers worldwide” per https://help.kagi.com/kagi/search-details/search-sources.htm... ), so it's still dependent on the weird business practices around…

These two issues are directly correlated. > I think it's a bit of a cultural defect that we consider a product with 29K (supposedly satisfied) users a failure. > More problematic (for Kagi) is that it's not a free-standing engine (“Our search results also include anonymized API calls to all major 20K subscribers can’t support the creation of a quality search engine.

Fair point. Creation might not be such a challenge, but sustaining operations might be. On the other, hand I feel like the amount of content that would be beneficial to me in search results has not grown as fast as computing and networking cost decreased, so maybe it gets easier over time.

Re: Perplexity got ads

#288

Post by nickpsecurity »

hoppyhoppy2 wrote: post #286
nickpsecurity wrote: post #265

Earlier quoted context omitted.

There’s also the idea of vote with your wallet. What we spend can be a tool to change the landscape. Enough people paying $10/mo. for Kagi might keep a good, ad-free option in business. Similar to why many people pay for OpenOffice and Ubuntu. They want them to keep existing.

FYI OpenOffice is pretty much a dead project; nearly all development effort shifted years ago to LibreOffice.

That’s the one I was talking about. Thanks for the correction. :)

Re: Perplexity got ads

#289

Post by ericd »

nunez wrote: post #104
ericd wrote: post #33

Just a reminder that you can run a fairly capable model like qwen2.5:72b locally on an 128 gig M4 Max Macbook Pro.

Lol who has a $5k computer lying around like that?

Well, I work on this stuff, but I’m mostly sharing this to spread awareness that you don’t need a multimillion dollar rack of nvidia gpu machines to do inference with surprisingly powerful models these days. Not that long ago, you’d need a much more expensive multi-kilowatt workstation to run this sort of thing at a useful speed.

Re: Perplexity got ads

#290

Post by ericd »

tomjen3 wrote: post #192
ericd wrote: post #33

Just a reminder that you can run a fairly capable model like qwen2.5:72b locally on an 128 gig M4 Max Macbook Pro.

Or you could buy a slightly less insanely expensive model, split the savings in two, and get many many years of credits on OpenAI.

The thread is about ads in language model results. Do you think OpenAI is going to resist putting ads into their results for the next few years? I hope they do, but I wouldn’t say it’s a given. You definitely won’t get (intentionally injected) ads on a model you run locally.
Post reply on HN