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

DuckDuckGo removed the ability to filter search results

github.com

Post reply on HN
Print view View original

241–250 of 341 posts

Re: DuckDuckGo removed the ability to filter search results

#241

Post by doubtfulcan »

sMarsIntruder wrote: post #209
sMarsIntruder wrote: post #202

Earlier quoted context omitted.

I know HN is against Bitcoin by design (it’s destroying the planet!11), but I really miss the opportunity to pay with Lightning TBH. It’s fine for me having an account, but not using a credit card linked to my name. I found a 2022 thread where someone asked to pay with BTC or Monero, but still. No news.

[flagged]

It's probably more about the very generalizing statement at the beginning that isn't adding anything to the conversation. The core of the issue seems to be that Kagi doesen't support anonymous payment, which kills it in my eyes unfortunately.

Re: DuckDuckGo removed the ability to filter search results

#242

Post by laniakean »

MagicMoonlight wrote: post #223
DanHulton wrote: post #70

I debated bringing up Kagi at the top level here, since I already commented on it in a reply, but honestly I really want them to succeed and shit like this is why. It's a paid search service, and I'm honestly really happy with that distinction and what it entails. Being paid means that they can deliver on and support features that users can use to get the best results for them, without worrying how those features may…

200 searches a month and then having to pay extra per search is a hilariously bad offering. This isn’t 2008 mobile contracts, I don’t want usage caps for basic functionality.

I recently gave Kagi a try after using DDG, and signed up for their $10/month plan. However, I found myself running out of searches within the first 20 days, which left me feeling hesitant about making any additional searches due to the extra cost. While I do want to support Kagi's success, I personally found the cost of $10-15 per month for search to be a bit steep. Perhaps I have unrealistic expectations on how cheap search should be, but for now, I'm not sure if the cost is worth it for me.

Re: DuckDuckGo removed the ability to filter search results

#243

Post by doubtfulcan »

npteljes wrote: post #215
jdironman wrote: post #107

Earlier quoted context omitted.

Why not just have a payment method that isn't linked to you?

This is not unrealistic at all. For curious people, here's how the Posteo email service does it: https://posteo.de/en/site/privacy#anonymisedpayment

Posteo relies on you trusting them to remove your identifying data after payment. There is no way to make sure they actually do

Re: DuckDuckGo removed the ability to filter search results

#244

Post by doubtfulcan »

pembrook wrote: post #225
xlii wrote: post #111

Earlier quoted context omitted.

I’ve been using Kagi for multiple months and due to advocacy I made few people pay for it too, but I’m completely disheartened by their business model. The whole explanation of the recent “we want to stay afloat” was weird, unexpected (price doubling with my usage, to 20$/mo) and felt untrue (ordinary user doing 30 searches per day?). I felt like it was written by Cut-Me-Own-Throat Dibbler - made sense but also very…

I think this is why advertising will remain the default business model of the mainstream internet and search. HN might possibly be the wealthiest collection of people on the internet. And if we’re complaining about $10-20/month here, there’s certainly no hope for the mass market.

High earning does not necessarily mean wealthy. Rich people in my eyes don't even need to work. People also often gain more money by not spending everything all nilly willy. Someone inheriting a lot of money is more likely to spend it on something superfluous than someone that meticulously grew it over their lifetime.

Re: DuckDuckGo removed the ability to filter search results

#245

Post by marginalia_nu »

thornewolf wrote: post #60
itake wrote: post #51

Can anyone explain why Google and DuckDuckGo made this change? I don't see how this impacts them selling ads. My only thought is its technically expensive to support?

Boolean search is pretty expensive computationally. Maybe they are so far disconnected from that initial implementation that there is no semblance of support anymore?

I'm not convinced this is true. It's more likely they're moving away from a keyword model of search toward a more natural language model because the AI hypetrain goes choo-choo.

It's genuinely difficult to do both of these at the same time. You can do keyword search well, you can do natural language search well, you can't do both at the same time. What you get in a mixed paradigm is confusing and unpredictable results.

This is mostly a limitation in query understanding. The traditional boolean index is still there.

Natural language search engines typically first do a series of traditional keyword searches to identify candidate results, and then re-score them using a natural language understanding of the query.

Re: DuckDuckGo removed the ability to filter search results

#246

Post by 1vuio0pswjnm7 »

doubtfulcan wrote: post #235
freediver wrote: post #193

Earlier quoted context omitted.

Searches are not tied to an account, Kagi does not even log searches. [1] What would be the point when your search history would just be a liability? That is the whole point of the paid search business model. [1] https://kagi.com/privacy

How would Kagi enforce quotas if they can't link searches to the account paying for them?

IP address. The classic example is scholarly/academic journals. Access by subscribing university libraries and corporations is still generally dependent on IP address. It has been this way for as long as I can remember. Early 90s at least.

An expected retort might be "But consumer IP addresses are subject to change." True, but sometimes a "static" IP is non-negotiable. For example, in the past if one subscribed to ICANN's zone file access program, which is public information open to access by anyone, not only university libraries or corporations, a static IP address had to be provided. Of course one could use a consumer IP address because those can change infrequently, sometimes on a timescale of years. If it ever changed then one had to execute a new agreement. Perhaps the requirement for a static IP has been dropped but I honestly doubt it.

Rate-limiting/access control by IP address is common and straightforward.

However so-called "tech" companies have shown they cannot be trusted when collecting information from users. For example, Twitter and Facebook each claimed they were requesting phone numbers from users for "security" reasons but it turned out they were using them for commercial purposes.

Re: DuckDuckGo removed the ability to filter search results

#247

Post by doubtfulcan »

freediver wrote: post #216
_8j50 wrote: post #213

Earlier quoted context omitted.

I don't want them to fail but I have 0 support for Kagi after trying them out. Main reason is they collect PII and have no privacy friendly payment options. They should have gone the way if Mullvad with that. Another reason is requiring an account for search is begging to be tracked but even if that is not an issue, it is sooo annoying having to sign in to search and even more annoying when auto complete for a search…

Kagi does not collect PII. You can register with a fake email if you want to. Payment information (if any) is collected by Stripe, not Kagi. How would you keep it simple while running a paid search service with non-trivial search cost?

As someone with high privacy focus, just having a user id given by stripe, that stripe then can connect to a real person is a liability. Mullvad allows payments in crypto and their site is accessible as an onion service. I know it definitely adds overhead to do both of those things, though personally I wouldn't mind paying a lot of extra for it. I'm not sure which fraction of users are like me though, I might just be an unviable customer

Re: DuckDuckGo removed the ability to filter search results

#248

Post by doubtfulcan »

1vuio0pswjnm7 wrote: post #246
doubtfulcan wrote: post #235

Earlier quoted context omitted.

How would Kagi enforce quotas if they can't link searches to the account paying for them?

IP address. The classic example is scholarly/academic journals. Access by subscribing university libraries and corporations is still generally dependent on IP address. It has been this way for as long as I can remember. Early 90s at least. An expected retort might be "But consumer IP addresses are subject to change." True, but sometimes a "static" IP is non-negotiable. For example, in the past if one subscribed to IC…

Wouldn't they have to validate a paying customer to tie it to an IP aswell? If there is a point where a user is identifiable, it can't be rolled back with a "trust us" in my opinion if the focus is privacy.

Re: DuckDuckGo removed the ability to filter search results

#249

Post by doubtfulcan »

mike_d wrote: post #150
plaguepilled wrote: post #137

Earlier quoted context omitted.

This is an interesting detail - I was not aware Microsoft's pricing increased. That said, would this really increase complexity substantially? And as a nitpicky aside, if this was done in response to Microsoft's cuts, I should really have found out about this from DDG themselves, rather than a github repo commit. Though it does change the intent quite a bit.

Bing buckets queries into pricing tiers based on complexity. Why on earth would DDG have any obligation or interest in telling you about their supplier pricing? You found out something that has little to no material impact on you from a post on a nerd site.

Transparency would be in a self-proclaimed privacy oriented businesses interest. Hiding changes and being found out is really bad PR
Post reply on HN