"The cause of the recent crash of flight ADZRUS-666 has been determined to be a badly scheduled ad impression which covered all screens in the glass cockpit to show aan ad of a dancing hippopotamus in a tutu selling skin care products while the plane was on final approach in IFR conditions."
Do Kagi users get paid for shilling the company? Nearly all threads relating to the subject of search has a few mentionings of the glory of Kagi, often including links to the site. I suspect this is not as effective as the Kagi crew thinks since there is likely to be a large overlap between their potential customers and those who are really turned off by such shilling.
Please don't post insinuations about astroturfing, shilling, brigading, foreign agents, and the like. It degrades discussion and is usually mistaken. If you're worried about abuse, email hn@ycombinator.com and we'll look at the data.
Flip side how much does Google pay you to defend their monopoly? Kagi is a solid product with a team that clearly cares about what they’re building. They’re transparent and post change logs when things update. I simply trust them infinitely more than Google.
> The effect is most likely to occur when there are no obvious reasons for performing the task. Because expending effort to perform a useless or unenjoyable task, or experiencing unpleasant consequences in doing so, is cognitively inconsistent (see cognitive dissonance), people are assumed to shift their evaluations of the task in a positive direction to restore consistency.
I interpreted to mean that using a search engine is “useless or unenjoyable, or experiencing unpleasant consequences...”, with attention given to the last two feelings. And I can't figure out what that has to do with people who like Kagi and why it’s wrong or irritating for them to do so.
Granted I’ve been annoyed by similar occurrences with other services, but not to the point of suspecting collusion between the service and the public like the GP comment did.
Searching on the web takes effort. I don’t think this sentiment is controversial. Especially not on HN.
But do you think that because/if searching on the web takes effort and because people have to pay for Kagi, they are compelled to exaggerate its usefulness in public to justify the cost?
Switching to a nondefault technology takes effort and switching to Kagi in particular also costs money, which is also effort for the purpose of the psychological effect known as effort justification. Therefore, people would be likely to rate switching to Kagi as a good thing even if it was exactly the same as Google (says the effect). Therefore, people who say Kagi is good find it exactly the same a Google (implies the commenter).
Kagi customer here. Not getting paid to shill. I think it's worth occasionally mentioning alternatives that are good enough to pay for so that other people know there are other people using other options.
But full disclosure, sometimes I'm using DuckDuckGo and it's also good enough most of the time that I occasionally forget until I go down some rabbit hole and realize that I'm using the wrong search engine.
Whenever I fall back to Google and see how terrible it has become I feel sorry for everyone still using it as their main search engine so I tend to link people to kagi because it's just so much better. Especially the customization aspects. I also like the idea of mainstreaming to pay for critical services like search. No paid shilling whatsoever. Back in the early 2000s people used to drop links to Google whenever search engines where discussed because the alternatives were mostly bad.
Today we have Brave and the alternative Bing frontends but Kagi is still unrivaled because how easy it is to remove shitty results.
Nope, it's just a nice thing I like. It is nearly the platonic ideal of a search engine for me. It causes me no problems and doesn't try to sell me garbage.
It's like discovering that there a better pair of shoes that're more comfortable. Everybody can use a slightly improved more comfortable pair of shoes, so it comes up frequently.
I just don’t understand people who get so upset that someone might like something enough to talk about liking it. So upset that they won’t ever try the thing. Like … ok I guess? You do you. It’s just a strange way to make decisions.
At least this is just a consumer product. Worse is when people here say they make technical decisions using the same process. They’d black list certain tech because they’ve heard people talking about how it solved their problems. Also ok, but now I know I should avoid them professionally.
I get the impression it's the volume of the folks who sing its praises. There was a web3 crowd for a while, Bitwarden champions would show up to any mention of a password manager, and (ahem) some AI champions can be over the top
In all of these cases, a reasonable counterpoint is that if it were that applicable for all audiences, one wouldn't need to sing its praises, it would sing its own praises
It sings its own praises... how exactly? Maybe by a bunch of happy users talking about how they like it and it's a better solution to the problem that the thread or article is about without being explicitly paid? Which is exactly what's happening here and some people are complaining about it?
I tried it, it's slow and bad and free tier is only 100 requests, and it's too expensive, and price is unjustified. I use gemini with google search grounding.
I understand skepticism in the age of LLM-generated content and CAPTCHA-solving bots. What I don't understand is why people choose such weird hills to die on and think that posting about it will accomplish anything. Do you think people will read your comment and go "gee, I was going to use Kagi but now I won't because this random person has a bad feeling about a series of comments they remember seeing"?
I signed up for a specialist forum not too long ago and posted an honest review of a product because I hadn't been able to find one anywhere on the internet. Immediately a bunch of people accused me of being a "shill" for a direct-to-consumer business that's been powered by a Yahoo storefront for the last 20 years, as though a business that's run by a guy with an AOL e-mail address is sophisticated enough to figure out Fiverr and astroturf their reputation on a phpBB forum.
Think about it for just a moment - do you really think that the Hacker News audience is large enough or full of enough tastemakers to sway an alternative search engine's market share? It isn't. If Kagi wanted to do that they'd hire TikTok influencers.
Just run your own XMPP server in your own domain, use OMEMO for encryption and you're set. You can communicate with others on other servers, none of them 'publically available' and the TLA's can stare at all that encrypted gibberish 'till the cows come home. Even if they break into a single server they won't get access to the cleartext, for that they'll need to access the terminal devices - phones, browsers, etc.
This is how I've been communicating for years now, it works fine and does not feed any of the data parasites out there.
And what are you going to do when they write a law that requires ISPs to drop any packet that lacks a digital signature from a trusted hardware manufacturer?
I don't think that's the issue, plenty are already. The issue to me is I'm not going to use something my friends/family aren't using. Maybe something matrix like where many clients are interoperable will work? I still think to take off it would need to support being a frontend for imessage/whatsapp/messenger too or no one will start using it, in a similar way to how imessage falls back to sms, this theoretical app could fall back to whatever shared app the two contacts have.
Matrix is overly complicated for the purpose and XMPP/OMEMO already are interoperable between many different clients. Just like all other communication systems it is the network effect which makes them usable. Tell your friends to install a client of choice and - for those so inclined - run a server or create an account on one. Keep your current W/app or Telegram or whatever active for now while you slowly move more communications to XMPP. Once you have contact with most of your friends and family via XMPP make it your default wat to communicatie, i.e. do not start conversations over the legacy apps and answer those who contact you over them through XMPP. You'll find that you'll end up using those legacy apps less and less. Keep them active if you want but don't initiate conversations over them and you'll be set for the moment using those services becomes untenable.
This is not just fiction, it is what I have done and am still in the process of doing, in my case moving from Telegram - I never used nor will I ever use things which requires accounts run by metafacebook or Google or Microsoft or any of the others.
Time to change your laws and/or prosecutors I'd say so those 'minor thefts' can and will be prosecuted resulting in fines which need to be paid - no ifs and buts. Get them early and get them (hopefully not that) often and you may be able to keep the majority of 'proletarian shoppers' on a somewhat less crooked path. If crime pays more people commit crimes, if shoplifting is not dealt with more people shoplift.
Could you give some examples which refer to the activities you accuse him of? Examples of actual violations of the Constitution of the United States, that is, not just examples of him deciding things you happen to disagree with.
Separation of Powers and Congressional Authority
• Trump repeatedly challenged or ignored the constitutional boundaries between the executive branch and Congress. This includes refusing to comply with congressional subpoenas related to investigations and attempting to block the release of his financial records, which resulted in the Supreme Court reaffirming Congress’s investigative powers.
• He has used executive orders to unilaterally override or undermine laws enacted by Congress, which the Constitution does not permit.
Use of Military on U.S. Soil
• Trump turned to the military as a “personal police force” against American citizens, particularly in response to protests, which has been accused of violating both federal law and the Constitution’s protections for civil liberties.
War Powers Clause
• He ordered military actions (e.g., bombing Iran) without congressional approval, contradicting the Constitution’s war powers clause that reserves the right to declare war exclusively to Congress.
First Amendment Violations
• Trump’s actions targeting peaceful protesters, threatening lawmakers for their speech, and attempting to punish academic institutions and law firms for political reasons have been labeled unconstitutional violations of free speech and due process rights.
14th Amendment and Birthright Citizenship
• His administration sought to deny citizenship to people of color born in the U.S., contrary to the clear language of the 14th Amendment, which guarantees birthright citizenship.
Punishing Political Enemies and Targeting Institutions
• Trump issued executive orders and directed government agencies to retaliate against law firms and universities perceived as opponents, which experts say violates the equal protection and due process clauses.
Speech and Debate Clause
• Attempts were made to investigate and potentially criminalize speech by Democratic lawmakers, threatening constitutional protections for congressional speech.
Power of the Purse
• Trump encroached upon Congress’s constitutional “power of the purse” by redirecting funds without congressional approval, thus violating the separation of powers.
Easy: the hundreds of millions of dollars he has extorted from universities, media companies and law firms for "saying things he happens to disagree with".
I get downvoted when in the last 24 hours, he has signed an executive order to go after people who burn the American flag and has threatened to revoke the licence of ABC and NBC. Imagine just for a second the reaction of Clinton or Obama did that, on top of having a loyalty list of companies and sending the FBI after your critics...
There is clearly a cabal of MAGA supporters trying to game this website to suppress anything that make Trump look bad.
Gott ist tot! Gott bleibt tot! Und wir haben ihn getötet! Wie trösten wir uns, die Mörder aller Mörder? Das Heiligste und Mächtigste, was die Welt bisher besaß, es ist unter unseren Messern verblutet.
The average teenager who reads Nietzsches proclamation on the death of God thinks of this as an accomplishment, finally we got rid of those thousands of years old and thereby severely outdated ideas and rules. Somewhere along the march to maturity they may start to wonder whether that which has replaced those old rules and ideas were good replacements but most of them never come to the realisation that there were rebellious teenagers during all those centuries when the idea of a supreme being to which or whom even the mightiest were to answer to still held sway. Nietzsche saw the peril in letting go off that cultural safety valve and warned for what might come next.
We are currently living in the world he warned us about and for that I, atheist as I am, am partly responsible. The question to be answered here is whether it is possible to regain the benefits of the old order without getting back the obvious excesses, the abuse, the sanctimoniousness and all the other abuses of power and privilege which were responsible for turning people away from that path.
...and as if on queue the narrative around climate "crisis" is woven into an article about a plane experiencing turbulence: experts say the issue is getting worse in an era of climate crisis. Do publications like the Guardian have narrative quotas they need to achieve?
That paper has nothing to do with the incident in question. You're referencing a BBC article that references a paper stating that Clear Air Turbulence is getting worse [1]
> Turbulence is unpleasant to fly through in an aircraft. Strong turbulence can even injure air passengers and flight attendants. An invisible form called clear-air turbulence
But in the incident in question, the plane flew directly through a convective storm.
> Climate models consistently project environmental changes that would support an increase in the frequency and intensity of severe thunderstorms that combine tornadoes, hail, and winds (high confidence), but there is low confidence in the details of the projected increase.
The models project it, but there is currently low confidence in the increase.
if you actually read the cited study it boils down to "we plugged in numbers and these neat heatmaps came out" and if you look at the dates it's 1979 and 2020 (in the heatmaps) and i wonder how much of that is actual location accuracy. i notice the word "accuracy" isn't in the study. that is, there were both less flights and less accuracy of actual location in 1979 (no GPS, etc); and more flights and actually accurate location information "today". It explains the heatmap differences without having to model climate at all. It would be more interesting if there was a similar, zoomed map over some coastal route during daytime for the two years where the pilots knew exactly where they were at nearly all times.
it's a fresh "model" and if you've used an LLM you know how useful models are; and the sorts of models used in these studies are about 1 billionth the size.
Further, their own dataset shows massive areas with decreased turbulence. I guess the sun and CO2 don't work there?
Page 34. They have a graph. "After normalizing the data by annual flight hours, there was no obvious trend over time for turbulence-related Part 121 accidents during this [30 year] period."
BBC article is citing some academics doing a modeling exercise. They never learn. Academics can prove the sky is green if they're allowed to play with R for long enough. That paper isn't measuring actual turbulence, they try to derive it from physical models, but their models must suck because they draw a totally different conclusion to the real world experience of accident investigators. Evidence > academic theories.
the "accident" refers to title 14 part 121 of the CFR, where an accident is after disembark with the intent to fly and before landing where a person is "seriously injured or killed" per 49 CFR section 830.2 (https://www.ecfr.gov/current/title-49/subtitle-B/chapter-VII...); and it's any injury where hospitalization is required for more than 48 hours, or a fracture of any bone except simple fractures of fingers, toes, nose.
it does not mean "crash", although a crash would be included. Specifically, bouncing off the ceiling and fracturing your arm or whatever would count as an accident per the definition.
it's completely valid as a refutation of the bbc article.
But that's not what the paper says. It says Clear Air Turbulence (CAT) has gotten worse, not all types of turbulence. In this case, the flight flew through a convective storm.
Even so, the paper says there's been a 0.2-0.3% change in CAT:
> The largest increases in both absolute and relative MOG CAT were found over the North Atlantic and continental United States, with statistically significant absolute increases of 0.3% (26 hr) and 0.22% (19 hr), respectively, over the total reanalysis period.
And if you found research contradicting that narrative, it would risk your career and/or would not be published. There is only one way to think, not open discourse.
For instance, wildfires. It's definitely climate change because something bad happened. Not classical phenomena about wind currents drying out vegetation. Or increased human development into wild areas, or a century of PUTTING OUT smaller fires, or environmental regulations against harvesting lumber. No, the only way to think is the way that leads to more regulations, taxes and grants and government waste, criminal charges for the unlucky SOB who starts the fire, higher prices (in energy, cars, buildings, and insurance), and more human suffering that they can sell you their next solution for. For this reason, I don't believe a single thing they say anymore.
The irony that you are claiming everybody else will not listen to conflicting evidence while not providing any of your own and saying that you won’t believe anything unless it passes your own world view.
The irony is that if the CA forest were logged and the timber used to build homes, the price of housing would be lower, there would be fewer homeless people, more people with jobs, and few or no devastating fires (less CO2, you know, the big bad "greenhouse gas") . But admitting that would be heretical to the pseudo-religion that is leftism. The party profits when suffering increases.
I legitimately cannot tell if this is a parody comment or not.
It shows a shocking lack of familiarity with the effects of deforestation, carbon capture, logging rules for sustainability, land ownership for building the houses or even what the bottlenecks for housing currently are.
All to blame the intellectually bereft bogeyman of “leftism” when forests exist in many right wing states, and the right wing currently runs the government and yet even they don’t do what you’re suggesting…
>It's definitely climate change because something bad happened. Not classical phenomena about wind currents drying out vegetation.
>[T]he extent to which this trend is due to weather pattern changes dominated by natural variability versus anthropogenic warming has been unclear... Our results show that for the period 1979 to 2020, variation in the atmospheric circulation explains, on average, only 32% of the observed VPD [vapor pressure deficit] trend of 0.48 ± 0.25 hPa/decade (95% CI) over the WUS during the warm season (May to September). The remaining 68% of the upward VPD trend is likely due to anthropogenic warming.[0]
Could be many things. Did they consider whether the incidence of seat belt wearing has decreased? It doesn't seem that far-fetched that compliance with the directive to fasten seat belts has decreased along with respect for authority.
Interesting! As a continued hypothetical, it's interesting that the "No Smoking" permanently-lit sign is next to the seatbelt one. It's a weird contradiction: by being an electronic illuminated sign in the most prominent area (like a passenger HUD - reserved for critical info) it is given an elevated importance that doesn't really align with user expectation (is it really on the same level as the 'alert' implementation of the seatbelt sign?). So, there may be some kind of "cries wolf" subtle psychological effect in play: the cigarette signage is so obviously unnecessary in place and prominence that maybe the seatbelt signage takes on some of that cognitive placement (and implied importance) in mind. I think it kind of plays into that "respect for authority" you noted -- not unlike the possibility that programs like DARE that tried to group drugs like marijuana with heroin may have caused an increase in harder drug use when people realized that they were misled by that initial 'noble lie'. (See also mask use during the pandemic)
Certainly XMPP is not perfect but then again, nothing is. The extensibility does make it a bit of a gamble whether any specific server - which can be used for all kinds of purposes, many of which not related to human communication - offers everything a client program expects it to. Then again if your communication/discussion partners all make sure to use servers which support the essentials for the type of use you want to make of it - usually that'll mean those XEPs need for OMEMO, muc and maybe Jingle, the Conversations project publishes a list of what is needed [1] - things work just fine and you'll be communicating without the need for centralised (monetised, censored, monitored, ...) services. I've been running it for decades now, first as a backup "just in case" communication channel but for the last 7 years or so as my main channel to family and friends. We're using mostly the Conversations (-derived) client(s) on mobile, Gajim on desktop and Converse.js on web with servers running on different types of hardware ranging from SBCs (RasPi etc) to ex-lease enterprise hardware. The maintenance burden on the server software is close to zero with Prosody, it hardly takes any resources and has never crashed on me.
A lot of that time was spent chasing the ever-changing "standards" which describe what a browser is supposed to do. It is easier for a new project to implement the distilled essence of those "standards" than it was to try to keep up with the house-of-cards-built-on-quicksand while its "architects" kept on changing and expanding its "design". This means Ladybird should be able to reach parity in a much shorter timeframe, it is the slog which comes after which decides whether they can keep up with whatever Blink does (which more or less is what those "standards" end up describing).
Getting to 99% of websites will be straightforward, but the last 1% will be tough. Everybody has an obscure website they rely on regularly. Every one of those warts grown during the torturous evolution of the web is depended on by some web site somewhere. https://xkcd.com/1172/
> Getting to 99% of websites will be straightforward, but the last 1% will be tough. Everybody has an obscure website they rely on regularly. Every one of those warts grown during the torturous evolution of the web is depended on by some web site somewhere. https://xkcd.com/1172/
I often mention that in Germany, Firefox has a much higher market share of the browsers than in most other countries:
See for example the statistic of browser market shares among desktop users:
Of course, because Firefox does not have this large market share in many other countries, there exist some (rare) websites that don't work so well in Firefox. But since the Firefox users (at least in Germany) often are very vocal about their browser
- If the website is "not important": ignore it; who would want to visit a website where the developer/company does not care about Firefox?! :-) Additionally, use this as a great opportunity for venting anger, and write some furious e-mail to the webmaster of the website why they dare to ignore Firefox users. :-)
- If the website is "important": use, say, Chrome for this single website - and then write some furious e-mail to the webmaster of the website why they dare to ignore Firefox users. :-)
TLDR: It is much more important that the users of the web browser are very vocal about using it than about getting the last 0.1 % of websites to work.
