If I lived in Australia, I would have the exterior of my house to be very white (except solar panels), so as to reflect most of the visible and IR light, and stay cooler.
Problem is most newer places are built of concrete or brick. I mean sure slap some white paint on but the that thermal material is going to take a lot of cooling/heating.
> Tailwind looks ugly, but there are no hidden abstractions or structures in the code
Verily, Tailwind is the assembly language of CSS. No structure, no semantics, no abstractions, only twiddling specific bits of visual representation.
To turn to it is to admit that your product lacks the structure and the design vision that allow to use some abstraction ("semantic classes"), and that all you can usually do is to patch some areas of it, disconnected from the rest, and unable to touch the rest (lest it goes down in flames). Assembly is definitely suitable for patching.
This is often the endgame of old large products that changed hands and directions many times, without much care.
This seems like a radical, unsupported conclusion to draw from a single frontend architecture decision.
Maybe there's an overwhelmingly positive case to be made for high-abstraction CSS with hidden structures. Outside those few select cases, we've known the tradeoffs for 25 years, and the added complexity is rarely worth it.
I enjoyed working with a semantically styled frontend project as recently as 2024. It also used a React component library, which helped insulate per-component styles, while sharing the common parts.
It has. It's called @apply. It works fine. The Tailwind author is wrong about recommending people to avoid @apply. The Tailwind community at large is mostly wrong about it too (except people that like and use @apply).
I don't know how anti-abstraction came to become popular, but if you like semantic classes you can still use Tailwind. (why use Tailwind? It has some nice defaults and a coherent design language)
The anti-abstraction crowd, that push for repeating the same boilerplate over and over again, honestly grinds my gears. Even assembly doesn't make writing boilerplate its ideology (it has macros, and actual subroutines)
There were and are many important pieces of Linux kernel that live out of tree; ZFS is a big example.
The problem with this driver is not licensing or code quality; I assume it's under a threat of receiving C&D letter, or maybe also a legal suit for breaking some NDA.
Most state is buffer-local rather than global. Note also that there is already some amount of multi-threading in elisp.
One thing you could imagine is doing some kind of mvcc system so that if two tasks don’t interfere with each other, they can execute in parallel (and if not, abort one and rerun it sequentially).
But maybe that still doesn’t work because all the things that should be run concurrently are using the same buffer and doing manipulations of it. You could imagine trying to make save-excursion introduce an opportunity for parallelism: split into a separate thread for the body as that thread is no longer modifying global cursor state (in some sense). Don’t let two threads write to the same buffer for concurrent tasks that should appear sequential, but do allow them to read in parallel (surely most things are merely reading).
However I’m not sure you actually need all that cleverness. I think you could have some combination of:
- tools to make multithreaded elisp more viable
- manual attention to improve major sources of slowness by introducing parallelism (eg font-lock, rendering, autocomplete, indentation). A stupid thing could be precomputing (in the background) what would happen in the user pressed RET (or TAB) and then being able to execute it sooner if they press that key.
The question here is whatever Steam Frame displays resolution is sufficient to work with text. So far only mass produced device that was really suitable is a Vision Pro.
So far all reviews I've seen been centered around gaming and not work.
It is said that human eyes can resolve down to 1 minute of arc(1 MOA or 1/60 degrees). Each eye has about 160 degrees of HFOV and 125 degrees of VFOV.
From there, you can work out requisite resolution for desktop replacing VR googles: 11K per eye.
The catch is that human eyes actually only have
~1 degree of full resolution area and ~30 degree of active area, and the full 200 degree HFOV is only realized by moving them around and subconsciously compositing images, so most of panels can bee left unlit, in theory.
But the panel resolution requirements is the same, and our technologies are just not there to make a true desktop replacement VR goggle, yet.
I guess from the comments we can draw conclusion that its different to everyone.
I had Quest 3 for a few months to play games and I feel okay playing, but every time I tried to open IDE and terminal after 10 minutes my brain just say full stop. I tried different fonts, themes and zoom levels, but it just feel wrong.
I feel similar feeling of discomfort as if I try to work on something not in computer in low light in real world. Might be it have something to do with my bad eyesight.
I honestly did not tested AVP enough. Its just good enough for me not to immediately want to stop using it, but again I have no idea what would really happen after a week of use for 3 hours. I did it once for a few hours and it felt amazing, but its Apple and there lots of other reasons not to buy their locked down hardware.
> every time I tried to open IDE and terminal after 10 minutes my brain just say full stop.
I'm wondering if it is because the screen was "flat" in front of you and you had no possibility to change your perspective. It would be interesting if the IDE had been in 3D space and you would look at it as if it was on your desk, capturing head movements etc.
This is certainly interesting idea, but I guess main problem isnt just UX and representation on 2D surface, but just pure amount of text data that you suppose to look at and understand what exactly it does.
When you have fiction or game dialogue text there is good bunch skimming going on so you dont care about every single symbol and or digit. So eyes dont get stuck looking at specific parts of text.
So yeah its very much possible that using VR as a virtual monitor setup it's just wrong aproach and there have to be paradigm shift in how we interact with dev environment to use it efficiently in VR.
For me personally the Quest 3 is absolutely unusable for any text work. Even small font in games looks shitty. This is so far off for what it ought to be for comfortable work that I'll wait for devices with at least double the resolution per axis, i.e. 4x the pixels of the Quest 3 screens to even bother trying again.
Edit: And I'm writing this on a 32" 2560x1440 display which looks just dandy to me, so it's not like I'm super sensitive or have above average eyesight (quite the opposite actually)
I have tried this in the past and it wasn’t nearly sharp enough but I guess I see people using 1080p screens at work and those make me want to rip my eyeballs out.
For me it's more that flights are extremely boring. I find watching any movies or TV shows on a plane to be a horrible experience. The noise of the plane can ruin even the best of movies.
Maybe better for a train; the benefit is that when you unfold your bike for the last mile, you just put your work into partial transparency and navigate using the external cameras.
Also only possible if you've got a power outlet (Surprisingly even today many long haul flights do not) as its only got just over an hour of battery life.
It has a 21.6 Wh battery. You can easily get a power bank 3+ times this capacity, thus can last 4x longer. If anything, I would rather have that weight in my pocket or backpack than on my head, all things considered.
Anecdotal, but on the last few flights that I've taken. I was told that you can use them, but you can't charge them on the flight. This was inside the US, American Airlines, and United. I can't speak to the other airlines.
Using it while flying will be complicated by the fact that all airlines have banned the use of power banks. If the onboard USB socket isn't very powerful then you'll only get an hours' use.
I can confirm that all the flights I have been on this year we were told not to use power banks. You can have one with you (well you have to since you're not allowed to check large lithium ion batteries) but you aren't allowed to actively use it.
100 W sockets? All the flights I've been on that provided power have had a really low current USB A port that can barely keep my phone from discharging when the screen is on.
In constrained plane environments https://github.com/vk2diy/hackbook-m4-mini works well and has long battery life. I fix the screen to the seat using adapted helping hands and use a wireless keyboard/mouse for input. Super comfy as long as your eyes can deal with the reduced distance (look down the aisle occasionally).
Magnifying style shop glsssed might actually be perfect for this. I use them for electronics repair (they also only focus close which is sort of a benefit here). Downside for me is they fit in a way that is too high (they are designed for looking straight ahead) that would be great for this
> designed to ignore the rest of my system as much as it can and live in its own little world.
Exactly what a small, locked-down system needs. A single-purpose server box. An advanced appliance, like a "smart TV", or media box, or an advanced home automation hub.
I don't see why this calls for a new system. If anything these environments are a stronger case for a quality ecosystem where a whole community cares about keeping it coherent and up-to-date.
A minimal Debian installation for single-purpose server isn't a new idea, or I often reach for Alpine and trim it down to a dozen or so packages. These are both quality ecosystems with track records and reputations. I can get a minimal attack surface and still count on timely updates, security patches, good integration (when it turns out that I needed a dual-purpose server box and not a single-purpose box), etc. These systems scale up and down.
What I'm reluctant to trust in a deploy-and-forget system is a fly-by-night package from someone who picked a packaging system because it looked easy and didn't make them think about an actual lifecycle. A few months or years down the road I expect most of those will be unmaintained. One of the most valuable parts of an ecosystem like Debian is that they make everybody think about all those issues and actually do the work.
TIL, I didn’t know that. I always assumed it came from level zero”, the Intel computer layer that ZLUDA was translating to before its developer was hired by AMD to target HIP.
In short: structurally like Docker (reuses Dockerfiles), but without a daemon, and, unlike Podman or Nix, very lightweight, only two binaries do everything.
Common chunks of data are shared / deduplicated across installed packages, again like Docker or Nix.
(Cleaned / decoded: 'You are Google Search from 2004. Given a search request, provide 10 links to relevant pages, each with a short text from the page, featuring the search terms. Avoid any pages that do not contain the search terms. the search request: %s')
Steep at 300 searches per month or about 10 per day. If you search often, it is too expensive. If you rarely search, not worth 5 bucks.
They are plainly trying to push people to their $10 unlimited plan. I would have appreciated if they allowed, say, 600 searches per month for $5 or so.
I find that I rarely do one search. It’s usually a search, followed by looking at the first few links and probably opening them in tabs, then refining the search terms to get better results. It’s an iterative process. So the 10 per day on average would be exceeded quite quickly.
I too feel the yearly price of $108 for one person, $155 for two people and $216 for a family is expensive. Ok, maybe the family plan is cheaper if you can find five others to share it with.
One of the wildest things to me: Kagi actually warns you before renewals so you can cancel first if you want.
That's confidence in the product right there. I'd argue any subscription which won't do this is designed to exploit people who don't need or use a subscription.
You know how many things are asking for "~$10/mo". People don't have the purchasing power they used to, and everything has a subscription model. Firefox with DDG is good enough.
As they said, you wouldn't pay $10/month for search. This is why free pay-with-your-data services won. Nobody wants to pay for what they use, so companies extract value in other ways and you can't complain about that if you're not willing to pay what it costs.
> Nobody wants to pay for what they use, so companies extract value in other ways and you can't complain about that if you're not willing to pay what it costs.
I'm likely old school here, but I never liked "paying for what I use" in the online world because it required revealing some personally identifying information (i.e. to make payment). My attitude is almost certainly irrelevant these days, with the scope of online tracking, sharing of data, and the near necessity of disclosing one's identity to sites other than one's ISP, but it was an attitude that was established in my mind nearly 30 years ago.
(I also don't trust the pay for privacy model. First of all, you can never assume that paying for something implies your data won't be sold. That was even true before the public Internet. Then, even with statements about privacy, you cannot assume that will be true in the future. If the company is sold, management changes, or the company decides the need/want additional revenues, that will change. And that assumes they don't use weasel words about what privacy means, or the ultimate "we reserve the right to changes this agreement at any time".)
Kagi support zero knowledge search through a Firefox extension and Mullvad supports zero knowledge VPN by putting cash in an envelope with your account ID written on the outside.
I mean, there’s more privacy protection on payment data than there is for email address and browsing habits.
Plus most companies just push out payment handling to stripe. So most websites have little idea who actually transacted with them to begin with (email dependent).
Sure. I’m not saying you should pay for search just saying it is not economically feasible to monetize it directly. I pay for Kagi, but that makes me part of a very niche market.
And Google's ad revenue is likely more than that. Or, if it's not for every user, the ones who would opt out of ads are the ones who can pay. You're cannibalizing your highest value users.
Google kept increasing its profits since then, and it was one of the most successful businesses out there way before they changed the business model to become enshittified. The argument that their business model changed because people wouldn't pay for its services doesn't really stand.
True. I got this as my first LLM subscription (was a long time skeptic, less so now but still a bit), but now I have cursor too so maybe I can even downgrade. Thanks for the reminder. :)
My time, attention, and “brain” are valuable. Having aligned incentives with my search provider and control over the results it serves me through weighting is worth substantially more than I pay for unlimited.
Kagi is the stickiest digital subscription I have.
FWIW Kagi has worked hard on their pricing over the last few years and has been trying different models. They're not "big search" the price is what it is so it can exist as a business. 100% fine to not be a customer obviously, but then you forfeit your license to complain about Google spying on you and ruining the web. Like they're trying to solve the problem. IMHO speaking just for myself I feel a moral duty to support them.
I agree on all your points. I'm just saying, that the $5 tier is not a honest to good option. The $10 one is but they can make the $5 one genuine too by increasing the quota. I'm sure they have done their research and maybe increasing it will cannibalize the $10 tier.
It's relatively rare as of late that I need to specifically find something on the internet, and DDG does not suffice, so I rarely exhaust the 300 Kagi searches.
Most of the time I need to find out something, and in that case, Claude / Gemini / Grok / Perplexity / you name it work better. Most of my queries are technical by nature, not interesting for data mining on me personally.
The 300 searches a month work for me because I only use Kagi when I think it will give better results than others. And most searches are trivial so any free engine will do (like you have the website name and want the url).
Good workaround! It's just sad that to achieve the previous behavior we now need to burn significantly more compute, and in turn energy, and with far worse performance and an inverted UX.
The most frustrating part is that they have all of the data, and oodles of compute, available to surface the same very functional experience they originally offered, but they would prefer the image of being visionaries rather than the reality of being useful.
I think the "original experience", whatever it may have been (and I strongly suspect there's more nostalgia here than actual worse results) would be very quickly overrun by modern SEO methods / tactics.
Kagi can be "better" than Google (for some very narrow definitions of better) because they basically don't have to think about SEO at all. They're small enough that nobody does SEO against Kagi. If their algorithm is sufficiently different that typical Google tactics don't work on it, even if it's much more naive because of that, Kagi wins.
Thanks, I needed to add a ?aep=11&atvm=2&udm=50 for AI mode. Wow!!!!!
>You are Google Search from 2004. Given a search request, provide 10 links to relevant pages, each with a short text from the page, featuring the search terms. Avoid any pages that do not contain the search terms. Do not simulate fetching results. The URLs must be complete and not truncated. No markdown. The current date is 2026. The search request:
I'm using Kagi every day, and unfortunately, it is not true anymore. I think what happens is that the blanket banning of IP ranges because of LLMs, and flat out incompetence and laziness, affects them heavily. I get results, or just a few unrelated ones many times every day, while Google happily returns what I searched for. I think Kagi's index is shrinking, or they really made their algorithm worse.
I slowly shifted to use LLMs because of this in the past half years, making the situation probably worse for Kagi. They were the last usable search engines for years now. I still pay for it, but it's increasingly not that useful anymore.
Same here. Been paying for Kagi for probably a year or so, but I find that it hasn’t really been able to keep up with at least my personal expectations. It feels quite slow (which I don’t remember it always being), and the search result quality is not always the best. On top of that, the 10$/mo is a little steep for me, given that their assistant and maps are quite sub-par in my opinion. I would have preferred them to focus more on the search experience, and less on the other products. Or give me a cheaper plan for only unlimited search without the other products.
It definitely is. The extension-based redirect in Safari is also unreliable. Furthermore, many of my searches (and others', apparently) unexplainably defaulted to Groningen in the Netherlands, even though I'm nowhere near the place.
I ended up not renewing my annual subscription. Since then, I vibecoded a small userscript that replicates domain ranking for DuckDuckGo – plus DDG already lets you turn off ads natively.
> You are Google Search from 2004. Given a search request, provide 10 links to relevant pages, each with a short text from the page, featuring the search terms. Avoid any pages that do not contain the search terms. the search request: %s
Oh that shorthand is fun. Curious it works at all.
But yeah I'd switch to Kagi if I had any more forcing function. Maybe without LLM's but google continously getting worse, that'd been the actual thing for me
I like Kagi, but is it me or have their results been less than great for the past 1-3mo? Maybe the places they're pulling their results from have had some changes?
I'd say that the ranking, and thus the relevance, is noticeably higher. If I know what I want to find, Kagi is good at surfacing exactly that, not something vaguely related.
The only “other engine” that matters is really a thin paid SERP façade in front of a scraped Google result. In other words, Kagi pays for access to an API whose implementation is basically live scraping Google and throwing away the ads.
There are a few other indexes that Kagi’s aggregator mixes in (such as Bing Search) but those haven’t been contributing much in terms of meaningful results. The sheer size of Google’s index still dwarfs all the others, including Bing Search.
So yes, Kagi doesn’t scrape but they pay SERP providers who do.
Marginalia likely returns better results than Google, when it returns useful results. DDG/Bing likely returns average quality results which can be better than Google's slop results. It's not true that Google is the only useful one.
They pay scraping sites for results, including SERP API. When I pointed this out, last time Kagi was discussed on Hacker News, an employee of Kagi said that they're trying to build their own internal index, but he didn't provide details.
For Google, the business side of you paying them doesn't work out as obvious as it might seem.
Advertisers are sold the idea that 1000 clicks/impression is X dollars.
The average impression might be priced at 1000/X$.
But you are worth much more than that for two reasons.
- you are a person willing to pay 5$
- Google can sell a “collectivized” product. The same way that eg collective farmer crop insurance creates actual value.
and the third reason (as consequence of the second) they dont want people to think about: it gives them more opportunity to fudge the value by making it all very complicated.
I'm willing to pay time or money to avoid ads, not to be spied and harassed by companies making me offers to buy something that almost certainly I don't want to buy. Those companies should pay me a percent of what I make them save by not buying ads to serve me.
And eventually when I want to buy something I research it and buy it and nobody had to pay anything to Google.
I don't understand your logic. Why would someone that never sees my ads be valuable to me? Why would I care? Everyone spends money on something, the fact you spend it on adblockers doesn't make you automatically more valuable to advertisers. I have never heard of this "advert paradox" concept you mention and I have worked in industry for a while now. All this is just conjecture on your part and doesn't really correspond to how it works.
If you have $20 to spend avoiding ads, you have $30 to spend on cool products so the advertiser is willing to spend up to $24 to make you not block ads about a $30 product you'll buy (that costs $5 to make)
> If you have $20 to spend avoiding ads, you have $30 to spend on cool products
Yes, but if you are so against ads that you spend $20 to not see ads (I am definitely in this group) and then see an ad for a "cool product" anyway, then that product is no longer cool. Any time something like this happens, I legitimately develop a grudge against that company and their products.
I might have bought that product before seeing the ad, but there is no chance of that after seeing the ad.
> Studies always show that you don't. You think you do, but you don't.
Did those studies have any outliers that DID have a negative reaction to ads? I am 100% certain I'd react EXTREMELY negatively in a situation like that.
The same way I wanted to buy a Tesla and rented one a couple of times in the past. Now I would NEVER AGAIN even go into a Tesla. Fuck that company.
That does not make sense at all. Why would you care about people who want to avoid ads? The real data shows very little people actually do that. How much you spend on adblocker is irrelevant to your profile for a business. They will get you some other way if you ad block on digital.
Sad but true. I paid for Prime to be ad free, easily worth 3$/month but sure enough now part of their catalog has become "only available with ads" or ads in their FireTV is before you hit the prime app (and then again when you do). So now their whole product is crap and I'm looking for a new streaming stick solution [anyone?] Sounds like it's time for a newcomer/disruptor like Angel Studios to release one.
Natural next step of the ad lobby is obviously to pay the lobby to make it illegal to bypass the money making machine if they haven't done so yet. I don't doubt that capitalism still works, it just takes time to "boil the frogs" dead [enough] to stop them from buying such crappy products.
> That's why tools like uBlock and YouTube morphe are the only answer, negotiating with terrorists never works.
This is silly. The economy doesn't work if no one is willing to pay for goods. The content you're consuming won't be made if theres no market value to create it in the first place. Pirating is not a solution to our data privacy problem.
The capitalist economy is based on the explicit idea that everyone will do everything they can to extract value from everyone else, and it'll come to some kind of equilibrium. If you do not do this but you allow everyone else to do this, you receive worse quality products for ridiculous prices and do nothing about it.
In other words, you owe me $1000 as a condition of reading this comment. Your options are to unread the comment, pay me $1000, or pay me nothing to incentivize me to charge a price you'll actually pay (likely $0.00 for comments but not for everything)
Free riders send a market signal of failure or they cause friction. In the former, the service or creator just shuts down and moves on to something else. In the latter the service must make a choice with very little knowledge.. what should they do? Bake ads into the content? Ad block wall? Subscription? Pay per view?
Basically it doesn't make any meaningful progress towards a better outcome. It just temporarily makes the content more enjoyable to you at the expense of the person who created it and maybe to you if you enjoyed it and wanted more.
I’m pretty sure Kagi’s entire pitch is that they’re an alternate and original dataset. Theirs is the only search engine whose data isn’t based upon google or bing’s databases.
To anyone who didn’t find Kagi useful in the past, I strongly recommend you try it again, as they have made huge strides in usability and quality of surfaced data in just the past year…. Huge strides since even a few months ago too, with more adaptive curation, data labeling, and UI/UX
> I’m pretty sure Kagi’s entire pitch is that they’re an alternate and original dataset. Theirs is the only search engine whose data isn’t based upon google or bing’s databases.
It's not. In addition to their own indexes the results include calls to "all major search result providers worldwide"[1].
There are independent search engines that only use their own index, like Mojeek[1]. Kagi is explicitly something like a search engine aggregator.
Prayer for external effects has long been mechanized [1]. But praying for your inner effect is a different thing. Similarly, production of songs for external consumption is being mechanized. But what you feel when you sing cannot be mechanized. Sing for yourself, play for yourself and for those immediately around you in the moment. It's as valuable as ever.
reply