An Update on Orion for Linux and Windows | Hacker News Reader

An Update on Orion for Linux and Windows

blog.kagi.com

185 pointspentagrama112 comments

I've never understood what niche Orion is supposed to fill. People who value privacy would be more likely to use an open-source browser. Firefox and Brave already serve that purpose. Maybe it's less important to Mac users, but on Linux, a proprietary browser is a major deal-breaker for most people. Plus, it's less convenient if the browser isn't simply included in the repository as a result. Orion should have been developed as open source from the start, with community involvement. Kagi can still benefit from this, though, by serving as the default search engine in exchange for funding the development, and by having the Kagi Assistant directly integrated into the platform.

For me the answer is quite simple: it's the only browser on iOS that lets you run Chrome extensions and Firefox add-ons.

It could run extensions a lot better if it ran on other platforms, so the extension developers could actually test it and support it properly.

I'd love to support Orion, but I can't because I don't have a practical way to run it.

Doesn't Firefox also run Chrome extensions? Not that Firefox doesn't have enough extensions by itself...

Firefox on iOS doesn't support extensions at all.

Firefox on any platform does not run Chrome extensions, because it is not built on Chromium.

having written a few extensions... they are basically the same API with a few differences you need to account for. there's polyfills that can unify the experience a bit more too.

They're referring to https://www.tomshardware.com/news/firefox-webextensions-depr...

The underlying APIs are very similar, but you can't just install an extension from the Chrome Web Store.

> I've never understood what niche Orion is supposed to fill

For me, as battery-efficient as Safari, but with better chrome and access to ffox extensions ecosystem.

Chromium-based browsers actually have had better battery life than Safari for 2 years (!) now[0]. And perform better. Which is a real egg on Apple's face.

The only reason to use Safari is because you're knee-deep in the Apple ecosystem, otherwise something like Chrome, Edge or Brave just makes more sense.

[0]https://birchtree.me/blog/everyone-says-chrome-devastates-ma...

> But this is not to tell every Safari user that they are irrational or wrong in any way for using Safari over Chrome (or any other browser, for that matter). Does Safari use more battery than Chrome? In these tasks it sure seems to, but it’s not by a massive amount, potentially costing you about 0.5% of your battery life every hour. Is that enough for you to switch browsers if you prefer Safari’s UI or reliability or privacy options? I suspect not.

The point is moreso that Chrome being a battery hog is a myth.

Safari is less private than Brave and does not have the breadth of extension that Chromium browsers (and Firefox) do. So again, the only reason to really choose Safari is if you're completely in the Apple ecosystem.

I suspect a lot of users don't care about Chrome's (hobbled) extensions

Safari's extensions are hobbled more than Chrome's and have been since 2018.

That's interesting, because that has not been my experience at all – for my usage, Orion empirically clearly buries Vivaldi & Firefox.

Granted, I used Vivaldi rather than Chromium, but I don't expect it to have such a noticeable difference w.r.t. Chrome.

Anecdotally, the niche it solved for me for a while was "Safari for optimal performance on mac, but with proper extensions like ublock".

I used it regularly for almost a year, eventually got off because of some UI bugs that came up and because the Safari dev tools feel so clunky/need relearning compared to Chrome & FF.

I’ve been using it on iOS since it’s near inception and I can report that it is much more stable now than it was since its original release. I tried to live on the desktop browser, however, and eventually had to declare it unusable and switch back to Firefox. Mostly around things like 1Password and similar extensions being pretty broken in the Orion desktop browser.

In addition to what other have said about extensions, ad blocking, no telemetry, and some nice features like tab management, it works in the Apple ecosystem. If you use a mac and an iphone, you get the native sync services for history, bookmarks, passwords, etc. Makes sense that they're focusing on that.

Developing internally initially, as opposed to making it open source, was probably the only practical way for a small team with a specific goal to get there. I don't see any problem turning it over to community development for secondary platforms. If it gets adoption, great. If it fizzles out on linux and windows, nothing is lost.

Browser diversity is a good thing. We've seen what happens when one huge monopoly dominates the browser landscape. Twice now, first with IE and then Chrome.

I use Orion because it's not a product of Apple or Google. There's no reason the browser should be a product of the same company who made the OS.

But it's not browser diversity in the sense people usually mean when they talk about IE and Chrome. It uses the same WebKit engine Safari uses.

> Kagi can still benefit from this, though, by serving as the default search engine in exchange for funding the development

This is not gonna work because Kagi is not free to use

Getting your product in front of more eyes will result in conversions. That's the whole reason they're doing this.

On Mac, a non Safari browser that moves faster, allows chrome extensions and more customizability. I stopped using it because they refused to implement DNS over HTTP, handwaving it as "use what the OS already gives you".

Firefox fits that description aside obviously from Chrome extensions, is there something inherently more appealing about the Chrome API? Didn't they just sunset it and replace it with some crippled version that can't even support a decent ad-blocker anyway? I know of a few commercial products that only offer a Chrome extension I guess, but I would consider that those product's problems.

> a proprietary browser is a major deal-breaker

If this was a real issue for most, people would look at Google closing up Android and shipping laptops that are locked to the Play Store and refuse to have anything to do with Chrome or Chromium.

I used it on iOS for ad blocking until uBlock's release https://apps.apple.com/pl/app/ublock-origin-lite/id674534269...

Pragmatically, on iOS the ad blocking you’re getting with paid apps like Wipr 2 is still better than uBlock origin lite.

So, in my opinion, there wasn’t a really good reason to wait for uBlock to land on Apple’s platform.

The ad blocking aspect never seemed like a flagship reason to use Orion if you ask me.

Some of the other features are.

The ad blocking in Wipr 2 makes the Apple News app tolerable.

It comes with adblocking out of the box, and is a reliable way to watch YouTube ad free on iOS.

The site made it difficult to even know what Orion was. Your comment answered that, thanks.

On iOS it integrates better with Kagi search. It is basically a skin for safari though.

Being able to swipe up on the address bar to close tabs almost makes it worth using.

You don’t think the native mimic of Safari to open up the tab list is better?!

Don’t get me wrong, options are always a good thing!

I don't get the impression Mozilla puts much effort into Firefox on macOS - whether its performance, or fitting in UX wise. So it's nice to have another browser on macOS that can run uBo.

I've been paying for kagi since it came out, use linux on my laptop and lineaege on my phone, and I don't use Orion.

In other words, as you say, exactly.

I have no idea if it's a good or bad or I'd like it or not. I never even considered trying. I don't mind that it exists. I don't think kagi are doing anything bad with it. But I also just have no interest in even trying it. Presumably it's good for soneone but for me (a paying kagi customer linux user) it's almost like it's invisible. A probably benign, utterly uninteresting thing.

I don't use any of their AI features either. I hope they cost extra seperately so that I'm not paying for that stuff I'm not using but I suspect I probably am.

The AI is pretty benign, too. They offer diverse models and they make it as privacy-preserving as possible, considering you can opt to use closed models. They are very open about training and retention: https://help.kagi.com/kagi/ai/llms-privacy.html. An interesting note there on models served by OpenAI: "The Assistant does not include a unique user identifier for these requests." I don't know if you can trust OpenAI, but Kagi is at least doing its part.

Also, All the AI integration into the product is gated by a single setting.

Afaict the niche they attempted to fill is the one Firefox already fills on desktop, a highly customizable browser with extension support. (Firefox does not support extensions on iOS, apparently has no plans to do so, and as I understand the support is somewhat limited on Android as well). I switched for about a year because they delivered what seemed like a serious effort at that with their initial release. After seeing basically zero effort at addressing very fundamental bugs in that support in that year, I switched back. As far as I’m concerned their battle is already lost.

The iOS codebase is the only one that potentially has any value to the OSS community; this kind of "enjoy our garbage!" source release really irks me.

So as a non-Apple-using Kagi subscriber since Beta, I kind of wish they'd just concentrate on search - the thing I'm paying for.

Yes, it's good, and I'll keep using it for the foreseeable future, but it's not an unassailable lead.

I feel the same way. I like Orion, and I use it on some of my devices. But if I could choose I’d rather the company focus on Kagi development. There are a good number of bugs and feature requests that have been “planned” for months/years that I would be very happy to finally get fixed.

Agreed, I'm a Kagi subscriber and a mac and iPhone user, I don't use Orion and have no desire to. Browser and search engines should be separate entities, but for some reason search engines really want their own browser, and in the case of Brave, a browser company really wants a search engine.

Lately I haven't been very happy with Kagi. A few searches have completely failed for me returning a blank page when both DuckDuckGo and Google had a response. The Kagi translate on/off switch made me think their services are unreliable. Then this happens and it feels like more that their word is unreliable. They expanded into too many domains and spread themselves too thin and that doesn't give me confidence.

My subscription is up for renewal in December and I think I actually have a decision to make. For now I'm very happy with the Kagi MCP for my research agent. But I'm definitely having a moment of "how reliant should I really be on this company?"

If you're mainly using the MCP, it could be worth making your own MCP that uses your own API keys from Brave, Mojeek & Marginalia (optionally also add Yep). For the most basic ranking, get it to group URLs that appear in more than one search engine and list those ones at the top. Your favorite AI will be able to vibe-build this all for you and give you an exe / binary you can run as an MCP that runs locally over stdio.

If you try it out, you'll either find a replacement that saves you money, or you'll at least have the makings of an escape plan if you need it later. You'll also have your own comparison data for where Kagi beats a generic metasearch, and if that difference is meaningful for you.

> They expanded into too many domains and spread themselves too thin and that doesn't give me confidence.

That's because imo there is no way to grow this bigger with a search service. Same for open source browsers, just look at Firefox instead of focusing on bugzilla and push for new web apis, they're trying very hard everything else than can be monetized

Search is super valuable.

Google managed to monetize the everyday person's searches, but at the detriment of everyone that wanted to get nontrivial work done that needs in depth search.

Lots of institutions want quality search. They could index high value things, and sell them directly to businesses, universities, etc.

> Search is super valuable

It is for Google there isn’t much else we are left with

They could do nothing but marketing and maintenance, and watch as their competitors continue to make the same mistakes that drive people to Kagi in the first place.

So... they shouldn't grow. There's nothing wrong with having a sustainable business doing one thing and doing it well.

I agree, searches return very often blank pages and I find myself more and more using google instead. I won't likely renew my subscription.

I dropped Kagi last year. It’s just not worth it. Search sucks no matter who is doing it and now the game has changed with agents that can search more links than I can.

I really tried to make Orion work for me on iOS but I was running into so many issues browsing websites. Performance issues, issues with certain interactive controls. Switched back to Safari and it was like night and day. Feel like it needs a lot of polish so on one hand I think this is a good move to help them focus. However, I feel like it is very much possible that in a year we will hear them shuttering Orion completely to refocus on search.

I wish they would cancel the Mac version as well. It really has no clear purpose and just further dilutes the "alternatives to Chrome" space. If they have money to spend on a browser, why not put that effort into a browser that actually has a chance to tilt the scales a bit, like Zen?

> why not put that effort into a browser that actually has a chance to tilt the scales a bit, like Zen?

Zen is a Firefox (skin? fork?), and the Firefox community is their own niche (and IMHO, more Linux/Windows-focused rather than Mac). I think Kagi's focus on Mac made WebKit a natural choice, even though Firefox would've been the more platform-independent choice (as in being able to make the one browser for all three platforms, rather than bringing one to Mac, then another to Linux and to Windows).

Firefox has never really felt at home to me on Mac. I've used Safari (and previously Orion) ever since letting go of Chrome for most personal use.

FWIW, I find Zen on macOS to have much less of that uncanny clunky foreign feel that Firefox and its brethren such as Thunderbird on macOS have never been able to shake. It’s no Camino[0], but it’s the closest thing I’ve seen since to a Gecko browser that feels like it actually belongs on macOS.

I’d rather see Orion continue to be developed alongside Zen though, because goodness knows we need more non-Chromium browsers.

[0]: https://en.wikipedia.org/wiki/Camino_(web_browser)

You might want to revisit Firefox once again now that they have done another major redesign.

The issue with Firefox isn’t so much its visual design (though it is a component) as it is things like interactions all feeling a bit “off” and little bits of lag scattered throughout (like frames for controls and windows being blank for a split second before drawing in). It gives a “made in my back yard” impression, and not in the good handcrafted way but rather the duct tape and bubblegum way.

> and the Firefox community is their own niche

I don't think most people care about the engine that powers their browser. They care if it's fast, renders the sites they need, has a pleasant user interface, and doesn't crash.

A Gecko-based browser can do all of that.

FWIW, for me personally, particularly with newer versions, there has never been a moment when I felt Zen didn't fit on my Mac just fine.

I’ve been a long-time subscriber to kagi, but also a long-time critic that they bite off more than they can chew. Relatively happy to see this, and I’m sure there’ll be an open-source community willing to carry on the Linux work.

As ex-Kagi subscribers, I could no longer see the value in paying $10, as the results weren’t significantly different from those on my self-hosted SearXNG.

I moved from a densely populated area to a more sparse one. I'll probably renew Kagi because its assistant can answer my questions, such as: "find me a location where I can be on my computer but also public where I can also be with a group of people". Google's Gemini hallucinated spots or imagined details -- at one point it included a shed in the middle of a road -- the shed existed, nothing else did. DDG did similar.

PS: did not test Copilot nor my own LLM accessing another Search Engine's result -- haven't got the headspace for it.

that is significant use case even for me, however i dont have that situation.

It's ironic (and sad) that they'd both abandon it and open source it at the same time — open sourcing a browser is key step in expecting it to gain any adoption on Linux, but abandoning it kinda kills motivation for anyone to contribute.

As other said: the target audience of the Linux version was never clear.

I'm not a kagi subscriber (yet), but am more and more becoming a fan of them and their approaches. I count myself as one of the folks who was thrilled about this web browser, because as much as i love firefox, we need diversity. Ok, ok, not too too much fragmentation...but still we need offerings to help people. So, while i was hopeful for kagi folks to keep working on orion i'm happy that at least they'll open source it and have a community continue it forward. My future hope now is that community members indeed take this browser and grow it further.

Kagi subscriber here. I think mature technologists should stop supporting proprietary browser technology workarounds, and I personally wont support them although I like the spirit of what Kagi tried to do with Orion on Mac OS.

We dont need ad blockers. We need advertiserproof/trackingproof browsers. Like we have waterproof shoes and coats and hats. We could also use some nicer browsers in terms of features.

But what folks who really care about these details should do is simply de-apple, and de-google (and obviously de-windows!). That will get you much further than using a particular browser, but a good browser helps.

> We need advertiserproof/trackingproof browsers.

Care to explain what do you mean by this and how would it work in practice?

Like others in this thread I’ve questioned why Kagi is developing a browser. My only guess is they at one point thought there would be some opportunities in integrating AI stuff and a browser. Because their CEO has said in the past that they are convinced of LLM usefulness but in a specific way.

I wouldn’t be surprised to see some agentic features in their browser at some point. Their assistant tool is already quite good - it has replaced search for me in a number of scenarios.

The way I remember it marketed, and the way I originally found it, was as a workaround to setting the default search on iOS. This was before I was conscious of AI and before I remember it being in any of Kagi’s products.

I ended up staying because the Adblock is good (search remains good). The browser is fine. It does have bugs and needs to be hard closed occasionally.

I agree on the assistant. It is very capable. I use the Kagi mcp with coding agents and I like the results, but they have done something with their assistant that makes it even more useful.

It wasn't just Kagi. Literally every search engine decided to have its own web browser. DuckDuckGo had one. Brave has one.

https://orionfeedback.org/d/13108-megathread-orion-memory-is... open issue since January. They should open source macOS too, they don't have the personnel to develop this.

I want the same browser (port) on each of my OSes. No Linux/AOSP port means I won't use your browser. But I guess I wasn't interesting enough of a customer (am subscribed to Kagi Search). No problem. For example: on my Steam Deck, I am happy with Mozilla Firedox and uBlock Origin. This company pivots too much to USA to my taste anyway.

As a Linux user, sad to hear, but totally understandable. As a paying Kagi user, I don't want them to spread their resources too thin.

Thanks for the honest attempt! Hope someone takes the torch and carries it forward.

Pretty much the same for me, except I'd say it's probably for the best. I'm also not thrilled about Chromium's total dominance, but I don't think Kagi, small as they are, are going to be the ones to break it. I don't have particularly high hopes for it as a pure open-source project either, but best of luck to whoever picks up that torch I guess.

> We also chose to build it the hard way by not forking Chromium.

Bad call, because it's genuinely too hard for someone the size of Kagi, as they've discovered.

There's room for a Chromium fork that has a better reputation than Brave. (I use Brave, and it's great, but I realise many people don't want to use it.)

I use Orion (because pay for Orion Plus to support them) specifically because it is not a fork of Chromium.

We need more rendering engines out there in the wild, as the situation with chrome's monopoly is already critically dire.

I agree. But I think it'll take someone bigger than Kagi to achieve that.

Mozilla have more momentum, experience, and funding, but still struggle.

The struggle is the blind opening of chrome, like most people probably just do without thinking. Unless you are really a hardcore frontend developer, you lose nothing by switching to Firefox (or non chrome/chromium browser). I operate both and for 95%, maybe even more, there is no difference. By switching you will put pressure on google to keep on the users good side.

They never made the rendering engine. It’s open source.

I don’t see any more difficulty between being a chromium browser or being a WebKit browser.

I think currently Helium[1] is trying to fill that niche of a privacy oriented Chromium fork without bullshit and focus on performance.

[1] https://github.com/imputnet/helium

There's also ungoogled-chromium and it uses similar set of patches: https://github.com/ungoogled-software/ungoogled-chromium

From their README, it seems Helium is a fork of ungoogled-chromium.

Vivaldi?

I would like my extensions to support every browser possible, but if a browser only runs on Mac, I won't be able to test my stuff in it. I've had users asking me for Orion support for ages, and I keep telling them I'll add it once I can run it... but now it looks like that's not going to happen.

Dropping the ports doesn't necessarily make the Mac version better. It makes the entire browser unsupportable by developers who want to make their sites/programs/extensions work in Orion but can't run it.

So my support list now is...

- Supported: Firefox, Chrome, Edge, Floorp, Chromium, Brave, Vivaldi, Librewolf, Waterfox, Ungoogled Chromium, Thorium, Mullvad, Tor, DuckDuckGo, Yandex, ... - Impractical to support: Orion - Impossible to support: Zen Browser

Dang, looking at this later, I realize I forgot again that HN doesn't support common list formatting. If I put one "- item" per line, it joins them all into a single paragraph. And now it's too late to edit. It should work if I put a blank line between each item though.

That last bit was supposed to say:

- Supported: Firefox, Chrome, Edge, Floorp, Chromium, Brave, Vivaldi, Librewolf, Waterfox, Ungoogled Chromium, Thorium, Mullvad, Tor, DuckDuckGo, Yandex, ...

- Impractical to support: Orion

- Impossible to support: Zen Browser

This probably won't be the last time I trip over that. I use lists a lot, and almost every other platform allows them to be one per line without anything between.

Out of curiosity, what prevents extension support in Zen? How is it different from the other Firefox forks? It has releases for all 3 major platforms.

The full answer for this is a long, wild ride which would fill entire chapters... but the short answer is: Zen's author broke the extension API so badly it's literally impossible to do a lot of what my extension needs to do... and the author is aggressively opposed to fixing it or even admitting that he broke it.

More generally, Zen Browser is the buggiest browser I've ever encountered... by a wide margin. Some of the issues are jaw-dropping. It's notorious for losing user data. And looking at its history, it has mostly just kept getting worse over time. Extensions which used to work don't any more, because he keeps breaking more and more stuff.

I tried to support it. But instead, I got a quick education on why nothing supports Zen Browser. They can't... because of the guy behind it. Sometimes due to his incompetence, sometimes because he actively intervenes when people try to support it.

So... my experience with Zen Browser rapidly went from "hey this seems neat" to "wouldn't touch this with a ten meter pole".

I used Orion, briefly. The experience on macOS, linux, and iOS was as bad as the Kagi Search experience is good. I hope this is the start of the Kagi team ditching distractions and sidebars and focusing on the thing that makes me a paying customer in the first place.

Sad that the other two platforms won't have the same stewardship as Orion on macOS. Especially since we're likely to see drift from the core WebKit implementation between them now.

I don't understand the decision to open source the Linux and Windows ports, but not the MacOS version. They should have moved their Orion development into the open, and said we're not going to support Linux and Windows going forward. Then someone else could have picked up the task of keeping the Linux and Windows forks in sync with their MacOS work.

Agreed, the open source Linux and Windows versions feel pretty much DOA without the ability to reference and port the macOS source code

It’ll be interesting to see the Linux source. It felt pretty close to GNOME Web. Maybe the gnome folks can back port some stuff.

Port what? The UI code is going to be completely different across platforms?

This is typically how quality native ports work.

If orion was open source and had a beautiful UI like Arc it could be the best browser for me. But it is not beautiful and I can't fix its bugs.

I don't understand the reasoning behind Kagi developing a browser. Such a small team would gain more from partnering with the likes of Mozilla/Vivaldi/etc IMO

Firefox uses Google as the default search engine, Vivaldi goes with Startpage. I doubt either would change to a stripped-down version of Kagi. Because that's how this would most likely be done, and Kagi wouldn't get much of such deals.

Browsers being broken is very closely related to Google/search being broken.

I assume it's the same reason that Google decided to develop its own browser

Google was vastly better resourced though when they started Chrome. They even waited for many years because Eric Schmidt thought the company was still too small to enter the browser wars.

Orion is great on iOS, good browser. Some jank on sites like Substack but for that I blame Substack for janky user hostile code.

I had to unsub from search after a year. Catalan searches are worse than Google's, unfortunately.

Maybe it is for better. I used Orion as my main browser for awhile on iOS, but then it started getting weird bugs. Now I have both Safari and Firefox open all the time and watch yt on Brave.

I try to see if they got Orion to behave again, but find a bit difficult to trust it as the main browser.

I wish these articles would have a link to what they are at the top.

I know they don't want more users at this point, but half the things I find out about are through their discontinuation press releases.

Good call by kagi, I hope someone else takes the lead!

I found the move to release Orion on Linux and windows weird.

If you were Google and you wanted to spread your search engine to the world, it would be fine but such a small company. And the market share would still have been abysmal on the platforms.

Great thing that they release it before abandoning it, truely a great move !

And hopefully they keep making it better on macOS and iOS.

Orion is my daily driver on my Mac. Frankly just don’t want them to break it while making it more and more secure.

> If you were Google and you wanted to spread your search engine to the world, it would be fine but such a small company.

I've done successful, safe cross-platform development at several small companies and it goes just fine. But you do need capable hands at it and some developers really worked their best to sabotage the work platforms due to sheer laziness.

Size of company is irrelevant, not hiring fanboys is the most important part at this process. I wonder how Kagi managed to fail at that.

Very disappointing. I have no reason to use Orion on my iPhone if it isn’t matched by a Windows version. Immediate uninstall for me. Orion on iOS is slightly subpar, so idk why anyone would use it over Safari.

I was very excited about the possibility of having some rendering engine competition on Windows.

That's unfortunate, though kind of expected

Huh -- this blog post was published on October 2nd and says updates stop as of October 2nd. Some more notice would've been nice, but I wonder if maybe they've been considering this for a while and just couldn't take it anymore.

Oh. They may not be glad knowing that I am planning to stop funding their small development team further. I just do not like how they form some usability details. Crucially and recently forcing link preview on in an update shows that their way of thinking misaligns mine. This one was breaking usability big time for me, if only for a brief period of time. Suddenly clicking on links stopped working. The number one thing of web, hypertext itself, got disabled. I had to chase down what it is and how to revert so having my browser being usable again. This is the opposite I expect from an 'improvement'. I do not like their way of thinking. It is fortunate that I never migrated to Orion completely. Due to some other - to me - weirdness.

nowhere except hackernews do I see kagi links

Hackers tend to have their own culture, their own blogs and so on. So you're unlikely to just find someone recommending Kagi on social networks the same way some HNers do. Maybe on Reddit (though not nearly as much as before). Generally, social networks for hackers don't turn out well, there are exceptions but you're usually more likely to find them using regular tools (like IRC) as a social network rather than being anywhere actually designed to act like a social network. You might see recommendations like this more often if you frequent IRC spaces or etc. where lots of hackers hang out; they share their tools a lot, including search engines.

Reply on news.ycombinator.com