I posted some of my experience with Kagi’s LLM features a few months ago here: https://literature.cafe/comment/6674957 . TL;DR: the summarizer and document discussion is fantastic, because it does not hallucinate. The search integration is as good as anyone else’s, but still nothing to write home about.

The Kagi assistant isn’t new, by the way; I’ve been using it for almost a year now. It’s now out of beta and has an improved UI, but the core functionality seems mostly the same.

As far as actual search goes, I don’t find it especially useful. It’s better than Bing Chat or whatever they call it now because it hallucinates less, but the core concept still needs work. It basically takes a few search results and feeds them into the LLM for a summary. That’s not useless, but it’s certainly not a game-changer. I typically want to check its references anyway, so it doesn’t really save me time in practice.

Kagi’s search is primarily not LLM-based and I still find the results and features to be worth the price, after being increasingly frustrated with Google’s decay in recent years. I subscribed to the “Ultimate” Kagi plan specifically because I wanted access to all the premium language models, since subscribing to either ChatGPT or Claude would cost about the same as Kagi, while Kagi gives me access to both (plus Mistral and Gemini). So if you’re interested in playing around with the latest premium models, I still think Kagi’s Ultimate plan is a good deal.

That said, I’ve been disappointed with the development of LLMs this year across the board, and I’m not convinced any of them are worth the money at this point. This isn’t so much a problem with Kagi as it is with all the LLM vendors. The models have gotten significantly worse for my use cases compared to last year, and I don’t quite understand why; I guess they are optimizing for benchmarks that simply don’t align with my needs. I had great success getting zsh or Python one-liners last year, for example, whereas now it always seems to give me wrong or incomplete answers.

My biggest piece of advice when dealing with any LLM-based tools, including Kagi’s, is: don’t use it for anything you’re not able to validate and correct on your own. It’s just a time-saver, not a substitute for your own skills and knowledge.

Thanks! I didn’t see that. Relevant bit for convenience:

we call model providers on your behalf so your personal information (for example, IP address) is not exposed to them. In addition, we have agreements in place with all model providers that further limit how they can use data from these anonymous requests that includes not using Prompts and Outputs to develop or improve their models as well as deleting all information received within 30 days.

Pretty standard stuff for such services in my experience.

I’m not entirely clear on which (anti-)features are only in the browser vs in the web site as well. It sounds like they are steering people toward their commercial partners like Binance across the board.

Personally I find the cryptocurrency stuff off-putting in general. Not trying to push my opinion on you though. If you don’t object to any of that stuff, then as far as I know Brave is fine for you.

Short answer: inserting affiliate links into results, and weird cryptocurrency stuff. https://www.theverge.com/2020/6/8/21283769/brave-browser-affiliate-links-crypto-privacy-ceo-apology

I don’t know if that’s “worse than Microsoft” because that’s a real high bar. But it’s different anyway.

If you click the Chat button on a DDG search page, it says:

DuckDuckGo AI Chat is a private AI-powered chat service that currently supports OpenAI’s GPT-3.5 and Anthropic’s Claude chat models.

So at minimum they are sharing data with one additional third party, either OpenAI or Anthropic depending on which model you choose.

OpenAI and Anthropic have similar terms and conditions for enterprise customers. They are not completely transparent and any given enterprise could have their own custom license terms, but my understanding is that they generally will not store queries or use them for training purposes. You’d better seek clarification from DDG. I was not able to find information on this in DDG’s privacy policy.

Obviously, this is not legal advice, and I do not speak for any of these companies. This is just my understanding based on the last time I looked over the OpenAI and Anthropic privacy policies, which was a few months ago.

Yeah, I wouldn’t be too confident in Facebook’s implementation, and I certainly don’t believe that their interests are aligned with their users’.

That said, it seems like we’re reaching a turning point for big tech, where having access to private user data becomes more of a liability than an asset. Having access to the data means that they will be required by law to provide that data to governments in various circumstances. They might have other legal obligations in how they handle, store, and process that data. All of this comes with costs in terms of person-hours and infrastructure. Google specifically cited this is a reason they are moving Android location history on-device; they don’t want to deal with law enforcement constantly asking them to spy on people. It’s not because they give a shit about user privacy; it’s because they’re tired of providing law enforcement with free labor.

I suspect it also helps them comply with some of the recent privacy protection laws in the EU, though I’m not 100% sure on that. Again, this is a liability issue for them, not a user-privacy issue.

Also, how much valuable information were they getting from private messages in the first place? Considering how much people willingly put out in the open, and how much can be inferred simply by the metadata they still have access to (e.g. the social graph), it seems likely that the actual message data was largely redundant or superfluous. Facebook is certainly in position to measure this objectively.

The social graph is powerful, and if you really care about privacy, you need to worry about it. If you’re a journalist, whistleblower, or political dissident, you absolutely do not want Facebook (and by extension governments) to know who you talk you or when. It doesn’t matter if they don’t know what you’re saying; the association alone is enough to blow your cover.

The metadata problem is common to a lot of platforms. Even Signal cannot use E2EE for metadata; they need to know who you’re communicating with in order to deliver your messages to them. Signal doesn’t retain that metadata, but ultimately you need to take their word on that.

A non-smartphone, that is, a cell phone like the ones that today’s parents had when we were young and with which we made calls and sent text messages, was enough for us, and it did not cause addiction.

That’s not the way I remember it. Texting addiction was a thing. That’s how Twitter became popular; it was basically a way to broadcast SMS to friends at first.

I guess it’s a matter of degrees.

Ad-based services are the real problem here, I think. You don’t hear people complaining about Wikipedia addiction.

Thank you for saving me the trouble of investigating this as an option.

No reason to tolerate proprietary licenses when there are so many viable FLOSS solutions out there.

Sadly, there’s no official LineageOS for the OnePlus 10 or 11 series. I remember back with the 7 series it was officially supported on launch, and OnePlus sent units to open source developers. I don’t think OnePlus cares that much about the development community anymore.

https://wiki.lineageos.org/devices/#oneplus

These are not “normal” tablets, but Boox’s line of ePaper-based readers are the only Android tablets that distinguish themselves sufficiently in my already-large family of devices. I’ve used “normal” tablets with full-color LCD/OLED displays, on both the Android and iPadOS side, but I rarely find a good use for them. I’ve found them to sit in an awkward space with neither the convenience of my phone, nor the utility of my laptop.

The ePaper-based tablets are ideal for reading, but I do not relegate them merely to the “e-reader” category because they allow you to install Google Play and run basically any Android app. This makes them more flexible and powerful than most e-readers.

It comes with a built-in browser optimized for monochrome, and you can also install third-party alternatives like EinkBro.

That said, it’s only for advanced users, and it’s not a perfectly smooth experience. Just getting Google Play running on it requires jumping through some hoops, and you will find that most Android apps simply don’t work well on a monochrome display (though Boox does offer color models, I have not used them myself).

I was hoping, for example, to use my Boox tablet to play Go, but despite the fact that Go is very much a “black and white” game, most of the apps use shading and colors that look like absolute ass on a black and white display. Some of them do not properly support the 4:3 aspect ratio either. So I don’t want to set unreasonable expectations here. These are niche devices.

Despite these drawbacks, I really appreciate having an ePaper device. It complements my device family (phone, laptop, etc.) in a way other tablets do not.

I feel this.

Back in the 90s, there was a fantastic paint program for Mac called ColorIt! (The exclamation point is part of the name, though this is the last time I will respect that because it’s obnoxious; lookin’ at you, Yahoo!*)

It was a commercial product, but ColorIt 2.3 was eventually released as freeware after newer major versions were released for sale. 2.3 was everything I needed, and while I did try ColorIt 4.0, it didn’t click with me the way 2.3 did. At the time I felt like they bowed to the pressure of Adobe’s success and instead of playing to their unique strengths, they made ColorIt’s UI a bit too much like Photoshop. So I stuck with version 2.3.

By the time Mac OS X came around, ColorIt was no longer in active development. But OS X had the “Classic” environment, something akin to an OS 9 VM tightly integrated into OS X. Classic apps didn’t look or feel like native OS X apps, and running Classic came with a heavy RAM burden. But I did it anyway, because ColorIt 2.3 was da bomb.

I continued using ColorIt 2.3 up until Apple killed support for Classic in 10.6 Snow Leopard.

At that point, the intrepid developers came out of hiding and created a Carbon port of ColorIt 4.5 that could run natively on OS X. It was Carbon-only, which meant that it it didn’t run natively on Intel Macs, but it did run thanks to Apple’s Rosetta compatibility layer — at least until Apple axed that as well.

If I ever get into pixel art again, I’ll probably run ColorIt 2.3 again in an OS 9 VM with Sheepshaver or whatever works best nowadays.

*That exclamation point is strictly to emphasize my disdain for Yahoo.