Use a custom domain on Protonmail (which includes Simplelogin) and you won’t have any issues. It’s a grand total of $5 per year for the domain.

Have you got an example I can test? I switched to Firefox mobile over a year ago and I can’t think of any time I’ve come across a site that didn’t work.

In Logseq, everything is a nested list. This feels like a limitation, but I’ve been preferring it. The decision is made for you: you’re going to jot this information down as a list. So then you just start writing it.

I really appreciate you posting this. I’m a long-time Obsidian user, and an Evernote user before that, and I never “got” Logseq. I just couldn’t understand what people saw in an app that didn’t let you “write” anything. I’ve tried to start using Logseq so many times and just given up because the interface made no sense.

Thanks to your comment I finally get it! I prefer to be using something open-source, so I’m going to give Logseq another go, now that I finally understand it, and see how that approach feels.

Obsidian, Zettlr, and Logseq live in the category of local plain-text file-based PKMs.

Trilium lives in the category of local database-based PKMs.

The reason the first category exists is that people wanted to get out of vendor and file lock-in.

Apples and oranges.

Having been through the enshitification of Obsidian, it was important to me and many others to be not beholden to any vendor’s file system. Your database requires Trilium to be instantly usable. My notes are useful and usable (and frequently accessed) from Logseq and VSCode.

The two options are simply not comparable, hence apples and oranges.

Not relevant to you, but relevant to others who might require local plaintext files, rather than a database.

Which brings us right back to apples and oranges 😘

You’re describing now a larger scope of requirement

I am not. I am saying data storage format is a basic, critical factor. And it is. And I already know you agree on this, which is why you choose FOSS options with known, open formats.

Lmao. No, I don’t agree that file format is the most critical choice

Local vs web-hosted, or open formats vs closed formats are part of the exact same choice. So I think you probably do agree that it’s a critical, basic component of your software decision. 😉

Yes obsidian supports various linking formats, but mainly uses its own.

But it doesn’t. The only two options are Wikilinks or original Markdown.

The only software that I’m aware of that is in the same camp as Obsidian - plaintext Markdown files and non-outliner - is Zettlr.

this is just a silly assertion to make.

It’s the most critical, most basic factor in determining what software to choose. I am specifically using software that works on plain-text Markdown files for many reasons, least of all that I need other software to be able to interact with those files. You can’t do that with Trilium.

Secondly, Obsidian does not use its own linking system, it supports both the widely used Wikilinks system and the DaringFireball/CommonMark markdown system.

Come on. At least have knowledge about the software you are trying to criticise.

But Trillium is not plain-text Markdown, so you’re comparing apples to oranges. They’re completely different approaches at their most base level.

Having been through the enshitification of Obsidian, it was important to me and many others to be not beholden to any vendor’s file system. Trilium notes require Trilium to be instantly usable. My notes are useful and usable in Obsidian, Logseq, VSCode, and others, because they use plaintext Markdown files.

Joplin stores its files inside a database. Obsidian stores all notes as individual plaintext Markdown files.

In the first instance, that’s clearly more future-proof and robust - your notes are immediately available in any application without a layer of abstraction. You can’t have a single file corrupt and destroy all your notes.

I vastly prefer it for that reason. I want to know these notes are still going to work fine in 10 years, and be easily accessible.

During the protests, Reddit was very high on both the Block and Lower lists. Quite interesting that this has changed. I still have mine set for Lower.

The best thing about Kagi is never again seeing Quora, W3Schools, or Pinterest in results.

You share it to the Omnivore app. Couldn’t be easier.

It does neat things like unfolds Twitter threads also.

Another vote for Omnivore. It syncs great into Obsidian also.

Thanks!

and I don’t actually think bitcoin is anonymous

You can pay with XMR:

I definitely agree with you though, it is a negative for Kagi. It would be nice if they let you pay direct via crypto or other methods.

If I managed to untrain myself from this and start using tools for their core-purpose, the limits of Kagi might indeed be more than enough. But currently I am too lazy for such a deep change in my daily workflows.

Exactly - exactly my problem. And why I’m probably going to reluctantly upgrade to the $25/mo unlimited. It just irks me that I feel like I’m getting ripped off :P

Imagine installing and opening a separate units conversions app just to find something that used to be an instant search away.

Sure, but the perceived value for my money. Hence why I appreciated the “search is the most important thing I use the internet for” comment.

Me too. That’s probably the easiest comparison and one of the reasons I struggle with Kagi’s pricing. I get Proton’s highest paid plan for less cost… and that includes all their products, mail, VPN, 3TB cloud storage, and clearly doesn’t sell any of my data since they don’t have any access to it. Not to mention that my paid plan subsidises free users. (Assuming I upgrade to Kagi Unlimited which it definitely looks like I will be.)

I use their email aliases function a lot. So you can one-click generate an email to use when you sign up on a service and when you don’t use that service anymore, delete the email address.

I do the same thing, but with a catch-all email. Only started doing that this year and it makes such a huge difference when signing up for services!

(I know that Proton has a similar one-click service, but I worry about some scenario 10 years in the future where they decide to shut it down and I have to migrate all my logins.)

To me, search is the most important thing I use the internet for

I like this framing. That might help me come to terms with their cost 👍

I couldn’t find anything about your claim that conversion would cost extra, not on the pricing page and not in the FAQ section. I also did a few conversation searches and there was no info about additional price. Can you link to where it says that?

Just look at your billing page and do a few of those searches. You will see they count as a paid search - nothing special you need to look for.

I’m not saying they charge extra for them, just that they charge for them like other searches. Doing math in the address bar is so second-nature to me now, and it seems a bit silly for Kagi to charge me for working out what 2 * 8 is.

Kagi has been very transparent about the reason for the costs - it’s what they need to charge to not lose money, since they don’t sell your user data or track you.

I’ve seen their posts on this, but the question is how accurate that data is. 80 searches costing Kagi $1 doesn’t intuitively feel reasonable, but perhaps it is the truth. Google’s search API is $1 per 200 queries, and you would assume they make a profit at that pricing: https://developers.google.com/custom-search/v1/overview#pricing

Of all the subscriptions I have, this one seems like the least value for money for me personally, when I can get for example 5TB of cloud storage for less cost.

It’s not that I’m comparing no-search to search, it’s that I’m comparing the incremental improvement from DDG to Kagi, and considering whether that improvement is worth $10 or $25 a month.