Same here. I'd feel a tad more comfortable if I could set an upper limit for the extra Chat credits that are used, but I guess it's hard to rack up 100s of $ when using it manually without noticing, so I'm not super concerned.
Overall I really enjoy the new functionality, and I've cancelled all the other GPT integrations I used so far in PyCharm.
There is Daniel Schmachtenbergers 'Metacrisis', which comes with its own little spin, but is probably close, and seems to gain traction. Could not find anything about him or the term on Wikipedia though, which I find a bit curious. Maybe just me being stupid and not looking right though
You can just use your digital documents directly, and augment it with the few paper receipts that you might (or might not) still have to deal with. The main selling point is really document management (to me, anyway), the 'branding focus' on physical documents is probably a little misleading.
Can't speak for physical documents in general, but personally I really appreciate paperless-ngx for it's general document indexing/storage. Being able to scan and ocr physical documents (usually using the camera on my mobile phone) is very nice, but I mainly use it with pdfs that paperless automatically fetches, ocrs (if necessary), and tags from my email inbox, or which I copy into a specific local folder which gets synced with paperless.
Getting all my invoices from last year to prepare taxes is now just a simple query in the paperless UI, the result would be about 95% digital and 5% physical documents, probably. Of course I could do all that old-school using filesystem folders, but having all my documents indexed and searchable in a single place was definitely worth the (small) effort of setting it all up and keep it running.
I don't understand what you mean with prepare taxes.
I just add all purchases/sales right when they happen in my accounting app and attach the invoice PDF. Then when I have to file taxes, I export the correct numbers.
Are you doing your bookkeeping in Excel or something?
This is just for my personal taxes, no accounting involved. I just get all the relevant stuff together once a year. Of course it's not 10s of 100s of documents, but still enough so it would take me some time to get everything together manually.
Also it was just meant as an example, paperless is generally useful (to me) in situations where I need to access somehow related documents, like traveling and such, or searching my documents for some information. As I said, there are other systems and ways to do this, but for me this is the one that stuck.
Edit: also, looks like you have to be in the 'Team' plan for access. But who knows, if there is demand I'm sure they'd open this up, so maybe just ping them.
I seems I am a bit daft for not finding that; thank you.
Anyone knows why the API would be so much more expensive? With Google I can understand because with the API you miss a lot of the tracking, targeting and advertising, but if you don't have any of that, why would it cost so much more than normal searching?
I understand unlimited is off the table for an API, but 1000 for $25 is quite a bit.
Thanks for sharing, very cool! There doesn't seem to be a whole lot of documentation on how to use it as a library, but looking at the code all seems to be nice and tidy, and it doesn't look too difficult to adapt to a use-case or two I have in mind.
Any plans to publish a release version to crates.io, or is it mainly intended as an internal tool for prefix.dev?
As far as I know the plan is indeed to publish a version to crates.io, and the only reason it hasn't been done yet is that it is still early days. You might want to jump in their discord server to ask, though, if you want to be sure (or maybe you can even create an issue on GitHub)
Yay! :) Thanks heaps! It's funny, 'rattler' is exactly what I need at the moment. I was about to write a rust app that downloads 'micromamba' and runs it as a subprocess. But this is so much better for what I try to do, being able to include everything in a single binary and calling rust code directly...
I have to admit, I'm a bit disappointed by the mostly negative responses here, I wouldn't have thought this would be received as badly as it apparently is, by this here crowd.
I haven't followed along that closely, but I dropped by their discord every once in a while, and from the discussions I saw I felt Kagi was as open as they could be, trying to communicate the upcoming change(s), why they are necessary, getting user input, changing their plans accordingly, etc.
So, maybe my expectations where already 'anchored', but I find those changes very reasonable, and only if you compare it with 'free' is it expensive. But as we all know, 'free' doesn't necessarily mean no money is involved, just who pays directly and who pays indirectly. I'd guess the 'actual' cost is close to the same in both cases, minus some scaling that Google can do and Kagi can't. I'm willing to pay for 'unscaled', personally.
I've been using Kagi for about a year now, averaging between 700 and 1200 searches per month, so none of the proposed plans would break the bank, compared to what I've been paying so far. Tempting to choose the legacy plan for a year, but I'll probably just go for the 'ultimate' one, the yearly plan, to support a service I really really don't want to go away.
During the time I've been using Kagi, I hardly noticed it. Which I think is brilliant. There was no annoyance, in probably 90% of my searches I don't even have to scroll and one of the first 5 results is what I'm looking for. No ads, at all. In my calculations, that are at least 700 interactions with software that are fast and to the point, with very little chance to annoy me, reliably, every month. And those interactions are very important for me to do my job. I value the time and piece of mind that are connected to those 700-1200 interactions.
So -- as a professional -- to me this is worth $25 a month, easily. I dislike subscriptions as much as the next person, but to me it seems obvious that the context matters here: if it's for a software, then I really want the option to pay for a permanent license, and I think it's fair to pay for (major) upgrades if I want them. Companies that lock you into their product and you can't use the data you created with it anymore -- unless you keep paying -- are acting morally reprehensible in my view.
But on the other hand, if the service that is provided has a considerable cost for every user-interaction, then, well, 'pay-per-view' is justified. If I use it more, then I have to pay more. I don't want companies to make a loss because I'm using their service, esp. if I like their service. And I even want them to make a healthy surplus. A business model where heavy users pay more and are not subsidized by low-volume users seems like a good thing to me.
If a search costs them, say, $0.02, then I'm happy to pay $0.04 for it, it is what it is. Of course I'd want them to be efficient, and save as much cost as one can reasonably expect (maybe some of my searches could be answered from my personal cache?), but from what I see, kagi hasn't been buying villas and yachts from my money so far, and I don't think they'll be able to in the near future with their new pricing structure.
What kagi was doing here ticks all of my boxes, I don't think they have been 'baiting & switching' me, and I trust their reasoning and calculations unless someone has good evidence that shows I shouldn't.
I pay $20 for ChatGPT which is WAY more useful to me than Kagi. Hell, I pay half of that for Office subscription which I use even more. There is no way I can justify paying $25 for a search engine. And I am a professional too, just one who cares about money.
If you're a Kagi fan, the use of Discord by the search engine may seem impressive, but as a regular user looking for an affordable and trustworthy search engine, it's unlikely that you would spend your time engaging with the search engine's Discord community.
It's a common occurrence among startups to rely on Discord as a means to measure software usage, but in actuality, the platform mostly attracts highly engaged power users and a few followers who may not represent the average user. As a result, the company's perception of their software's typical user may be skewed.
I don't consider myself a kagi fan, nor do I think their use of discord is 'impressive'. That would be quite silly. So you are wrong if you assume that is where my positive judgement comes from. I just make different value judgements than you -- ones I feel align well with my own, personal priorities and specific context. That doesn't mean I don't care about money, I hope you can see and accept that.
I just stumbled upon their discussion about pricing, and found it interesting how such a problem could be approached, and seeing their point of view and reasonings, and pros and cons for several options, and how its a really difficult problem for a young company like them. Compared to that, the discussion here feels a bit one-dimensional and flat, tbh. Anyway, just wanted to give a different, bit more emphathetic perspective to most of the sentiments in the rest of the comments.
Basically, I can understand why they felt the need to make the decisions they made, but I acknowledge that I'm in the minority with this. Doesn't mean I don't think I'm right and you're all wrong :)
Their costs have been soaring as they add more and more features which are literally meaningless to me.
I didn’t care about that, because they promised me my reasonable plan was grandfathered forever. That was a lie, so I canceled. Cause and effect. I don’t trust companies who lie to me.
I'm happy to pay the money for search, but what's happening here seems to be Kagi jacking up the prices to pay for their other non-search stuff that I'm not interested in.
Me too. I think the cost/benefit calculus is pretty simple. The pricing is around 1.5 cents per search. Let's say someone does 600 searches, which is about what I do in a month, or around 20 per day. That costs around $10.
For someone with a net income of $20 per hour (a lot of people in the US), that's 30 minutes of time. Can Kagi save 30 minutes over 600 searches in a month, or 1 minute per day over 20 searches? I'm pretty sure it can easily, yes. The result quality for me is far better than DDG (which I'd realistically be using otherwise, for privacy reasons) or even Google.
I'm doing the same. It's also quite nice for de-duplication, a lot of operations on our data happen on a column basis, and we need to assemble tables that are basically the same, except for one or two computed columns. I usually store all columns in a separate file, and assemble tables on the fly, also memory-mapped. Quite happy with being able to do that. Not sure how easy that would be with parquet.
Overall I really enjoy the new functionality, and I've cancelled all the other GPT integrations I used so far in PyCharm.