smodo

You have to retrieve it through the Vibe interface (I used the CLI) and then you can use it with any harness.

I agree. I’m a customer out of principle (EU) but they make it hard to find out where to get a subscription and what that gets you. But their 18 eur per month offering is actually good with 255 eur worth of tokens included. Take care though: you need to create a specific kind of API key to avoid being charged for per token use besides your subscription.

This is on par with European stereotypes about French engineering so I’ll just go along with it :)

I’m a casual user out of principle and it’s fine. Nothing extraordinary but capable of assisting coding sessions. It won’t one shot a cancer medicine but it handles my Rust projects well. Mind you I don’t like the style of LLM use where people just brute force their way to a working state from a very high level prompt. I mostly ask it to implement tickets/issues I write beforehand.

Hah I knew that was you from the first mention of DS and game remixing :) Love your contributions on LI.

Great to see support for the M-chips catching up. The chips are obviously very powerful and efficient and I find the latest MacOS releases to kind of drag them down.

With the build quality these machines have I can see Linux support extending their usable lifetimes well past 10 years. By which time we’re back to a mainframe-terminal model anyway since all compute is AI and it is all done in the cloud. (?)

It can still be true though, because entry is nothing without sales and a continued search for product market fit. I don’t know about 1% but if you say ‘for every 100 apps launched in 2015-2020 only 1 turned into a sustainable business’ I’d say that’s in the ballpark.

If code becomes cheaper that may change the business fundamentals but it also means capturing customers becomes way harder.

Openrouter stats seem to indicate that there is a lot of demand for ‘good enough at a good price.’ After all, not everyone is decompiling PS5 games. In fact, most people are just speedrunning notetaking apps in JavaScript frameworks.

I was a little cranky when I replied. What I meant was that this flood of LLM-noise already started around GPT-4.5 for many people. I was surprised to read about this now. But it’s good to have a discussion about it at large, anyway.

I wholeheartedly agree but how can someone only come to this conclusion after years? It is so obvious that writing and reading were for centuries bound in a social contract between reader and writer. Post-modernism if anything reinforced that notion by rebelling against it.

I assure you anyone in the business of administrating stuff through paperwork (government contracting!) is well aware of the complete obliteration of the basic assumption that ‘someone wrote this so it is at least worth a cursory glance’.

For all its other problems this is my main issue with LLM’s. Writing let people communicate across vast distance and time. Given no metadata, today one must initially assume all writing to be noise.

Ideally you use so many different compute providers that your footprint isn’t as complete for any of them. If you commit to a single party they have total information which is more dangerous. If you’re into that sort of thing. Obviously just don’t put anything in a http request you don’t want anyone else to know…

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.

There are some indications that make Russian saboteurs a very likely culprit. It would be unwise to look at every occurrence of sabotage as a singular incident in the context of the ongoing campaign against European political stability.

https://www.universiteitleiden.nl/en/news/2025/01/research-e...

https://www.sueddeutsche.de/politik/sabotage-deutschland-bka...

So walking back some of the Liquid Glass insanity and better search. Okay, glad I skipped Tahoe.

The kill switch is M365 account management. You take that offline, many SME’s and local governments just stop working. At least for a while.

Still baffles me that Outlook won’t let me tie together emails, calendar items and documents. It makes so much sense to combine these things. All my meetings are about documents and their related correspondence but you never have a coherent information space for them.

Yes. For some people product owners don’t want to hear this. If having access to email means you can access the account then don’t prance around that with complicated recovery steps.

Well maybe you lose your ability to focus or your sanity or your privacy but yeah.

It’s a notetaking app.

Well if we just store the inputs and outputs of the LLM on the blockchain we’ve just created a secure end to end AI-managed tender process. In fact we can automatically pay out the contractor in AlbaCoins now.

Click here to prebuy AlbaCoins and be ahead of the curve in money laun—- I mean AI governance.

Us Europeans are just baffled by the fact that this ‘administration’ wants this. The EU is a big economy that’s relatively easy to deal with. Why would you alienate us?

But yeah so far Trump has been relatively true to his word, as far as it goes. Not really practically but going further down the road of a dare I say fascist outlook. I think Europeans still can’t believe it’s happening, much less intentionally so.

My colleagues at the head of a company. I’m one of four bosses. One of us is pushing for AI every single meeting. The other is ignoring her. The last one is starting to ‘see her point.’ I’m considering quitting if this goes to far but unwilling to make that threat yet, as it’s a bridge I can only cross once.

Anyway. To me it just speaks to the disdain for semi-intellectual work. People seem to think producing text has some value of its own. They think they can shortcircuit the basic assumption that behind every text is an intention that can be relied upon. They think that if they substitute this intention with a prompt, they can create the same value. I expect there to be some kind of bureaucratic collapse because of this, with parties unable to figure out responsibility around these zombie-texts. After that begins cleaning up, legislating and capturing in policy what the status of a given text is etc. Altman &co will have cashed out by then.

The recent blog post discusses moving into the email space and suggests some interesting uses of LLM’s for handling mail (sorting and labeling rather than writing). I’ve been quite pleasantly surprised with the Kagi products so far and would probably consider their e-mailservice if it’s as good. However the hassle of switching email providers is of course around 100x the hassle of switching search…

I’m still hurting from when I didn’t listen to Yamaha’s warnings about the dangers of piano music.

Prose like an adventure novel, content like a space opera. Lots of overconfident crystal ball gazing. Yep, must be a “future of AI” piece.

Looks good. I liked Notion but hated that I can’t migrate my data in the future.

Are there any good ways to attach files to notes? I often want to tie a memo or email to a note but have not found a convenient way.

Yeah and now people who can’t even write and never put in the effort to learn it are flooding the zone (my inbox) with useless 10 page memo’s.

Thanks. Your team has been busy! :)

Putting aside the GPT-vibe I get from this comment.

> Integrated Development System:* Makepad offers a unified environment encompassing a GUI framework, a development system, and an IDE, streamlining the development workflow.

What? There’s an IDE?

Gaaf. I’ve written a piece of software using egui and while it let me iterate fast it has its limitations. There’s also gpui but that’s very Mac centric for now… I’ll give this a go in the near future. I’m a little curious about the DSL though. What is the advantage of a/this DSL over the builder pattern used by gpui? Does the fact that you can insert new code during runtime mean that there is in fact an interpreter between my code and the ui? How do you manage performance/efficiency in that case?

Excel is not even Excel anymore. It’s different on Windows, Mac, web, mobile… And there’s different bugs across all them! My favorite ones have to do with regional defaults.