SoKamil

And that’s their problem.

Don’t be afraid to make grammar mistakes or misspell stuff. Others will understand. You’re a human after all. That’s okay to make mistakes and feel uncomfortable with that.

Since Big Sur redesign, light mode on macOS is borderline unusable.

I need contrast in order to differentiate content. I need contrast on buttons to know where to click and what is clickable. I don’t need to depend on muscle memory. On Catalina it was automatic. Chrome in moderation is not bad.

Authority washing.

a) LLM’s don’t have introspection capabilities

b) in my observation, the longer context window, the more unhinged/pessimistic LLM output becomes

What is your opinion on the recent Cloudflare outages?

It seems they trained the model to output good svg’s.

In their blog post[1], first use case they mention is svg generation. Thus, it might not be any indicator at all anymore.

[1] https://blog.google/innovation-and-ai/models-and-research/ge...

> thanks to EU Digital Markets Acts

Is it official integration then? Do you have source for that claim?

I always forget about it, thanks for the tip.

This was the only article that also highlighted recent removal of desktop apps. This is a calculated move to reanimate facebook.com

No, it is not. You have whole Facebook bar on top with the notification bell, with unread notifications dark pattern. You need to have your Meessenger account tied to Facebook account. This is clear indication that facebook.com is on a drip. The site is dying.

Next, expect to have to install Facebook app to access Messenger.

Am I the only one who finds monospace font barely readable for articles? Good for code, bad for longer forms of text.

I always used browser address bar for that. But giving it a second thought, I uploaded the data to Google servers.

This post is discussing more specialized databases, but why would people choose Oracle/Microsoft DB instead of Postgres? Your own experience is welcome.

I thought it’s about Lynx Browser, a text based browser that lives in terminal

It’s funny that almost all vibe coded software have this detailed tree project structure in README. If I recall correctly, this was not common in pre-LLM era. It was too much burden to maintain.

Could you cite that claim from the paper?

It’s like not only the technology is to blame, but the culture and incentives of modern world.

The urge to cheat in order to get a job, promotion, approval. The urge to do stuff you are not even interested in, to look good in the resume. And to some extent I feel sorry for these people. At the end of the day you have to pay your bills.

Dot folders contribute to the „System Data” section. Maven/Gradle caches can take a lot of space. Take a look at your home folder first and press CMD + SHIFT + . to reveal hidden folders. You might not need 3rd party software.

Interesting, I didn’t know that it was immediate mode. According to the article [1] though, it now uses rendering engine with Retained Mode due to performance issues.

[1] https://medium.com/@0s.and.1s/flutter-part-iv-skia-vs-impell...

Why?

The thing is, someone will collect rent from IP anyways. LLMs shift rent collecting from decentralized individuals to a handful of big tech companies.

What is the UI SDK used for Windows version?

Swift

The last 12 months? The last decade.

I’m more afraid that some manager will read this and impose rules on their team. On the surface one might think that having more test coverage is universally good and won’t consider trade offs. I have a gut feeling that Goodhart’s Law accelerated with AI is a dangerous mix.

It is the same thing though?

Charles is a http proxy, Requestly judging by the landing page is a http client like Postman.

I hope they will allow me to download e-books that I uploaded through their upload site.

I do backups but better be safe than sorry.

> and tracking methods through mouse (without JS).

How?

They are also not supported on iOS native apps. We use pdfs for vectors.