Leandro Favarin

@leandrofavarin
60 Followers
68 Following
186 Posts
@cketti congrats! Let us know how it goes 💚
@mg @ianlake very excited to watch this! Sadly I won’t be attending this year.
@markv @stroughtonsmith newer models will have more recent knowledge cutoffs. Also, LLMs will use the (few) examples they find and are good at testing their own output to cover for their gaps. See https://simonwillison.net/2026/Mar/9/not-so-boring/
Perhaps not Boring Technology after all

A recurring concern I’ve seen regarding LLMs for programming is that they will push our technology choices towards the tools that are best represented in their training data, making it …

Simon Willison’s Weblog
@ironicbadger I’m using a dockerized mbsync to archive on a Linux box. Here’s the project: https://github.com/JakeWharton/docker-mbsync
GitHub - JakeWharton/docker-mbsync: A Docker container which runs the mbsync tool automatically to synchronize your email

A Docker container which runs the mbsync tool automatically to synchronize your email - JakeWharton/docker-mbsync

GitHub
@jessewilson thanks for sharing your appearance on the podcast. I agree that serendipity for pairing is much welcome, specially in remote companies like the one I’m currently in.
Also love NetNewsWire!

@leandrofavarin yep!

Here's our motivating use-case:
1. An EventListener collects trace IDs from a ThreadLocal when a call starts. It attaches them to the Call.
2. An Interceptor propagates the trace IDs as request headers.

Step 1 always runs on an application thread. Step 2 runs on either an application thread (with Call.execute) or a dispatcher thread (Call.enqueue).

@jessewilson This is really cool! I'm trying to think of what to pass through it.. Maybe a function that refreshes the auth token when a 401 comes up in an interceptor?
@atpfm you mentioned in the show how you have a smart albums to easily separate photos by camera/lens? Could you explain what is the smart filter used to do that, for us, non-photographers?
@brentsimmons nowadays Apple and Google should allow a interpreted languages to run on iOS and Android, but political and platform control are too strong for this change to be self-served.
@brentsimmons back when mainframes were far more expensive than a developer’s time, it made sense to use highly performant programming languages. Development time wasn’t a priority because labor costs were negligible compared to hardware costs.
In the early years of smartphones, power consumption was a critical concern, so efficient code was vital.