Daniel Lombraña

@neynar any new drama

On another side note, it is interesting how the drama with the keys for this schedly thing is happening in 2026. In 2023 I built something similar that you could run yourself. Easy, simple and free: https://daniellombrana.es/blog/2023/06/18/farcaster-bot

Really interesting approach from @cassie for allowing anyone to build on top of the protocol for free. This was always a problem for me. In the early days, the Neymar team always told me, "How many calls will you do? We will give you some for free." For me, this was a problem, and I stopped developing on top of Farcaster. If this fork happens, I will reconsider coding for Web3 social again.

@neynar what drama has been going on here

I'm still here. I haven't left. I will start sharing again soon.

when agents write the code, security becomes a bottleneck

When agents write the code, using non-vague language becomes a bottleneck.

With all the bans on social media regarding teenagers, I wonder how many platforms will go through zk proofs so you can verify I'm over 16 years old, and I don't give you all my personal data. This pressure from the govs should translate into attestation proofs without revealing personal information. I know, that's a utopia.

So good

Change pumpkin for AI or your next Crypto coin, and you will have the same outcome

It is a pity that they are closing because I know how much effort people put in these projects. I lost interest when everything went on coins and not nfts. I don't know what's going to happen with Zora and their coins as well, but I think people are getting tired of all this. I want to collect like in the good old days.

Clearly, using AI is about being SPECIFIC. If you are not specific you allow a ghost machine to imagine things for you.

Change writing code for writing prompts 😊

You would never give access to your personal life to a completely stranger, but here we are in 2026 giving full rights and access to AI bots that you don't know at all to all your digital life. What could go wrong?

I have been working with Claude the last days in order to do something "difficult" migrate a very old nuxtjs 2 version with custom stuff and old libraries to nuxtjs 3. So far, third time redoing everything because it never gets it done on the very first run. Learning my process also around how to ask things. I'm kind of happy with the result, but it is frustrating to see the same errors or going nuts with over engineering (I didn't explicitly state that claude should be KISS driven).

Not sure about what could happen next in the USA. It feels pretty bad.

We have had enough quantity, but not quality, around here. How are we going to rise quality vs quantity when now is easier than ever to create anything? How do we actually get there?

Does anyone here by accident have an invite to lobste.rs?

Just got mine!

I did my PhD with David Anderson, as I needed a faulty distributed system and there was nothing better than home grid computing :D (or how we called it volunteer computing). I worked at CERN with the LHC@Home project allowing for the very first time to run real particle physics simulations in home desktops using BOINC (the framework used by SETI@Home) and VMs (VirtualBox was the one to go!). After all these years, it looks like SETI has finally settled down in 100 promising signals that should be reviewed again. I really loved this outcome and I still admire David for his work and for his support when I was at UC Berkeley working with him. https://news.berkeley.edu/2026/01/12/for-21-years-enthusiasts-used-their-home-computers-to-search-for-et-uc-berkeley-scientists-are-homing-in-on-100-signals-they-found/