@mattpocockuk I keep coming back to time being the issue. You had time to make mistakes and learn from them. This is how seniors are „born“. Now we have AI, everything gets trivialized and ppl get pushed to churn out more & more. Remember what @yacineMTB said about outsourcing understanding!
@kunchenguid I didn’t notice less Claude speak.
I had one answer in which every paragraph contained the words „load bearing“. And it still says stuff like „two halfs to this and one is even more relevant to what you mentioned earlier“.
It’s funny though, so I don’t mind.
Introducing TimesFM-3, a state-of-the-art time series foundation model that enables accurate multivariate time series forecasting in a single forward pass, significantly outperforming other forecasting models across major benchmarks.
More on the blog →https://t.co/uSlnIdUJ4Q
@theplgeek Can it filter for skills that the model can invoke itself? Tried superpowers, tried pocock's skills, both have skills that the model can either call itself or that are hooked into every request. Every little thing becomes a huge ceremony. I just want to remove the auto-stuff...
@MParakhin The ADHD way of work is what causes burnout. I juggle up to 5 projects at once, get constantly interrupted by meetings and colleagues, thus keep forgetting the context of each agent convo. The strain from Zeigarnik effect + attention residue is real. Deep work is the way to go.
@kunchenguid@bcherny It didn’t work for me. Concise will only make the responses in Claude Code be just that. If you let it write code or a markdown, it’s still writing long verbose comments and sentences every step of the way. I tested it just a few hours after they released „concise“.
Spent 3 weeks building a texting app slower than email. Messages fly at pigeon speed — LA to NYC takes 22 hrs. Sometimes the bird dies & the message isn’t delivered.
@SheaAllan@mattpocockuk I agree. But I think it’s right for the part of a readme that explains how to get a program running. But for requirements it‘s hurting because of its sentence structure.
Ambiguity is the biggest issue with reqs. For this you could consider EARS, created by Mavin at RollsRoyce.
@mattpocockuk Use it for the part of a readme that explains how to get the software running but not to convey what it does. Say you have code that‘s like grep but for json (as catchphrase). STE would be like „This program reads json data. This program finds patterns in json data“. Not catchy.
@mattpocockuk I would advise against this. Afaik STE has a vocabulary of ca. 900 words. It’s purpose is to convey HOW to do, simple terms, no ambiguity. Hedging, analogy, conditionals, what you want if you want to talk i.e about architecture and implementation, is actively suppressed by STE.
For my first post, I’m sharing a letter @NVIDIA signed on why open models matter.
AI will transform every industry, power every company, and be built by every country.
Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty.
The world needs both frontier closed models and frontier open models.
https://t.co/AUKzoQ5Ikb
@SaidHaschemi SAP and Telekom released the technological marvel which was the Corona app. Apparently, it cost 214 million. Because that went so well....here we are.
The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees.
The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance.
Access to all other Claude models is not affected.
We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible.
Read our full statement: https://t.co/bwn0sximKZ