Hacker News: dworks

New comment by dworks in "Cursor's Composer 2 model may be rebranded Kimi K2.5"

dworks — Fri, 20 Mar 2026 16:20:05 +0000

funny how its not on the front page despite the score

Cursor's Composer 2 model may be rebranded Kimi K2.5

dworks — Fri, 20 Mar 2026 15:55:08 +0000

Article URL: https://twitter.com/thdxr/status/2034978341594231245

Comments URL: https://news.ycombinator.com/item?id=47456411

Points: 3

# Comments: 2

New comment by dworks in "Source code of Swedish e-government services has been leaked"

dworks — Sat, 14 Mar 2026 00:28:21 +0000

Go back and edit your original comment because it is irrelevant and misleading.

New comment by dworks in "Source code of Swedish e-government services has been leaked"

dworks — Sat, 14 Mar 2026 00:00:37 +0000

You can trivially purchase the data from Bisnode Dun & Bradstreet Sverige.

New comment by dworks in "Source code of Swedish e-government services has been leaked"

dworks — Sat, 14 Mar 2026 00:00:06 +0000

Yes, you can buy the database for the entire population. There are commercial vendors for this, one of them is Dun & Bradstreet (Bisnode Dun & Bradstreet Sverige).

New comment by dworks in "Source code of Swedish e-government services has been leaked"

dworks — Fri, 13 Mar 2026 23:41:35 +0000

The OP didn't claim it had happened to you. What they said is that it is possible to use the information about regular individuals that is publicly available to cause harm, and there are no attempts to stop this.

New comment by dworks in "Source code of Swedish e-government services has been leaked"

dworks — Fri, 13 Mar 2026 23:39:21 +0000

they don't handle it at all. they let it go on. you for example have hundreds of people falsely registering their place of residence as somebody else's home, which causes massive problems for that home owner or apartment resident, and there is nothing done about it at all.

These types of laws are designed for the 1950s where there were natural barriers to acquiring and disseminating information. There is no attempt whatsoever to update them and to reduce harm caused to the average citizen today.

New comment by dworks in "Launch HN: Spine Swarm (YC S23) – AI agents that collaborate on a visual canvas"

dworks — Fri, 13 Mar 2026 23:24:08 +0000

What his landing page shows is a workflow builder like n8n. It is indeed better than a window manager. I'm working on something related.

New comment by dworks in "Willingness to look stupid"

dworks — Fri, 13 Mar 2026 05:25:56 +0000

No, it does. Even if the audience knows that your English or other languages is perfectly professional, speaking Chinese at a lower level does leave a certain negative impression.

New comment by dworks in "Willingness to look stupid"

dworks — Fri, 13 Mar 2026 04:57:11 +0000

A willingness to look stupid is a core requirement for learning languages. I look stupid everyday.

New comment by dworks in "“Design me a highly resilient database”"

dworks — Fri, 13 Mar 2026 02:16:13 +0000

Maybe globally spread out POS servers if that's how it works?

New comment by dworks in "High fidelity font synthesis for CJK languages"

dworks — Fri, 13 Mar 2026 01:45:09 +0000

This is awesome. I have a real need for a suitable Chinese UI typeface that pairs well with a sans-serif.

New comment by dworks in "Malus – Clean Room as a Service"

dworks — Thu, 12 Mar 2026 22:33:10 +0000

This could work out great, because the OSS devs can focus on building their project instead of marketing to businesses, running sales processes, consulting on implementation and supporting the implementation. No need to find corporate sponsors either.

New comment by dworks in "Kotlin creator's new language: talk to LLMs in specs, not English"

dworks — Thu, 12 Mar 2026 21:30:55 +0000

You need to lock the specs and implementation plan and verify the implementation about the previous phase docs.

https://github.com/doubleuuser/rlm-workflow

New comment by dworks in "Kotlin creator's new language: talk to LLMs in specs, not English"

dworks — Thu, 12 Mar 2026 21:27:08 +0000

>I do think there are opportunities in this space, but what I'd like to see is:

>* write text specifications

>* model transforms text into a formal specification

>* then the formal spec is translated into code which can be verified against the spec

This skill does just that: https://github.com/doubleuuser/rlm-workflow

Each stage produces its own output artifact (analysis, implementation plan, implementation summary, etc) and takes the previous phases' outputs as input. The artifact is locked after the stage is done, so there is no drift.

New comment by dworks in "Claude now creates interactive charts, diagrams and visualizations"

dworks — Thu, 12 Mar 2026 20:56:24 +0000

I think the problem should be defined as "why does it not loop back the errors from the first attempt so it can fix it on the second attempt" rather than why it fails to produce a fully correct implementation on the first pass.

New comment by dworks in "Agents that run while I sleep"

dworks — Wed, 11 Mar 2026 10:54:46 +0000

Sorry about that. Let me push an update.

New comment by dworks in "Agents that run while I sleep"

dworks — Tue, 10 Mar 2026 22:56:22 +0000

I built rlm-workflow which has stage gating, TDD and sub-agent support: https://skills.sh/doubleuuser/rlm-workflow/rlm-workflow

New comment by dworks in "Agents that run while I sleep"

dworks — Tue, 10 Mar 2026 22:55:26 +0000

rlm-workflow does all that TDD for you: https://skills.sh/doubleuuser/rlm-workflow/rlm-workflow

(I built it)

New comment by dworks in "The death of social media is the renaissance of RSS (2025)"

dworks — Mon, 09 Mar 2026 12:46:05 +0000

thanks appreciate it! you can boost sources and categories (youtube and substack have a boost by default since they're published less frequently than reddit posts), create priority views, filters etc.

planning to set up 'topics' which can be a feed you can subscribe to that combine different source types, that are tagged and associated to the topic based on previous content.

also, email send outs per topic.