2026-05-25T05:03:58Z https://yabu.me Brian "bex" Exelbierd https://yabu.me/npub1yu0q89v3mz8jtmj8zqdmdnhn4znjx88t35gq5g7j9u00kp4m3j6s8f27ev https://s3-toot-io.b-cdn.net/accounts/avatars/113/010/512/519/992/743/original/1a95ae8af674328e.jpg https://s3-toot-io.b-cdn.net/accounts/avatars/113/010/512/519/992/743/original/1a95ae8af674328e.jpg https://yabu.me/nevent1qqsyfsd2rpgh9ad48mg0fyyuwhgtf7ze3tyhzegjpzk3v6d932mhvvqzyqn3uqu4j8vg7f0wgugphdkw7w52wgcuawx3qz3r6gh3a7cxhwxt20lytzt In reply to <a href='/nevent1qqsd4tdzjqk9mukhsfacglx9reqmuaqv8c5tczteta6tcnmxqtammycwmq8zz'>nevent1q…q8zz</a><br/>_________________________<br/><br/>I use podman for all my unorchestrated containers. Love it. How we I’ve stayed away from podman secrets as they used to be written to disk in plaintext. Did that get fixed? 2026-04-06T19:17:48Z https://yabu.me/nevent1qqsy7zdug3q8a2k7k86gpkg5qr3pvcqv0d6cngv2y3c6pvx4wquj0kgzyqn3uqu4j8vg7f0wgugphdkw7w52wgcuawx3qz3r6gh3a7cxhwxt23lxj2k <span itemprop="mentions" itemscope itemtype="https://schema.org/Person"><a itemprop="url" href="/npub1gv26rplqyjqcfyhxryuqjwaxs0dwve3y6gpv6smn3hjm3wse3s8squtlwl" class="bg-lavender dark:prose:text-neutral-50 dark:text-neutral-50 dark:bg-garnet px-1"><span>Simon Willison</span> (<span class="italic">npub1gv2…tlwl</span>)</a></span> Have you looked at Fara-7B? I’m wondering if I missed a comment from you. It claims to operate as a local only computer use model at the pixel level. 2025-12-04T16:39:59Z https://yabu.me/nevent1qqstzvht2j0wgpcawkngw0q64yrhggn9av98ey2h4tk9lpyttxwn4cgzyqn3uqu4j8vg7f0wgugphdkw7w52wgcuawx3qz3r6gh3a7cxhwxt2v2ktkk In reply to <a href='/nevent1qqswsx693nehvcfeuv0xcj0je7fhwwnhmg6axzgdj3uwratx8va7uss28y473'>nevent1q…y473</a><br/>_________________________<br/><br/>interesting. So you’re still driving the creation and curation of the tags and not, for example, having the LLM suggest tags for specific posts or that have emerged as themes unbidden over time? 2025-05-19T17:02:56Z https://yabu.me/nevent1qqsp6f99sq64t65hpm3aks7qn0au04n0qvnxn5xynxlns8mpsppfndgzyqn3uqu4j8vg7f0wgugphdkw7w52wgcuawx3qz3r6gh3a7cxhwxt2tlcxjr In reply to <a href='/nevent1qqs03yyc83qewvaaa8teslnd0fqywdng4x068s6nr9sutyjq6hzmhwgpp2dff'>nevent1q…2dff</a><br/>_________________________<br/><br/>did you do the backfill by hand or with LLM assistance in identifying the appropriate posts? 2025-05-19T06:50:15Z https://yabu.me/nevent1qqst5vmff2ncf6fhg9sj05yytt6l848e3940d5rm7jwgxl3z3g7n6cszyqn3uqu4j8vg7f0wgugphdkw7w52wgcuawx3qz3r6gh3a7cxhwxt29qrl7k I&#39;ve been thinking about &#39;vibe coding&#39; as I use LLMs, and <span itemprop="mentions" itemscope itemtype="https://schema.org/Person"><a itemprop="url" href="/npub1gv26rplqyjqcfyhxryuqjwaxs0dwve3y6gpv6smn3hjm3wse3s8squtlwl" class="bg-lavender dark:prose:text-neutral-50 dark:text-neutral-50 dark:bg-garnet px-1"><span>Simon Willison</span> (<span class="italic">npub1gv2…tlwl</span>)</a></span> &#39;s recent post on the topic pushed me to write down some thoughts. I&#39;ve used AI to build apps in unfamiliar languages, but how much accountability for every line of code is required? I share my experience with Swift, Kotlin, and a Go-lang 1Password tool. Is this just assisted development, or does it cross into vibe coding? What’s your take?<br/><br/><a href="https://www.winglemeyer.org/technology/2025/03/11/did-i-vibe-code">https://www.winglemeyer.org/technology/2025/03/11/did-i-vibe-code</a><br/><br/>#AI #SoftwareDevelopment #Ethics 2025-03-11T09:14:00Z