Iβm intrigued by the idea that old codebases can be better than newer codebases b/c of decades of work squashing bugs. I kinda imagine like old growth forest vs a new garden. The old ecosystem is mature and complicated and tangled. It works (maybe better?) but it appears messier than new code.
31.03.2025 17:36
π 2
π 0
π¬ 1
π 0
Just applied! Iβve been inspired by @newpublic.substack.com on Bluesky and I would love to contribute!
27.03.2025 21:22
π 1
π 0
π¬ 0
π 0
At some point will LLMβs be so normalized that people learn βLLM-nativeβ coding skills without prior coding experience? I bet so. Iβm curious what itβll look like.
For now, knowing how to code already is just a massive head start to learning how to effectively do human-in-the-loop LLM coding.
11.03.2025 21:13
π 2
π 0
π¬ 0
π 0
Iβm a big fan of thoughtful takes on how LLMβs are influencing coding. My fav section title here: βLLMs Amplify Existing Expertiseβ.
My take: having coding expertise allows you to develop this new human-in-the-loop LLM coding skillset. Coding with LLMβs is a skill you have to develop.
11.03.2025 21:11
π 3
π 0
π¬ 1
π 0
This seems like a good time to reflect a little on Judith Butlers writings about nonviolence. Part of her analysis involves questioning who gets to define violence, and when states use this semantic power as a form of oppression. lithub.com/judith-butle...
11.03.2025 20:48
π 0
π 0
π¬ 0
π 0
What. The. Actual. Fuck?!???!!
11.03.2025 20:35
π 0
π 0
π¬ 1
π 0
βI think society starts to reflect the structure of its dominant form of communication and so you need communications infrastructure that is democraticβ - @jay.bsky.team πππ
11.03.2025 20:22
π 2
π 1
π¬ 0
π 0
Great thread about how lack of social-media scheduler support for @bsky.app is an obstacle for larger orgs to move here. Hopefully this will continue to change as the Bluesky community grows
11.03.2025 20:09
π 10
π 4
π¬ 0
π 0
As a billionaire, the best things you can do are: (1) redistribute your wealth in a way that you cede control and give power back to the people, and (2) use your platform to deconstruct myths that billionaires are uniquely talented + smart (sorry if that hurts your ego, but look at Elon - itβs true)
11.03.2025 20:01
π 0
π 0
π¬ 0
π 0
@mcuban.bsky.social Wealth is power, and Iβm skeptical of the amount of power billionaires have. Weβve seen billionaires become a threat to democracy. Weβve seen billionaires move towards becoming oligarchs. But that doesnβt mean that you canβt individually use your wealth + power for good
11.03.2025 19:54
π 0
π 0
π¬ 1
π 0
Alternative business models not only exist, but can also function successfully within capitalism and help improve our systems from the inside-out
11.03.2025 19:52
π 0
π 0
π¬ 1
π 0
And many smaller utility companies are great examples of co-opβs that serve their communities better than massive utility conglomerates.
11.03.2025 19:51
π 0
π 0
π¬ 1
π 0
Oh, and #communitytechalliance as well!
11.03.2025 19:50
π 0
π 0
π¬ 1
π 0
There are successful examples of tech nonprofits running at least partially revenue based business models (#enveritas and @watttime.org come to mind)
11.03.2025 19:49
π 1
π 0
π¬ 1
π 0
And while youβre at it, why not just buy out the @bsky.app investors and turn Bluesky isnβt either a tech non profit or a co-op?
11.03.2025 19:48
π 1
π 0
π¬ 0
π 0
I know @mcuban.bsky.social loves @bsky.app. Itβs time for him to fully fund @freeourfeeds.com and take a massive step towards securing blueskyβs future as a better social media.
11.03.2025 19:47
π 0
π 0
π¬ 2
π 0
While Iβm at it - @mcuban.bsky.social I might not like billionaires as a class, but Iβd certainly reconsider you as an individual if you fully funded @freeourfeeds.com π
11.03.2025 19:42
π 0
π 0
π¬ 0
π 0
I believe @jay.bsky.team and her team are genuinely working for social good, and fighting for social media without billionaire control. Itβs good to be clear-eyed about the risks of βenshittificationβ that @bsky.app still faces, while also celebrating how amazing this app + the ppl working on it are
11.03.2025 19:32
π 0
π 0
π¬ 1
π 0
Until then, itβs critically important to secure the future of the AT Protocol independent from Bluesky. Donate to @freeourfeeds.com! The AT Protocol can serve as a key check-and-balance on the power of a for-profit @bsky.app
11.03.2025 19:27
π 1
π 0
π¬ 1
π 0
@jay.bsky.team is at her weakest when talking about monetization. History suggest VC wants returns, not social good. If Bluesky investors want social good, they should have a clear exit plan that allows @bsky.app to become a nonprofit, a co-op, or another innovative org structure.
11.03.2025 19:23
π 1
π 0
π¬ 1
π 0
Great talk, and highlights how Bluesky is giving users control and letting users own their data. These are massive wins! #FreeOurFeeds
11.03.2025 19:19
π 0
π 0
π¬ 1
π 0
As someone looking for work as a SWE and not changing careers I guess my bet is: (1) I can still do a lot of important stuff gen ai canβt do, and (2) I have a specific skill set that lets me use gen ai better than non coders. Itβs human-in-the-loop, and the human is the lynchpin.
11.03.2025 19:17
π 0
π 0
π¬ 0
π 0
I heard gen ai is to software engineering as the calculator was to accountants. It drastically transformed the field. But it turns out accountants do a lot more than just calculate ledgers. But, calculators could never do the other stuff accountants do. TBD if gen ai can do the other stuff SWE do.
11.03.2025 19:11
π 1
π 0
π¬ 1
π 0
Real codebases involve work not needed in vibe coding (e.g. story refinement, buy-in from dependent teams, hyper-specific implementations, complex side effects, code style, code reviews, testing, deployment, monitoring, ...). Gen ai is great at coding. Itβs just not great at being a SWE.
11.03.2025 19:08
π 1
π 0
π¬ 1
π 0
Mundus sine Caesaribus
11.03.2025 03:45
π 0
π 0
π¬ 0
π 0
Well worth watching the full talk!
11.03.2025 03:44
π 0
π 0
π¬ 0
π 0
π―π―π―
11.03.2025 03:42
π 0
π 0
π¬ 0
π 0
Shout it from the rooftops! Algorithms can be better! #holdtechaccountable
07.03.2025 01:57
π 2
π 1
π¬ 0
π 0
I cry.
28.02.2025 16:36
π 3
π 0
π¬ 0
π 0