Two things accompanying the rise of formal verification that make me uneasy:
1. Right now, frontier labs already flag (or refuse outright) FV tasks even on your own code, because FV can be used for attacks.
2. The (still nascent) tendency to gatekeep FV RL datasets to sell fine-tuned models as a service. FV experts need to earn a living but it’s such a pity if openness has to be sacrificed in the process.
This is why I’m placing a lot of hope on type theory. Using a language that is correct by construction is absolutely above board, and I don’t think type theorists will start to gatekeep type theories anytime soon.
This Diwali, I would strongly recommend that when you're considering giving, please consider giving to organisations like @internetfreedom and @SFLCin (and I'd add supporting @medianama as well. though we don't take donations), because the work we all do helps protect our freedoms online, especially in an environment in India that is negatively impacting free speech, privacy, and the ease of doing business online.
Our role is to hold governments and companies to account and help maintain the balance of power, and things look dire for safe harbor right now: the removal impacts us as citizens, and businesses we work with, not just Meta and Google, alike.
What happens when that support for these organisations vanishes?
Well look at what happened to the fount of digital rights research in India: CIS India, that helped support the learning that many of us gained related to digital rights.
Drag-and-drop makes the digital graspable, but building it well is fiddly enough that many don’t bother. Declarative Dragging (with Josh Horowitz & Jeff Heer, #uist2026 best paper award), lets you make an element draggable just by listing the states dragging it can reach.
Looks like this was picked up by an influencer on Instagram. 80k+ views, 1k+ likes and comments!
instagram.com/reels/DfDKNo6s…
I see where the engagement is :-)
FP Launchpad is running a PL Winter School between Dec 7th and 11th at IIT Madras: fplaunchpad.org/winter-schoo….
Application to attend will open early next week. Please share widely.
That’s a wrap on our October meetup!
Thanks to @FunctionalIndia for collaborating with us, our speakers Sanchayan, Kartik & Benjamin, and everyone who joined us.
Thanks to @EPAMSystems for hosting us!
Announcing our third speaker — Kartik Sahu from EPAM!
Topic: Lumen – AI for BI
A conversational BI agent that turns business questions into analysis, explainable insights, evidence-backed reasoning and interactive visualizations.
24 Oct | @EPAMSystems Office, Bengaluru
Haskell + Elm + Nix + deploying to a nixOS VPS is the most pleasurable experience ever, and the best part is LLMs are good enough to nail this stack already… haven’t yet seen a weird bug in @konekokanji and the stupid bugs are 1 line UI fixes! 🙌🏻💯
Announcing our next speaker for the October Meetup — Benjamin Philip!
He’ll be speaking on “The BEAM Odyssey: From Industrial Telephony to Modern Domains”,
In collaboration with @FunctionalIndia .
Thanks to @EPAMSystems for hosting us.
24 Oct | 10:30 AM – 2:00 PM | Bengaluru
Announcing our first speaker for the October Meetup -Sanchayan Maity!
He’ll be speaking on “Introduction to Haskell”, exploring Haskell and some of the ideas behind functional programming.
This edition is in collaboration with @FunctionalIndia.
You wrote "Remember IBM's UML and Rationale Rose, etc.? Those were created to communicate bugs and defects from US based analysts to Indian bases devs".
You have a remarkable relationship with the truth.
I'm the inventor of the UML.
I created the Booch method which morphed into the UML, and both were designed to visualize, reason about, construct, and document the artifacts of software-intensive systems.
Furthermore, the UML and Rational Rose existed about a decade before IBM acquired Rational.
I suggest you reconsider your life choices.