Elektrine lite

← Feed

Ramkumar Ramachandra

artagnon@mathstodon.xyz

<p>Upstream LLVM contributor and Formal Methods researcher. Expect a mix of technical posts interspersed with cute real-world musings.</p>

Posts

  • Post #1667295

    The latest in the hunt for great kefir.

  • Post #1667294

    The latest in the hunt for great coffee.

  • Post #1667293

    Happy to report that my milk misadventures are behind me. I never expected it, but the best milk is available at an express supermarket (Tesco) near me! Better than any premium milk you get at large fancy supermarkets. This milk is thick with chunks of cream. Not for the weak.

  • Post #1667292

    Brilliant essay on the human tendency to constantly use metaphors to explain what we don’t even remotely understand: the human brain. We’ve been stuck on the information-processing metaphor for decades now. (via @linguacelta) https://aeon.co/essays/your-brain-does-not-process-information-and-it-is-not-a-computer

  • Post #1667291

    In the latest piece, I tease apart intelligence and ego, and argue that we successfully synthesized ego with our language machine. Spent nearly two days on it. https://artagnon.com/art/ego

  • Post #1667290

    My overall vision for what #LLVM VPlan should ideally be is becoming clearer. It&amp;#39;s going to be huge design overhaul, and the work has to be done piece-wise with no unit of work causing regressions. The hard part is convincing all stakeholders to align on the vision, which I cannot explain in natural language sufficiently well.

  • Post #1361967

    How apt: the headline mixes politics with technology 🤦

  • Post #1361966

    Ugh, what a security loophole! &amp;quot;FBI managed to extract Signal messages from a defendant’s iPhone by accessing the phone’s notification database, where incoming messages were viewable even after the app was deleted. Signal users may want to turn on the app’s settings feature that hides message content in notifications.&amp;quot;

  • Post #1361965

    Someone recommended that I read &amp;quot;Learned Optimism&amp;quot;, and wow, what a flaming pile of self-help pseudo-scientific garbage it is!

  • Post #1361964

    I found this 2015 interview of Hanya Yanagihara, author of A Little Life and The People in the Trees quite interesting. Few people understand humans as well as some fiction writers. https://electricliterature.com/a-stubborn-lack-of-redemption-an-interview-with-hanya-yanagihara-author-of-a-little-life/

  • Post #1361963

    I don&amp;#39;t have any Ali Smith lying around at home, and a friend sent me photographs of pages from his copy of Winter. It was a much-needed reminder of exactly how atrocious my writing is :)

  • Post #1361962

    Hm, the layout of Dublin is a bit strange: looks like the effect of some heavy zoning laws? Very large parts of it are highly residential.

  • Post #1361961

    Had a blast over the last couple of days at EuroLLVM! Flying back to London in the morning, but the current challenge is falling asleep!

  • Post #1361960

    I think what people often forget when discussing AI is that the thing is simply a computer program you run to get some output, not unlike git or tar. Unlike git and tar, which are highly reliable and useful programs, this unnamed program is highly unreliable and has questionable utility. Why are we all getting so riled up about a poorly-written computer program?

  • Post #1361959

    RE: https://nondeterministic.computer/@mjg59/116424732252818630 Look writers, we&amp;#39;re not coders. We have no skills to work on this machine that generates volumes of beautiful prose in a few minutes -- something that would take us a few months to produce. There is no way to turn our prose into precise instructions to a machine -- the skills of these coders lie out of our reach. It always has, and always will. Writing is an outdated craft. Let it go.