Hacker News: loganmhb

New comment by loganmhb in "Building better AI tools"

loganmhb — Wed, 23 Jul 2025 16:08:51 +0000

Especially concerning in light of that METR study in which developers overestimated the impact of AI on their own productivity (even if it doesn't turn out to be negative) https://metr.org/blog/2025-07-10-early-2025-ai-experienced-o...

New comment by loganmhb in "Show HN: Semantic Calculator (king-man+woman=?)"

loganmhb — Thu, 15 May 2025 14:09:43 +0000

I once saw an explanation which I can no longer find that what's really happening here is also partly "man" and "woman" are very similar vectors which nearly cancel each other out, and "king" is excluded from the result set to avoid returning identities, leaving "queen" as the closest next result. That's why you have to subtract and then add, and just doing single operations doesn't work very well. There's some semantic information preserved that might nudge it in the right direction but not as much as the naive algebra suggests, and you can't really add up a bunch of these high-dimensional vectors in a sensible way.

E.g. in this calculator "man - king + princess = woman", which doesn't make much sense. "airplane - engine", which has a potential sensible answer of "glider", instead "= Czechoslovakia". Go figure.

New comment by loganmhb in "AI 2027"

loganmhb — Fri, 04 Apr 2025 15:55:20 +0000

I respect the forecasting abilities of the people involved, but I have seen that report described as "astonishingly accurate" a few times and I'm not sure that's true. The narrative format lends itself somewhat to generous interpretation and it's directionally correct in a way that is reasonably impressive from 2021 (e.g. the diplomacy prediction, the prediction that compute costs could be dramatically reduced, some things gesturing towards reasoning/chain of thought) but many of the concrete predictions don't seem correct to me at all, and in general I'm not sure it captured the spiky nature of LLM competence.

I'm also struck by the extent to which the first series from 2021-2026 feels like a linear extrapolation while the second one feels like an exponential one, and I don't see an obvious justification for this.

New comment by loganmhb in "Clean Code vs. A Philosophy Of Software Design"

loganmhb — Tue, 25 Feb 2025 15:31:07 +0000

Plenty of people are ragging (justifiably) on Clean Code, but I really admire by contrast Ousterhout's commitment to balanced principles and in particular learning from non-trivial examples. Philosophy of Software Design is a great and thought-provoking read.

New comment by loganmhb in "Range Partitioning: Zero to One"

loganmhb — Wed, 27 Mar 2024 14:29:42 +0000

Martin Kleppmann's book Designing Data-Intensive Applications is a great starting point if you're not familiar with it.

New comment by loganmhb in "Governments should compete for residents, not businesses"

loganmhb — Wed, 08 Mar 2023 18:01:28 +0000

The roads might actually cost more than that:

https://www.strongtowns.org/journal/2019/4/30/my-city-has-ma...

New comment by loganmhb in "GPT-3 can run code"

loganmhb — Tue, 29 Mar 2022 20:05:08 +0000

I wonder how much of this is an illusion of precision that comes from pattern matching on content from filler sites like https://www.free-hosting.biz/division/16-divided-7.html (I do not recommend clicking the link, but the result appears there).

New comment by loganmhb in "What’s so hard about understanding consciousness?"

loganmhb — Fri, 04 Feb 2022 21:41:46 +0000

The question is not whether perceptions are true -- that's irrelevant. Undoubtedly perceptions present a skewed and unreliable view onto reality. It's whether they exist at all. You can't trick someone who isn't looking.

> Except everybody knows there is no "I", you're just a bundle of atoms, and a bundle that's changing from moment to moment.

This is begging the question in the other direction.

New comment by loganmhb in "Simplifying Board Games"

loganmhb — Tue, 02 Jun 2020 18:17:12 +0000

I have heard excellent things about Crossfire but alas have not had the chance to play it -- hopefully one of these days!

New comment by loganmhb in "Simplifying Board Games"

loganmhb — Tue, 02 Jun 2020 17:51:21 +0000

I have this feeling sometimes too, but I think there is an important aspect of complex board games, in particular strategy games that is missing from computer games. When you play a board game, you are forced to understand the rules (because you are the one executing them) so you are able to more fully consider their implications on strategy. (Of course, the mechanics must be tasteful in addition to complex in order for this to actually be a benefit.) In a computer game, my experience is that it's much easier to revert to playing by feel and lose that effect, and much harder to design a game where the full mechanics are obvious to the players. As a wargamer this is the main reason I prefer playing board wargames, even though they are not able to simulate in nearly as much detail as computer wargames.

Digging Up Troy (2017)

loganmhb — Fri, 07 Apr 2017 11:55:25 +0000

Article URL: http://laphamsquarterly.org/roundtable/digging_troy

Comments URL: https://news.ycombinator.com/item?id=14058912

Points: 6

# Comments: 0

New comment by loganmhb in "Revisiting 64-bit-ness in Visual Studio and elsewhere (2015)"

loganmhb — Fri, 10 Mar 2017 15:26:50 +0000

When reading about Elasticsearch I was fascinated to learn that the JVM can often use 32-bit object pointers to address about 32GB of RAM when running in 64-bit mode: https://wiki.openjdk.java.net/display/HotSpot/CompressedOops

A Memory Allocator

loganmhb — Sat, 11 Feb 2017 19:55:44 +0000

Article URL: https://bckly.com/code/a-memory-allocator.html

Comments URL: https://news.ycombinator.com/item?id=13624421

Points: 1

# Comments: 0

New comment by loganmhb in "Programming Is Forgetting: Toward a New Hacker Ethic"

loganmhb — Sun, 18 Dec 2016 15:10:03 +0000

My impression was that the implications of sexism were directed at Levy, for glorifying the hackers who made Hamilton's work more difficult while trivializing her work to some degree, not so much at the hackers themselves, who indeed just broke something because they had an incomplete understanding of the system they were working on.

New comment by loganmhb in "Facebook's Filter Bubble Is Getting Worse"

loganmhb — Mon, 04 Jul 2016 15:06:05 +0000

Just because Facebook is making these decisions based on profitability doesn't remove them from the moral sphere, though. If the side effects of the decisions Facebook makes are substantial (which seems common to me with ad-based business models) people are perfectly justified in being outraged because of those side effects, regardless of whether the decisions are good for Facebook's profits.

New comment by loganmhb in "There Is No Handbook for Being a Writer"

loganmhb — Wed, 29 Jun 2016 11:58:47 +0000

I second the recommendation of Style: Lessons in Clarity and Grace. That book did more than any other to help me both write more clearly and explain to others how to do so (was a writing tutor in college). Many writing books give concrete advice on trivialities and retreat to vague prescriptions on subjects like elegance and clarity. This book is the only one I've read that actually studies those aspects of writing and offers actionable advice on how to make writing clearer, more graceful and more elegant.

New comment by loganmhb in "Simple Ways of Reducing the Cognitive Load in Code"

loganmhb — Tue, 28 Jun 2016 12:28:23 +0000

One of the nice things about the Clojure standard library is that most everyday functions actually do fit the first/last category (by design). Functions operating on sequences (map, filter, reduce, etc) take the sequence as the last argument and are suited for use with the ->> macro, while functions operating on data structures in a non-sequence context typically take the data structure first (assoc, conj, update) and are good for ->. So you get either:

    (->> (range 10)
         (map inc)
         (filter even?)
         (take 2)) ;=> '(2 4)

    (-> {:body {:some {:json :data}}}
        (assoc-in [:body :some :more-json] :more-data)
        (assoc :status 200)
        (update :body json/generate-string))
    ;;=> {:body "{\"some\":{\"json\":\"data\",\"more-json\":\"more-data\"}}", :status 200}

It doesn't work all the time, obviously, and it can be easy to get carried away with 15 threaded map/filter/reduce calls that should be factored into separate functions, but most of the time I find it to be a nice idiom that substantially improves readability.

New comment by loganmhb in "I wrote a small piece on planned death"

loganmhb — Wed, 04 May 2016 12:11:07 +0000

> One would think curing cancer ought to be amongst humanity's greatest concerted challenges.

One of the reasons "curing cancer" is so difficult is that cancer itself is not really one disease but many, with diverse causes and treatments. What works for one kind cancer doesn't work for another, so some kinds are fully curable and others unstoppable, with of course many in between. New research tends to chip away at variations of the disease without being able to cure it wholesale. If you're interested in the recent history of cancer and cancer medicine, the book "The Emperor of All Maladies" is a fascinating treatment of the subject.

New comment by loganmhb in "Composing Programs – Python 3 in the tradition of SICP"

loganmhb — Sun, 10 Apr 2016 18:42:12 +0000

That does not match my understanding of first class functions, nor Wikipedia's.[1] Python's inner defs are not anonymous functions (which is what people usually mean by lambdas) but they are definitely first class - they can be returned from a function, stored in a data structure and so on. I agree with the grandparent that the need to label the closure is just an inconvenience, not a disqualification.

You can't have anonymous lambdas that aren't first class (how would you reference them?) but you can have first class functions that are not technically anonymous.

[1]https://en.m.wikipedia.org/wiki/First-class_function

New comment by loganmhb in "NPM and Left-Pad: Have We Forgotten How to Program?"

loganmhb — Thu, 24 Mar 2016 12:17:15 +0000

If you like Javascript and you want to try a language with a good static type system, you might like Elm (http://elm-lang.org/). As a bonus, it has fantastic documentation and examples of small in-browser projets -- a clock, Pong, and so on.