Hacker News: rp1

New comment by rp1 in "Retailers say thefts are at crisis level. The numbers say otherwise"

rp1 — Mon, 20 Dec 2021 19:09:18 +0000

The article makes no mention of changes to insurance claims.

New comment by rp1 in "Everything I googled in a week as a professional software engineer (2019)"

rp1 — Mon, 20 Dec 2021 04:20:15 +0000

My searches are mostly for esoteric error messages that I tweak a dozen times to get Google to stop returning unrelated results. Then I usually give up, get lunch, and the answer comes to me when I sit back down.

New comment by rp1 in "Connecting the iDOTs (2020)"

rp1 — Sun, 19 Dec 2021 05:39:48 +0000

True, bandwidth is a concern. It might be possible to copy memory directly from disk to the GPU, which may be more efficient.

New comment by rp1 in "“With those changes we're up to a 94% pass rate for dEQP-GLES2”"

rp1 — Sun, 19 Dec 2021 04:29:23 +0000

Ultimately, the developers on Asahi are trying to play catch-up with massive GPU driver teams comprised of dozens of talented engineers. It’s not feasible to catch up, but I’d be happy to be proven wrong.

New comment by rp1 in "Connecting the iDOTs (2020)"

rp1 — Sun, 19 Dec 2021 02:18:22 +0000

Bitmaps take memory, but are quite efficient from a runtime perspective because they aren’t encoded.

New comment by rp1 in "JPMorgan hit with $200M in fines for letting employees use WhatsApp"

rp1 — Sat, 18 Dec 2021 17:13:34 +0000

You still haven’t given an example of how coordinating the two agencies can lead to a bad outcome. If all that can be done is enforce the law, then there is no issue. You keep implying something extra-legal can be done through the coordination, but aren’t providing any specifics.

New comment by rp1 in "Man dies after he is shot by the police with a stun gun and catches fire"

rp1 — Sat, 18 Dec 2021 17:10:19 +0000

Having 10 officers beat an individual with clubs doesn’t exactly seem good either…

New comment by rp1 in "JPMorgan hit with $200M in fines for letting employees use WhatsApp"

rp1 — Sat, 18 Dec 2021 16:58:37 +0000

Normally people say that power is being wielded dangerously when the power is used to do something disagreeable. Simply enforcing the law correctly doesn’t seem dangerous to me.

New comment by rp1 in "YouTube TV loses ESPN, ABC and other Disney channels in fee dispute"

rp1 — Sat, 18 Dec 2021 16:56:52 +0000

This is just a small microcosm of what’s going on in the cable TV world. Many channels are raising their fees, but the cable providers are very reluctant to raise the price charged to end users. Instead, cable providers are opting to drop channels. This is leading to an increasingly fragmented cable TV market where it’s impossible to get all the channels you may want to watch. For instance, Comcast just dropped MSG, which is a channel for NY-based sports. The only alternative many people have is to use FuboTV, but Fubo has dropped Adult Swim for the same reason. So now it’s impossible to get both MSG and Adult Swim.

New comment by rp1 in "JPMorgan hit with $200M in fines for letting employees use WhatsApp"

rp1 — Sat, 18 Dec 2021 16:47:44 +0000

This would be more convincing if you had specific reasons, unrelated to the enforcement, that JPM shouldn’t be punished for this.

New comment by rp1 in "Man dies after he is shot by the police with a stun gun and catches fire"

rp1 — Sat, 18 Dec 2021 16:19:37 +0000

That’s not really an answer to the question. It would be nice if all situations could be de-escalated, but that’s obviously not the case. De-escalation also has its own risks for the officer. Again, not saying I support what happened, but it seems so hard to manage a situation with a belligerent person when I imagine trying to do it myself. Even doing something relatively benign, like removing a loitering person, is not something I’d want to attempt.

New comment by rp1 in "Man dies after he is shot by the police with a stun gun and catches fire"

rp1 — Sat, 18 Dec 2021 16:04:36 +0000

How else would you subdue someone? You can’t expect all cops to be able to go 1-1 or even 2-1 with someone to subdue them physically. Cops also have guns strapped to their sides that can be reached for. I’m not justifying their behavior, but when I try to imagine subduing someone, I’m not really sure how I would do it. There is definitely room for tools that make it easier.

New comment by rp1 in "Why Wolfram tech isn’t open source (2019)"

rp1 — Sat, 18 Dec 2021 15:19:46 +0000

Almost none of the reasons have anything to do with releasing the source. You can release the source but still maintain central control. They shouldn’t even have posted this. Obviously they want to make money from their products, and that’s fine.

New comment by rp1 in "Group of monkeys kill over 250 dogs for 'revenge' in Indian town"

rp1 — Sat, 18 Dec 2021 15:12:05 +0000

Why are they evil? Or at least more evil than humans? Humans have eradicated wolves almost everywhere due to the danger they pose humans, particularly children. How is this situation different? The monkeys are even targeting the very same species that humans have long targeted.

New comment by rp1 in "Log4j 2.16: Certain strings can cause infinite recursion"

rp1 — Sat, 18 Dec 2021 15:05:22 +0000

I think you’re generalizing way too much. Java is old. The libraries are old. There is a lot of bad old code out there, and very little glory in fixing it.

New comment by rp1 in "The empathy gap, or why good-seeming ideas fail"

rp1 — Fri, 17 Dec 2021 02:55:44 +0000

Maybe… but the article had an anecdote of asking Ian Goodfellow for ideas and none of those ideas working. I would assume Ian Goodfellow would have the requisite amount of experience for sufficient intuition.

New comment by rp1 in "Polars: Fast DataFrame library for Rust and Python"

rp1 — Fri, 17 Dec 2021 02:28:31 +0000

This question was asked last time the author posted this few months ago. I’m surprised they didn’t update the benchmarks. Kind of makes me think Vaex is faster.

New comment by rp1 in "Animal protein intake is inversely associated with mortality in older adults"

rp1 — Fri, 17 Dec 2021 00:50:07 +0000

Mortality is often used when talking about the likelihood of dying.

New comment by rp1 in "Can “Distraction-Free” Devices Change the Way We Write?"

rp1 — Fri, 17 Dec 2021 00:21:34 +0000

You could write the above easily with a typewriter. The point isn't to use the most primitive method of writing possible. Rather, it's to acknowledge that writing on a computer has a certain set of drawbacks in addition to its benefits.

New comment by rp1 in "The empathy gap, or why good-seeming ideas fail"

rp1 — Thu, 16 Dec 2021 23:52:17 +0000

Your experience rings true to me. It's one of my biggest frustrations with ML at the moment. There are so many ideas I'd like to try, but I know only a small fraction of them will work, and discovering which of the ideas fail and which succeed is a herculean task. Your conclusion that empathy helps may also be true, but I have a different take.

It currently takes way too much time to explore ML-based ideas. I compare this to the early days of computer programming where programmers needed to manually fill out punch cards, and doing anything took days of full time work. There is lots of room for improvement along every step of the ML pipeline, from data wrangling, model choice, training, and evaluation. Good ML tooling will likely bring huge gains in the field.