Hacker News: HybridCurve

New comment by HybridCurve in "US Court nullifies FTC requirement for click-to-cancel"

HybridCurve — Wed, 09 Jul 2025 13:51:31 +0000

The Asian Financial Crisis in 1997: deregulated capital flows allowed speculators to rapidly pull money out of countries like Thailand, causing their currencies to collapse. The IMF stepped in, but their 'rescue' packages demanded strict conditions- forced privatization, and further deregulation, which often made things worse. And let's not forget Black Wednesday, when speculators broke the Bank of England. This was called "a textbook case of a speculative attack enabled by capital mobility" which is a core neoliberal policy. Just like all politics: never trust the meaning or identity of something derived from it's headline, title, name, or label- those are always the first lies we are told.

New comment by HybridCurve in "Ask HN: Anyone struggling to get value out of coding LLMs?"

HybridCurve — Mon, 26 May 2025 13:29:30 +0000

I've only had great luck with the LLMs(chatgpt 3o) generated Perl code. It was able to synthesize code for a GTK2/3 application fairly consistently, without generating any syntax errors. Most of the code worked as described, and it seemed to make more mistakes misunderstand my descriptions of features rather than when implementing them. My colleagues suggested it was because Perl's popularity had fallen significantly before 2016, and the training data set might've had much less noise.

New comment by HybridCurve in "Ask HN: What skills do you want to develop or improve in 2025?"

HybridCurve — Wed, 25 Dec 2024 16:53:56 +0000

Definitely Andrew Ng's courses, especially for concrete explanations.

New comment by HybridCurve in "Large Text Compression Benchmark"

HybridCurve — Fri, 20 Sep 2024 17:26:55 +0000

I think it might be a moot point since the transformer run times scale very poorly and the algorithm has a symmetric run time.

New comment by HybridCurve in "Linux kernel contributor swatted and handcuffed live on stream [video]"

HybridCurve — Fri, 13 Sep 2024 15:59:23 +0000

I noticed that as well. Is there actually any reason he might've pissed off bunch of rustbros or are these just more baseless accusations?

Also, everyone should know if you have some type of online exposure or receive threats you should be able to contact you local law enforcement, inform them of the situation, and warn them someone might try to have you SWAT'd. I couldn't guarantee all police would respect a heads-up call but I've heard it has worked for many in the past.

New comment by HybridCurve in "We know 'Linux is a cancer'; could CentOS chaos spell opportunity for Microsoft?"

HybridCurve — Tue, 10 Sep 2024 20:09:01 +0000

Seeing that Microsoft hired Lennart Poettering a while ago, I don't think they ever intended to compete with the Linux market share by pushing Azure. It seems more like EEE from the Ballmer era. https://en.m.wikipedia.org/wiki/Embrace,_extend,_and_extingu...

Those of us who remember the SCO bullsh*t have not forgotten what they are capable of.

New comment by HybridCurve in "A game theory behind the dark forest strategy?"

HybridCurve — Mon, 09 Sep 2024 02:52:17 +0000

It seems to me that any technologically advanced civilization that is capable of completely annihilating another at such distances would likely be equally capable of avoiding detection, thereby reducing the potential threat level posed by the less advanced civilization. Avoidance would seem to be the best strategy unless the civilization posed a direct and imminent threat. Otherwise, the more advanced beings risk exposing themselves (by deploying a weapon) and declaring to the universe that they are both hostile and aggressive to others.

New comment by HybridCurve in "Four Thieves Vinegar Collective – Harm Reduction for the Living"

HybridCurve — Sat, 07 Sep 2024 19:22:53 +0000

While I can appreciate some of the intent to make certain healthcare more accessible, it is never a good idea to have anyone inexperienced attempt to perform some of the reactions required to synthesize medicines. This should always be done by experienced individuals with quality reagents and the proper lab equipment. While it might be easy to substitute a mason jar for a proper glass reaction vessel it is not so simple to find a substitute for a gas chromatograph/mass spectrometer in the back of your pantry.

IMO, The test equipment required to analyze the results of the reactions is generally most cost prohibitive aspect of this type of 'research'. And this is where I have a problem with these guys: I don't see any plans available for building any of that. Building many of these devices is not out of reach for a skilled individual, and it makes more sense to me that this equipment should more readily accessible than a glorified Keurig machine for drugs. This kind of arrogance and lack of respect for the discipline required in organic chemistry is going to result is someone getting hurt.

New comment by HybridCurve in "5 Russian GRU Officers, 1 Civilian Charged: Conspiring to Hack Ukrainian Gov't"

HybridCurve — Thu, 05 Sep 2024 23:10:57 +0000

Russia is not formally at war with NATO, and the indictment specifies crimes committed against 26 other member nations.

New comment by HybridCurve in "Zuckerberg claims regret on caving to White House pressure on content"

HybridCurve — Tue, 27 Aug 2024 23:38:06 +0000

For a while it seemed like they might have it out in a cage match, can we just do that instead?

New comment by HybridCurve in "NSA releases 1982 Grace Hopper lecture"

HybridCurve — Mon, 26 Aug 2024 21:25:36 +0000

With the type of work the NSA does, I can't imagine many of the didn't know who Grace Hopper was. I expect they did it out of respect for her, rather than for the benefit of the general public.

New comment by HybridCurve in "What good are whizzy new drugs if the world can't afford them?"

HybridCurve — Fri, 21 Jun 2024 19:34:25 +0000

I just paid $5.3k in the US for four 350mg doses of Albendazole (antiparasitic). This medication is something like $2 or less anywhere else in the world. This is not simply a problem with new drugs, but lack of pricing regulations from the federal government enabling pharma companies to fleece patients and insurance providers.

New comment by HybridCurve in "Using AI to spot edible mushrooms could kill you"

HybridCurve — Thu, 21 Mar 2024 05:44:49 +0000

This is exceedingly bad advice as there are several spongy mushrooms which are toxic in the US: Bolete Huronensis, Sutorius Eximius, Boletus Sensibilis to name a few. There are a few in genus Leccinum as well.

New comment by HybridCurve in "We were wizards – a foreword to Learning Perl (1993)"

HybridCurve — Sun, 17 Mar 2024 21:52:56 +0000

Some of the arguments about Perl and bad engineering practice are not entirely without merit. Neglecting 'use strict;' or 'use warnings;' in your script typically makes this more apparent. The language itself however does not attempt to constrain developers in the same way modern languages do, instead the burden of responsible engineering is placed on the developer. The result is people that deal with badly engineered Perl typically become sour to the language itself.

I reserved my criticisms of Perl for things like signal handling and it's implementation OOP(to name a few), but overall it's a good language.

New comment by HybridCurve in "Operation Triangulation What You Get When Attack iPhones of Researchers [video]"

HybridCurve — Wed, 03 Jan 2024 16:47:20 +0000

Necessity is the mother of all invention. This could just be related to socioeconomic factors.

New comment by HybridCurve in "Fidelity marks down X valuation by 71.5%"

HybridCurve — Tue, 02 Jan 2024 23:25:46 +0000

This is why I hate when anyone puts this guy on a pedestal. I can't be certain about his level of knowledge in most things, but with comments like this he regularly demonstrates lack of credibility due to either complete ignorance a highly self-centeric view of the world. Either way, he is not to one be trusted, admired, or emulated.

New comment by HybridCurve in "Mathematical Introduction to Deep Learning: Methods, Implementations, and Theory"

HybridCurve — Tue, 02 Jan 2024 04:26:07 +0000

Ah, I think I remember bookmarking this when it was posted before. You really don't have to go very far in computing to find a frontier where most everything in described pure mathematics and so it becomes a substantial barrier for undiversified autodidacts in the field. The math in these areas can often be quite advanced and difficult to approach without the proper background and so I appreciate anyone who has made taken the time to make it less formidable to others.

New comment by HybridCurve in "Mathematical Introduction to Deep Learning: Methods, Implementations, and Theory"

HybridCurve — Tue, 02 Jan 2024 03:03:11 +0000

I appreciated the notation in Goodfellow book as well, it was easy enough for me to follow without having a strong mathematics background. I'll agree however with others that this text is instead focused for a different audience and purpose.

New comment by HybridCurve in "Mathematical Introduction to Deep Learning: Methods, Implementations, and Theory"

HybridCurve — Tue, 02 Jan 2024 01:27:49 +0000

This is the probably the most succinct explanation, and as an experienced perl developer, I admire your brevity.

New comment by HybridCurve in "Mathematical Introduction to Deep Learning: Methods, Implementations, and Theory"

HybridCurve — Tue, 02 Jan 2024 01:21:18 +0000

This is essentially my problem. I started writing programs at a young age and was introduced (unknowingly) to many more advanced mathematical concepts from that perspective rather than through pure mathematics. What was it that helped break this paradigm for you?