Hacker News: hderms

New comment by hderms in "A Portal Connecting NYC to Dublin Opens in Flatiron Today"

hderms — Wed, 08 May 2024 21:25:45 +0000

is Dublin becoming more of a cultural center since Brexit?

New comment by hderms in "The Cloud Computer"

hderms — Thu, 26 Oct 2023 16:33:39 +0000

can you elaborate a bit? What you're saying sounds pretty interesting but I'm too ignorant to read between the lines

New comment by hderms in "Google assigns a CVE for libwebp and gives it a 10.0 score"

hderms — Tue, 26 Sep 2023 17:17:56 +0000

but isn't the whole point of io_uring that it's more performant?

New comment by hderms in "Blue-light glasses may not reduce eyestrain from screens, study says"

hderms — Mon, 21 Aug 2023 16:12:23 +0000

Yeah it's considerably more harmful in professional environments, imo. I forgot to mention in my initial comment that I feel like people underestimate how effective it is to drop the pretense that you already know enough about what's being discussed to learn more, and how 'disarming' it is to people to be asked instead of told.

Disarming in a sense that they will usually be inclined to react in a magnanimous way, instead of possibly adopting some other frictional communication pattern. It really just makes everything easier if you actively try to leave your ego at the door.

New comment by hderms in "Blue-light glasses may not reduce eyestrain from screens, study says"

hderms — Mon, 21 Aug 2023 16:06:44 +0000

yes, there are so many frustrating communication archetypes I feel like there should be a compendium/dictionary of them. Unfortunately, most of them seem ego-driven, so it takes a lot typically for people to grow out of them.

New comment by hderms in "Blue-light glasses may not reduce eyestrain from screens, study says"

hderms — Mon, 21 Aug 2023 10:34:55 +0000

Well, engineers are kind of the archetypal model of the "but ackshually..." guy, so I'd venture to say you're right. I used to be more bold in my pronouncements when I was younger, now I think I like to repeat stuff I've heard like "interesting, I had heard that X was a factor with this, do you know if that's true or not?" In these situations. Having some intellectual humility and actively seeking opinions of people who might be more informed goes a long way.

New comment by hderms in "Forget ‘quiet quitting’ – ‘loud laborers’ are killing workplace morale"

hderms — Wed, 16 Aug 2023 14:53:57 +0000

Bravo. I had a laugh from reading this which quickly turned to horror after I realized it might actually come true

New comment by hderms in "US urged to reveal UFO evidence after claim that it has intact alien vehicles"

hderms — Thu, 15 Jun 2023 14:22:32 +0000

I personally believe FTL is likely impossible so I think the only realistic hypothesis is if we encounter aliens, that they're von Neumann drones on the vanguard edge of a colonization wave from an alien race that's been doing it for millions of years.

If we survive for long enough as a species, either we'll meet races in this manner or we'll be the ones doing it.

For the record I still only assign a fairly low probability that this is what is happening, it's more likely that UAPs are craft from terrestrial origin.

New comment by hderms in "Programming with natural language is going to work (2010)"

hderms — Wed, 24 May 2023 11:29:50 +0000

Why does it suck, because it's doing a lot of allocations?

New comment by hderms in "Google DeepMind CEO says some form of AGI possible in a few years"

hderms — Wed, 03 May 2023 01:19:01 +0000

Well in real terms, isn't all it takes is something that's a convincing enough forgery? Like why the turing test was considered interesting at one point. I don't think drawing a line is necessary for us to quickly become caught off guard by developments in this space.

For the record, I don't believe we're close to AGI but I'm also pretty far from knowing anything about that field.

New comment by hderms in "Virtual Threads Arrive in JDK 21, Ushering a New Era of Concurrency"

hderms — Wed, 12 Apr 2023 13:30:59 +0000

Maybe io_uring?

New comment by hderms in "Things I learned after getting users"

hderms — Tue, 14 Mar 2023 13:22:42 +0000

Hey there! Thanks for your comment. What do you specifically mean by rescale/watermark/scrub content to protect users from themselves?

New comment by hderms in "FTX’s Sam Bankman-Fried cashed out $300M during funding spree"

hderms — Sat, 19 Nov 2022 04:13:58 +0000

Well replace "intelligent" with "competent" and what do you get? Should we not expect the people telling us all to give them money because we couldn't hope to make more efficient use of it to be able to exercise the critical judgement necessary to not get scammed by a blatant scam?

New comment by hderms in "The good delusion: has effective altruism broken bad?"

hderms — Wed, 16 Nov 2022 12:56:14 +0000

Maybe you can totally rank order goals or principles of morality within the confines of a single person, but how could you hope to do that for large groups of people without forcing everyone to assume the same things as you? Long-termism being a great example, because I personally, think the amount we should care about future people decays rapidly as we move into the future. It's not as if their lives don't have value equal to our own, but how can we hope to assess how many of them there may be or what aspects of the realities of human life will change over the next century even? This is like trying to predict the future by reading tea leaves, except you have to pledge 10% of your income to whichever leaf talks about the risk of general AI the loudest.

So should everyone be forced to value all future human lives as being as concretely valuable as our own? In order to put them into a total ordering of morality one is forced to make a judgement call on how many people we think are likely to exist in the future as well as how relevant our musings of today actually capture what life will be like in that future.

Effective altruists like to use exponential growth to beat people over the head with the idea that the future is more important than the present. Do they ever acknowledge the fact that the far future is also exponentially more unpredictable than seconds into the future?

It just reeks of hubris to me. To be a really effective altruist you must first create a time travel device.

New comment by hderms in "Build the Modular Monolith First"

hderms — Mon, 14 Nov 2022 23:21:22 +0000

Imo there's a huge distinction between microservices in a monorepo or microservices where there are 100s or thousands of git repos.

Monorepo can be much easier as you can at least verify the given build is consistent.

Monoliths don't completely avoid the problem of version compatibility, they just reduce it to a more clear layer of external clients and compatibility between N-1th release and Nth release. That's certainly worth it in the early stages of a project

New comment by hderms in "The Future of the Web Is on the Edge"

hderms — Thu, 06 Oct 2022 18:08:35 +0000

A relational system with the guarantees usually provided by them seems really difficult to do. Like cockroach db but considerably harder or with considerably worse performance.

I always thought it was more common for stuff like KV stores with looser guarantees

New comment by hderms in "Why there's no 'Dijon' in Dijon mustard"

hderms — Fri, 26 Aug 2022 02:32:54 +0000

Yeah, that's a great one.

New comment by hderms in "Why Does the U.S. Have the Best Research Universities? (2020)"

hderms — Thu, 25 Aug 2022 16:03:35 +0000

There are huge obstacles for foreigners to move to the US and work, but higher education is also one of the most straightforward paths to the immigration pipeline (afaik as someone who's never dealt with it themselves). I believe that indicates there are two opposing forces, so the relative proportions being off would likely be expected as we don't know which force is stronger.

New comment by hderms in "Woman ‘dehumanised’ by viral TikTok filmed without her consent"

hderms — Thu, 14 Jul 2022 16:38:29 +0000

Why are those questions the end all be all of the discussion? Many good things are done in the world that have a variety of motivations. He could get views in a variety of ways but the fact that he's doing it by helping people shouldn't be dismissed just because he wouldn't spend millions of dollars with no one watching.

Not to suggest he's operating in any way like a non-profit, but they all receive money and then distribute that money in ways that either increase the future money they can use for their mission or directly on the mission itself. The fact that he's built a self-sustaining model for helping individuals without merely forking over his own assets is impressive.

New comment by hderms in "How fast are Linux pipes anyway?"

hderms — Thu, 02 Jun 2022 14:51:03 +0000

with Rust you could also avoid using a lock on STDOUT and get it even faster!