mvdtnz's comments

On: A few words about FiveThirtyEight

[mvdtnz]

Are American elections decided on the popular vote?

[intermerda → mvdtnz]

Nearly all of them, yes.

[lurk2 → mvdtnz]

Not presidential elections, no.

On: In Severance, Office Perks Couldn't Be More Sinister

[mvdtnz]

I loved season one. I'm struggling with this season. [Minor mid-season spoiler ahead] Since they entered the goat room I just feel like I don't want to continue. Ever since Lost finished I have become extremely skeptical of shows that raise more questions than they answer, especially when they are really, really weird questions.

[ericwood → mvdtnz]

When it was screened to critics it was shown as the full season. I get why now; the latest episodes show so much while conjuring up more questions. The most recent one especially; the cinematography stands on its own. It's very rewarding to continue watching!

[klipklop → mvdtnz]

I was going to make a parallel to Lost as well, but you beat me to it! I am considering dropping off this show too because it seems like they are just presenting us with mysterious gibberish teases instead of actual plot.

[viraptor → mvdtnz]

I feel like it really depends on how you introduce the questions. Lost made a big thing out of some of them in a way, which can be annoying. On the other hand, twin peaks made some things just a part of the world. What are owls? I don't know, but the fact they're something else is interesting and that's fine.

On: James Harrison, whose blood donations saved >2M babies, has died

[mvdtnz]

That doesn't answer the question.

On: Show HN: FlakeUI

[mvdtnz]

I don't think it's very likely that his 48gb machine is over a decade old.

[yencabulator → mvdtnz]

I have a 64GB i7-3930K from 2012 next to me. It still worked in late 2024, and was faster than many modern computers.

[tracker1 → mvdtnz]

Looks like it's an 8th Gen with mixed ram running single channel mode. The point stands though, the brand segment is meaningless over time.

It's 5-6yo just before Ryzen got competitive with Intel. A current option should be noticeably faster.

On: Show HN: Sonauto API – Generative music for developers

[mvdtnz]

Why would a hobbyist need an unlimited plan?

[zaptrem → mvdtnz]

E.g., in the case of a future "LibreMusic" open source UI or an integration into their DAW they work with on the weekends. I'd get pretty annoyed if I had to keep putting a coin in the machine to adjust Logic Pro effects.

On: Show HN: Recommendarr – AI Driven Recommendations Based on Sonarr/Radarr Media

[mvdtnz]

Don't LLMs usually have knowledge cutoffs from months or years ago? Does that mean it came recommend new shows?

On: Permacomputing

[mvdtnz]

I wonder what that has to do with "anarchism, decoloniality, intersectional feminism, post-marxism, degrowth, ecologism".

I don't think I have anything to learn about computing from a website that takes multiple seconds to load plain unstyled text.

[makapuf → mvdtnz]

Depends on how many joules are needed for that. And maybe you won't learn anything technical from it, but that the intent is laudable and the approach worth knowing/spreading ?

[volemo → mvdtnz]

Hmm, it loaded pretty instantly for me and I don't see much bloat in the source. May be you're experiencing the HN effect? :/

On: Mozilla is trying to backtrack on Firefox's controversial data privacy update

[mvdtnz]

It absolutely is not faster. It just isn't. I have no doubt you can find some bollocks benchmark which proves your point but in real world usage it's just not.

Memory is there to be used. Have at it.

[capitainenemo → mvdtnz]

At least on our ram limited work VMs I literally can't launch Chrome without it crashing on launch. Firefox runs with no issues dozens of tabs, although occasionally I do get a tab OOM crash, which restarting Firefox resolves.. And yes, this is even if I completely quit Firefox to ensure Chrome gets as much RAM as possible.

If I've just restarted the machine, sometimes I can launch Chrome.

Chrome print preview never renders about 90% of the time when I am able to launch it. Firefox, no issues.

On a large complex page, when Chrome does manage to launch and print preview does load, it takes tens of seconds to render. Firefox, no issues.

I pretty much just use Chrome when I absolutely have to do a cross-browser test these days.

BTW, a Chrome-tangential annoyance is that the code process in VSCode which is basically an embedded browser sometimes runs wild sucking up gigabytes of RAM on another VM dedicated pretty much just to VSCode (and a couple of other minor tools on a lightweight desktop). The irritating thing there is that due to, apparently, a limitation of the blink embed they are using, you can't restrict the RAM available to VSCode to any number (even 100% of system ram) so you basically have to wait for the oomkiller to kill it, or run a parallel monitor to kill it once it sucks up too much ram.

[j_maffe → capitainenemo]

What you're describing is <0.1% of browser use case though.

[johnea → j_maffe]

This is not a good argument.

Many different users have many different use cases.

What you're basically saying is: It works OK for me, so the fact that it doesn't work for you is insignificant.

[j_maffe → johnea]

No, I'm saying it's working for >99% of the users.

[capitainenemo → j_maffe]

shrug This is a Windows VM with 8GB of RAM - if 8GB is no longer enough for Chrome, that's kinda sad. But does mean it is useless for at least the longer tail of lower end hardware.

[capitainenemo → capitainenemo]

(oh, and I suspect about:memory might fix the demanding sites that OOM, but restarting is easy enough too)

[bigstrat2003 → mvdtnz]

Memory is there to be used by all my applications, not to be hogged by one application when it could get by just fine with less.

[seba_dos1 → mvdtnz]

In my experience, Chromium can feel somewhat faster when you keep a low number of open tabs. It doesn't deal with many tabs as well as Firefox though.

That said, these days I only use Chromium occasionally, so this could have changed over the least few years.

[jay_kyburz → mvdtnz]

Keep in mind that Mozilla is heavily focused on Windows. It's very fast and responsive on Windows, but I fund it unusable on Mac and Linux.

[bossyTeacher → mvdtnz]

I use both and I think they are similar in speed.

On: Making o1, o3, and Sonnet 3.7 hallucinate for everyone

[mvdtnz]

That's not just small, it's utterly miniscule. It's most certainly not large.

[ForTheKidz → mvdtnz]

Nah this is miniscule: https://github.com/coreutils/coreutils/blob/master/src/yes.c

You can fit a hell of a lot of functionality in 3k statements. Really whether it's considered large or small necessarily must rely on the functionality it's intended to provide.

[stavros → ForTheKidz]

You're thinking on "lean" vs "bloated". "Large" and "small" have meanings of their own, and a 3k LOC project wouldn't be accepted as "large" by anyone. "Small", maybe.

[ForTheKidz → stavros]

I certainly never described 3k as large, so I'll assume you replied to the wrong comment. If not, let's just agree to use the terms differently.

[johnisgood → mvdtnz]

Depends. 3k is pretty much enough for a fully-featured XY.

So no context, and differences of the definition of "large".

Perhaps if you come from Java, then yeah.

shrugs

[magicalhippo → johnisgood]

I'd still call 3kLOC quite small in all mainstream languages.

I worked on a Python project which I'd consider as medium sized, ie not small but not large, and it was around 25kLOC.

My $dayjob is a Delphi codebase with roughly 500kLOC, which I'd say is large but not huge.

Though if you wrote it in something like K[1], then yeah ok, I'd agree 3kLOC probably counts as large.

[1]: https://en.wikipedia.org/wiki/K_(programming_language)

[mvdtnz → johnisgood]

I have spent my career working on software that measures its size in MLOC (millions of lines of code). Not because it's Java, but because it's big.

On: Making o1, o3, and Sonnet 3.7 hallucinate for everyone

[mvdtnz]

I'll take boring scaffolding code over libraries that perform undebuggable magic with monkey patches, reflection or dynamic code.