Hacker News: tom_

New comment by tom_ in "A sleep-like consolidation mechanism for LLMs"

tom_ — Tue, 26 May 2026 16:31:22 +0000

New comment by tom_ in "You Only Use 10% of Printf() – Here Are Things They Didn't Teach You [video]"

tom_ — Tue, 26 May 2026 10:24:47 +0000

Tough crowd, but I for one hate this stuff too and thank you for the heads up. "Engraved invitation to chaos", fucking hell. I think I just lost 3 IQ points. Who's the target audience here?

Everything I needed to know about printf, I learned from the reference manual. Anybody could do the same. Here's a reasonable one: https://en.cppreference.com/c/io/fprintf - look, it's like 5 pages or whatever, and the last 2 are examples and xrefs. You sit there and read it 2 or 3 times and you'd be done faster than Task Manager Man can read his script.

New comment by tom_ in "CVE-2026-28952: Apple macOS 26.5 Kernel Vuln found by Claude"

tom_ — Tue, 26 May 2026 02:39:17 +0000

> The affected releases include iOS 18.7.9 and iPadOS 18.7.9, macOS Sequoia 15.7.7, macOS Sonoma 14.8.7, and macOS Tahoe 26.5.

Where does this quote come from? I can't see it in https://support.apple.com/en-us/127115, the article link at time of writing. It mentions CVE-2026-28952, but we're forced to guess why. I'd take the reference to mean that this issue is fixed, but I'm just some internet rando, so what the hell do I know?

If I do a google search for "CVE-2026-28952", it points me to various pages. Here's one, for example: https://www.cve.org/CVERecord?id=CVE-2026-28952 - which is a bit more explicit, though of course this is not from the horse's mouth:

> This issue is fixed in iOS 18.7.9 and iPadOS 18.7.9, macOS Sequoia 15.7.7, macOS Sonoma 14.8.7, macOS Tahoe 26.5

New comment by tom_ in "sp.h: Fixing C by giving it a high quality, ultra portable standard library"

tom_ — Sat, 23 May 2026 23:17:47 +0000

Because it's const void *pa, you don't need the cast. A void * pointer will convert to any other kind of pointer. Now you only need to mention the type once. (I forget the const-related rules, but since the consts match in this case I don't think it'll be relevant.)

New comment by tom_ in "FatGid: FreeBSD 14.x kernel local privilege escalation"

tom_ — Thu, 21 May 2026 18:08:23 +0000

Is that even the fix though? The problem sizeof*groups expression has already been removed by that point. This fixes something but it's not obviously related to the vulnerability description.

git log -S suggests 4cd93df95e697942adf0ff038fc8f357cbb07cf9, which looks more likely: https://cgit.freebsd.org/src/commit/?id=4cd93df95e697942adf0... - though not to say you don't want the later commit too. I'm sure you do.

New comment by tom_ in "Google Declaring War on the Web"

tom_ — Wed, 20 May 2026 22:41:32 +0000

What if there is no long game? Just people at Google optimising for their current KPIs.

New comment by tom_ in "GitHub is investigating unauthorized access to their internal repositories"

tom_ — Wed, 20 May 2026 02:51:56 +0000

Ahh, that's a good point, and I actually hadn't thought of that angle! I was thinking of it purely from the point of view of the attackers using LLMs to generate interesting new exploits, with a side helping of letting myself get mildly annoyed, possibly incorrectly, by the writing style.

But yes, it's also possible the defenders have been kind of forced into having the slop machine shit out a huge pile of shit-ass changes, one way or another, that end up making the attackers' job even easier. (Even assuming no mechanisation at their end! Which is of course in nearly-June of 2026, probably unrealistic. And LLMs do appear to be really quite good at that side of the equation...)

New comment by tom_ in "GitHub is investigating unauthorized access to their internal repositories"

tom_ — Wed, 20 May 2026 02:43:24 +0000

Apologies to all - I am British. The phrase "non-zero" does cover every case other than zero, but the intent is that it covers some cases more than others. What I'm trying to say is: yes. My intent was just to push back on this specific (and slightly bizarre to me) instance of kind-of-vagueposting, to my eyes written to imply that it might be some sort of unnoticed conspiracy, detectable only by the most enlightened of observers, attuned to the subtle signals that most people miss: that people are using LLMs to find security exploits.

New comment by tom_ in "GitHub is investigating unauthorized access to their internal repositories"

tom_ — Wed, 20 May 2026 01:04:38 +0000

It's more likely that it isn't coincidental at all: software development-oriented LLMs became a lot better towards the end of 2025, and so there's a non-zero chance that people are using them to find new security exploits.

(People are not sleeping on this and it is not something people have failed to notice. I don't use LLMs at all and even I have noticed it - largely because there is approximately nobody that isn't talking about it.)

New comment by tom_ in "Dumb ways for an open source project to die"

tom_ — Tue, 19 May 2026 23:43:28 +0000

Whatever age range you have to be in to remember the most viral videos from 2012, some measurable percentage of HN's readership probably aren't in it.

New comment by tom_ in "The American Rebellion Against AI Is Gaining Steam"

tom_ — Tue, 19 May 2026 03:51:06 +0000

Things take time to play out!

New comment by tom_ in "Mode collapse has a name, and he's selling cancer treatment advice on Amazon"

tom_ — Sun, 17 May 2026 02:32:42 +0000

I don't, as my point applies regardless of the article's contents: that any downvotes of the submitter's comments are perhaps due to people not liking LLM-generated content, such as this article appears to be. (Being a person made of meat, rather than an LLM, I had to get my own opinion of LLM prose squeezed in there too - but it's a side issue, and you can ignore it if you like.)

Regarding the style, no concrete feedback I'm afraid, other than the repetition of "upstream" sounding like the LLM tic of the day. Other than that, I just got that LLM vibe from the writing, and, as above, perhaps I wasn't the only one. Some discussion here of some LLM tropes: https://news.ycombinator.com/item?id=47291513 - which may apply.

New comment by tom_ in "Stochastic Parrots: Frequently Unasked Questions"

tom_ — Sun, 17 May 2026 02:08:30 +0000

Perhaps actual thinking is not automatically necessary for that either! - and the LLM is proof.

New comment by tom_ in "Mode collapse has a name, and he's selling cancer treatment advice on Amazon"

tom_ — Sat, 16 May 2026 18:08:29 +0000

I assume it's because the article appears to be, if not AI slop, then certainly something that reads very much like it has gone all the way through an LLM's digestive tract and come out the other end. Perhaps the odd piece of sweetcorn or pepper seed can be found, but I for one would prefer to dine elsewhere.

I vouched for the sibling comment, which seemed innocuous and contained (I felt) the most interesting part.

New comment by tom_ in "The Emacsification of Software"

tom_ — Thu, 14 May 2026 03:23:36 +0000

For whatever it's worth, I've always only ever used the native Windows build of Emacs, and I've never had any awful problems sharing my config between Windows, Linux and macOS. I'm sure I had to expend at least a bit of effort to make this work initially, but it wasn't enough for the process to stick in my mind, and the ongoing effort doesn't feel like it's added up to much.

(I admit it's added up to more than zero though! Keeping (require 'cmake-mode) working reliably on Windows and macOS has proven a minor annoyance, and fonts seem to require a degree of system-specific attention.)

New comment by tom_ in "The Emacsification of Software"

tom_ — Thu, 14 May 2026 01:33:56 +0000

I've always found it easier to collaborate using markdown. It's much more popular than org-mode! And you can get org to export text as markdown, so interop isn't necessarily even all that difficult.

And as per the general theme of the post anyway: whatever. Because you can just work around it, whatever it is, by cobbling together some code, one way or another.

New comment by tom_ in "Running local models on an M4 with 24GB memory"

tom_ — Wed, 13 May 2026 01:59:54 +0000

I didn't click through the full UI to get the lead time or anything; I just looked at the options presented on the UK site. Maybe there's a stock of laptop types here that have all sold out elsewhere? Or maybe they were just teasing me, and I'd have been hit with a 6+ month delivery time if I'd gone all the way.

New comment by tom_ in "Screenshots of Old Desktop OSes"

tom_ — Tue, 12 May 2026 20:48:52 +0000

Over 75% (well... just...) of the screenshots are 1024x768 or greater - though admittedly typically not by much. Over half (well, again: just!) are 1152x864 or larger.

New comment by tom_ in "Screenshots of Old Desktop OSes"

tom_ — Tue, 12 May 2026 20:29:56 +0000

Use portrait displays! You'll never complain about the height again. Use 1440p (or scaled equivalent) and the width won't be an issue either. Many displays make this easy (all the Dells I've had can rotate), and adequate rotating stands aren't necessarily expensive.

There's even maybe some Actual Scientific Evidence to justify the switch: https://news.ycombinator.com/item?id=47638928

New comment by tom_ in "Screenshots of Old Desktop OSes"

tom_ — Tue, 12 May 2026 13:54:39 +0000

If you haven't tried it already, I've found it useful to get Windows to use the accent colour in the title bar and window borders: https://support.microsoft.com/en-us/windows/personalize-your...