This is the type of text that tells more about the person writing than what it's written about. It feels to me his defending so hard his views because he doesn't even believe in it, as if he needs validation. Like, of course there are people who thinks vim is better because of macros, but I'm pretty sure many (if not most) of vim users don't even use that. Vim's main selling point is that you can do _anything_ without a mouse. And it's very customizable. Now, you can be more productive with any tool you want than the average vim user. Heck, I'm pretty sure there's at least one person in the world that's more productive using Notepad++ than the average vim user.
His whole rant on Linux and "highly configurable software" vs good defaults is just plainly nonsense to me.
> “Highly configurable” is often just an excuse for shipping no opinion at all and calling the resulting work your problem."
I couldn't disagree more. You can argue that Linux or other open-source software don't have "good defaults" is mostly because there's way less investment in 1) user experience; 2) quality assurance; mostly because there's no product logic involved in it.
Especially if you think that Linux, for example, is the mostly used in servers, and it works usually fantastically well in most server VMs. Maybe it needs a lot of fiddling to make it work on your old dell laptop because it's not where the work is put at. Windows will run well on it because Microsoft puts people actively working on making it run on most commercial user-end hardware. Apple machines will work perfectly on their hardware because that's what they're made for.
Arch Linux is not just better than any other Linux distro. It's better at one thing, just like Ubuntu is better at something else.
> I don’t want my tools to be “fun”. I want my tools to be invisible.
> A good tool is and ought to be invisible—striving to make such tools is the goal of a toolmaker.
Bit of a wrong take, on my opinion. Every tool has its quirkiness, and you should embrace it. No tool is invisible. It feels less "visible" as you build more "muscle memory", but it's still there. We have to embrace the tools as part of the craft, not pretend they don't exist.
The thing is it's easy to define free, unused memory. But a lot of the used memory is your system caching stuff that would be free if you needed more than what's actually free. So you can see you have 1g of free memory out of your 4g, but then you allocate 3g and it will do without a sweat and you'd be confused. So you have to go and dig for what those caches are and report that they're effectively free too.
> The “cached” memory includes tmpfs and ramfs for seemingly no reason.
If you're curious why that is by the way, it's because that's actually how these are implemented (tmpfs/ramfs is just a mount to a filesystem where the files never get marked clean[1])
AFAIK the only way for you to figure out how much of your disks is actually cached involves enumerating all tmpfs and ramfs mounts, summing their sizes, and subtracting the sum from the cache size reported by the kernel.
Well, what's the alternative? Invent a new type of memory reservations specifically to account for tmpfs/ramfs mounts? That'd violate your own stated desired goal of
> The OS should just expose a counter for available memory instead of having applications understand every type of memory reservation.
I think a memory pressure indicator is useful. For example, if you’re writing a garbage collector, you can choose to hold off from returning pages back to the kernel when the system is under low pressure, and do this more often under high pressure.
That's very, very marginally useful. If your application recently ran GC when the system was under low pressure and freed a bunch of pages (and still holds to them), and then some other application suddenly acquires a lot of memory, your application won't free those pages until next GC happens. Which most likely won't until you've used up all those freed pages, and at that point a lot of swapping will already have had happen.
Explicitly set per-application upper bounds à la GOMEMLIMIT and Java's -Xmx, together with cgroup's enforcement, seem to be much more useful in practice.
Or there was a proposal for some sort of SIGLOWEM signal that would be sent to all processes in the system (with default disposition of "do nothing") that'd allow applications to release some of its non-vital memory holdings: also a much more timelier notification.
That metric would give you a number of bytes which can be used for pages not backed by files, but it won't give you actual memory usage statistics:
It won't count executable pages and memory-mapped file use as "used" memory, so your system might display gigabytes "free" when it's starving, executables getting paused when code pages are paged-in from disk.
It's just less useful than what's displayed now. "Everyone is doing it wrong" is usually a signal that you're missing something.
The kernel keeps track of active/inactive pages by scanning page tables during the reclaim process. It only scans until it finds enough inactive pages to reclaim. Scanning all pages all the time is very expensive from a performance point of view.
Then most people on this comment section would have to find something else to preach and declare absolute knowledge about. Ah wait, that’s just every comment section around here.
Chill down. It's just someone who has a lot of experience in the field making an analysis of the current landscape of the career, using their own as an example.
> Telling people to stay away from graphics programming is not how to entice tomorrow’s John Carmack.
John Carmack was one of the _first_ graphics programmer to ever exist. The next John Carmack can't be in the same field. The same way we can't expect the next Beatles to be playing rock music. :)
> "These haven't been tested, validated, debugged, or verified!"
I really don't understand what the point of it is, then. It's not anymore "I put a lot of effort into something because I have the knowledge, experience and time to do so, hope you enjoy", it's like "I paid AI tokens to to that. Everyone could've done, but I paid with my own pocket. And it's untested.". That's it?
> "Yes, I used Google Anti-Gravity to convert the programs from GW-BASIC to 'C', but what a better learning tool than to debug a program?"
Debugging a program is an excellent learning tool. It's just not better than another learning tool: coding the program yourself. :)
Well, let's be careful about that. The owner of the repository probably didn't ask for it to be submitted to HN and probably wasn't looking for public attention; perhaps they were merely playing around. The onus is strictly on the submitter here.
that's just objectively not true. code is not just the thing that "builds" the final program, it's also the blueprint of that program.
it's both the most detailed _description_ of what the program does and also the instructions on how to build that program. and it's deterministic, which means that the same compiler will *always* generate the same program based on that code (search for bootstrapping, and you'll understand better why that is). you can think that a very detailed prompt is like a "code" in the sense that you'll always be able to build the same thing, but that's simply not the case because LLMs are not deterministic.
the other thing I like to comment is that you mention _"product"_ a few times, but you never mentioned a _"program"_, which is what it becomes after it's built (or executed, for that matter). I think those are in different levels of abstractions. if you're looking for a way to come up with a product faster than your competitors, good for you. code is waaaay more than that.
Hi Vivian. Thanks for coming up and explaining things. I apologize if I sounded too harsh, because I really want to be fair here. But I do think we have to push for a more conscious content consuming in the era of highly capable LLMs. There is a threshold where I can't really tell anymore if something is pure AI Slop or AI assisted, as pretty much everything is and will be from this point on. And I think that having that critical thinking is important before spending 60~90 bucks on a book.
Whoever owned your domain 10 years prior is not important to this matter, I was just pointing it out that it was very likely not you.
> One aside, in genuine astonishment rather than complaint: it is striking how multifaceted a picture of a stranger can be drawn from a WHOIS record, an Amazon sample and a GitHub page. I have read it with more curiosity than dismay, and learned things about myself in the process.
Well, in my opinion that's your fault entirely. All your descriptions are rather vague on all your networks, which makes this whole thing more suspicious.
> Slop in the AI-slop sense it is not. Heavily edited and translated, yes. Authored, structured, fact-checked and re-read line by line by a man in Germany having rather more fun with this than he had expected, also yes.
Thanks for being transparent with this. I didn't find any mention of that before in your site, so that's good to know.
> Now it is up to you to dig further and keep me on my toes. Ask away. One small request: stay fair.
Once again, I want to be fair and this is nothing personal. I'm not digging any further either. I did change my original comment from "This is 100% AI Slop" to "This seems AI Slop" minutes after I posted it, because I want to acknowledge that I might be wrong.
However, I want you to take my comment more like the one from a possible costumer (after all, I have huge interest in the subject and that was what drove me to do it) who wanted to do some research on an author before spending $90 on their book. This is a platform for discussions, I also raised that concern so other could chime in.
- All his other domains are recently registered too, the oldest one `byvoss.tech` is from May 2025.
- GitHub has way too little projects and contributions for someone with such experience. The first project dates on the same day as the domain registration: https://github.com/VivianVossNet?tab=repositories
- As other people mentioned, too many blog posts lately (almost one a day), and a very weird writing style. Which doesn't seem at all that's because it's English as a second language. It's just plainly vague and disconnected sentences.
upon further review, this is a complete LLM-driven attempt at trying to be AI Banksy
aside from the complete lack of background like he's Jobriath except without talent, it isn't hard to find festering boils like the UN link (https://www.un.org/en/observances/book-and-copyright-day) being a sham coming from Russia
This person appeared out of nowhere 6 months ago and started posting on linkedin. With 6k followers and so much experience I would expect a couple of recommendations.
Also gitlab and Github account appeared around the same time...
Maybe it's just my paranoia in the AI slop age.
Things do seem a little off. If you check both the website and the LinkedIn you'll notice a strange motif of AI-generated women with cat ear headphones.
Have you considered that "very weird writing style" might be due to the fact that a German person is writing in English?
One post a day over the last short period of time does not seem unusual for someone who is suddenly interested in a least a part-time career as an author.
Your random online persona assassination aside, I am curious about the book. The Devil Book kept me occupied for years when I was in university; I'd like to see what I've missed since then.
I want to believe that Vivian made a bootcamp and turned herself from an elite business trained elite scort [0] into an elite greybeard, empowering her former "nerd" and tomboy spirit [0]
I'm not sure I'd put it at full slop, but the book is clearly AI assisted. You can view samples on retailer websites, and it features 'interjections' from an AI avatar. Further, in the opening pages, the author feels it pertinent to point out that every fact in the book is verified. Not my cup of tea.
Counterpoint: most AI slop authors have multiple books/posts on random subjects. This is their only book on Amazon and their posts sound like draft chapters. Could this just be new author with an unconventional writing style and a disciplined daily schedule?
"Github has way too little projects"... I had PRs against stinkpad wireless drivers in 1999/2000 time frame - FreeBSD specific. Do I expect those to be around? No.
Do I think 20+ years of working for commercial companies means I have a splendid GH resume - absolutely not.
Using your logic for HN, 140 karma, ignore.
VV has been in the community awhile (I'm not active, but watch). AI slop isn't a thing, trolling for top of HN is.
His whole rant on Linux and "highly configurable software" vs good defaults is just plainly nonsense to me.
> “Highly configurable” is often just an excuse for shipping no opinion at all and calling the resulting work your problem."
I couldn't disagree more. You can argue that Linux or other open-source software don't have "good defaults" is mostly because there's way less investment in 1) user experience; 2) quality assurance; mostly because there's no product logic involved in it.
Especially if you think that Linux, for example, is the mostly used in servers, and it works usually fantastically well in most server VMs. Maybe it needs a lot of fiddling to make it work on your old dell laptop because it's not where the work is put at. Windows will run well on it because Microsoft puts people actively working on making it run on most commercial user-end hardware. Apple machines will work perfectly on their hardware because that's what they're made for.
Arch Linux is not just better than any other Linux distro. It's better at one thing, just like Ubuntu is better at something else.
> I don’t want my tools to be “fun”. I want my tools to be invisible. > A good tool is and ought to be invisible—striving to make such tools is the goal of a toolmaker.
Bit of a wrong take, on my opinion. Every tool has its quirkiness, and you should embrace it. No tool is invisible. It feels less "visible" as you build more "muscle memory", but it's still there. We have to embrace the tools as part of the craft, not pretend they don't exist.