barnacled

Reasonably politely pushing back on these has got my post not only downvoted to oblivion (after upvotes) but then flagged (!!) while of course the guy who was abusive in return, no problems.

I know you have written eloquently about HN in the past, but I think that site is gone. And I remember because I've been on it for over a decade. I joined an early stage start up on it.

Sadly, they give no means for deleting your account. But thought I'd flag to you anyway.

Anyway, I will be abandoning this account, but it's maybe something for other HN old timers to think about.

I had similar experiences suggesting covid had natural origins. I'm not sure the culture that HN possesses is the same as the one it alleges it desires.

Anyway, last post from this account. I am sure this one will equally be nuked... :)

Each of these points are false in my opinion but I genuinely do not understand why you think I'd continue to engage with you after you condescend with that ludicrious 'garbage in garbage out' comment?

The fact this was upvoted despite that makes me feel like returning to commenting very infrequently is the right choice.

Hope you have a nice day.

I am writing a book I intend to release commercially, and a good sized one at that (likely 1,500 pages+). I am absolutely stubbornly determined to release it as both print and DRM-free PDF.

It'll get pirated anyway (if anybody cares enough to do so), so why bother punishing paying customers?

Not out for a long while yet, but https://linuxmemory.org/ :)

That it wasn't completely insane to write a book on the linux memory manager.

Incredible work! I'm writing my own book (obligatory mailing list link [0] and description [1]) and I wondered how you tackled the mental side? It is a bit of a rollercoaster I am finding. That constant fear gnawing in the back of your head 'is this really any good at all?' :) I suppose it is the price of caring.

[0] https://docs.google.com/forms/d/e/1FAIpQLSen0gefOrPWi6ZtEQ25... [1]:https://ljs.io/book.html

This really chimes with me, I'm writing a book about the Linux memory management subsystem [0] and my primary motivation is to learn the subject more deeply and what Paul says here really aligns with my experience.

The effort to explain its machinery demands that I look very deeply - trying to answer questions like 'how does a page of memory get reclaimed?', or 'how does a userspace allocation propagate through the kernel?' - and following the white rabbit all the way down the burrow, forces a far deeper level of understanding than simply figuring something out for a patch.

(Obligatory plug!) For anybody who's interested in the topic, I plan to launch the book in a year or so, to be notified when it's released, sign up at https://docs.google.com/forms/d/e/1FAIpQLSen0gefOrPWi6ZtEQ25...

[0] https://ljs.io/book.html

You do realise there are countless people who have 'read the papers and thought about the conflicting evidence' about climate change right?

The reality is unfortunately that these people ARE nutty conspiracy theorists. They are not 'thinking about the papers' and providing a reasonable push back, they're reiterating long debunked points (did you read the thread I posted?).

Ebright already claimed the papers were scientific fraud until he thought better and deleted the tweet. This is the kind of person you're dealing with.

I'm not going to both-sides something which now has very clear evidence in one direction and not a shred of evidence in the other, I'm sorry I'm just not.

Hacker news of old would be more open to that. Again,this is why I no longer post here.

No, you literally said 'all his claims are facts until proven otherwise'.

I am not being vague, I literally posted a link to a rebuttal thread.

This is why I don't post on hacker news any more.

The 'counter' is incredibly weak sauce and she's been rebutted several times. Again, as I said, if you just ignore rebuttals you can keep acting like you have a point.

The second post is a scurrilous and nasty smear job from a thoroughly unpleasant individual and I'm quite appalled you posted it. Shameful.

Ah I expected to get downvoted into oblivion, I know it's a cliche to say it but HN isn't what it once was.

A key point to take away is that the fact there are 2 lineages means that lab leak is super unlikely. It would require somebody from the lab to come to the same relatively small market to give lineage A, then somebody else who got infected through a totally different evolutionary route at the lab to happen to come to the market just after and both to not spread it anywhere else.

Of course lab leakers are working hard to try to deny this or claim the data is wrong or yada yada. It's an ongoing battle and they have several highly dedicated 'independent scientists' (lol) working on it seemingly 24/7.

It's quite dispiriting to see.

'Claims are facts until proven otherwise' is quite the take, I think Karl Popper would take issue with that.

I linked a thorough thread from a domain expert explaining in detail why the claims are false. These things have been addressed/debunked many, many times but still get repeated because the untruths get amplified and the rebuttals not.

There is now evidence firmly against lab leak and firmly in favour of natural origins, covered in depth by a domain expert here - https://twitter.com/angie_rasmussen/status/15519378265808240...

I feel like this isn't publicised enough. The lab leakers are louder and people definitely seem to want it to be true more than natural origins and they are not being honest in their attempts to dismiss these papers.

This is not evidence. It's hand waving 'how do you explain' stuff that has been addressed many times by experts in the field. No credible evidence whatsoever has been presented. If you ignore experts and just keep repeating the same old nonsense of course you claim to have evidence in the same way 9/11 truthers do.

The problem with this is it is full of utterly debunked nonsense that Ebright knows full well has been addressed.

It's Brandolini's law [0] in action, and no matter how many times virologists rebut these positions the same old tripe gets wheeled out again, and again, and again. If you ignore rebuttals and just repeat yourself + get 20k likes on twitter how many people are going to see a rebuttal like the excellent one from Dr Rasmussen here - https://twitter.com/angie_rasmussen/status/15669764417361715...

The recent science papers which have been closely peer reviewed and none of these guys have been able to rebut (but oh have they tried, since they are VERY inconvenient for lab leak), instead it's the same tired nit-picking and misunderstanding.

For a break down of the science papers read https://twitter.com/angie_rasmussen/status/15519378265808240... which goes into painful levels of detail to explain what the paper discovered and why it's so compelling and why it all but rules out lab leak.

It's sad to me that somebody with a name and position and blue tick abuses that position to spread what amounts to misinformation, but that's where we are in society I guess.

If you want a test, see how these lab leak advocates have reacted to any paper or data that pushes back against their claim. It has been a mad scramble to attack, nit-pick, dismiss and smear (Ebright especially likes confict-of-interest implications). These are not honest people.

[0] https://en.wikipedia.org/wiki/Brandolini%27s_law

Interesting this came out just as I rejigged and updated a visualisation tool I made a decade ago (!) which shows sorting algorithms in a really pretty rainbow colour form. I think it gives a good insight into how these work.

https://ljs.io/rainbow/

This reminds me somewhat of golang - ostensibly open source project but in reality all meaningful decisions are made by those at the company (including shameful throwing away of community work). Perhaps we need a different term to differentiate between open source projects that truly have open governance vs. ones that are essentially bolting on some open source shaped stuff.

As somebody who has started collecting ultima box sets recently (pure nostalgia loved them as a kid) it's been mad to see how ultima 5 has gone from £50 when I bought it a few years ago to maybe even £250+ now. I was hoping I could got all of the ultimas but this makes me realise that 1 - 3 are probably out of reach (let alone akalabeth!)

I am at the start of writing a book and had some major freeze at the start before, much like the author here, realised that all that matters is that I write the damn thing.

I love how LaTeX lays things out, so I'm using LaTeX. I'd like to have a web version possibly but I can think about that later.

And now, I better get back to it...

While I agree in spirit with your point that being true to yourself is very important, as somebody who's been consistently honest (over ~15 years or so career) I'm afraid I think you're far too optimistic on that front.

By all means do what I do which is -- be honest so you can sleep at night knowing you tried to do the best for the company -- but don't always expect good outcomes.

That last point is the crux of why things don't always work out because, who decides whether you're being constructive or a whiny, critical, negative contrarian? Might it be the manager you're criticising? :)

I've had the 'it's the way you say it' thrown in my face before despite having said things a number of different ways each getting the same response. Manipulative, gaslighting managers absolutely thrive on those kind of blurry lines.

Often middle managers are playing an entirely different game than you are. They want to look like they're not only essential but the reason things turn out well. Some decide to adopt a meta of bashing down people who call any of that into question (an underling raising concerns might risk them taking credit rather the manager).

I've also had the bad luck of having a truly evil manager in the past who made sure to cause actual psychological harm to me and other ex-colleagues. I am glad you haven't experienced that but they do exist.

I guess the correct test for 'is it me?' is to assess what others think. If many independent employees are experiencing the same thing then it's probably real. If it's only you, then question whether maybe you're the cause.

You are taking out the human element of a bad manager wanting revenge for you bad-mouthing them. You might say 'well it's not professional to do that!', however a well motivated manager with an axe to grind can find all sorts of ways to cause trouble for you within 'the rules', which is often precisely how bad managers operate on the job. In fact, the article goes into detail on this.

It's actually something of an archetype for a manager to be snide and 'keep within the rules' while absolutely making employee's lives miserable. That's how they survive politically. I should know, I've had it done to me.

As somebody who works in London, you are right that infra is much better here than at least some places in the US (was shocked when I visited Austin) but the commute is brutal, overcrowded and often unreliable. For example the train line I used to rely on would fail 1-2 times a weeks such that I has to find another way into the office.

Unless you're rich you will be living 40 mins - 1hr out, I am more like 1hr20 so I can own a home that isn't tiny or in a high crime area.

The commute costs, if I pay a year in advance, ~£5k/yr, set to go up by 10-15% next yr.

Generally speaking, and especially taking into account the climate crisis, it really doesn't make much sense to do all this even when there is more infra.

It is time for more companies to accept that WFH is largely a win/win.

huge companies do not make changes like this purely on the basis of newspaper columns.

The same newspaper ran similar tactics against YouTube which resulted in colossal changes. These kinds of stories have an enormous impact, magnified I think by company's (in the case of YouTube, advertisers, in the case of OF, payment providers and investors) being terrified of snowballing bad PR in social media.

It's strange but in the modern world yes, 'old media' can have transformative impact on corporations.

I really hate that people call Linux distributions 'X OS' and you get sites like this which at no point clarify where it is actually a distro or an entirely new OS.

It is confusing and somewhat disrespectful towards Linux.

I work in the exact same industry as him and earn considerably more. As always coding jobs are not fungible and finance coding jobs are not fungible either (a .net back office programmer is not going to earn the same as a front office quant dev doing C++).

I should know, I worked my ass off to transition from one to the other via a startup (which I sold shares in for $500k) and tech firm.

I didn't just accept my initial job was the end of my career (my first job I was on £19k, first finance job £24k or so), I made things happen in spite of being set back massively due to serious issues at uni which was not my fault (long story).

So unfortunately my sympathy is zero. If you don't like it, work at changing it, or shut up.

Also the idea that programming is not in demand is truly laughable. I interviewed 200+ people throughout my career. Trust me the issue is not on the demand side.

EDIT: Removed as getting downvoted into oblivion...

Thank you! I wanted to reply last night but got rate limited. I've been questioning my side project recently (hence why no updates for a month) so hearing positive feedback from people does help motivate.

Thanks, I am reading the kernel source as I go and using that to answer questions as I figure things out.

Actually no, the malloc doesn't allocate any memory it just updates the process's VMA to say that the allocated virtual range is valid. The pages are then faulted in on write. This is where things like OOM killer become very confusing for people.

In linux (in sane configurations) allocations are just preorders.

EDIT: I can't reply below due to rate limiting:

I'd argue that overcommit just makes the difference between allocation and backing very stark.

Your memory IS in fact allocated in the process VMA, it's just the anonymous pages cannot necessarily be backed.

This differs, obviously, in other OSes as pointed out. Also differs if you turn overcommit off but since so much in linux assumes it your system will soon break if you try it.

They have to be same as the maximum addressable bit, i.e. in the case of 48 bit virtual address size the 48th bit.

This is actually kind of a cute way of dividing kernel and userland space as you just set the upper bit to 1 for kernel addresses and 0 for userland.

EDIT: Specifically talking about x86-64 here.

https://github.com/lorenzo-stoakes/linux-mm-notes/blob/maste...