caspper69

I'm sorry sir/mam, but that is simply not true.

All you have to do is read comments from members of the Rust community online, in every public forum where Rust is discussed in any way.

Understand, I am not trying to villainize an entire community of software developers; but for you to say something that's blatantly false is to just stick your head in the sand.

You should try and read the words people write. Opinions are not formed in a vacuum.

Edit: to be clear- I have no problems with Rust the language beyond some ergonomic concerns. I am not a Rust hater, nor am I a zealot. I do advocate for C# a lot for application code though. But I do not deride others' language preferences. You should not dismiss my observations because I used hyperbole. Obviously not every Rust dev thinks you can't write a secure C/C++ function; don't pick out the one hyperbolic statement to discredit my entire post. Bad form.

Because many of them are still relatively young. There is nothing wrong with youth, but it can contribute to over-zealousness.

Also, for many of them, Rust is the first systems language they've ever touched. And that fact alone excites them. Because now they can "dream big" too.

But they have bought into the whole C/C++ are by default insecure and therefore garbage. In their mind, no mortal could ever write so much as a single safe function in those languages. So their points of view are always going to be based on that premise.

What they fail to recognize is that an operating system kernel, by virtue of the tasks it has to perform- things like mapping and unmapping memory, reading/writing hardware registers, interacting with peripherals, initiating dma transfers, context switching, etc.- have effects on the underlying hardware and the runtime environment; effects that neither the type system nor the temporal memory safety of Rust can model, because it happens at a level lower than the language itself. Rust's safety guarantees are helpful, but they are not infallible at that level. The kernel literally changes the machine out from under you.

They further fail to appreciate the significant impedance mismatch between C and Rust. When one language has concepts that are in fact constraints that another language simply does not have, there is going to be friction around the edges. Friction means more work. For everyone. From planning to coding to testing to rollout.

So you have well-intentioned, excited, but still self-righteous developers operating from what they perceive to be a position of superiority, who silently look down upon the C developers, and behave in a manner that (to outsiders at least) demonstrates that they really do believe they're better, even if they don't come right out and say it.

Just read the comments in any thread involving Rust. It is inconceivable to them that anybody would be so stupid or naive as to question the utility of the Rust language. To them, the language is unassailable.

The petty drama and social media brigading on top of it, along with the propensity to quit when the going gets tough, it's pretty easy to see why some people feel the way they do about the whole thing.

A programming language is not a religion. It is not a way of life. It is a tool. It's not like it's a text editor or something.

Just what the Federal Bureaucracy needs: $400M worth of armored Teslas.

I can't wait to hear the spin on this one. Noses will be growing without bound.

Can you imagine the uproar from Conservatives if a Democrat had bought $400M of EVs with government funds? It would have been a meltdown of epic proportions.

We have gone through the looking glass.

Edit: I see that Tesla was the only one that responded to the RFI that was indeed issued last year. So probably nothing underhanded going on. BUT- I still stand by my assertion that the right would have melted down and spun this like a top if a Dem had done it.

That's not what he said.

From the horse's mouth (lkml; Hellwig's headers chopped for brevity):

On Thu, Jan 16, 2025 at 02:17:24PM +0100, Danilo Krummrich wrote: > Since there hasn't been a reply so far, I assume that we're good with > maintaining the DMA Rust abstractions separately.

No, I'm not. This was an explicit:

Nacked-by: Christoph Hellwig <hch@lst.de>

And I also do not want another maintainer. If you want to make Linux impossible to maintain due to a cross-language codebase do that in your driver so that you have to do it instead of spreading this cancer to core subsystems. (where this cancer explicitly is a cross-language codebase and not rust itself, just to escape the flameware brigade).

---

Hellwig was abrasive and unreasonable. But there is no need to perpetuate, repeat, and repost absolutely one-sided, self-serving misrepresentations of the words he used.

You don't need to paraphrase. You don't need to guess. You don't need to distill or simplify.

He wrote English so we could read it; stop paraphrasing. It's unhelpful at best and nefarious at worst.

Edit: I think it's very telling that there is a crowd here that would literally downvote the actual quote. Actually it's more sad than anything.

Hey, I have a healthy respect for Ballmer (1600 SATs, advanced math degree from Harvard).

I don’t hate MS at all. They made me a good living when I was getting started. They actually spread the wealth (for small and medium sized shops at least).

My post was meant to be an indictment of Google.

This whole html/css layout shit with js sprinkled on top is nothing but a giant moat. Complexity for complexity’s sake.

You could render resolution independent vector art & text with pixel perfect layout orders of magnitude more efficiently. It’d be printable too.

It just laughable to be honest.

Well, he’s an associate and we’ve always been on good terms, diverging politics notwithstanding (I lean more dead center with a healthy distaste for authority and as I said, he’s MAGA), and he knows exactly why I don’t like DJT (I see right through him, because I’ve watched people exactly like him my whole life). Having grown up adjacent to, but not part of, the elite, his type is not uncommon. The ones that had family money, but never actually put any effort in to learn anything or do anything on their own. Or worse, pretending that they were doing big things by talking loudly and showing off, but actually failing at everything they did try. It’s not uncommon. Plus having worked in gaming, I would be downright ashamed if I went bankrupt not just once, but twice running a casino. That’s like an easy-mode business. I mean, come on.

Anyway, I know deep down he got it. But it’s always that moment of realization that I find satisfying.

I did it to eff with his head. It’s that moment when the cognitive dissonance pulls apart and the lucid thought makes a connection, but before the dissonance sets back in. It’s like watching someone melt for a split second.

Yes, I’m an ass.

To think, they have advocated over the years to keep adding to the specs to the point that they’ve become so complex that even Microsoft, the king of this rube goldberg type shit, threw in the towel and said eff it, we’ll just use Chromium.

We should write an app browser with sensible layout and a qemu virtual machine backend. Pay the 5% hit to run actually performant native code that’s 100% sandboxed.

Couldn’t be any worse, that’s for sure.

Had a talk with a pretty dyed in the wool MAGA supporter about the birthright citizenship situation.

I was like you can’t just knock down a constitutional amendment with an executive order, even if you do agree with it.

He was incredulous. The president has that authority and the courts can’t stop him.

I was like ok, by that logic, a Dem president could sign away free speech and the second amendment the same way.

Still don’t think he could reconcile my point.

When the rationale for cutting spending is that the debt is out of hand and that our deficits are unsustainable (read many HN comments over the last few weeks), it sort of flies in the face of all reason to turn around and increase the deficit by $2.5T in year 0.

In fact, I would go so far to say all the talk was an outright liar liar bold faced lie of epic proportions.

It’s like catching the guy from an AA meeting at the bar an hour later with a bottle of Jack and him saying he’s only drinking it to keep it out of the hands of children.

Either we need to get deficit spending under control, or we don’t. There is no middle ground.

These guys have always been hypocrites. Scream bloody murder when out of power, spend like drunken sailors while in power. Never fails. 30+ years, always the same.

So just casually glossing over the additional $2.5T in deficit spending?

I almost had a stroke when I saw this plan.

Are they kidding? All this noise to spend EVEN MORE?

You're absolutely right, they can.

And they have.

But the reality is more like this: https://xkcd.com/1172/

Your complaint is more related to how the OS you're running on handles dynamic libraries, rather than anything inherent to dynamic libraries themselves. It is possible to version libraries and serve up the correct version(s) to different applications simultaneously.

Now, consider the case of a Linux distro with a few thousand binaries. Should a defect be found in a common library, the burden of updating, say, 10,000 servers might rapidly become a headache. In this case, a shared library probably makes sense.

Also, if you want to create extensible software (like a plugin system), dynamic libraries are one of the most expedient and well supported ways to do so. This is not necessarily the traditional use case for shared libs, but it's not uncommon either. And this use case normally doesn't have those versioning problems either (as it leans more to the dynamic side than the shared side).

All that being said- static linking also has its advantages; as you stated, you can be guaranteed that your users are executing the library code you intended and that nothing breaks. It's also faster, because you're not waiting for ld.so to resolve your imports at startup & first use (nit: yes, I know you also can force all resolution to finalize at startup at the expense of slightly longer startup time). Further, you can be sure that someone hasn't hooked your library calls to do something nefarious. The downside, of course, is that you have to re-build and re-deploy your entire application on every library's bugfix/security updates.

As with anything related to computing, everything has tradeoffs. Some tradeoffs may be dealbreakers for your use case, but they might not be for someone else's.

Right, I understand that. You can do the same on Windows too.

My point was that most languages' runtimes don't do that, not that they can't do that.

The article states that /bin/ls is statically linked. That means that the parts it uses from libc have long-since been taken and merged directly into the ls binary. So this randomization has no effect.

Even if ls were dynamically linked, the ls binary itself wouldn't change at every boot; instead its import/jump table, generated dynamically by ld.so as part of the OS's executable loading process, would simply point to different (and random) offsets within libc. It would have a different in-memory representation after each boot, but its on-disk representation would remain the same.

Not related.

The blog post is a mini-rant complaining about how other people are complaining about having to link "all of libc" on BSD when their standard libraries are not based on libc (libc is the interface to the kernel on BSD). He's talking about static linking.

Honestly, I'm not terribly sure I get it, as almost all "other languages" link to libc in some way on most OSes, and don't call native kernel interfaces directly at all. But maybe it's more common than I think. Famously, Go went this route and it caused them no end of consternation with their FFI, and they've gone back in the other direction of linking with libc directly.

In your case, reordering has to do with ASLR (address randomization) in dynamic libraries. OpenBSD takes ASLR a step further and so not only randomizes the libraries themselves, but the archives inside the library and the functions themselves- they call it KARL. That's what it's doing when you see that message.

Yeah, I didn't notice the site switch to English link; If I had seen it, I might have also seen the source, lol. Sorry about that.

Terry Davis, the author of TempleOS, was fairly well known online. He had a lot of struggles psychologically, and met an unfortunate end a few years ago. It's an interesting case, but I honestly wouldn't take too much from the project. He was a gifted developer, but he did things for reasons that were decidedly non-technical.

For building up your knowledge (you can't only read the reference manuals, after all), there's a ton of resources online, but a particularly good one is the OSDev wiki: https://wiki.osdev.org/Expanded_Main_Page. I can't speak for it as it exists today because I've been out of the game for too long, but it was maintained by helpful folks. I didn't check to see if the forums are still up, but there used to be a wealth of information in them back in the day too.

As a note, relying on BIOS is ok, but the whole write a bootsector in 512 bytes to switch to 386 protected mode, to then switch to long mode (64-bit) is kind of outdated (the modern computer boot process is handled by the UEFI firmware, and by the time your OS binary is loaded, the CPU is already in long mode with paging enabled and setup with a flat linear mapping of physical memory).

Not to say that the method/approach is worthless (it's all learning), it's just that you spend an awful lot of time doing things in the whole mode-switching dance that are basically tossed in the bin almost immediately, and most of it has nothing at all to do with OS development. UEFI at least allows you to start on the real meat & potatoes without spinning your wheels on code that isn't going to stick around very long.

Prime example: we leased 3 Nissans in a row (dipping our toes in with an Altima and then 2 Rogues in a row), so 9 years worth, but prior to the last turn-in, they released a new Rogue that was smaller on the inside (but I believe may have been slightly larger on the outside), sapped the power out of the base engine (our Rogues had some pep in their step), and as the final nail in the coffin, raised prices by 15% or so and lowered the lease residuals- the net effect was a worse automobile with a ~30% higher monthly payment.

Going from under $350/mo to over $500/mo on a 36-month low mileage lease made what had been an easy decision one way (just get another Rogue) into an easy decision the other way (get a different vehicle from a different manufacturer).

When you venture into the pricing tiers of higher-quality automobiles, you need to be equipped to play in that market. Nissan wasn't, at least in our situation, and it cost them a loyal customer.

The page isn't in English, but Google helpfully translated it for me.

Without source, and only those commands available, it's kind of difficult to see what you've implemented and how far you progressed.

Did you do anything with the MMU yet? Any hardware detection? Do you ever jump to a user space context?

Do you have any specific design goals you want to explore, or was this just a learning exercise?

Like Groundhog Day, I have been doing this over and over and over for the last 25 years.

So many toy kernels, so many ideas and tests and watching A's and B's and C's and D's flash on the screen. Memory dumps. Writing executable loaders. Poring over the Intel/AMD manuals (and later AARCH64). ACPI (ugh). Trying to implement the techniques from the latest papers (usually in an effort to maximize the performance of a microkernel-ish design).

It's all great fun of course, so I'm not lamenting that, but when you start to get to the userspace infrastructure, unless you're YOLO'ing it with libc and POSIX compatibility, everything is so opinionated, and there's SO MUCH TO BUILD, I just kind of peter out.

It's been about 18 months or so since I last had the itch, but it's inevitable that I'll do it all over again.

Best of luck!

Well, I'm certainly glad that you have figured out the answer to the universal question.

I have spent many years pondering it myself, as I'm sure most of us have. I've even spent some time pondering it while not under the influence of mind altering substances (again, not unique, just wanted to throw that out there).

"The question" is, by design or by accident, unanswerable. At least not with any technology that we have yet contemplated.

Honestly, I could probably write for hours about the many possibilities I have pondered, and some of them are quite interesting (and some are just stupid), but I don't want to ramble and HN is not the place for it.

However, there is a good one that gives me some comfort from time to time. So I will at least share that particular thought:

Assume there was an entity with the characteristics one would normally ascribe to a god: the power of creation, omnipotence, all-knowing intelligence, and being eternal yet simultaneously timeless.

At the moment such a being were to come into existence, certainly in an instant, it would experience the "ultimate realization": it would be forever alone.

In my feeble human mind, that is indeed a twisted fate for such a powerful being. Something straight out of the Twilight Zone.

So then I pondered what I would do in that scenario. And the answer seemed fairly obvious: Everything, with a capital "E".

If it could be done, created, built, lived, experienced, dreamed, loved, hated, you name it.

Where would I do it? Everywhere.

For how long? As long as it took. Even if it required doing it all just a bit different 1000^1000^1000^10000000000 times over.

Why? Because then I would have memories. I would have friends.

My great grandmother, who was 83 or 84 at the time, had had a brain tumor (which was removed once it got to between golf ball and baseball size) and was pretty far along into Alzheimer's as she got got close to her death.

For the last several months, she wasn't able to feed or clothe herself, and she was basically immobile.

The day she died, she got up, went to her room, made her bed (unthinkable in the state she was in), put on her "Sunday's best" and laid down peacefully in the middle of her bed and passed.

Her daughter (my grandmother), was floored. She just said she must have just known it was her time, and that she had a few minutes of lucidity before dying.

Whether there's any truth to that I'm not qualified to answer, but if you had seen the state she was in for quite some time prior to that day, her actions would certainly have been surprising to say the least.

Don’t want to go down the Musk rabbit hole, but wasn’t this a tactic started by Tesla?

This idea that every product must produce a recurring revenue stream has permeated way too deep. Buying a luxury automobile should entail luxury features.

I almost accidentally bought a scanner off Amazon that had a monthly fee in the fine print. As in the damn thing wouldn’t work if the subscription stopped. Wtf?

Frankly it’s tiresome and pathetic. It’s bad enough that pennies are pinched at every step of the manufacturing process to the point hardly anything lasts even 5 years anymore.

We get screwed over and over again so some asshole can hit their quarterly targets, or so the company’s stock can tick up a quarter point.

As if it wasn’t bad enough they were so out of touch they thought $90k for their Wagoneers was a good base price. A vehicle literally no one was asking for.

With median US income around $43k/year, these geniuses thought there’d be a huge market for these monstrosities (and I am in no way anti-SUV).

This was after axing their Dodge sports cars that were actually selling, and dicking around with the Ram pickup. The bread & butter of their lineup.

But ARU! Listen you c-suite morons, sell the products your customers actually want to pay for. Stellantis’ stupidity is going to bankrupt a lot of dealers if they can’t execute these re-rollouts successfully and finally clear this glut.

And now I see shit like this. Pure, unadulterated enshitification.

How do these ideas not get shot down 2 seconds after they leave someone’s mouth? I guarantee you, not one person who heard such an idea thought it was a good one. It is impossible to witness such stupidity happening and think this will accomplish any business objective or be a good direction for the brand.

Brawndo is coming- it's got electrolytes and IT'S WHAT PLANTS CRAVE!

Life imitates art indeed.

I am sympathetic to wanting to automate complex workflows. Hell, I'm sympathetic to wanting to automate simple workflows. In fact, I bitch about the stupidity of the things I do at least once a week (no, you see, I take the numbers that show on this monitor, and I type them into a box on that monitor; why no cut & paste? faster to re-type the numbers; sigh).

But people provide context. Sure, an AI might tell you utility costs were up last quarter, but they won't know it was because of a water leak that went unnoticed and tripled the bill. Or it will tell you that wages were up, but not that it was because Bill from Operations had hernia surgery and we had to bring on a temp for 2 months. And it certainly won't tell you that Jim's back on the sauce, so we should probably begin putting out feelers for a new salesman.

So much of what business does is tracking metrics, yes, but the numbers never tell the whole story. There's always a backstory. Things that just can't be captured in raw data and hence can't be summarized by an AI. And AIs can't keep the ship sailing. Every small business has the guy/girl that does all the little things for everyone that absolutely holds the whole damn thing together. I'm not a BigCorp guy, but I imagine most departments are similar.

How about customer feedback? How can a model distill valuable (actionable) meaning from disparate communication mediums other than superficial high-level conclusions?

Expectations are just not realistic right now. There's going to be a lot of disappointment.

That site gives me flashbacks to my 1994 O’Reilly HTML book. That and the Perl one (was that the Camel book?)

I am glad to see something like this though. You might reach some people.

You see folks, when you had to live through doing this shit by hand, and if memory serves, this was even pre-css (or maybe pre separate css files, shit, maybe just pre-css2?) and your only dynamism was through cgi with C or Perl, because Cold Fusion cost waaay too much for mere mortals (and the Sun server costs were even more eye watering), you welcomed things like Linux, PHP, MySql and Apache with open arms. It was like being rescued from a refugee camp and being given prime rib. And think about this- if someone thinks PHP back in the version 3/4 days was salvation, that person has seen some real shit.

Point being, we came OUT of the woods. We knew better than to build rube goldberg machines that would lead us right back in, lol. We have a word for that- masochism.

Take what you know now, and simplify. Untangle the knots. Spring clean. Throw away the cruft. Free your minds.

And I know y’all can. Don’t be pretending like you don’t rewrite all your shit every year anyway ;)

My initial reaction was to be horrified for these guys, and ooh boy the comments didn’t disappoint.

But this is an open telemetry data aggregation and analysis tool (at least that’s what my half hearted perusal of the post led me to believe).

Probably not too heavyweight for such a tool, although my default response would be a Tannenbaum-like admonishment. Or maybe the Jurassic Park response.

I will say I have seen worse wastes of resources.

Let them have their fun. Either they learn from this implementation for v4, or it works out and there is no need for v4, in which case, as they say, the proof will have been in the pudding. Win-win.

They probably won’t like their egress bills tho.

Not everyone does their work in a web browser.

And even still, you don’t have to parse raw markup to grab properties from DOM elements. That could be handled by a browser plugin coupled with some some user guided training.

PDF is another beast entirely. I think there’s already a whole thread about that going on now. I’m going to zip my lips. I’m still waiting on Adobe to return my call from two years ago inquiring about the licensing costs of their parsing library for a small shop. Good thing I wasn’t relying on them to get that project done, and thank goodness for oss.

I think my quip about the toy flew over a lot of heads, so I appreciate that someone got it.

The reality is that most applications and websites don’t expose enough context about the what of what you’re actually doing for AIs to be able to meaningfully infer from natural language the steps required to complete a given task.

We humans are very good at filling in the blanks based on if we’re working in Photoshop or VS Code or Excel. We infer a lot of context from the specific files we’re working on or the particular client or even the files’ organization within the file system, or even what month or day it is.

I am skeptical that models will be able to replicate a complex workflow when there’s very little in the way of labels and UI controls even visible.

I know a weekly spreadsheet from a monthly and quarterly, etc. I know the minutiae about which options to use to generate the specific source reports, etc.

Workflows can be quite complex, no matter your role.

I mean I can just see it now: gift receipts being sent to the recipient before their birthday, internal draft proposals prematurely sent to clients, mixing up clients or commingling their data, overwriting or losing data; this whole thing just screams disaster. And I’m not even thinking about people involved with safety, or finance, or legal/regultory, or medical. Law enforcement?

This kind of thing can be done properly with well defined interfaces, common standards, and reasonable and prudent guardrails.

But it won’t be. It’ll be YOLOed on a paper thin training budget and it’ll be like your own little personal chaos monkey on ketamine.

Is it already 1997 again?

Funny story for you.

Maybe this was just a GenX thing, and isn't really done anymore, but in elementary school we were all fingerprinted.

They told our parents it was in case we were ever kidnapped it would help them find us.

You might be surprised about how many people's fingerprints the gov't has.

I propose the Open Agent Interface.

We can call it OpenAI.

I'll see myself out.