Once again, people are grappling with an axiomatic definition of "memory safety" and avoiding the fact that it's a term of art with a very specific meaning: comprehensive protection from The Memory Corruption Vulnerabilities, which include overflows, the lifecycle vulnerabilities like UAF and type confusion, and uninitialized variables.
To the extent Fil-C and Rust both address these vulnerabilities, and don't include design features that in any practical way admit them, they're memory safe. What always feels like is missing from these kinds of analyses is that there are lots of memory-safe programming environments. Almost every Java, Python, Ruby, Javascript, and Go program is memory safe, in the real meaning of the term.
The competition to lock in and promote adoption of the "most" memory-safe language is a category error... unless you do what every language-war argument does, and redefine the term.
The really silly thing is that Fil-C and Zig here have chosen a definition of memory safety that specifically excludes a bunch of overflow vulnerabilities and type confusion. At least it’s been specifically defined, but such a narrow definition makes the end result to me almost wholly uninteresting. UAF no longer turns into a RCE, but your average parsing packet code is still just as likely to be a buggy mess.
Even within Rust some folks do this to themselves, misusing unsafe with code that is 100% safe Rust, only to taint code that should only be called in specific cases, completely unrelated to memory safety.
I never seen this happen with GC/RC languages that also support unsafe code blocks.
That seems a very strange claim that it's a reference to one minor use of a vastly larger tragic cultural phenomenon [1] without any other contextual indications that's the case?
I'm not dying on this hill. I punched up this post and that paragraph is mostly mine, and I'm telling you it's a random Seinfeld reference and not a joke about an actual tragedy, any more than me joking about the Andrea Doria would be. I'm not arguing with you about whether the actual event is something worth making fun of.
> it's a random Seinfeld reference and not a joke about an actual tragedy,
lol ok so I wasn’t deeply offended by the dingo reference for what it’s worth, but this attempt to distance yourself from it is weird. Both this post and Seinfeld are referencing the event directly.
And that’s fine with me, to be clear! I’m not clutching my pearls, I tend to have fairly edgy humor, not trying to put you on blast or die on this hill either, etc - I mostly just find it odd. Most companies, on their corporate blog, wouldn’t feel comfortable joking about most tragedies - would you make a similar quip about 9/11? Charlie Kirk? Casey Anthony? I’m guessing not. But the dingo thing is fair game. People say it without even considering it. What’s different? Genuinely asking
sucks: fly backburnering the app platform to build sprites.
> [...] Sprites, and focusing the company on [Sprites] and the problem they solve.
Then later
> first path [the thing I like/advocate/use] and [Sprites] ... We do one thing or the other. We don’t limp in on both. And if we fail, we fail with our full asses.
Hard to interpret that as anything but the thing I like being backburnered. Which is how it goes, but sucks.
Ah, that's fair! I just got the impression that other people on this thread were axiomatically deriving that we were killing Machines, rather than just shifting the roadmap towards Sprites.
Well, I do interpret this is step one towards a forthcoming post titled "Our Incredible Journey [with Fly Machines]"...
Whether that comes to pass depends on sprite uptake and so forth, but I believe it to be a lot more likely than I did a month ago. So, like I said, bummed, because fly is genuinely nice and I dislike using aws.
That's nice of you to say, and I appreciate it. I'd be sad in the event you're describing too! I spent a lot of time on Fly Machines (not nearly as much as JP and Peter, though).
Why are you so sure "DX" is the right way to compete with commodity full-stack hosting? Have you had Claude deploy on AWS recently? It one-shots deployments just fine.
I never met anyone who likes Oracle ($331bn mkt cap), SAP (€187bn) or Salesforce ($134bn) either.
However as businesses they sell, they know how to sell and know what they customers need. They need a lot of compliance and IT maturity for sure.
Enterprise (and even SME) customers don't want to fuck around with Claude and use hero shadow IT. They want to buy shit like Jira because it turns out cheaper than DIY even with AI. They know it will be HIPAA and SOC2 and most other things they desire like that.
AI does kill the tail end but Atlassian was always practically free to the kind of person who can replace it with a vibe coded weekend project (try it! sign up for a Jira!).
While you are there, you can go and set up myriad AI workflows, build a forge (plugin) app using AI in studio then set up Rovo AI automations. Yes it is very AI native now.
Some day you can type "make the next Google" into AI and it will do the paperclip thing. Until then... SaaS is not dead. I repeat SaaS is not dead. Not dying. Disrupted yes. Not dead.
I think I'll be the 10,000th industry person to say that Oracle, SAP, and Salesforce are deeply at risk at the present moment, and if they have a chance at retaining value it's only because they're Systems Of Record, and that very few companies are going to make the mistake of letting them lock up their data and keep it away from agents in the future.
I'd rather discuss first principles than use ad populum arguments.
Going back to Jira (you may be right about Oracle but I doubt it) they are well position to ride the AI wave and do well. I gave reasons and details in my last comment. It is easy to slap on "AI pilled" baseball cap and use the same talking points. I'd love to hear specific arguments about why Jira is not well placed and I happy to change my mind (coffee mug extended out).
I would like to take this moment to apologize to you, personally, for not giving a shit about "AGI". I wish I could tell you I've come to see the error of my ways, but I continue not to care about it.
this was amazing post. i enjoyed reading it, it was relevant to some work stuff i was battling with. the linked articles at the end were amazing too. i had tonnes of respect for fly.io and yourself because of these posts. technically informative, clear, engaging. great stuff.
the llm post? complete polar opposite.
> I would like to take this moment to apologize to you, personally, for not giving a shit about "AGI". I wish I could tell you I've come to see the error of my ways, but I continue not to care about it.
thanks for the snark, and kinda proving my previous point about the tone of the article that was supposed to be a personal blog post and then ended up on a company bloghttps://news.ycombinator.com/item?id=49052217
I'm not being snarky. I genuinely do not care about AGI.
It's weird to watch people write as if I'm trying to back away from that article or hedge or whatever. I expect to wrap my career up at some point and look at that post as one of the better and more valuable things I've ever written.
I can't judge how snarky I'm being overall. I'm writing like I would on any HN thread. What I was saying was: my lack of concern about AGI is not snark.
It's weird to me that anybody would call that out. Are you super worried about AGI?
okay, well, then this is me telling you that overall, the comment with the fake apology came across as really snarky.
>Are you super worried about AGI?
no. i typically agree with your takes on ai. i usually just happily ignore your tone and enjoy the substance of your comments but i thought this particular exchange from you was exceptionally odd so i commented on it.
I am paying attention to what you're saying but I think if you look at the context of this thread and take a beat you might sense, like I do, that the bar for charity is a bump or two lower than it normally is.
> this was amazing post. i enjoyed reading it, it was relevant to some work stuff i was battling with. the linked articles at the end were amazing too. i had tonnes of respect for fly.io and yourself because of these posts. technically informative, clear, engaging. great stuff.
i deliberately pointed out past praise and respect i had for your writing. i made an effort to go out of my way to provide an example of good stuff. as someone said in the andrew kelley black pill youtube comments -- people are a package. i too can be abrasive, dismissive, arrogant etc. i am a flawed human being too. so, in the spirit of not being a complete fucking dick i decided to point out something good you have written.
i'm highly critical of the article in question. i have the freedom and right to be critical of it, whether or not you think it's one of your best pieces of writing. and your general tone in this thread (and previous threads) has only solidified my view.
You gotta understand the context we're discussing this in, where a post from a year ago that raised a mob of anti-AI people got brought into a discussion about this, a post that is by itself a tricky needle to thread. That was an unpleasant and unproductive thing to do, but more important a cringe thing to do: there are apparently people still in their feelings about that post despite everything that happened in the year since to moot the debate.
You're of course free to have your own opinion! I'm just relating mine.
This is what I mean about the static equilibrium in 2025. That's what everyone was doing: just writing the other side off. That post of mine did a lot of things, but it did not do that.
The title was the point of the post. I have acquaintances who are still sore at me about it. But the problem wasn't that people disagreed with me about LLMs; the problem was that we were locked in an unproductive and static equilibrium of people talking past each other and hedging their way out of any direct, empirical confrontation over what was happening.
It's easy to forget how different early 2025 was to today. You would generally feel like you were in an inverse Emperor's New Clothes situation if you believed what your own eyes told you about how LLMs wrote code. That's not the case anymore; you hardly ever see people talking about (or, really, extrapolating wildly about) hallucination today.
(For a lot of reasons I think a lot about how that post was received in 2025 and what the world is like today. "Vindicated" doesn't capture the feeling.)
I still find it incredibly troubling. Not because the substance was wrong, but because a person of your caliber has stooped to the level of writing provocative headlines and language to attract attention. I strongly believe world-class technologists are supposed to be above that sort of thing. Leadership matters: not just thought leadership, but attitudinal leadership.
The funniest thing about this post is that I was writing it, in my head, to a specific group of developers I've worked with, and who I know to be smart and more talented than me. What I should have seen coming was that when I described a group of people as the smartest developers I know, everybody who read it would think it was about them.
> consider how difficult it would be to find a nice way of writing an article
Writing is often difficult. That's what makes good writing so distinctive and valuable. Just because something is hard doesn't mean it's not worth putting in the effort.
> "You are all fucking idiots" is sometimes a reasonable starting point.
It might be a good starting point in one's head (which, admittedly, it often is for me, but I then try to convert that into something more persuasive and less divisive). I think we'll just have to agree to disagree on that.
> I think we'll just have to agree to disagree on that.
In a saner timeline I'm right there with you. Certain types of policy and cognitive errors, though, remove the floor on communication decency--some popular conceptions are just so incredibly backwards that it's seemingly a disservice to treat them as anything worthy of more gentle correction.
reply