Hacker News new | past | comments | ask | show | jobs | submit login
[flagged] Theo de Raadt 'Re: new: lang/zig' (marc.info)
66 points by DyslexicAtheist on Feb 28, 2021 | hide | past | favorite | 51 comments



> > Zig is a general-purpose programming language and toolchain for > > maintaining robust, optimal, and reusable software.

> I am sure both csh and php aspired to do the same.

And I am sure they didn't. Such a bad take...

Languages are very often designed with a use-case in mind, or with a set of goals or core tenants.

Since he highlighted PHP specifically, there's a nice quote on wiki:

> Early PHP was not intended to be a new programming language, and grew organically, with Lerdorf noting in retrospect: "I don't know how to stop it, there was never any intent to write a programming language [...] I have absolutely no idea how to write a programming language, I just kept adding the next logical step on the way."

If you look at different languages, they have different goals. Haskell was initially created to explore lazy-evaluation, one of Python's drives was to be extensible and to be friendly to use for people not familiar with programming (largely inspired by ABC)[0].

[0]: https://www.artima.com/intv/python.html


Opinions from Twitter and mailinglists are not "news".

Please let people have their opinions, so don't try to overexpose them!


TdR's opinions, even thought controversial, often turn out right, even thought others thought otherwise. See for example https://news.ycombinator.com/item?id=20865492


The porter obviously just copied Zig's elevator-pitch from the website instead of describing Zig in his own words. The latter would be more helpful to the community.


Can somebody tell me why this is "[flagged]"? I don't get it...


Funnily enough, this has nothing to do with Zig... People will flag this because you simply can't challenge Rust, the poster child, where it's due; You have to understand that such thoughts are deemed almost "counter–revolutionary", i.e. the fight against the unsafe, evil, dirty hegemony of C/C++ must continue at all costs.


Would you please stop using HN for flamewar, including programming language flamewar? One of the core principles here is not to let the forum to burn itself to a crisp like other communities which got consumed by such flamewar.

If the flamewars were more interesting, it wouldn't be such a big deal. But the comments in a thread like this are such weak sauce.

https://news.ycombinator.com/newsguidelines.html


So how did Rust acquire this reputation to begin with? I've read about it and have written some. It's a nice language but not above criticism. Seems to have the usual modern language problems such as its own package ecosystem that's incompatible with everything else.

The creator of Zig got this right. If your language's binary interface is too complex, nothing else will even attempt to interface with it. Even some C features are too annoying to interface with. Newer languages either have insanely complex ABIs or are fully virtualized within a virtual machine written in C.


I wish dang would remove the flag function or modify it in some way. I understand it's there to stop spam, flamewars, etc but it can be abused and kill any chance of a decent debate.


The issue is the expected value of the thread [1]. Decent debate is great but requires a more substantive starting point than usenet/mailing-list flamebait.

The OP is fine in its original context of an OpenBSD mailing list—that's a focused-enough system to respond coherently. A large, weakly cohesive forum like HN is more like all the mailing lists put together in a blender. Posting the same message to this context amplifies the provocation by 1000x, if not 100000x, and the thread will fill up with comments by the self-selected most-provoked on all sides. This leads to boring, shallow comments, an effect which compounds as more comments appear. This is a failure mode for HN.

We've learned over the years that threads are sensitive to initial conditions: the same forum can respond either substantively or with meltdowns on the same topic, depending on small perturbations. If the goal is high-quality discussion (or, let's be reasonable, as quality as possible under the circumstances), the place to focus on is the starting conditions.

When a provocative post like this gets flagged sufficiently to remove the thread from circulation, the flags are usually coming from a coalition of users: one set who dislike/disagree with the author, and one set who recognize the thread as a failure mode for HN. That goes for all provocative topics, not just programming ones.

[1] I've written a lot about this in terms of comments—i.e. a comment's quality is the expected value of the subthread that it is the root of—but as in many things, you can treat a submission as a kind of root comment and the argument works the same way: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...


It gets flagged because it's fairly shallow ("programming language advertising blurb is an advertising blurb, news at 11", without any useful in-depth argument that hasn't been rehashed to death) and pulling individual mailing-list or bug tracker quotes up seldom does any good, especially if comments like yours will try to turn it into a flamewar.


Interesting thought, I'll keep my eyes open.

Funny, because Rust is rather a side-note/example in Theo's post...


The thread topic is now: top the linked email’s exaggeration and make a programming language’s community sound even more like a revolutionary force. Go!


This got 30+ points, flagged, dead, and revived in a matter of minutes. God, this whole ordeal is in fact ideologically charged, I guess he maybe onto something!


I wonder why this was flagged...

Was it about his view on zig? Or his comment about rust??


100% his comment about Rust.

See my comment below: https://news.ycombinator.com/item?id=26293035


Emotional attachment to tools is a human thing.

But Sturgeon's Law applies even to what we fancy.


"Today on the internet: Angry man rants"


Being angry doesn't make him wrong.


Being right isn't enough to be interesting. The key thing is the amount of information in the message. High-indignation, low-information messages produce uninteresting discussion on HN.

Actually Theo's message may not be so low-information in its original context, where there's a lot of shared state i.e. implicit information. But that doesn't carry over automatically. Posting a mailing list thread to HN is a lossy operation that way.


Maybe he needed to get that off his chest. Maybe that’s part of what people do on mailing lists, public or not. I don’t think everything that one posts is supposed to be rigorous, even though anything a person with a certain amount of fame writes in public can draw the ire of the whole internet against them with one off-hand sentence or paragraph.

There’s not much to discuss here. Just accusations of propoganda.


Don Quixote comes to mind. Tilting at windmills.


Windmills like spectre and meltdown?


Languages other than C as imaginary enemies.


He railed against the breathless descriptions not the languages themselves. I still agree with his put-up-or-shut-up position, and with throwing cold water on hype. My trade is engineering, and while I like whimsy as much as most humans, I will fight it when precision is critical.


"He railed against the breathless descriptions not the languages themselves."

He said Rust obviously had nothing and that proper privsep was exceedingly rare in languages other than C.


Is he wrong?


About "Rust has nothing"? That wouldn't be a productive conversation.

On privsep being rare outside of C? What's the measure? I suspect privilege separation is present or not more based on the type of software, not the language used. And I'm sure you can find bad/good examples in software written in C, Go, Rust, Zig, etc.


Not willing to use something others suggest doesn't make it your enemy.


There's all sorts of things in the OpenBSD ports collection that Theo probably doesn't use. A public rant about something you characterize as inconsequential is a bit odd.


Sebastien Marie wanted to make Zig work on OpenBSD. I don’t know why Raadt then respondend by saying that Zig hasn’t proved that it can do what it claims to do. But maybe I don’t understand this mailing list’s culture (or what “import Zig” means in this context other then to port to OpenBSD).


> I don’t know why Raadt then respondend by saying that Zig hasn’t proved that it can do what it claims to do.

I know why. It's because Zig hasn’t proved that it can do what it claims to do.


Does an application need to prove it can do what it claims to do before it can be ported to OpenBSD? Seems like a high bar. :-)


It's not a high bar, just plain honesty. I remember a while ago someone was promoting a language called v, making many claims of which some turned out completely untrue. This might be a norm in commercial software - some people believe that you can't sell anything if you don't exaggerate - but the open source world in general prefers a more honest approach. Hence many projects always remaining at 0.x release, for example.


It is a high bar that makes OpenBSD in security-sensitive roles way more appealing than a distribution with a lower bar. That high bar paid for itself repeatedly when I ran OpenBSD in the late 90s and early 2000s.


That's true, but not for the ports collection. There's all sorts of software there with historically bad track records in security. That's the point...it's just ports of a bunch of popular software.


Fair enough.


"One thing Rust has done, is to teach everyone they must lead with propoganda"

Sounds like somebody is bitter.


While I agree that de Raadt sounds bitter, I also agree with the sentiment. When I look at the software development scene these days (may it be compilers or libraries) it genuinely feels like there is more advocacy than substance.

I am not going to blame Rust for that. Rust is a language I have been meaning to look into and they have a need to be heard over the ruckus. Yet the preponderance of propaganda begs the question: is this real or is it going to be yet another language that fails to reflect the claims? While learning something new is wonderful, I would rather expend my energy on a language that has something genuinely interesting to offer.


There is no substance for most of it.

But taking Rust as example?

Rust is the one language that delivered in terms of performance and safety.


He's right, though.


Hard agree!

Not recognizing how right TdR (and OBSD by extension) have been is a giant red flag to me. It would be a whole different deal if they didn't have years of track record showing their conservative approach was 100% justified.


That doesn't justify unrelated rants in the mailing list though.


He is exactly a cultural leader in the space of operating system distributions. I think this is a prime example of using organizational system leverage for profound technical impact.


If that "culture" means that someone (but not everyone) is allowed to say unrelated things in unrelated venues, sure, I agree that it would have a great impact, negatively.

Seriously, you can instead say that the description for newer programming languages tends to be subjective (which is IMO true). Then the discussion about regulating descriptions can follow. Instead you have this useless rant. The original link is correctly flagged, not because it is anti-Rust, but because it has no substantial values to discuss at all.


Hasn't he always sounded like that?


I think he and I suppose many OS developers are sick of the constant calls to rewrite their OS in the new wonder language. Rust has had the hype train jacked (check all the threads on HN with 'just rewrite it in Rust').

He has had much the same criteria for years. Do the work to prove it. I've often thought it would be interesting to map out the OpenBSD userland and figure out exactly what the work would be.


The frustration is fair and understandable, but his characterization of Rust is wrong IMO.

Side note, has anyone read a Java textbook? They are notable for not being able to go more than a page without extolling Java's virtues in a way that is repetitive and annoying.

(My experience with Rust textbooks was not like that. They tend to explain the drawbacks of Rust's "radical wager" with the stipulation that for the authors, the payoff is worth it.)


Theo often sounds like that if somebody is using "handwavy" arguments.


That's the reality of mozilla

Propaganda, how do you think you hear about firefox everyday, everywhere, despite it having only just 3% user share

It was the same with rust

Propaganda machine works, just look at the USA


The popularity of a tool among masses, more so a technical tool, doesn't have anything to do with its merits and capabilities.

Firefox does offer unique privacy features which are not even a remote priority for Chrome and Edge (and IE) which do have the lion's share of the browser market.

And you are hearing about FF, mainly in tech-related forums and in places that people care about their freedom.

Sadly, the recent actual propaganda pushed by the Mozilla executives, especially that talk about "de-platforming" is beginning to question Firefox's main and only selling-point over its competitors.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: