52 comments

[ 1.7 ms ] story [ 93.0 ms ] thread
Comparing Browser speed with JS disabled is like comparing car speed without tyres.
(comment deleted)
V's main draw is absurd vaporware claims.
> Vox: Upcoming open-source...

So no source yet, only the HN demo render. I might pay attention when the open source is published and can at least pass Acid3 tests and works on all the site I visit, not only the simplest one.

It is surprising the amount of claims (well, future promises) on here for a project that is clearly getting off the ground.

Compare it to Ladybird's [0], where it is nice and simple. They're building a browser, but they're not ready yet. There's no claims as to the future.

It should already be cool enough that you're building a brand new web engine. There is no need for claims about performance, ad blocking or vertical tabs, when you're nowhere near there.

[0]: https://ladybird.dev/

What a beautiful car analogy.

Anyway it uses V8 just like all the other Chrome-alikes, so it should have the same performance there.

> What a beautiful car analogy.

Almost but not quite.

Web worked just fine without JS and to some extent still does (this site for example). So a better analogy would be something cars were fine without, but now have it and it improves experience (and sometimes bloats it eg one more thing to break).

Electric windows, seat heaters, GPS, come to mind.

But you can drive on any road without those, while you simply can't use much of the web without JS. I don't think there's a good car analogy.
My comment may have been a little tongue-in-cheek reference to outdated /. humour.

Anyway for the record Karl Benz first produced his Patent-Motorwagen in 1885 whereas the pneumatic tyre wasn't to be invented for another two years and even then in faraway Ireland. So in some some sense cars do predate tyres.

Things are more complicated than that, I'm afraid. Lots of CPU time (and browser complexity) is spent on DOM/CSSOM bindings and the way browser engine handles updates/redraws.
Not only that, but from the screenshot of Hackernews (which itself is an extremely simple website), it seems like it can't even do very basic HTML/CSS rendering either yet. It probably doesn't even pass ACID 1.
My goodness, calm down. The author just said it is in 'alpha'.

What did the author ever do to you?

> What did the author ever do to you?

Made incredible claims, did not fulfil them. Ergo, heightened scrutiny.

(comment deleted)
But what's the point of posting numbers for it then?

I can make you a browser engine that loads in 0ms by showing a blank page.

Built in V: https://github.com/vlang/v

> [the V language is a] simple, fast, safe, compiled language for developing maintainable software. Compiles itself in <1s with zero library dependencies.

So it's an another Chrome-based browser?
Not according to the link. But I didn’t look at the code.
Per the bottom of the article, there is not yet any code that anyone (besides the current author) can look at:

> Source code release in 2024.

It's not a browser, it's a browser engine. So it would be a replacement for Chrome's engine (Blink) or Firefox's (Quantum).
It's going to be a browser engine and a browser.
> Using Hacker News as a test (JS disabled):

    20x faster rendering than Chrome (3ms vs 60ms)
    4x less RAM usage than FireFox
    40% less CPU usage than FireFox
I would hope that the browsers tested against also had JS disabled.

Although regardless, I don't think it's an amazing comparison tool, being that most of the web uses JS.

> Although regardless, I don't think it's an amazing comparison tool, being that most of the web uses JS.

Add to that that most features to render Hacker News correctly are not implemented.

Nice to see another browser engine project!

For all the apparent diversity in full-featured browsers, right now they're all running on one of Gecko, WebKit, or Blink. And Blink started as a WebKit fork. If this succeeds we'll be up to four!

(Possibly you could also count Flow, though I haven't heard anything from them recently.)

There's also SerenityOS LibWeb/Ladybird[0], which is very active, though I wouldn't say they've made it to being daily drivable yet.

[0]https://ladybird.dev/

> source code releases in 2024

Can anyone from the Vox team provide a rough date?

Looks [1] like it's created by Alexander Medvednikov, who is also the creator of V.

I have no dog in this fight, but V and its author have a somewhat storied history because the language's documentation made very strong claims around performance, safety, and the state of the language that may or may not have reflected reality.

Some previous discussions:

https://news.ycombinator.com/item?id=20229632

https://news.ycombinator.com/item?id=20250990

https://mawfig.github.io/2022/06/18/v-lang-in-2022.html

https://github.com/vlang/v/issues/35

https://christine.website/blog/v-vaporware-2019-06-23/

https://christine.website/blog/v-vvork-in-progress-2020-01-0...

Again, I'm not taking a side one way or the other, nor am I endorsing any particular interpretation, but I think it's relevant history.

[1]: https://github.com/vlang/vox-browser

I'd love it if we could have a single story that's even somewhat tangentially related to V not devolve into rehashing the past. Let it go.
V is for Vendetta, I've been told.
I think if VLang was indeed stable or if it delivered on some of its promises it would be justified to stop nitpicking flaws.

But the language unfortunately remains unstable to this day. There are memory leaks with very basic things (e.g. readline, reading from maps) that are still being fixed as recently as half a year ago: https://github.com/vlang/v/issues/18009, https://github.com/vlang/v/issues/19454

V is stable for a pre 1.0 lang and delivers everything promised on the home page and in the docs.

A fixed memory leak in os.input() in a 0.4 language, really?

The creator of V repeatedly appears on HN with incredible claims about a new thing they're going to supposedly release soon.

I don't think it's "rehashing the past" to point out that a person's new incredible claim is suspect, given a history of making incredible claims in the past without actually delivering.

> I don't think it's "rehashing the past" to point out that a person's new incredible claim is suspect, given a history of making incredible claims in the past without actually delivering.

Yes, it is. You literally used the words "in the past." If you don't like his new browser engine, fine, but repeating this two minutes hate every time one of his projects gets posted on HN is self defeating and toxic.

I would agree that people are giving Medvednikov an unreasonably hard time if he learned from his mistakes, but everyone kept bringing up his past anyway.

But it just seems like Medvednikov keeps repeating the same patterns over and over. He's making dubious claims about software without sharing how he arrived at those claims, and he's calling something open-source when he hasn't released the source.

Why should people who know about this pattern stay silent and allow others to be misled by the same patterns?

> He's making dubious claims

That's your opinion. I don't agree with that characterization of him or his projects at all. It's a convenient way to give yourself some cover for attacking a creator you don't like.

> he's calling something open-source when he hasn't released the source

never happened

Isn't it what's happening now with vox?
V has been open source since its release on June 22, 2019.

Vox is not released yet, and the website literally says "upcoming".

Well considering he's still making wild claims like rendering 20x faster than chrome, it doesn't seem like it's in the past unfortunately.
You can download the browser, press D (debug) to see how fast it's rendered, and compare to Chrome (dev tools). You'll get similar numbers.
> A settings option to have vertical tabs.

How come it's the browser engine who deals with that and not the browser itself?

They also say things like "Efficient built-in ad blocker Based on Ublock Origin" and "Keyboard navigation Vim bindings", so I think they're developing both a new browser and a new engine for the browser to use, and kind of conflating the two when discussing features on their home page.
If this browser does what it claims it will spread like fire. Congrats on the initiative!

Edit: Tried it in Proton on Linux. Renders 20x at least slower than any other browser.

Won’t have a lot of Spanish users (Vox is a far right Spanish political group…)
That would be ridiculous. Vox is Latin for "voice" and has been used widely for a lot more than your reference. I am fluent in Spanish and lived there in the late 80s, but I have been out of touch, so this is the first I've heard of it. People use the word 'trump' all the time. Context and intelligence are needed in discourse.
And its also the name of American news site (vox.com), Polish furniture company (vox.pl)...for every major gtld, the vox.* is taken for something...Do not event start with the fruit nammed apple.

Context is the King, like @eggy said.

well, the V hello world example still leaks memory like a sieve so, that's probably bad news for a browser engine
I wonder the developers' chat app from what feels like 5 years ago will ever come out [0]. (edit: it's 6 years, according to the HN thread I found it on! [1]).

By the way, you should really close that Patreon (which in fairness isn't generating much income), considering Volt is clearly never coming out or being developed on, and you have another Patreon for your language!

[0]: https://volt-app.com/

[1]: https://news.ycombinator.com/item?id=14778263

Another laud marketing attempt in converting attention economy into donations.

Deliver value first, then waste everybody time as it is probably will suck anyway.