Hacker News Reader

Comment by satvikpendem | Hacker News Reader

I've used it, it's fine I guess, nothing super outstanding. I don't want to be that guy on HN, but the way they market themselves as being some revolution in browsing while simply wrapping Chrome like others like Vivaldi (which also has a lot of customization), Edge, Opera etc is not that interesting. It's even less so when I would lose adblocking due to Manifest v3 because, really, how long will these browser makers continue keeping a fork around before eventually giving up?

A true browser revolution would be something written from scratch like the recent SerenityOS one. Of course, the average person won't use it, but that's what it would take to truly be revolutionary. In the meantime I'll continue using Firefox.

View on HN

A browser revolution isn’t gonna come from remaking an engine to do exactly what Chromium does. Ie render websites.

If they brought some new ideas to it, then maybe. But it’s likely that if a new browser engine did bring some useful new ideas it would be incorporated into chromium well before the other browser engine could achieve a fraction of Chromium’s market share.

Browser revolutions will most likely be built on the front end. By focusing resources on the parts where the browser makers can actually differentiate and not on the part that by definition must have the same output.

I envision a browser inspired by the likes of foobar2k in which core engines are open source and interchangeable, and any browsing experience is achievable through a rich plugin and layout system. The browser itself would be little more than scaffolding and an SDK.

That's more or less what manifest v2 allows for. But, you can use that to block ads. It simply had to go!

Isn't this what XUL was? Admitedly I'm aware its arch was too complex to compete with Chromium's speed and security features, but way before Electron, I was using desktop apps built with XUL like Songbird for my music app, as well as a Twitter client whose name I can't recall now (of course, Thunderbird as well). Maybe they threw the baby out with the bath water a little bit by electing to concede to WebExtensions API since now Electron (read: Chromium) is filling exactly the niche that XUL provided.

Yeah, Electron and Chromium fill a close enough niche that I don't think the market is really demanding what I envision.

Try giving foobar2000 a spin or checking out some screenshots to get an idea for what I mean. Unlike Chromium, the browser should be extendable, at runtime, via a component system and editable layout tree. Runtimes, backends, common libs, fonts etc should be managed using a repository system.

Chromium requires recompilation (and know-how) in order to make any big change. Comparatively, a user should be able to completely reskin their browser, switch JavaScript engines, etc. on a per-domain, per-group or per-tab basis in seconds without restarting the browser.

There are two things that become harder with such architecture: integration, and security.

Integration is relatively easy to solve. I did not use foobar2000 for a long time; I can compare instead with Emacs, where various packages are aware of various other packages and will integrate if a counterpart is present, but won't require that.

Security is much harder; the need to isolate pages means relatively smaller customizable API surface, and, most importantly, greater trust to the components. It's much like web extensions now: you have to think well before giving an extension access to your web mail or online banking web pages.

XUL was an attempt to build a very flexible architecture, and it was great, but the (lack of) security killed it.

You can make PWAs of nearly every page. I do this to make 'apps' of Netflix and such. Don't have offline storage though, but at least you don't need the Electron stuff and you decide the extensions it runs (uBlock is a welcome candidate). There's also containers in Firefox.

How do you do this? Do you wrap the website with a PWA-friendly HTML template, then load the desired website in an iframe?

I can see Arc somewhat doing this with their "boosts" feature. I imagine they will eventually have something like a marketplace for these, e.g. youtube/twitter without the distracting sidebars.

But to your point the core browsing experience would remain the same. Would be cool if they opened that up to variability though.

From where I sit, browser engines are commodities. There's not notable differentiation to be had there. Are four actively-developed, open source, mainstream browser engines enough? I think so, given that we haven't seen a downside to a relative monoculture in open source operating systems.

Which are the four? I count Chromium, Gecko and WebKit.

One day I discovered an independent engine: LibWeb (Ladybird Browser) which is developed from scratch. Quite brillant guys and effort.

Ladybird isn't mainstream. So it doesn't count in this list.

Blink (Chromium) and WebKit are also closely related as the former is a fork of the later.

So there are effectively only two "independent" engines.

But in reality there is actually none. Both engines exist through Google money.

Given that most people can't even tell the difference between the internet and the web (hello, Arc developers!) and the web became the new all encompassing platform (like MS Windows was before) we have here a very dangerous concentration of power. Almost all private computing is controlled end-to-end by Google these days.

If Microsoft got almost split up for having way less power Google is long long overdue for a divestiture. Should this ever happen (I suspect the US would not give up control of personal computing all over the world so it's imho unlikely, but if) everybody would recognize quite fast how bad the mono-culture in the browser space really is…

Blink has forked from WebKit a long time ago and only diverged since, they should be considered two engines.

WebKit itself forked from KHTML.

I counted Goanna: https://en.wikipedia.org/wiki/Goanna_(software)

> Are four actively-developed, open source, mainstream browser engines enough? I think so, given that we haven't seen a downside to a relative monoculture

We have a near-monopoly with Chrome, a very distant second Safari, and a nearly non-existent Firefox. Anything else (like Goanna or Flow) have literally zero impact.

Chrome's outsized influence is such that it no longer even pretends to care about consensus or processes at w3c and just ships random features at neck-breaking speed.

What features is Chrome shipping that it shouldn't?

The question isn't "should or shouldn't". The question is how.

Web standards are supposed to be based on a) consensus and b) at least two independent implementations.

As an example, let's take a recent Chrome-only non-standard that made HN frontpage, WebTransport: https://web.dev/webtransport/

The status of this API? "Editor's Draft". That is, scribbled on a napkin [1]

--- start quote ---

These drafts have not received formal review and are not endorsed W3C.

These drafts MUST NOT be cited as W3C standards and may or may not become W3C standards.

--- end quote ---

Guess what its status is in Chrome? Already shipped, and advertised as available.

As with other such non-standards expect HN comments decrying other browsers as being "the new IE" because they are in no rush to develop this.

[1] https://www.w3.org/standards/types#ED

And the worst thing is when you call out Chrome for being the new IE instead[1] you get jousted by your web development peers because their dev tools are so great and other browsers are too slow to implement.

[1] Pushing through (user hostile) apis without consideration about others

> It's even less so when I would lose adblocking due to Manifest v3

You don't lose adblocking with MV3. I have been using uBlock Origin Lite since it came out and I have not noticed any difference in actual blocking. I like having adblock extension that has significantly less security or privacy exposure, and I like MV3 provides that.

You already lost some adblocking with MV2 because you are using Chrome where uBO doesn’t support CNAME uncloaking.

EasyPrivacy gets a lot (most?) of the CNAME-cloaked stuff even on Chrome and MV3 because of efforts to detect and incorporate the subdomains into filter lists. AdGuard's public DNS server watches for CNAME cloaking and the subdomains get pushed to a list [1]. Or one could set their DNS to AdGuard and get the CNAME blocking in real time without filter list updates. uBO on Firefox is of course the more flexible solution but along with cosmetic filtering, this kind of thing would make it hard to notice the difference between a really good MV3 blocker and uBO on Firefox.

[1] https://github.com/AdguardTeam/cname-trackers

It’s next to unusable for me, still using Safari + Firefox.

> It’s next to unusable for me [...]

May I ask why?

It’s bloated, the light purple colour is too opinionated for a browser, and it makes me think.

Btw, it's improved a lot since I used it for the first time! Cool, can use it for a few activities now :)

Reply on news.ycombinator.com