I guess `www.` (and only `www.`) is always special, and the only TLDs with two components are `"co.uk", "co.jp", "com.au", "co.nz", "com.br"`?
I don't know how critical this "Manager" is (what even is a "boost"?), but a web browser should absolutely have a proper list of TLDs!
What is wrong with such structure? How would you structure this code? Genuinely asking
This typically creates artifacts like each "Manager" owning too much of its implementation (not benefiting from or contributing to shared structures, such as a proper domain suffix list), inconsistency between different parts of the app (since different "Managers" don't necessarily share common patterns between them), and tons of hooks into random "Managers" all over the code.
To me, it feels a bit like an "emotionally driven" architecture, where the organization of the code is based on the list of features of the app, and not based on the implementation of those features. So rather than having, for example, a drag and drop component for the tabs to use, you would have, for example, a ReorderingTabsManager, and the implementation may behave differently than drag and drop in other places. So rather than factoring out code into modules for deduplication, you're making modules ("Managers") based on where they are in the product, and duplicating functionality across each module, to varying standards of completeness and/or quality.
Now I don't know if this project is quite that egregious, but it hopefully illustrates why I raise an eyebrow when I see a project architected this way.
Is this sarcasm? The public suffix list will give some ideas for omissions: https://publicsuffix.org/list/public_suffix_list.dat
"I've been experimenting with old UNIX systems recently and have come to somewhat similar conclusions. (Regarding software like window managers becoming more simplistic and some programs having to poorly attempt to pick up the slack themselves)
It feels like open source software projects shifted from making 'program' and instead tried to make "alternative version of windows program". Looking at these old systems I see all these options and intuitive ideas, even down the metaphors used to describe actions. Last time I used a modern UNIX desktop environment it felt like everything was just trying to be a simplistic Windows alternative instead of a good operating system."
* Fancy logo
* Blink engine so it's basically Chrome like every other alternative browser
* Mention of AI somewhere on the website
* Minimal UI clearly inspired by Safari
* Heartfelt promises of speed and privacy
* macOS only
Thank you, I looked all over their site trying to figure out which engine they were using.
More like Arc † nowadays.
For example, most of the key differentiators of Brave could be accomplished similarly in Firebox with a litany of extensions -- such as UBlock Origin as just one example -- or Privacy Badger if you'd like to be less 'heavy handed'.
The only other differentiator I see is the use of cryptocurrency as a way of compensating users for watching ads and the use of a crypto wallet; which if your not interested in such functionality is meaningless.
Yet I see very educated, competent, and intellegent people I've known for years be advocates and at some points "zealots" over the browser.
I would love to understand this. I'm honestly open to discussing this in good faith as I would like to understand the benefit here, and if I am somehow missing something will be the first to admit I was ignorant.
Firefox-based browsers do not support macOS automation (AppleScript/JXA). Safari lacks features/extensions. Orion/Vivaldi had bugs any time I tried them.
From the Chrimium-based browsers I tried, Brave blocks ads, supports PWAs, the crypto stuff can be turned off, and is stable. Brave does not excite me, but it's good enough.
https://old.reddit.com/r/browsers/comments/1j1pq7b/list_of_b...
The lobste.rs site has taken hostile speps towards Brave:
https://lobste.rs/s/iopw1d/what_s_up_with_lobste_rs_blocking...
Still, Brave does offer a few unique advantages.
- it is equivalent to Chrome on sites that require it, and does not have the compatibility problems of Firefox
- Ad block is built in
- it is easily available if you are not running Play and GMS
- it is a mature browser, where most everything works as expected
- the bad aesthetic choices that have been introduced to Brave so far are easily undone
No, it's not perfect, but there are use cases.
The trouble with Chrome is that it is deliberately configured to maximize Google's ad revenue. The omnibar does not show you recently visited websites when you start typing something because they want you to do another Google search so they can serve you more ads. The new extension model deliberately neutered the most effective ad blockers available.
Brave is Chrome without the perverse incentives. Their developers take a security-first approach to everything, to the extent of explicitly _not_ having a centralized sync service for bookmarks, passwords, etc. They have an excellent content blocker built in, thereby doing an end-run around Chrome's new extension model. The crypto wallet and Brave ads are optional - you can disable both in the settings very easily. And since it's a Chromium variant, you can use all of the existing Chrome extensions for third party software like 1Password and the like.
- vertical tabs
- maintained by more than a single person
- support for extensions
- not owned by China
- not Firefox
- not Edge*
All the AI and crypto slop can be turned off completely, so I don't care at all about features I never see after initial install.
*Edge is fine if properly configured via GPO, which I can neither be bothered to figure out how to do under Linux nor have the patience to do on my private Windows machines. Works great at work though.
I assume we would both already exclude the likes of Chrome, Edge, Opera, Safari, etc.
This will be a long reply though.
The TLDR is: Security is number one, so extensions are bad and built-in features are good. I hate the cryptocoin/adware/AI features but the degrading act of disabling it all is mercifully short. It also has to run on Linux, so I can't even consider browsers like Nook. Most important to me are the (1) Chrome features and (2) the Shields feature tacked on. I use profiles and shields very extensively.
The TLDR TLDR is: Shields good
---
Caveat with the below is that Brave is full of bullshit to disable, with a new piece of bullshit added every year or so. That disparaging term is not one I use lightly!
The bad aspects are made worse by the fact the CEO of Brave is a person who I generally don't trust. I've been using Brave for years with the understanding I might have to jump ship at any moment.
Onto the good things:
One of the necessary things it provides is a browser which I can use to browse the internet, including captchas. For my mileage, Firefox has been broken for me on every platform I've used it on, every time I've tried to get back into using it, for years. I've exhausted all the time I ever wish to spend trying to fix a browser. Since I could not use Firefox to browse, it was not an option for me.
A second necessary requirement is that the browser should be available on the major desktop and mobile OSes, especially Linux. So, Orion, Nook, etc. don't count as browsers to me.
A third necessary requirement are timely security updates. Last I checked, Brave got security fixes from Chrome on a timely basis. Nice.
Then, there are a bunch of nice to haves. Brave has the Chromes profile which I use heavily (although Firefox is set to get a clone of Chrome's profiles soon-- the existing 'profiles' and 'containers' solutions were not usable alternatives.)
A second nice-to-have is telemetry - how often is my browser making requests unrelated to browsing, and to how many parties? I last checked this years ago, but I remember Brave performing well here.
The third nice thing is the Shields feature, which I've come to rely on. (If Firefox copied this wholesale, like they're doing with Chrome profiles, that would be a major improvement.) It's an easy-to-use interface to block ads and JavaScript. It works on mobile as well, which is a huge advantage.
Shields can be replicated with extensions, but I try to minimize the extensions I use. Each extension requires permissions for every site (!!!) So, if just one of these extensions developers were compromised, or the extension itself had a vulnerability, then I would be compromised too.
plus, you get synchronization across desktop (Zen) and mobile (Firefox for iPhone/Android). since Google limited theirs only to official Chrome, this feature is basically exclusive to Firefox and forks, Arc <-> Arc Search, and Chrome for desktop <-> Chrome for mobile.
the only missing from the sidebar thing is Library as a central place to manage downloads, spaces, and history. and although the downloads window looks a bit unsexy, it's totally enough
I find this sentence to be a little odd. Who are “they”?
I'm interested in seeing all the new browsers that will come out in the next few years that are based off Ladybird. Or alternatively what Ladybird will enable in terms of customization. I think the days of Chromium/WebKit/Gecko forks are numbered.
I'm going out on a limb here and betting they're numbered in the high thousands minimum.
Nook is a well-known brand in consumer tech, ereaders aren't that far removed from Web browsers, Nooks have a Web browser, and B&N also has a "Nook for Web".
> Open-source forever
> Transparent code, permissive license, and a community-driven roadmap.
Which I was going to mention is contradictory, because the point of permissive licenses is that it does not have to be Free forever. But the license is actually GPLv3 instead. So still contradictory wording, but the "permissive" is the part that isn't correct :-)
No, the point of permissive licenses is that third-party derivatives, which have no impact on the licensing of the original, don't have to be free ever, while the point of copyleft licenses is that they do.
Neither has any effect whatsoever on what future first-party licensing can be; a commitment to "open source forever" by the copyright owner is mostly orthogonal to what kind of open source license the copyright owner offers. (Now, its true that if the owner accepted contributions under a copyright license rather than under a CLA, they would likely have no practical choice but copyleft now and forever, but that's an issue of the license they accept on what they can offer, not an effect of what they offer itself.)
(OTOH, using "permissive" for GPLv3, a copyleft license, is actually contradictory, as you correctly note.)
> No, the point of permissive licenses is that third-party derivatives, which have no impact on the licensing of the original, don't have to be free ever, while the point of copyleft licenses is that they do.
This should read "Yes, [...]".
The point of permissive licenses is for people to sublicense it. You can use this to sublicense the software using a license that actually enforces it must remain Free (see Redict for an example) but it is almost always the case that someone uses this headlining feature of permissive licenses to lock the code up and extract rent.
Your next paragraph frames single contributor or CLA as the two primary development patterns, when that those are absolutely the exception, especially if we exclude repos for things like AoC our homework, which are mostly single contributor.
Only the code that is yet to be released is not.
From an end user perspective BSD/MIT and GPLvX licences offer the same guarantees. It only is different if you are a contributor or if you intend to distribute modified or unmodified code yourself.
It's guaranteed to have at least the terms of the permissive license (usually requiring attribution), but no, it does not guarantee code released under a permissive license will remain available under permissive terms. That is literally the point of the permissive terms: so people can apply more terms under a sublicense.
> From an end user perspective BSD/MIT and GPLvX licences offer the same guarantees.
No they don't. I can decide to stop distributing a BSD/MIT licensed application in both source and binary form, in favor of only distributing it in binary form under a sublicense. As a user, this is not "open source forever". This is "open source until we use the distinguishing feature of the license to make it not open source".
GPL, assuming multiple contributors and no CLA (both of which are extremely common) ensures this cannot legally happen (unless they somehow get all contributors to agree against their exercised rights).
