Maybe it's the soulless cynic in me speaking; but the obvious snow-theme around NixOS notwithstanding, 'snowflake' MIGHT NOT be the best name for a distro aimed at 'casual users'. It's as though they're saying, 'LOL! You snowflakes can't be assed to figure out how to install nixos, but still want to reap its benefits? We got you precious snowflakes LOLOLOL'
... I mean, when you're this clueless, maybe don't put out 'articles' for others to read -- it's wasting everyone's time.
I thought the title of this article was intriguing; because in the Linux community certain aspects of the desktop experience do get hyped; & there's a tendency in general to sweep various usability issues under the rug, with the unwarranted confidence that we're already "better than everyone else" in every way; though the article doesn't address any of those.
Not sure what the question is -- are you looking to port extensions over yourself, or are you just exclaiming, "it can't be so hard, so why won't someone do it!".
For ground-level, basic stuff (managing a popup, communicating between popup & a 'background' script, between content loaded on the browser & your scripts, managing a context menu, etc.) writing an extension is straightforward once you develop some degree of understanding of the sometimes convoluted paths the data needs to take, the permissions you need to have in order to pass messages through, etc. Larger extensions are full fledged applications in their own right, though, so tackling them introduces difficulties of a different order of magnitude.
The Falkon browser is extensible (in its own way) through QML; and the Nyxt browser is extensible in common lisp. These aren't 'webextensions' in the precise sense of the term, though they could be just as useful. I wrote a basic bookmark manager that I use mainly on Firefox; but I ported its core functionality (just send the current page's title, url, & selections from the <head> tag over to my database (postgresql via the postgrest http frontend, to which I just make a fetch request)) to QML, and it was pretty straightforward. Falkon is based on Qt's QtWebEngine, which is Chromium-based; Nyxt is based on WebKit.
edit: There's also luakit and qutebrowser . The former is extensible via lua 5.1 scripts, the latter, python; there isn't a wealth of documentation & examples, though (at least there wasn't last time I checked) so the API can be a bit of a mystery. Luakit as webkit as its engine, qutebrowser is built on QtWebEngine just like Falkon.
xfwm4 could work w/o Xfce, though I doubt that it would be worth the effort to script the missing bits by hand. Xfce is pretty modular; once you turn off the tracker/indexer, and whatever useless package manager gui the distro may have included (e.g., 'dnfdragora'), it's pretty lightweight. You can also turn off the compositor. The stock xfce4-panel is also miles ahead (IMHO) of various independent panel programs, both in functionality, as well as looks -- and its widgets are also entirely modular.
labwc is a window manager in the vein of openbox; I guess under wayland a window manager has to be a compositor too (?); but it's no different from sway in this regard.
Kotlin 'built by communism'? Because the founders of JB are Russian? Is that it?
Swift is 'greed' how? It's open source since 2015 or so; & available on Linux. Apple's graphical toolkits are 'closed down'; & obviously restrict users' freedoms; though not sure how that implies 'monopoly'. 'Monopoly' would be trying to dominate all toolkits, not have one's own.
Those are straightforward; it's the remaining 900 options that are confusing. I always need to look up --excludes and always get --directory wrong, somehow.
Yeah, I'm really happy with my Leopold which I've been using for the past 3 months. I used to have Unicomp before that; and while the typing feel was a little better than the brown switches I currently have on the Leopold, its build quality was lower, and eventually it just died on me thanks to what I later found out was a notoriously failure-prone controller they used to use back then. I'm told that Unicomp's build quality has improved a lot since then.
.... though the frustrating thing is that I was able to get the Unicomp only because I was living in the US at the time; and the Leopold I got thanks to relatives in S. Korea. Where I live, 'mechanical keyboard' is treated like a synonym for 'gamer keyboard', and all the BS associated with that.
So excellent off the shelf brands exist, though one has to do some local research first.
For reference purposes, the manpages. For a more conversational, 'guide for the perplexed'-type book, though, I'd wholeheartedly recommend Michael W. Lucas's 'Networking for sysadmins' book: https://mwl.io/nonfiction/networking#n4sa -- this one helped me understand many of the fundamental concepts.
True; but many guides & tutorials that came out long after that still give their examples/instructions in iptables. Esp. those that involve ifup/ifdown/etc. scripts.
(although nftables is considerably easier to read, IMHO)
He/she runs off to online forums to bicker about which {distro, WM, DE, text editor} is best, and how all others are unfathomably inferior, fundamentally broken abominations?
Or ... oh, sorry, I thought you meant something else.
Just as a side note -- if the user is ok with (or actually prefers) handling the archival & indexing of the material himself/herself, SingleFile works great as a 'read later' tool. It can save into a WebDAV share; can be imported into a self-baked minimal webextension as a library; & has an almost feature-complete mv3 port.
Works on Fennec (based on FFox Beta, I think) though a bit slow, unfortunately. Available otherwise on desktop FFox & Chromium.
Maybe it's the soulless cynic in me speaking; but the obvious snow-theme around NixOS notwithstanding, 'snowflake' MIGHT NOT be the best name for a distro aimed at 'casual users'. It's as though they're saying, 'LOL! You snowflakes can't be assed to figure out how to install nixos, but still want to reap its benefits? We got you precious snowflakes LOLOLOL'