ebuku VS s6-man-pages

Another option is to use something like ebuku.

i remember looking at that Google style guide a while back, and not being enthused about it. It's true my style across my POSIX scripts isn't yet entirely consistent, as i'm new to writing POSIX shell scripts, and am still working out what's best for me, as the person who's going to be the primary maintainer. Still, i believe my style to be basically consistent within a given script. Having been programming for a few decades now - although i only started coding-for-pay in the late 90s, starting with Perl - i've developed my own preferences regarding code layout (such as in my ELisp packages, e.g. Ebuku), and nowadays take the approach: I'll follow others' style in others' projects, and will generally try to follow common style standards in my own projects, but will modify them as needed when i find they're not conducive to my work.

Well, indeed, i certainly wasn't expecting you to do so! But i already have more than enough volunteer FOSS stuff on my plate as it is (e.g. Ebuku, pulseaudio-control, s6-man-pages, execline-man-pages and guides, amongst various other things), and not using mold with gcc 11 is no problem for me at this point. So i've just noted the issue with the patch on the wiki page, and the patch will have to be updated by someone for whom it's more important.

Fair point. Still, as a FOSS dev myself, i feel that the title might serve as a useful reminder to a number of their readers. Particularly when so many people use FOSS developed by large corporations, and see themselves as 'customers' entitled to 'responsive service'[a] that those of us who aren't a corporation aren't necessarily in a position to provide (even if we do try to provide 'best effort'). The contents of the screencap for one of my FOSS projects is not random. :-)

Based on your post headline, i was going to suggest my Ebuku package, an Emacs frontend to the buku Web bookmark manager - i use it myself, and it has support for both tags and comments on links. But it doesn't meet your requirement of capturing from the browser; i copy-and-paste the URL from the browser into an Emacs prompt.

Me asking for actual data about the the extent to which various arguments to the -O flag is me "trying to argue with you" and "having a bad day"? Er, what? i'm saying your assertions might well be correct, but as a dev myself (here's some of the stuff i've been doing during our exchange), i want some concrete data in support of this, because "premature optimisation is the root of all evil" (cf. "Rob Pike's 5 Rules of Programming").

Not beyond playing around a bit with Squeak. i like Smalltalk's message-passing approach to OO, and i like how thoroughly modifiable the environment is. i haven't spent enough time to know how easily it can interface with the system on which the image is running; one of the things i like about Emacs is that, in addition to being so modifiable itself, it provides lots of support for interacting with software outside of itself. (E.g. i wrote an Emacs package which provides an Emacs UI for the CLI-based buku Web bookmark manager: Ebuku.)

Finally, as someone who has put, and continues to put, a lot of volunteer effort into providing quality documentation (e.g. my work thus far on the Gentoo wiki, my work on the Void docs, my work porting the documentation for the s6 ecosystem), i can tell you it's frustrating and discouraging to put in all this work to provide helpful documentation to others - rather than having to write the same things in discussion threads again and again - and to continually improve it, only for people to ignore it in favour of some random (and often inaccurate) tutorial. So personally, i'm fine with the fact that, since the OP's question quickly got succinctly answered, people here subsequently wanted to (non-condescendingly) make sure the OP primarily used the Handbook.

Ultimately, my perspective is "let's try out a plethora of different approaches, and learn from experience what does and doesn't work in which situations". So i'd be all for a 'MiniHTML' effort (even if i don't end up getting involved myself), which i've seen many people suggest - but there doesn't seem to be any such effort that has gained momentum. Partly, i suspect, because of the common ICT phenomenon of "I want this, but someone else will have to work out the implementation details" (where the devils are). One personal example of this is of lots of people demanding man page versions of the s6 documentation, and lots of people talking about how this could be done, but no-one actually rolling up their sleeves - until i just got the bit between the teeth and did the work.

Well, indeed, i certainly wasn't expecting you to do so! But i already have more than enough volunteer FOSS stuff on my plate as it is (e.g. Ebuku, pulseaudio-control, s6-man-pages, execline-man-pages and guides, amongst various other things), and not using mold with gcc 11 is no problem for me at this point. So i've just noted the issue with the patch on the wiki page, and the patch will have to be updated by someone for whom it's more important.

About the man pages issue, Laurent Bercot (the creator of s6) already mentioned why he does not include man pages with s6 (here)[https://skarnet.org/cgi-bin/archive.cgi?2:mss:2583:202008:liednclklokjhnokhdhg]. You can find s6 man pages (here)[https://github.com/flexibeast/s6-man-pages] (those are made by flexibeast, using the online s6 documentation).