JdeBP
Joined 12 years ago
Joined 12 years ago
Then you need to learn a lot more about control sequences.
You can start with CSI, value 0x009B, not being an ASCII character. Then proceed to ITU T.416 (ISO 8613-6) where the SGR control sequences for indexed and direct RGB colours are standardized. Then you can learn that the "E" in ECMA-48 does not stand for "ANSI", but for "European". And finally you can discover that ECMA-48 standardizes exactly 8 colours. 16 colour SGR sequences beloved of CGA-like and VGA-like art are not in the standards and are in fact the product of IBM's AIXterm.
Even if one took it to mean using ECMA 35 and ECMA 48 escape and control sequences, which wholly forgets the likes of Avatar and PCBoard, "ANSI Art" usually did not actually use any ECMA 35 character set switching, instead relying upon SAUCE data and suchlike and using IBM's (not ANSI's nor ECMA's) rather different code page system and 8-bit character sets, and usually furthermore misinterpreted ECMA 48 SGR sequences when it came to blinking, bright, faint, and bold.
If people are preferring "XTerm Art" or "Terminal Art" for what is made nowadays, that seems to be a nomenclature change for the better.
That said: please do not become yet another abuser of the 'xterm' terminal type ('xterm-16color' in this case). Thomas Dickey has a lot to say on the subject of things that pretend to be XTerm and are not. Windows NT has first-class terminal types of its own, and has had for about 30 years at this point. Pick whether you're going to be cloning the old CSRSS or the new Microsoft Terminal, or even something else like the old POSIX subsystem or MS/PC/DR-DOS.
* https://invisible-island.net/ncurses/terminfo.src.html#tic-i...
* https://invisible-island.net/ncurses/terminfo.src.html#tic-m...
* https://invisible-island.net/ncurses/terminfo.src.html#tic-w...
* https://invisible-island.net/ncurses/terminfo.src.html#toc-_...
* https://invisible-island.net/ncurses/terminfo.src.html#toc-_...
* https://invisible-island.net/ncurses/terminfo.src.html#toc-_...
* https://invisible-island.net/ncurses/terminfo.src.html#toc-_...
Reading that chapter, it turns out that it does. Maybe it should stress the point more, but the actual doco is the likes of these:
* https://cdn.netbsd.org/pub/NetBSD/NetBSD-11.0/evbarm-aarch64...
* https://cdn.netbsd.org/pub/NetBSD/NetBSD-11.0/i386/INSTALL.h...
* https://cdn.netbsd.org/pub/NetBSD/NetBSD-11.0/amd64/INSTALL....
* https://cdn.netbsd.org/pub/NetBSD/NetBSD-11.0/amiga/INSTALL....
* https://cdn.netbsd.org/pub/NetBSD/NetBSD-11.0/sparc64/INSTAL...
… and so forth. Plus the manual pages for postinstall(8) and the like.
* https://gnats.netbsd.org/60897
I think that it is specifically the interaction between ihidev and ACPI. The hardware devices themselves are PS/2, with a vendor-specific extension to PS/2 activated by a special handshake, plus I2C. (On machines from 2024 that come with 9 USB root hubs; such is the inertia of laptop design.)
(The legal frameworks that allowed computer systems to be exempt from copyright restrictions, if they were just making copies in the ordinary course of transmitting things around computer networks, did not exist until the turn of the 21st century. Under U.K. law, for example, the way that the WWW worked was technically a massive violation of copyright until The Copyright and Related Rights Regulations 2003 came along.)
The law was not different. People's attitudes were.
This is entirely a packaging choice. There is no reason that parallel cannot be in its own package, or renamed. In fact, do not believe the old Hacker News discussions where people point to a bug filed in 2010 being closed as 'WONTFIX' by Joey Hess. In the intervening years, Debian has instituted a wild system where it does rename the moreutils parallel, but only if GNU Parallel is also installed. Because Debian made its GNU Parallel package fix the name clash.
* https://sources.debian.org/src/parallel/20240222%2Bds-2/debi...
There is nowadays an open bug report (against moreutils rather than parallel) on this from a user's perspective:
* https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1107381
In contrast, Arch Linux simply changed its packaging to eliminate the name clash back in 2012, by fixing moreutils and not parallel, making it always the parallel-moreutils command.
* https://gitlab.archlinux.org/archlinux/packaging/packages/mo...
* https://news.ycombinator.com/item?id=44916853
The real challenge is not finding another channel for transmitting personal information. It is finding something else for the FINGER protocol to do. (-:
(Back when the headline was "Arc is Sunsetting". I am so 2 minutes ago.)