4 - add, remove, list jobs by dedicated commands: `qadd`, `qrm`, `qls`
5 - user decides when and where to run the next job by `qrun`
6 - may put `qrun` in scheduler (cron) or at the end of each submitted jobs;
7 making the queue "automatically running" this way
9 - default queue control dir is per-user
10 - record command, environment, PID, exit status, etc. in separate files
11 - pending jobs are files in the queue control dir
12 - leave log management to each job
13 - queue IDs are sequential numbers
15 # Similar tools/projects
17 - [at](https://pubs.opengroup.org/onlinepubs/9699919799/utilities/at.html)
20 - run jobs at a given time
21 - run as a daemon (atd)
22 - queue control dir is system-wide, multiuser
23 - queue IDs are sequential numbers
24 - [nq](https://github.com/leahneukirchen/nq)
26 - start jobs right away
28 - default queue dir is the CWD
29 - record command, output, and exit status in a single file
30 - list jobs by `ls -F ,*`
31 - pending jobs are processes waiting for a lock
32 - queue IDs are timestamps+PID
33 - [task spooler](https://vicerveza.homeunix.net/~viric/soft/ts/)
34 - per-user daemon, start on-demand
35 - [pueue](https://github.com/Nukesor/pueue)