| golinux | @jaromil . . . Ahoy! | 02:54 |
|---|---|---|
| joerg | gnu_srs: also untested, but a info quote >> If the '--from-file=FILE' option is given, the number of file names is arbitrary, and FILE is compared to each named file. << | 17:09 |
| joerg | i.e. aiui the trick is to explicitly use --from-file=/dev/null | 17:10 |
| joerg | manpage suggests similar: --from-file=FILE1 compare FILE1 to all operands; FILE1 can be a directory | 17:19 |
| * joerg idly wonders if there's a difference in diff's output generated, between from-file not existing in an existing from-dir, and from-file is (/dev/)NULL content | 17:28 | |
| mason | joerg: Tell me if I'm misunderstanding the question: https://bpa.st/YRLA | 17:32 |
| mason | Which is to say, I think the thing you're using as a base has to exist. The /dev/null thing is a funny idiom to say "I want to use patch to introduce a whole new file" generally. | 17:33 |
| mason | s/patch/patch(1)/ | 17:34 |
| golinux | This post is intended to provide some historical context about when and how systemd was wormed into Debian for those here who came after the fact. It is an analysis of the coup de grace that irreparably fractured the Debian community. Please have a look at: | 17:43 |
| golinux | Combatting revisionist history by dasein ยป 2015-02-25 20:23: https://forums.debian.net/viewtopic.php?t=120652 | 17:43 |
| mason | golinux: I absolutely love that. Very few arguments include the fact that they're trying to replace a lot of good work other folks have done over decades, and this article *leads* with that detail. | 17:48 |
| mason | Finished it, and the comments. Good piece. The only thing I'd want to add would be a note about harm to the free software ecosystem arising from a very small, insular group of developers trying to control more and more of our systems, thus leaving less and less opportunity for new free software development, not least because of the strong anti-standards stance, where standards are how competing | 17:54 |
| mason | implementations historically have met. Throw out the standards, and it dramatically increases the pain to swap out any particular part. But anyway. | 17:54 |
| golinux | mason: That was written nearly 10 years ago! A lot has come into focus since then that was not that obvious back then. | 18:12 |
| golinux | Like the MS connection . . . | 18:13 |
| golinux | As they say . . . hindsight is 2020. Dasein certainly put the pieces that were available at the time together in a masterful way! | 18:14 |
| mason | Agreed. | 18:29 |
| gnu_srs | Hi again, I've bee looking on how many packages in need to fork in Devuan not to be forced to install usrmerge. I volunteer. WDYT? | 18:44 |
| mason | gnu_srs: It's going to be hard to gauge. Upstream Debian packages are relocating things into /usr, but the trick will be understanding what's looking in the wrong place for a resource that's only available in one place. | 18:48 |
| mason | If all of upstream relocates into /usr then there's no further need for usrmerge. It's the interim state that's painful. | 18:48 |
| mason | (I'm not speaking to whether or not UsrMerge is a good idea, here, just the tactics of what ends up mattering.) | 18:49 |
| gnu_srs | I've found a limited number of packages to fork. Most of them move systemd files /usr/*, which does not matter to us :) | 18:56 |
| gnu_srs | I think we easily can monitor the move of files to /usr: some we have to move back (fork), others don't matter if they move to /usr. | 18:57 |
| DelTomix | gnu_srs: IF it were feasible (I don't believe it is) and IF Debian completes merge with a stable release, what would be the benefit? Devuan would then permanently add continual burden of keeping its forks functional on a major additional criteria, and an extra point of potential breakage to look after. | 19:13 |
| fsr | $ ls -1 /bin/ | wc -l | 19:17 |
| fsr | 154 | 19:17 |
| fsr | $ ls -1 /sbin/ | wc -l | 19:18 |
| fsr | 235 | 19:18 |
| gnu_srs | fsr: I don't agree with you. We adapt to their movements to /usr of course, and in due time we will not have to fork packages when they release. | 22:19 |
| gnu_srs | And still no usrmerge will be installed. Remember we also have dpkg-fsys-usrunmess :) | 22:19 |
| gnu_srs | LeePen: See #devuan | 22:22 |
| fsr_ | gnu_srs, I guess I don't understand what you are suggesting. I thought you meant we should un-do every package that moves its programs from /bin to /usr/bin or /sbin to /usr/sbin. | 23:26 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!