[aur-requests] [PRQ#21898] Orphan Request for sge
daimh [1] filed an orphan request for sge [2]: Here is the Orphaning request again. If this Orphaning request is denied again, please consider reason 4 below in terms of Copyright. I won't submit any request as long as reason 4 is taken care of. ### Here is some new reasons since my previous request was denied. 1. PKGBUILD file fails. 1. Current maintainer releases package while knowing it fails 2. Current maintainer gave advices without any testing. Two examples, - He said "after setting '-j' in /etc/makepkg.conf, the building process is already greatly accelerated on a multi-core machine.". However, this fails too, see the comment I made at 2020-10-19 14:28 . - He also said it was caused by fakeroot, but the '-j' error is caused by SGE itself instead. -lshced is SGE's own library, and it has nothing to do with OS environment. 3. Current patching style is a hassle. Current maintainer split my single source code patch file into many small files. While this splitting is already an issue because he did so without any acknowledgement of my work, other Linux distribution administrator will have trouble to use this SGE patch. It used to be one patch command, but now they have to run it many times and patch the files in the deep sub-directories one by one. Further, what if more files need be patched in future? 4. Copyright. The one single C patch file in my last commit needs to have the same COPYRIGHT disclaimer as the one in https://github.com/daimh/sge . This was not an issue when I was maintainer as I can always respond to user. Or please acknowledge it in a file in the top directory. I spent quite some time to make SGE work under Arch Linux. My work should be acknowledged. If we programmers don't respect ourselves, nobody else in the world will do so, and plagiarism will be everywhere.
Developed by Manhong Dai
Copyright © 2020 University of Michigan. License GPLv3+: GNU GPL version 3 or later"
5. I am more than qualified to maitain this packge, reasons are listed at https://github.com/daimh/sge ### Here are some known facts. 1. I am the original AUR SGE submitter and have been maintaing this package since 2019. the PKGBUILD file works just fine on all my cluster nodes in University of Michigan for the past decade. 1. I modified the SGE code and put all my modification in one patch file. Otherwise the compiling will fail under modern Linux. 1. On Aug 10, current maintaner commented and asked me to pull his work. I checked it but didn't reply for not wanting being negative. I also checked that he is not a Trusted User. 1. At 12:24, Oct 10, I received an email saying this package was disowned. 1. At 12:43, Oct 10, I received another email saying this package was adopted. Before these two emails, I didn't receive any Orphaning email notifications. 1. At around 1pm, Oct 10, I submited the first Orphaning request 1. User Archange (archange AT archlinux.org) confirmed that I didn't receive the Orphaning requests because of a bug at https://gitlab.archlinux.org/archlinux/aurweb/-/merge_requests/6 1. I registered AUR-general in Oct 12, 2020, as I never know such email list exists. 1. I top-posted quite a few emails to the list. I am sorry for the trouble it caused, but I really didn't know the rule. This mistake won't happen again. 1. I accidentatlly removed one comment while tring to modify makepkg output to make it cleaner. That comment is just makepkg output. ### Anecdotal evidence, all quoted from the comments of current maintainer unless staed otherwise. 1. 2020-08-10 14:45, "the PKGBUILD is at a very bad state", Current maintainer's first comment when I was maintainer. 1. 2020-10-13 02:59 "In terms of ability and obligation, I do not agree that you can maintain this AUR package" 2. 2020-10-13 02:59, "It’s really intolerable for pkgver to fill in this incorrectly", according to Google translation. 2020-10-13 06:41, After the package was taken over, User a821 commented "The pkgver used to be 20200527 but now is 8.1.9. However 20200527 > 8.1.9 so epoch=1 might be required." 4. 2020-10-17 16:12, "a one-big patch file obviously make it hard to check the changes and follow this package on other distributions either" 5. 2020-10-17 17:08, "after setting '-j' in /etc/makepkg.conf, the building process is already greatly accelerated on a multi-core machine. Further small acceleration by changing the documented compilation method may not worth it." However, a simple test with '-j' fails 6. 2020-10-17 17:19, "As the maintainer of sge on AUR, I do have the authority to recommend you that please don't use 'sge' as the name, the short name or part of name for 'some grid engine' on AUR." 7. 2020-10-17 19:51, "I know that there is no error in build() of this PKGBUILD in a clean chroot". When he recommended '-j' and I told him '-j' fails. 8. 2020-10-17 19:51, "Besides, I know that there is an error in package() when building in clean chroot. I'll fixed that a few days later." 9. Many of his comments don't exist anymore. Or we will see quite some comments that I was called 'zero knolwdge', 'know nothing', etc.. All my comments are there except the one I mentioned above. If my words are offensive, let me know and I am more than happy to fix it. [1] https://aur.archlinux.org/account/daimh/ [2] https://aur.archlinux.org/pkgbase/sge/
Request #21898 has been rejected by Foxboron [1]: Not interested. The sge package is not going to point at your fork. [1] https://aur.archlinux.org/account/Foxboron/
On Sat, Oct 24, 2020 at 6:02 AM <notify@aur.archlinux.org> wrote:
Request #21898 has been rejected by Foxboron [1]:
Not interested. The sge package is not going to point at your fork.
Thanks a lot for the reply! But I never ask the current maintainer to point to my fork. I guess those markdown syntax messed it up. This package is in clear text-book copyright violation right now! The current maintainer is using my original SGE source code patch file without mentioning I am the patch's author and University of Michigan's copyright. Further, he split it into many patch files, now those modifications look to be his contribution as nobody will dig into git history, not to mention patching with many files is the worst way to promote a software. I believe I am the first one in the world who made SGE work under the latest SSL and GLIBC. I created the single source code patch file, and initially put it on AUR sge. I also tried to settle with the current maintainer, I proposed that he can keep the package but he has to use my single patch file with the copyright and author on the top, then remove my name from the PKGBUILD file. He rejected me. Now here are all the facts, could you please reconsider it again? 1, I am the original author of the SGE source code patch file, I put my source code patch file on AUR sge. 2. I didn't receive any email notification before the package was taken over, This is confirmed by a TU 3, I didn't join AUR-general before the package was taken over, and the same TU confirmed that this is not required. 4, I tried to settle with the current maintainer, and asked him to respect Copyright and original author, he rejected me. 5, The current maintainer's PKGBUILD doesn't work, and he 'git push' three times anyway, while knowing it fails. It took him 7 days to fix a 'cd' error. Everything above are true facts, and it can be verified in court, and I am willing to take any penalty if they are not fact. Now, the simple question is, can any original work put on AUR be taken away like this? Best, Manhong
On Sat, Oct 24, 2020, 7:33 AM Manhong Dai <daimh@umich.edu> wrote:
On Sat, Oct 24, 2020 at 6:02 AM <notify@aur.archlinux.org> wrote:
Request #21898 has been rejected by Foxboron [1]:
Not interested. The sge package is not going to point at your fork.
Thanks a lot for the reply! But I never ask the current maintainer to point to my fork. I guess those markdown syntax messed it up.
This package is in clear text-book copyright violation right now! The current maintainer is using my original SGE source code patch file without mentioning I am the patch's author and University of Michigan's copyright. Further, he split it into many patch files, now those modifications look to be his contribution as nobody will dig into git history, not to mention patching with many files is the worst way to promote a software.
I believe I am the first one in the world who made SGE work under the latest SSL and GLIBC. I created the single source code patch file, and initially put it on AUR sge.
I also tried to settle with the current maintainer, I proposed that he can keep the package but he has to use my single patch file with the copyright and author on the top, then remove my name from the PKGBUILD file. He rejected me.
Now here are all the facts, could you please reconsider it again?
1, I am the original author of the SGE source code patch file, I put my source code patch file on AUR sge. 2. I didn't receive any email notification before the package was taken over, This is confirmed by a TU 3, I didn't join AUR-general before the package was taken over, and the same TU confirmed that this is not required. 4, I tried to settle with the current maintainer, and asked him to respect Copyright and original author, he rejected me. 5, The current maintainer's PKGBUILD doesn't work, and he 'git push' three times anyway, while knowing it fails. It took him 7 days to fix a 'cd' error.
Everything above are true facts, and it can be verified in court, and I am willing to take any penalty if they are not fact.
Now, the simple question is, can any original work put on AUR be taken away like this?
Best, Manhong
Request #22028 has been rejected by Foxboron [1]: Patches has been removed. [1] https://aur.archlinux.org/account/Foxboron/ Thanks! @Foxboron. it is all good as long as my original work is not involved now. As programmers, I always respect and acknowledge other people's work and expect the same from others. Truly appreciate those who helped me, including a few users who forwarded me the emails on this list. Or I probably will never know such an email list exists. So sorry I cannot mention your names as I just got 1000+ mail list subscription confirmation in one day. I also want to say sorry to those who feel these are spams. The thing is this kind of issues will happen from time to time as AUR doesn't associate a project with user name. I hope a little waste of your time will set a good AUR standard from now on. This is a good lesson for me too, as I will put copyright/license on all my work, no matter it is MIT, GPL or BSD, and should have published my work originally on github. SGE will be continued on my github project as I will make it more Linux friendly and totally ditch some unnecessary features, such as Windows support, whenever I find nothing better to do. As I added cmake compiling support recently, followup development will be much easier. Have a good day! Feel free to send me email directly at daimh@umich.edu. Best, Manhong PS, If someone wants to subscribe me on another 1000+ email lists again, please find some truly interesting ones. My hobbies are hunting, fishing and racing. Thanks in advance!
participants (2)
-
Manhong Dai
-
notify＠aur.archlinux.org