Bug#986709: rsnapshot is stable, not dead
Dear maintainer, could you please give some hints, why you actually think the package is unmaintainable or whre we can find information about this? This would be usefull for everyone considering to adopt it.
Bug#986709: rsnapshot is stable, not dead
Hi Michael, On Fri, 1 Oct 2021 10:29:27 -0500 Michael Lustfield wrote: > [ moving back to rsnapshot ] > > > [...] > > Debian package. The only bug of "serious" severity classification is > > this one. But when my uninformed assessment is at odds with an actual > > Debian maintainer, I have no choice but to assume that there is an > > important factor which I am blind to.<< > > There are definitely options; I'm just one person with an opinion. It's > entirely possible all of my previous reasoning has been permanently fixed and > I'm just too jaded to see that. If such a scenario were to be our present case, > then it would be very easy for someone else to just hop in, grab this, and > maintain (own) it indefinitely (... or until such time it must be retired). > > ^ This could be you, anyone that commented on this thread, etc. > > If, however, my $super_notsosecret reasoning still holds water, > then... that won't be so easy and it becomes a self-solving problem. > > >>I understand that it's not your > > responsibility to teach me just to satisfy my idle curiosity, so we can > > leave it at that. > > It's actually very difficult for me to not launch into a long-winded rant, so > thank-you for prompting me to provide this additional explanation. I heard of this issue around rsnapshot in Debian in recent months from various information sources. While I completely understand your opinion, this looks like another unexpected consequence due to Debian's strong package maintenance ownership. I am not against your decision, but I am wondering if the following actions would work for you: 1) Package the latest rsnapshot release 1.4.4 as-is, but still keep this RC bug open since it is not considered suitable for Stable release, or 2) Orphan package rsnapshot since you find this software not maintainable, or 3) Remove it from Debian archive as you originally planned. My personal thought is that some actions would be better than getting stuck here, and I am also interested in the next step. At least I believe doing nothing does not fall into the category of package maintenance. Thanks, Boyuan Yang signature.asc Description: This is a digitally signed message part
Bug#986709: rsnapshot is stable, not dead
Note: This is a general response, not meant to address rsnapshot specifically. On Thu, 30 Sep 2021 19:24:43 -0400 John Brooks wrote: > [...] > And that's about where my ability to contribute usefully ends. My belief My offer to mentor prospective debian maintainers stands. I might not be the bestest teacher, but I can also teach people where to find smarter people to teach smarter things. ;) > that the Debian organization and its contributors are generally > intelligent and sensible leads me to believe that you and the QA team > have good reasons for removing the package, even if I don't understand them. If you want to continue believing this, I encourage you to avoid any open source development, especially WRT distributions. :P Seriously, though... we're all just humans driven by various motives. Although rare, changes like this /do/ sometimes come with malice. Other times it's best of intentions, and sometimes those intentions are flawed. https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=964139 https://www.theverge.com/2021/4/22/22398156/university-minnesota-linux-kernal-ban-research https://www.theregister.com/2021/06/16/debian_11/ https://arstechnica.com/information-technology/2015/05/debian-8-linuxs-most-reliable-distro-makes-its-biggest-change-since-1993/ ^ one of these clearly intends to do harm [ moving back to rsnapshot ] > [...] > Debian package. The only bug of "serious" severity classification is > this one. But when my uninformed assessment is at odds with an actual > Debian maintainer, I have no choice but to assume that there is an > important factor which I am blind to.<< There are definitely options; I'm just one person with an opinion. It's entirely possible all of my previous reasoning has been permanently fixed and I'm just too jaded to see that. If such a scenario were to be our present case, then it would be very easy for someone else to just hop in, grab this, and maintain (own) it indefinitely (... or until such time it must be retired). ^ This could be you, anyone that commented on this thread, etc. If, however, my $super_notsosecret reasoning still holds water, then... that won't be so easy and it becomes a self-solving problem. >>I understand that it's not your > responsibility to teach me just to satisfy my idle curiosity, so we can > leave it at that. It's actually very difficult for me to not launch into a long-winded rant, so thank-you for prompting me to provide this additional explanation. Cheers, -- Michael Lustfield
Bug#986709: rsnapshot is stable, not dead
On Fri, 1 Oct 2021 16:26:58 +0200 Dirk Heinrichs wrote: > [...] > Esp. when compared to dirvish (see my previous mail), which is > unmaintained for 16+ years, but still available in bullseye. What's the See my note about whataboutisms and strawman arguments ... and thanks for highlighting a perfect example.
Bug#986709: rsnapshot is stable, not dead
John Brooks wrote: > I don't know precisely what criteria of stability and quality are used > to judge whether a package is suitable for inclusion; my outside view > is that this package is no more broken or unmaintained than the > average Debian package. Esp. when compared to dirvish (see my previous mail), which is unmaintained for 16+ years, but still available in bullseye. What's the point in keeping that one while at the same time removing rsnapshot, which is unmaintained for just a handful of months now? Bye... Dirk -- Dirk Heinrichs Matrix-Adresse: @heini:chat.altum.de GPG Public Key: 80F1540E03A3968F3D79C382853C32C427B48049 Privacy Handbuch: https://www.privacy-handbuch.de OpenPGP_signature Description: OpenPGP digital signature
Bug#986709: rsnapshot is stable, not dead
On 2021-09-30 6:13 p.m., Michael Lustfield wrote: On Sun, 26 Sep 2021 13:49:36 -0400 John Brooks wrote: [...] So... My first response was a wordier version of the message you replied to, emphasizing the bit where my opinion is moot. What's written below is as much as I'm willing to dip back into #debiandrama. While reading, please remember this point (and don't expect further response). My original request was for a removal, which is a stance I whole-heartedly still stand by, and which draws from experiences after adopting the package. A removal like this is basically orphan++ ("I'm afk4eva" vs. "bad package"). That changed slightly with zeha's bug modifications, but the effect is still largely the same, with a touch of stability added. (Thanks zeha!) (sensible action, but likely helps with that "limbo" perception?) ^ https://tracker.debian.org/pkg/rsnapshot side note -- > Additionally, in response to this very bug, a new upstream release has > now been issued. In light of this, do you plan to upload the new version You very correctly point out that a number of fixes and a new release came directly in response to certain actions. Unfortunately, we draw very different conclusions. (a hint, perhaps?) I appreciate that you responded to that particular (#30) message of mine, where I say that I don't intend to stand in anyone's way, and offered to help anyone interested in package maintenance, while also maintaining my position. This is important to me because some people have indeed taken a stab at rsnapshot maintenance; however, they very quickly disappeared when they learned that it would require more effort than just slapping an updated tarball onto the packaging. and continue to fill the role of maintainer for the rsnapshot Debian package, or is another maintainer still needed going forward? ^ "continue" stopped at the RM-RoQA (note: this tag was not an accident) The root of why I claim how I feel does not matter is because the end result is the same. The only thing that's required to override my (strong) opinion is for someone to pick it up, understand it well enough to confidently claim it's ready for release (start w/ debian bugs), and that'll be the end of this thread. Thank you for your reply. I admit I'm rather a dilettante in this area. I'm only a user and have had little or no exposure to the Debian development process. I didn't even see "RoQA" until you pointed it out, and then had to look up what it means — "Requested by the QA team". And that's about where my ability to contribute usefully ends. My belief that the Debian organization and its contributors are generally intelligent and sensible leads me to believe that you and the QA team have good reasons for removing the package, even if I don't understand them. I don't know precisely what criteria of stability and quality are used to judge whether a package is suitable for inclusion; my outside view is that this package is no more broken or unmaintained than the average Debian package. The only bug of "serious" severity classification is this one. But when my uninformed assessment is at odds with an actual Debian maintainer, I have no choice but to assume that there is an important factor which I am blind to. I understand that it's not your responsibility to teach me just to satisfy my idle curiosity, so we can leave it at that. Thank you for your service. John Brooks
Bug#986709: rsnapshot is stable, not dead
On Sun, 26 Sep 2021 13:49:36 -0400 John Brooks wrote: > [...] > Michael, > > I think it is important that you clarify or modify your stance given > that upon further inspection by others here, there are no serious > outstanding functional or security issues with the program. Even > self-asserted justification (i.e. "I just don't want to maintain it > anymore, so find someone else") is acceptable; that is your right as a > volunteer. But it would have been prudent to either defend your initial > assessment of the program as no longer suitable for inclusion, or > acknowledge that you may have been incorrect. Otherwise the issue is > just stuck in limbo. > > Additionally, in response to this very bug, a new upstream release has > now been issued. In light of this, do you plan to upload the new version > and continue to fill the role of maintainer for the rsnapshot Debian > package, or is another maintainer still needed going forward? > > I don't seek to impose anything upon you, I just want to see that this > doesn't fall through the cracks. > > Thanks > John Brooks So... My first response was a wordier version of the message you replied to, emphasizing the bit where my opinion is moot. What's written below is as much as I'm willing to dip back into #debiandrama. While reading, please remember this point (and don't expect further response). My original request was for a removal, which is a stance I whole-heartedly still stand by, and which draws from experiences after adopting the package. A removal like this is basically orphan++ ("I'm afk4eva" vs. "bad package"). That changed slightly with zeha's bug modifications, but the effect is still largely the same, with a touch of stability added. (Thanks zeha!) (sensible action, but likely helps with that "limbo" perception?) ^ https://tracker.debian.org/pkg/rsnapshot side note -- > Additionally, in response to this very bug, a new upstream release has > now been issued. In light of this, do you plan to upload the new version You very correctly point out that a number of fixes and a new release came directly in response to certain actions. Unfortunately, we draw very different conclusions. (a hint, perhaps?) I appreciate that you responded to that particular (#30) message of mine, where I say that I don't intend to stand in anyone's way, and offered to help anyone interested in package maintenance, while also maintaining my position. This is important to me because some people have indeed taken a stab at rsnapshot maintenance; however, they very quickly disappeared when they learned that it would require more effort than just slapping an updated tarball onto the packaging. > and continue to fill the role of maintainer for the rsnapshot Debian > package, or is another maintainer still needed going forward? ^ "continue" stopped at the RM-RoQA (note: this tag was not an accident) The root of why I claim how I feel does not matter is because the end result is the same. The only thing that's required to override my (strong) opinion is for someone to pick it up, understand it well enough to confidently claim it's ready for release (start w/ debian bugs), and that'll be the end of this thread.
Bug#986709: rsnapshot is stable, not dead
On Fri, 28 May 2021 15:39:28 -0500 Michael Lustfield wrote: > On Fri, 28 May 2021 19:56:47 +0100 > David Cantrell wrote: > > > [...] > > So what, exactly, is unmaintained about it? Looks to me like it has > > exactly the amount of maintenance that is required for mature software. > > I'm not going to strawman my justifications; it's not terribly relevant anyway. > Absolutely anyone is free to disagree with me and continue maintenance of the > package. If needed, I'll even sponsor the upload. > > https://mentors.debian.net/intro-maintainers (read 1-2, start at 3) > > Michael, I think it is important that you clarify or modify your stance given that upon further inspection by others here, there are no serious outstanding functional or security issues with the program. Even self-asserted justification (i.e. "I just don't want to maintain it anymore, so find someone else") is acceptable; that is your right as a volunteer. But it would have been prudent to either defend your initial assessment of the program as no longer suitable for inclusion, or acknowledge that you may have been incorrect. Otherwise the issue is just stuck in limbo. Additionally, in response to this very bug, a new upstream release has now been issued. In light of this, do you plan to upload the new version and continue to fill the role of maintainer for the rsnapshot Debian package, or is another maintainer still needed going forward? I don't seek to impose anything upon you, I just want to see that this doesn't fall through the cracks. Thanks John Brooks
Bug#986709: rsnapshot is stable, not dead
On Fri, 28 May 2021 15:39:28 -0500 Michael Lustfield wrote: > On Fri, 28 May 2021 19:56:47 +0100 > David Cantrell wrote: > > > [...] > > So what, exactly, is unmaintained about it? Looks to me like it has > > exactly the amount of maintenance that is required for mature software. > > I'm not going to strawman my justifications; it's not terribly relevant anyway. > Absolutely anyone is free to disagree with me and continue maintenance of the > package. If needed, I'll even sponsor the upload. > > https://mentors.debian.net/intro-maintainers (read 1-2, start at 3) > > Hi Michael, I don't understand this. Wouldn't it be easier if you orhpaned the package since it's already in stable? Thanks, Sam.
Bug#986709: rsnapshot is stable, not dead
On Fri, 28 May 2021 19:56:47 +0100 David Cantrell wrote: > [...] > So what, exactly, is unmaintained about it? Looks to me like it has > exactly the amount of maintenance that is required for mature software. I'm not going to strawman my justifications; it's not terribly relevant anyway. Absolutely anyone is free to disagree with me and continue maintenance of the package. If needed, I'll even sponsor the upload. https://mentors.debian.net/intro-maintainers (read 1-2, start at 3)
Bug#986709: rsnapshot is stable, not dead
Of the 11 open PRs, several are marked as "needs tests", and all but three look like new features, not bugfixes. Two of the three bug fixes are for rsnapreport, a tool that to be blunt I don't think is very important. One is for LVM- and BTRFS- specific issues so is hard for people without a very specific configuration to test. There are three broad themes in the open tickets. 1. help requests, which ought to be on the mailing list. 2. feature requests 3. minor problems with argument parsingin some unusual situations, especially when args contain whitespace. The only one of those themes that is even slightly important is the third, and they are mostly unfixable without breaking existing working configurations. Finally, the most recent release is almost completely up-to-date with the master branch: https://github.com/rsnapshot/rsnapshot/compare/HEAD..1.4.3 So what, exactly, is unmaintained about it? Looks to me like it has exactly the amount of maintenance that is required for mature software. -- David Cantrell