Looking at distrowatch site we get a vague feeling and some insight of how relative popularity are changing. As to what direction things are moving. However it does not really seem to give such good metrics beyond that.
But what breaks the experience for me on a distro is broken pkg dependancies in the pkging system. So this is why I am on ubuntu. And switched from RHEL to ubuntu for that reason.
Now that arch* based distros have come along they are more modern and have what i would deem to be a more advanced packaging system. Pkgbuilds being like FreeBSD ports system (or gentoo?). It is easier to work with and less overhead than debians pkging system.
However I have realized that actual stability and reliability of pkgs themselves is not ideally where it could be. At least not yet. So it would make sense to want to get some monitoring going over time and actually have some hard metrics on broken pkgs or dependancy conflicts. Which should be shared and gathered online.
This way over a longer term period of several month or years. It would be possible to track as trends for a specific arch based distro (or rather their pkging repos). Be it upstream arch or derivatives like manjaro, or whatever else.
Because really it is a bit of a deal breaker for me otherwise. At least compared to ubuntu which has been almost entirely painless. So being able to reduce those pain points and minimize the time overhead spent fixing things etc. Really it goes a long way.
I suppose this is also might be a reason why MX Linux ha recently become more popular on distrowatch than arch ones. Being debian based. Also i talk to people and a lot of the ones who try arch end up switching back to debian based distros (whichever arch distro and whichever debian based distro). Its always the same reason cited, their pkgs got screwed up and they could not recover from that situation.
Problem is i dont know arch. SSo how to get better visibility on this online? And to do so without installing a specific arch distro myself. Since that only gives visibility on 1 version of arch, not the other ones. Because the ultimate goal is to be sure to choose between and select those most reliable arch distro(s). And not randomly try my luck then get some bad experience. Which then deters me away from all of the other arch more generally speaking.
And no - anectodal comments don’t necessarily help. When that individual user experience is limited to only their individual specific subset of pkgs. So unless you are really a long time user and installs everything under the sun. It isnt going to help much this quest. It is not a substitute for a more general metrics. However perhaps the google trend / sentiment analysis COULD be a useful metric. When searching for a carefully chosen set of keyword search terms. And then comparing distro flavors that way on bug reports / issue trackers. But it would also have to take into accound the relative popularity and size of the respective installed user bases. Since specific distros has relatively different number of users than other ones. Which can skew such results
But what do you guys think?
Is it a worthy quest to embark on? And how difficult will it prove to be to gather a useful and reliable set of metrics? Will they be different enough to show a distinct difference between those targeted distros, or tracking over some longer timeframe? And can we also use the debian distros like ubuntu as a kind of baseline reference point?
Necause talking casually with other people who have been trying out distros… I know I am not the only person who has some level of concerns about this. And perhaps even the more seasoned arch or manjaro users. They still need to spend some time managing pkg updates and fixing things manually. It seems to be an overhead that is expected with arch. But how much of an overhead? Without knowing that well enough it makes life harder. But especially to manage expectations about the experience and not to be surprised and then find yourself feeling let down. Wheras with a better picture you can at least feel more prepared and then it becomes less of a pain point.
Really hoping the general stability and reliability of pkgs on arch* family will continue to improve. I just also want to point out it might plateau or not increase forever due to other reasons like the number of distro maintainers… the policies or proceedure for revieiwing and fixing pkgs. Whether they are core or aur. Theres a whole bunch of factors at play. And I just want to see and check that things keep going in the right direction. Hopefully to eventually become as rock solid as ubuntu feels these days. Lets not forget ubuntu was once really shaky and bad. It was not even so many years ago! Like 14.04 and before. Ubuntu only became really great around 16.04+ in my memory. Im sure others will want to contest that, it was just my own individual anecdotal experience based on the specific sets of pkgs i needed for my own usage. Anyhow I think you get my point. Lots of time for arch to keep improving over time. I just want to know when I can finally switch over! Hehe.
