Someone I met on Twitter is working on an open source distro-picking website/utility; seems like something the Level1 crowd could be interested in checking out.
The code is on GitLab:
Very basic and I got “no matches” several times.
It sounds cool but I think distrowatch’s top 100 list already contains all the distros a new user might care about.
“free software certification” description isn’t 100% accurate.
“Timely updates” is wrong when checked against there db. Timely updates is marked as “LTS” but described it as “once in 6 months”. Fedora fits once in 6 months (distro update 6 months, app updates slightly faster), but isn’t marked LTS.
Extreme security is marked as 0 (not) on Fedora but in comparison to most other distros fedora has a much higher set of security and would imo come under this description as it is described.
Technical expertise seems to reflect the amount of effort required to get a distro working, not the ability of a technical expert to get things done. For example Arch is marked as Advanced for linux expertise, but Fedora is marked as Medium. so it will discount distros just because you know linux well.
As like the other site that has a similar distro picker, imo this site is shiny but has zero reliability on categorising and matching distros.
It’s pointless until some serious research and effort is put into the back end.
2 Likes
Also, If I answer all the questions, it should at least provide something as an answer. The current default does not suffice.
Instead, I feel like it has cherry-picked some magic numbers without doing a proper wideband-Delphi calculation to determine technical expertise required for a distro.
You can check the db. Technical expertise for a distro for example is 0,1,2.
Distros that say they are for expert linux users (2) are
- parabola
- Guix
- Kali
- Tails
- Qubes
- OpenSUSE Leap
- Arch
- Gentoo
- OpenSUSE Tumbleweed
Gentoo for example doesn’t have 1 for security but its probably the only distro which can be locked down the most.
1 Like
Arguably, a new user might not know what to look for in comparing the Distrowatch pages.
@Dynamic_Gravity
We just talked on Twitter about some changes that should solve the no-results problem.
This is just his personal project, so the values are just magic numbers to start with; I’ve asked him to consider joining the forum, since it sounds like it would be a good fit for the Blog category.
2 Likes
It’s something worth working on. The other distro picker suffers from a similar problem. theres two main parts to consider i think (someone mention any more if there are).
- there needs to be a good set of categorys with good definitions for what they are so the dev can research and submit the distros into the correct categories.
- there needs to be good language used for the end user so that categories dont end up marked false-positives when they use the site.
the expert user question for example, while it makes sense if you put it as are you experienced with linux and can use distros requiring advanced skills, its more likely to be taken as are you an experienced linux user with no bearing on actual distro choice, and its phrased as the latter to the end user while being used as the former for picking a distro.
im an experienced linux user and would never use arch for example. But being an experienced user im more likley to use centos/rhel/debian because i use them for projects. i use fedora to get work done on my personal computers (done with distro hopping because ive done it for 15 years).
Id probably say that experienced linux users, who wand to get work done and have their distro out of the way for the most part are probably more inclined to fedora/debian/suse/etc. where as experience linux users, who enjoy and want to get into the gears of the system day to day would prefer arch/gentoo/funtoo
1 Like