Is Fedora's KDE Spin Too Bloated?

[spstarr]

Maybe that is part of the reason why GNOME has Code of Conduct now.

It helps, even if I hate Code of Conducts myself - when abused -

[TheLexMachine]

I can't speak as to whether or not Fedora KDE is bloated, but from my own personal experience with Kubuntu, I can tell you that KDE has a lot of broken and outdated software in it's portfolio, installed by default. That is definitely a problem and it contributes to bloated installs, especially if you go with the installation of the big ole KDE Applications package during the install. Is that still a thing? I don't know if it is, cause I moved to MATE a while back.

[rtfazeberdee]

Can we drop the psychology analysis and get back to the software? Thanks.

So how is it justified to keep two stacks when there is only the need for one? What we got today is a test case explosion that hurt both stacks.

I thought it might help your growth towards adulthood. Speak to Gnome and ask them why they unnecessarily duplicated stacks as they are the "johnny come lately"

[GreatEmerald]

Or someone needs to fix the KDE startup. It has become noticably slower. I think it is simply blocking the screen until all services has reported back to have started. It could probably stop blocking the desktop long before, because when it restores browsers on youtube, I can hear they have already launched and are playing videos several seconds before the splash screen lets me see anything.

Edit: And probably at least one of those services is also broken as it shouldn't take that long anyway.

Yea, that was already planned a long time ago: the point is to abolish the hacky startkde script and replace it with systemd --user units. Unfortunately, the progress on that seems to have stagnated: https://cgit.kde.org/plasma-systemd-integration.git/

[bug77]

I can't speak as to whether or not Fedora KDE is bloated, but from my own personal experience with Kubuntu, I can tell you that KDE has a lot of broken and outdated software in it's portfolio, installed by default. That is definitely a problem and it contributes to bloated installs, especially if you go with the installation of the big ole KDE Applications package during the install. Is that still a thing? I don't know if it is, cause I moved to MATE a while back.

I don't know about broken software, but outdated you should expect if you stick with Ubuntu (and family). Ubuntu is not even trying to be cutting edge, they're trying to be stable and traditionally do not ship with the latest of pretty much anything.
Kubuntu's problem lately was that staying 2-3 versions behind KDE meant staying with many unfixed bugs. This has improved since Neon (I think Kubuntu somehow leverages Neon's work), but Kubuntu still sits on an older Qt. Meanwhile Neon sticks with Ubuntu LTS, so no newer kernel. Take your pick

[gbcox]

80% of downloads is proper Fedora Workstation. That pretty much sums up how people go with The Standard Desktop.

Fedora kde is a minority of a minority.

You win the award for non sequitur of the day. BTW, what is the source of your data. AFAIK Fedora has no such metrics - and even if they did it wouldn't be accurate to correlate Workstation downloads to DE usage. There is no Fedora KDE Workstation download - Workstation is GNOME only - and what many people do is install their DE after the fact or have anaconda do it. Also, it doesn't account for upgrades.

It would be nice if there was somehow a metric based upon hits to the repository for a particular package, but I'm not aware of one - but even if there was one that would also not be accurate due to the way Fedora bundles GNOME into just about everything. Due to requisites it's a PITA to remove GNOME once it is installed. Myself I have GNOME installed along with KDE because in Fedora that's the path of least resistance. I just don't use it - but a metric would claim it was installed and being continuously upgraded. I suppose you could subtract the number of other DEs from the total GNOME number - making the assumption that in Fedora GNOME is always installed, no matter what - since most people don't have the wherewithal to delete it.

[gbcox]

I see your denials and non-sequitur hand waving. I raise you FACTS.
https://eischmann.wordpress.com/2017...ed-flock-2017/

There is a good correlation between betting on The Standard Desktop and marketshare growth.

You need to re-read my message above and look up the definition of non sequitur. Hint - the subject of this is the fedora kde spin being bloated - not how many people use different DEs. Second, the link you sent doesn't include the workstation stats you quoted. It just says: "It’s a pity that Matthew didn’t include the slide with ISO download shares of Fedora editions and spins. But last time he did Fedora Workstation amounted to ~80 % of all ISO downloads." - No mention of what last time was - and again, you can't do a correlation between Workstation downloads and DE usage due to the factors I mentioned above. If you do have something that would, that would be an interesting different topic, but doesn't really have anything to do with the subject at hand - which was someone complaining because they weren't happy with the choice of default applications.

I did find some stats from arch which were interesting. It would be interesting if something like this could be replicated in Fedora, but again, due to the issues I outlined above, the accuracy would be somewhat problematic. https://www.archlinux.de/statistics/fun

[aht0]

At least in one point of yours everyone who knows english could do something about it, translating what's not translated. In past i would add "easy" to such task, but it's everything but easy..., english is simple and efficient language, immagine someone who needs to translate new term on, I don't know, some language that have 12+cases and gendered words (like some baltic languages, I think estonian have 12 cases or something, or finish), now that's nightmare, often, I find it easier to "invent the term" on english than on native language, especially words that combine two terms into one. So if you see something untranslated, do it yourself, and if you come up with something you can't translate, leave it blank and there might be someone else who could do it better.

Estonian has 14 cases, no genders at all and also lacks certain time forms (no future for example). You would pass necessary information in the context. Plus bunch of other oddities.

Problem is in the different system of grammar, in need to keep translations as short , concise and precise as possible - all of it without deviating too far from the original. It's something like translating poetry, difficulty-wise.

Quite maddening while you are trying to translate computer UIs.