Yep, just like i've said on upstream there - by continuing this way they will inevitably fail, it's just a matter of time and how hard it will be.
well you just havent looked outside .
Arch / Manjaro + Deepin == Square Peg + Round Hole.
Deepin is buggy enough without the additional complication of retrofitting it into the rolling Arch ecosystem.
Add the haphazard developement practices and the proposed "customer information experience" functionality and Manjaro should run away without looking back.
RIP Manjaro Deepin, I barely knew ye.
Prove me wrong....I don't see anything prettier
Beauty is in the eye of the beholder ...
I suspect the initial way forward is to have an addon/third-party repo e.g.
[manjaro-deepin] which tracks only the official DeepinDE releases, i.e. not the Arch package releases.
This would require someone to step up and maintain those packages, rebuilding as necessary for e.g. new Qt5 versions.
If this approach works and produces a stable DE then those packages can be included in the main Manjaro repos.
That would make total sense, because at least it would be as stable as original Linux Deepin (which ain't perfect, yet still better than Arch).
+ Also getting rid of Customer improvement program...
+ Fixing GTK theme for open/save dialog on Arch / Manjaro https://github.com/linuxdeepin/developer-center/issues/613
Very true but I really like the look of the control panel haha
Well, seems like these things happen again and again when a company-driven open source project hits the open-source community...
I'm mitigated. Personally, I would have dropped support for Deepin a long time ago considering how an hassle it has become (especially for oberon, poor him). It won't be such a big loss IMO since there is some other DEs that has a similar workflow like Cinnamon or KDE for example. Removing support for Deepin means that the team will be able to give more time on other things, thus improving the quality of Manjaro overall. Anyway, I have never been a fan of the large amound of choice Manjaro offers, I prefer to have less editions but better quality editions.
However, there is many users using Deepin currently, and stopping support for Deepin completely means they will have to install another DE and do the cleanup (at that point, it will be faster to simply back up and reinstall Manjaro with a different edition) since they won't receive any updates anymore.
So it is not an easy decision. What I would do is the following:
- Immediately, remove the link to the ISO on the website (anyway, the lastest ISO is like 18.0.2, which starts to be really old), but do not remove the Deepin profile for now;
- Take a decision;
- Whatever decision you takes, be honest and write an announcement explaining the current situation with Manjaro Deepin, what is your point of view and what decision you took and why you did.
- If you decide to remove support for Deepin completely, do not do it immediately, but select a "date of deprecation" and warn your users that "Manjaro won't support Deepin in any shape or from July 1st 2019 (example) and onward." and tell Manjaro Deepin users that they will have to switch to another DE.
- If you decide to remove overlay package only (which seems to be the worst and most awkward solution with unpredictable consequences), mention that in the announcement and tell that regressions may happen.
Don't know about Manjaro Deepin DDE, but my deepinOS works flawlessly. No issues whatsoever. On the other hand, I do agree that KDE is also equally beautiful with unmatched configurations/options.
This may be a solution. While I personally enjoy Deepin, both on Manjaro and the Distro itself, I think it is time to depreciate it for a little while. I moved my wife's computer to another Distro when they started talking about the "customer experience" BS. Even before that time, the experience was starting to get buggy and my wife gave me an ultimatum, "Make my computer work right every time when I turn it on." Because the distro lost the important WAF (Wife Approval Factor) she is now on Kubuntu 18.04 LTS.
I moved on from Deepin on Manjaro about 6 months ago when I could not get it to retain my touchpad settings after each reboot. Maybe because I am getting older, I just got tired of having to fight the DE to be able to even to use it. Now I am back on Gnome and things are going smooth.
All that being said, I will be keeping my eye on the Deepin DE and hopefully in the future it will become stable enough to make a return. I really enjoy the overall elegance of the design and the work flow. However, those are not replacements for a stable and usable system.
in addition to that if manjaro drops deepin, i think average users wont know how to switch desktop environments without re-installing, or even if they do this forum will be flooooded with "i tried to switch to kde from deepin, it's broke, how fix?" .
so maybe a set of dummy-proof instructions for switching from one to the other. the arch wiki will only confuse them because they will be expecting a configured desktop comparable to what they would get from installing from a manjaro iso.
The last thing we need is for Manjaro to acquire a bad reputation. Could you imagine a vlogger on YouTube saying how buggy Manjaro Deepin is?
Better safe than sorry!
Manjaro is better off by not supporting the Deepin desktop environment.
That is an enormous task though ... a high level of commitment and technical know would be required ... and I'm not sure the result would be any better.
On top of existing issues you've got additional regressions related to upstream dependency updates, that I assume would require patching, which would involve cherry picking upstream commits (if they exist), which would lead to a [manjarodeepin] package mishmash that is probably not any more stable than Arch's currently.
Retrofitting an upstream DE developed against an LTS *buntu package base into a rolling distro is difficult enough, retrofitting a poorly developed & bug ridden DE really is pissing into the wind regardless of the approach.
Debian Stable right now, but that doesn't change line of thought.
I just posted this at deepin development:
I don't know if this is possible, does anybody here have an idea?
Like current KDE-Unstable repo in Arch, but Deepin-Unstable?
Is there a large enough Deepin user base to make this approach effective?
Is this just rearranging deck chairs on the Titanic? Will bugs found and reported be adequately patched in a reasonable time frame?
If it is possible than it would make sense, because single person realistically can't do what felixonmars currently doing for Arch / Deepin.
At least some degree of testing required there.
That is the most important question.
Because even if me & @ant will provide additional testing layer - obviously we need more people on Arch to actually being able to fix stuff / make pull-requests for quickifxes (like felixonmars is doing now), because certainly this alone will not change Deepin devs practice of prioritizing clients demands above bugfixes.
Several good ideas has been aired.
I know Deepin is the desktop for some users but from reading this thread and the github thread and from the multiple threads on this forum - I think the wise step to take is to discontinue the ISO.
This is to keeping Manjaro clear of bad press which - I think - ultimately will emerge from the state of converted Deepin packages.