Perhaps this could be some reasons why Jolla is moving away from OBS:
https://specs.openstack.org/openstack/fuel-specs/specs/7.0/replace-obs.html
As far as I understood, the reason for OBS removal is primarily legal. People are uploading whatever and Jolla may be held responsible for hosting illegal content or something along those lines. You can’t beat that sort of argument
IMO it should stay around at least to allow outside developers to make reproducible builds of their Jolla Store apps. I don’t think my opinion weighs much, though…
@slava, I presume it is not an official reply and from your response it sounds like the reasoning was not fully discussed in house among Jolla developers.
I am not aware of anyone pushing something illegal at OBS. I presume we are talking about Android binary blobs here. Or anything else?
Again, if it is for legal reasons, those have to be stated and not plastered over by something irrelevant. When we get the reasons behind considering closure of OBS, we can start the discussion regarding it and how to find the solution that would work for the all parties involved.
Just conspiracy theory: Last year I saw increased amount of attacks to build environments and other internal systems in various technology companies. When such attack is successful, you have to shut down such environment and don’t release details until investigation ends.
I’ve created a separate topic of the OBS shut down and next steps I’ve tried to describe the rationale behind the decision. Please use that thread for ideas, questions and discussion related to this topic.
The mer wiki (https://wiki.merproject.org) also seems to be regularly down. Is this also because of this change? Parts of the merwiki contained useful information, such as the current community adaptations, information about how Spectacle works and such.
If this is related to this topic, are there plans to make the content of the wiki available once again, albeit on its original location or somewhere else?
wow, wiki with https://wiki.merproject.org/wiki/Adaptations/libhybris table was really shut down? Are you…? Can’t believe this.
There’s just a problem with the disks on the machine hosting the wiki. I’ll try and get it back up ASAP.
aaaaand it’s back up
There’s a typo in your link - the word “project” should have an ‘r’ in it. So don’t be surprised if the site seems to be down if you try to follow your link
Is there some time plan for migration to https://github.com/sailfishos ? I would like to send some patches to connman, but I don’t have account at git.sailfishos.org…
You can ask @lbt on IRC (#mer or #sailfishos on Freenode) if you would like to have one there.
I don’t think we have any time plan for the migration - please don’t wait for that to happen if you have patches to send.
please don’t wait for that to happen if you have patches to send.
Well, in the past that did not make any difference.
In hope that this may have changed (you tend to sound so authentically enthusiastic @vige), as one example, see https://git.sailfishos.org/mer-core/udisks2 with a few issues and MRs nobody ever reacted to (I would have to look up what else I filed where).
Have you pinged the developers? It’s quite possible that no one has noticed your MRs, that happens quite easily with gitlab.
The way I usually do it: Look at the git logs, see who has been making most of the commits lately. Ask them to review.
I can think of at least two reasons why this does not work in practise: 1) When you get ~50 emails daily from various git repositories, it’s quite easy to miss the actually important one. 2) There usually isn’t only one person who maintains a single repository. When there are several people doing it, it’s quite easy to think that someone else will take care of this MR.
@vige, trying to express my point better:
- I am beta-testing SailfishOS for 5 years now, while paying (a little) for a couple of SailfishOS-licenses.
That is fine for me, including the fact, that SailfishOS was never stabilised to a non-Beta release, and probably never will, because that is not a requirement of Jolla’s primary and only real (i.e. substancially financing) customer.
The take-away is: I am a (minor / lesser) customer of Jolla, paying to be a beta-tester (as all other Jolla customers with a personal license do), and accepted that long ago. - I reported a couple of 10 bugs to Jolla, received feedback from Jolla for less than 20% of them (usually acknowledging the bug) and less than 5% were resolved.
- I believe I already did way more that a typical beta tester:
In most cases I also submit some proper analysis together with the a bug report, often accompanied with an MR or at least a workaround.
Jolla obviously has or deliberately created a big structural issue here (since SailfishOS’ beginnings):
As you point out, Jolla is apparently deaf for bugs reported over conventional channels, which exist for this very purpose: reporting bugs.
If somebody denotes that, Jolla’s usual reaction is to single-out that person and issue, by requesting to jump to further hoops:
- “Please bring it up at the open IRC session”
- “Please report it to somebody personally”
- “Please fill out our internal bug-tracker form (in addition to an extant bug report)”
- etc.
Jolla is systematically and consistently shrugging people off by this mindset and course of actions.
Still new people (like you) sailors are usually replicating this behaviour right from their start at Jolla, so it looks like they are internally primed to act so.
One only wonders, what the purpose of Jolla’s public beta programme (“Sailfish X”) really is, if not testing and gathering feedback?
P.S.: I once modelled, that our testing and reporting is used by Jolla for enhancing SailfishOS to become more appealing for big licensees, but Jolla never seems to have utilised it this way.
Fact is, that Jolla deliberately leaves many resources and much energy untapped, due to this very non-integrative testing and development process.
P.P.S.: @vige, while rereading your reply for a third time, I realise, that your central statement is: “Jolla has (developed) no working processes for bug reports and MRs, yet.”
While this understanding may be the basis for designing a change for the better (after more than 7 years, but “better late than never”), you really do not sound like this is ever going to happen (rather the opposite).
if you know the maintainer, the most efficient way to notify him/her is to @ message.
@olf I am sure @vige is long enough with Jolla/Sailfish, but he can speak for himself. I do not know how you conclude he is new here.
I think you are over-driving a little bit the situation. Of course there is a lot of room for improvment, but the baby learns to walk. I expect they’ll gain speed soon and they must, because markets are changing and it would be a pitty to loose such a good product.
Regarding the structural issue I agree and I admire how these decisions are taken. I wish we could have the freedom to get a copy of the code so that even Jolla dies some day, we could still have the ability to develop this system further.
Why I am writing negative? Because I do not trust western politics anymore and it could be they shut down Jolla, because it is cooperating with the Russians.
IMO Jolla should keep it’s proprietary code on own servers and the rest should be hosted on some open source servers and there should be a way to automatically download everything you need to build the project, flash it on a device and start using it.
Why it was possible 10y ago and now it is so hard with all those fancy tools gitlab, github etc.?