LineageOS guys are talking about /e/ and they are not happy with us…
Maybe we will have to start keeping statistics of what LineageOS bugs we inherited and fixed
Also maybe we should start porting the fixes back to the source… Read the discussion here
If it’s ok to share a newcomer’s take on that reddit thread (I don’t really know how Lineage or /e/ are run internally)…
To be honest, I’ve read that sentence during a short coffee break at work and decided to not expose myself to the negativity it implied.
Now that I’ve read it, I don’t think it wasn’t that bad.
I don’t think this is a good idea.
That sounds better.
Having said that, I think Gael’s response was pretty good. I can imagine everyone has their assumptions about the “other side” and it helps to clarify some details that could be the source of a misunderstanding.
indidea = Gaël ? https://www.reddit.com/r/LineageOS/comments/eq19l2/your_tips_and_ideas_to_make_lineageos_the/ff5ytvl/
I am still using an old Lineage version (CyanogenMod 13.0-20160323-2046-UNOFFICIAL-espressowifi) on a tablet Samsung Galaxy tab 2 7.0 which is no longer supported by LOS. This unofficial build is all I could find for this device. Howto build LOS or even better /e/ on unsupported devices? I also have an old French device KL-950 (Klipad 5.0 pouces) what isn’t supported anywhere, still running Android 4 KitKat. Sustainable hardware…
I know it is not a good idea but to suggest that the 2073 defects that are recorded on Gitlab are all /e/ ‘created bugs’ in the code and there are no issues being carried from the source is hard to believe.
We should have gone ahead and patched the fixes back. Hopefully this will be done moving forward.
Let me just say that I am ALWAYS skeptical regarding LOS official comments regarding forks from their project. Look at the way they treat the microg devs and if I remember correctly they even banned “microg” from their subreddit.
I don’t hold them in high regard.
I find all this to be pretty constructive. Gael’s explaination is factual and clear. And Lineage probably also has a point. I’m the type of user Gael is mentioning that hardly knows what flashing a device means. And my impression is that /e/ relies on LOS as a middleman. I noticed this with my Fairphone about a fix that Fairphone had made available but that LOS had not implemented and therefore never came to /e/ either.
Yes, but are those fixes submitted back into LOS? We really need to work together here in order to move forward and build a sustainable ecosystem - which IMHO does definitely include submitting patches back to the upstream project (LOS in this case).
yes – participating by reporting patches to the upstream sources resp. sharing the development forces is a very natural kind of mutial cooperation and support in the open source world and IMHO the only chance to succeed in the long run.
but another aspect, which was expressed by the LOS side, looks even more importent to me, if /e/ wants to be taken serious:
“We provide the best possible support for our devices, and this also includes actually dropping them from being supported for users and our own care. We thrive to provide something good, and letting the user know something isn’t up to our standards ( not anymore ) is the best way.”
that’s IMHO the only realitistic and responsibe answer to this end of life problem resp. prohibitive support, which shouldn’t be denied flatly by /e/ and its public announcements and promisses.
We thrive to provide something good, and letting the user know something isn’t up to our standards ( not anymore ) is the best way.
Dropping support for devices suddenly and without notice can be damaging for the end user, especially when you are on your own to find out about it and know what’s happening (it has happened to me).
/e/ is a way more user-friendly approach than LineageOS. They even have a forum where we can discuss and share our experiences, whereas for LOS is pretty much the wild west. Their community and developers (for some devices as I’ve seen from XDA) are not particularly helpful to those who don’t know how to flash a device.
So I agree with @cedricoola that this is a rather constructive thread. LineageOS has done a great job but /e/ is taking it one step further by providing an easy and functional alternative to stock Android, including cloud infrastructure and better support, which equals a much, much larger audience.
Saying that they are “leeching” is ridiculous, it’s an open-source project, that is the entire point! Go cry me a river elsewhere with this nonsense…
I’ve used LineageOS on 4 devices and always within about a year there is no longer any updates. I want my devices to last as long as possible, the hardware is still good so why drop support? I understand the maintainers are volunteers, and appreciate their efforts. But the support model of LineageOS is not conducive to long term support as Gael pointed out in his reply.
The maintainers of LineageOS devices are developers. Once they have completed the work to port LOS to a device it is natural to move onto the next one. That is natural, developers look for the next challenge. Unfortunately this dependence on volunteers does not include and ongoing support mechanism, so LOS eventually drops support for formally supported hardware. That’s just their model, LOS is more of a project by developers for developers who are willing to switch to new hardware devices frequently.
Those who are looking for more stable support for the long term need something like /e/ which commits to not dropping support of devices once they have been added to the roster. Different models for different target audiences. Gael explained it very well.
And I do agree there should be more effort to back port. But all changes are freely available, so LOS maintainers also have the ability to “leach” changes to /e/ back to their project. If they want to make the effort.
no – there’s only one part of the game in the hand of the /e/ and LOS developers.
the vendor provided baseband drivers and all it’s included bugs and security vulnerabilities are still unaffected by this alternative modification attempts, and even on the system level it’s rather unrealistic, that /e/ or LOS would be able to backport more fundamental security improvements of newer android major releases to older versions. you simple have to estimate the actual development man power behind this open projects from a more realistic point of view.
the mentioned statement about LOSs practice how to fight this unavoidable issue looks rather exemplary to me. it’s IMHO better to communicate the real limits and impossibilities in a responsible manner to the actual users than to lull them into a false sense of security by questionable promises and announcements. it’s simple an unsolvable challenge, to keep devices save and adequate supported on a similar quality level as in the time span, where they are supported by the original base os maintainers and vendors.
IMHO it’s therefore more useful to concentrate the actual development power on efforts, to make new devices sooner available in LOS and /e/ to profit from a longer period of remaining official android and vendor support, than to make them available much to late resp. often rather close to their foreseeable official end of life.
I don’t know too much about the LOS community, but my experience is different. For Fairphone there are LOS people staying with the brand (and only two models) since 6 years. One is backporting current security patches to Android 4.4.4 for the FP1 and Android 10 for the FP2 is under development, too.