[lubuntu-devel] [lubuntu-users] Lubuntu Next Blockers and Suggestions (WAS: Re: Lubuntu Artful Aardvark Beta 1 has been released!)

Simon Quigley tsimonq2 at ubuntu.com
Sun Sep 3 00:15:14 UTC 2017

Hey Alf,

On 09/02/2017 08:13 AM, Alf Gaida wrote:
> On 02.09.2017 09:06, Simon Quigley wrote:
>> One thing I can point out for sure is this -- LXQt makes releases too
>> slow for the amount of new features that are being added and in the
>> development state that it is in. While I do see the importance of
>> waiting until the right time to do a release, doing an LXQt release once
>> every year or so is not, in my eyes, the best way to go. It would be
>> great (in my opinion) if LXQt upstream could either:
>>  1. Set out a time-based schedule (even if it's general (give/take a
>> month)).
>>  2. Once a new LXQt release has happened (and not long after), set out
>> clear goals for what features want to be included and what bugs want to
>> be fixed, and work towards getting those done. Once those are done,
>> actually release. If bugs are found, fix them along the way.
> The ground work for having a LXQt with smaller changes that fit better
> to a possible release schedule is done with the latest changes in pcmanfm-qt
> and libfm-qt - so if these huge changes are finally finished we can release
> more often with smaller changesets - Simon is right with that. But please
> keep in mind that LXDE is mature and LXQt is not. That will change in the
> next two years.

That's great to hear.

> Next big thing will be wayland. I admit that i don't follow the wayland
> development at all that close, but the framework integration has grown over
> time.

I am personally hesitant to ship Wayland in Lubuntu given the problems
with the NVIDIA graphics drivers. Once there's a good solution for that
and Wayland has been tested a bit more, I think it'll be a good option.

To be clear, I hope LXQt doesn't drop X support any time soon. ;)

>> One way that could be accomplished is having some sort of Continuous
>> Integration for LXQt. I know Alf has been working on Siduction (I don't
>> know much about it, though) and I've been hacking up some scripts of my
>> own. But having some sort of central CI and platform for testers could
>> really help.
> CI is overrated in that case - it will be nice and we should work on it. But
> the best ci don't help with things like new versions (travis or circle with
> current Qt or KF5 will prevent us from shooting us in one or two feets with
> available Frameworks. Thats great and help a lot.)

Have you seen the work that Kubuntu has done with KCI? In their
packaging they have kubuntu_stable, kubuntu_unstable, and
kubuntu_$RELEASE_{archive,backports} and it works pretty well for them.
While that is another topic in and of itself, here's a link if you'd
like to check it out: http://kci.pangea.pub

<snip />

>> I am personally excited for when Lubuntu Next is production ready and we
>> can then look at the possibility of sunsetting the LXDE spin of Lubuntu.
>> While this might be a year or two in the future (I'm thinking between
>> the release of 18.04 and 18.10 Alpha 1 will be when we have those
>> discussions and make final plans, 18.04 cycle will be focused on
>> polish), I can see it happening. The problem is, Lubuntu has been saying
>> this for so long, that we're going to replace LXDE with LXQt, etc. etc.
>> etc. 
> Imho the current release of LXQt was the first release that fit the needs of
> so called normal users and fit into debian stable - so the 0.12 will top that,
> being more polished, less rough edges but still a way to go. 

I've sort of gotten that feeling personally when using it. I've been
using it on my desktop computer on and off for about a year now, and for
the past 3 months or so, it's felt polished enough and had little enough
breakage that I thought it was perfect timing for getting the Lubuntu
Next spin ready.

> To be clear about - 0.11.x fit the most of my needs as a DE, 0.12 will do
> better. And i guess the most users want things that just run. So looking
> forward i would say that 0.12 will be fine for lubuntu next 18.04 - did i 
> mention that we can release packages now as point releases without pain as
> long we don't break qtxdg and liblxqt abi? Seems we become more mature.

Here's what's probably going to happen with 17.10 and 18.04:
 - With 17.10, things won't be 100% polished yet, so while we still may
release it, we'll recommend that people go with the LXDE Lubuntu.
 - With 18.04, unless things change a lot in the next 8 months, may ship
with Lubuntu LXDE as the LTS (3 years of support) and LXQt as a non-LTS
(9 months of support). I'll have to talk to the Ubuntu Release Team to
see exactly how to implement it, but I know it's possible.

Again, just my opinions, this is still open for discussion.

Thanks Alf!

Simon Quigley
tsimonq2 at ubuntu.com
tsimonq2 on freenode and OFTC
5C7A BEA2 0F86 3045 9CC8
C8B5 E27F 2CF8 458C 2FA4

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 801 bytes
Desc: OpenPGP digital signature
URL: <https://lists.ubuntu.com/archives/lubuntu-devel/attachments/20170902/ec875984/attachment-0001.pgp>

More information about the Lubuntu-devel mailing list