Modularity Fedora Objectives

[Martin Curlej]

Hi,

I am the new Product Owner of the Modularity project in Red Hat. I want to
present to you the proposal for the Modularity objective for Fedora[1].
This follows the process as is described by the docs[2]. We already started
the work on some of the points described in the objective. Mostly fixing
bugs and making Modularity more stable and user friendly. The objective is
here to highlight the changes and give an estimate on the timeline in the
near future.


[1] https://fedoraproject.org/wiki/Objectives/Modularity_Improvements_2020
[2] https://docs.fedoraproject.org/en-US/project/objectives/

-- 

Martin Curlej

Software Engineer, Product Owner

Modularity

Red Hat

<https://www.redhat.com>

mcurlej(a)redhat.com
<https://red.ht/sig>

[Matthew Miller]

On Tue, Sep 08, 2020 at 11:13:17AM +0200, Martin Curlej wrote:
>  I am the new Product Owner of the Modularity project in Red Hat. I
> want to
>  present to you the proposal for the Modularity objective for Fedora[1].
>  This follows the process as is described by the docs[2]. We already started
>  the work on some of the points described in the objective. Mostly fixing
>  bugs and making Modularity more stable and user friendly. The objective is
>  here to highlight the changes and give an estimate on the timeline in the
>  near future. 
Awesome. Thanks for sharing. I think this all looks great.


-- 
Matthew Miller
<mattdm(a)fedoraproject.org&gt;
Fedora Project Leader

[Ben Cotton]

Thanks for sharing this Martin. I have a few questions, comments, etc
below. I like the overall direction, so these should be read as
opportunities for improvement, not as criticism.

* When this was circulated on devel[1], some people raised concerns
that MBS isn't represented. Since that's going to be a key component
of this Objective's success, it would be good to have someone from
that team explicitly part of the Objective.
* Given that the F33 release is getting close, it's probably better to
focus on F34-35 (or 36 if you're looking at 18 months). Specifically,
it would be good to see at least some rough targets for the F35
release timeframe, with the understanding that a lot of the
research/planning work for that hasn't been done yet.
* For goal #4, do you have an idea of what the architectural changes
are or what you want to accomplish with them?
* Generally speaking, 12-18 months from now, how will we know if this
Objective has been successful? Number of modules available? Improved
maintainer sentiment? Improved user sentiment? Something else?

[1]
https://lists.fedoraproject.org/archives/list/[e-mail].o...

-- 
Ben Cotton
He / Him / His
Senior Program Manager, Fedora & CentOS Stream
Red Hat
TZ=America/Indiana/Indianapolis

[Martin Curlej]

Hi Ben,

sorry for the late reply. First let me summarize what we are going to
accomplish in the near future and then I will go into

details on each of your points.

We are aware that the release of Modularity in Fedora was rough and we are
working to fix that. What we are trying to

do now is to use what we learned in Fedora and RHEL and make
adjustments/features/fixes in the codebase so Modularity

would be more user friendly and simpler.  It is a complex task which needs
a lot of time and effort.

Now to address your points.

With regards to MBS, we are in touch with the EXD team which is supporting
MBS in Red Hat. Currently Mike McLean

and Kevin Fenzi are working on the deployment in Fedora infra. EXD people
will announce MBS support on Fedora

devel-list soon.

For Fedora 34, we have recently planned fixes which will help Fedora as
well as RHEL. There is a critical issue with the

upgrade path of the modules. This concerns both Fedora and RHEL. The fix
for this is the introduction of static contexts

and phasing out of modules stream expansion. As you said, we can only
provide you rough estimates for Fedora 35. We

want to revisit the implementation of the default streams as this seems a
feature created controversy inside the community.

We would like  to make it more straightforward for the consumer and the
packager.

With regards to the deeper architectural changes, we want to get to the
drawing board with Modularity. We already started

with the upgrade path fix and continue with default streams with respect to
feedback and experience.


In the long run, the success criteria would be that we phase out module
stream expansion from Fedora 34 and revise default

module streams so we can use modular RPMs in non-modular repositories in
Fedora 35.


We need more resources to work on long term tasks. Due to the COVID-19
situation, we do not know when we could extend

the Red Hat team assigned on the Modularity project. Anyway, if there are
volunteers from the community who would like to

contribute, they will be welcome to join the project.

We would also like to improve the communication with the Fedora community.
Any help and feedback from the community

is really appreciated.

On Thu, Sep 10, 2020 at 10:49 PM Ben Cotton <bcotton(a)redhat.com&gt; wrote:

>  Thanks for sharing this Martin. I have a few questions, comments,
> etc
>  below. I like the overall direction, so these should be read as
>  opportunities for improvement, not as criticism.
>
>  * When this was circulated on devel[1], some people raised concerns
>  that MBS isn't represented. Since that's going to be a key component
>  of this Objective's success, it would be good to have someone from
>  that team explicitly part of the Objective.
>  * Given that the F33 release is getting close, it's probably better to
>  focus on F34-35 (or 36 if you're looking at 18 months). Specifically,
>  it would be good to see at least some rough targets for the F35
>  release timeframe, with the understanding that a lot of the
>  research/planning work for that hasn't been done yet.
>  * For goal #4, do you have an idea of what the architectural changes
>  are or what you want to accomplish with them?
>  * Generally speaking, 12-18 months from now, how will we know if this
>  Objective has been successful? Number of modules available? Improved
>  maintainer sentiment? Improved user sentiment? Something else?
>
>  [1]
>
> https://lists.fedoraproject.org/archives/list/[e-mail].o...
>
>  --
>  Ben Cotton
>  He / Him / His
>  Senior Program Manager, Fedora & CentOS Stream
>  Red Hat
>  TZ=America/Indiana/Indianapolis
>  _______________________________________________
>  council-discuss mailing list -- council-discuss(a)lists.fedoraproject.org
>  To unsubscribe send an email to
>  council-discuss-leave(a)lists.fedoraproject.org
>  Fedora Code of Conduct:
>  https://docs.fedoraproject.org/en-US/project/code-of-conduct/
>  List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
>  List Archives:
>  https://lists.fedoraproject.org/archives/list/[e-mail]...
>  

-- 

Martin Curlej

Software Engineer, Product Owner

Modularity

Red Hat

<https://www.redhat.com>

mcurlej(a)redhat.com
<https://red.ht/sig>