Hi,
@Jason, I think this is a very valid point. While I have more experience with the Gerrit flow and really like it, I do see that way more people are familiar with the
Github PRs and would prefer these. I can’t tell for sure if moving to Github would result in getting more contribution, but I believe this will one burden less for people willing to contribute to MAT.
The foundation started already moving projects in waves away from the Eclipse infra, and I guess we should be pro-active and volunteer after we release MAT 1.15.0.
Another thought I have – as we started the project in long ago, we tried to keep the repository for code only and put all supporting docs (like “how do I build the stuff?”)
“around” – web page, Eclipse Wiki, etc… Meanwhile I see that most projects on Github have all resources next to the code within the repository and all developers are used to this. In our case one has to explicitly search for how one thing or another could
be done. It is all there, but I think this is another burden which we should remove. And this fits well to the plans that the Wiki is deprecated and will be removed (in many years from now). I think I’ll start to collect all the pieces I have as .md files
within the repo and make our Wiki obsolete much sooner than the deprecation.
We’ll also need to move away from Bugzilla - another piece I really like
😊, but also there I think people would prefer having the code on Github
and using Github issues within the project.
So, my suggestion would be to start preparing for moving everything we have to Github and start making concrete changes to achieve this after the Dec. release, i.e.
the beginning of next year. Any thoughts on this?
I’m pretty sure this alone is not enough to attract more contribution, but hopefully combined with the other suggestions Andrew had we may see some positive effect
😊
Regards,
Krum
From:
mat-dev <mat-dev-bounces@xxxxxxxxxxx> on behalf of Jason Koch via mat-dev <mat-dev@xxxxxxxxxxx>
Date: Tuesday, 26. September 2023 at 20:24
To: Memory Analyzer Dev list <mat-dev@xxxxxxxxxxx>
Subject: Re: [mat-dev] Memory Analyzer 1.15.0
@Andrew,
I suspect that shifting to Github will help too. Even though I have made a few contributions, I still find it a challenge to think through the flow. I wonder if this is putting up a challenge in front of people
who want to contribute but maybe don't have everything in place with Gerrit, looking through bugs/issues, etc.
Hi team,
Thank you for all your work on this project.
As a lurker who has not yet contributed, yes I believe Andrew's ideas #6 ( keyword=’helpwanted’ ) and #8 (documentation on how to help) would be valuable to me.
Direct recommendations on how I can help are also appreciated. I have experience writing a standalone plugin using MAT API, but have not yet touched the core tool.
I think we should plan for a new release of Memory Analyzer.
There is a placeholder target field 1.15.0 for Git which I have used for some of the current defect / features.
https://bugs.eclipse.org/bugs/buglist.cgi?classification=Tools&list_id=21370820&product=MAT&query_format=advanced&target_milestone=1.15.0
The full unresolved bug and enhancement list is:
https://bugs.eclipse.org/bugs/buglist.cgi?query_format=advanced;order=Importance;bug_status=UNCONFIRMED;bug_status=NEW;bug_status=ASSIGNED;bug_status=REOPENED;product=MAT
I was thinking of December 2023, to coincide with SimRel 2023-12. Does that give enough time for us, and is that soon enough for users?
Other things to do might be:
-
Check how MAT works with Java 21 (LTS), from OpenJDK, Temurin and Semeru (when available), for running with, and analysing dumps from.
-
Check we are up to date on dependencies, to avoid security warnings.
-
Write a release record at the project page.
https://projects.eclipse.org/projects/tools.mat
-
Follow
https://wiki.eclipse.org/MemoryAnalyzer/Contributor_Reference#New_version_development_process
-
Follow
https://wiki.eclipse.org/MemoryAnalyzer/Contributor_Reference#Simultaneous_release_policies
I know that no one is working full time on the project – but how and for what can we encourage help from others?
-
Documentation – it’s in DITA which might be a bit unclear but is sort of like HTML.
-
Do we need an explanation and images of all the views?
-
The acquiring a heap dump section needs updates for more recent version of Java.
-
Build – is there a Tycho / Maven person who could review our build process? Can we get a dependency check for vulnerabilities, or a software bill of materials?
-
Translations – does anyone want MAT translated to another language. MAT is enabled, and translations updates can be made via
https://babel.eclipse.org/ and the results can be installed into MAT. However, all the translations are incomplete, and the best, Japanese, is only 33% translated. We would need people fluent in the target
language who also understand English and MAT well.
-
Do we need an automated regression test build?
-
Do we need an automated performance test build?
-
Should we tag some bugs with keyword=’helpwanted’ ?
-
How well do the ‘Leak Suspects’ and ‘Top Components’ reports work for users? Do they actually show the problems?
-
Do we need a section on the website explaining how to help the project?
Thoughts, anyone,
Andrew Johnson
Unless otherwise stated above:
IBM United Kingdom Limited
Registered in England and Wales with number 741598
Registered office: PO Box 41, North Harbour, Portsmouth, Hants. PO6 3AU
_______________________________________________
mat-dev mailing list
mat-dev@xxxxxxxxxxx
To unsubscribe from this list, visit
https://www.eclipse.org/mailman/listinfo/mat-dev
--
This message and any attachments are solely for the intended recipient. If you are not the intended recipient, disclosure, copying, use, or distribution of the information included in this message is prohibited -- please immediately and
permanently delete this message.
_______________________________________________
mat-dev mailing list
mat-dev@xxxxxxxxxxx
To unsubscribe from this list, visit
https://www.eclipse.org/mailman/listinfo/mat-dev
|