OpenWebStart and IcedTea-Web

[Michael Heinrichs]

Hi Jiri,

Good news! We just finished our meeting where we discussed the outcome of the hackathon. We decided that we want to continue with IcedTea-Web.

Next week, we will start to create clean PR for the main repository. But as we already discussed this will not work in the long run, therefore we need commit-rights rather sooner than later. We will start with two engineers and later a third one will join. What is the process to get commit rights?

Our suggestion is that all changes have to go through a PR and need at least one approval before they can be merged. What do you think? Do you have the rights to set this up in GitHub?

Thanks,
Michael



> On 27. Mar 2019, at 17:53, Jiri Vanek <jvanek at redhat.com> wrote:
> 
> On 3/27/19 5:49 PM, Michael Heinrichs wrote:
>> Hi Jiri,
>> 
>> This sounds great. I totally agree, PRs and code reviews are a must nowadays. AFAIK all projects at Karakun are done that way and we also plan to establish these practices for OpenWebStart.
>> 
>> Do you discuss somewhere how the process is going to be set up?
> 
> Yes. With you,right now :)
>> 
>> Thanks,
>> Michael
>> 
>> Von meinem iPhone gesendet
>> 
>>> Am 26.03.2019 um 17:27 schrieb Jiri Vanek <jvanek at redhat.com>:
>>> 
>>>> On 3/26/19 9:08 AM, Michael Heinrichs wrote:
>>>> Hi Jiri,
>>>> 
>>>> Yes, there is only one code base at this point. But we are currently evaluating if it makes more sense for us to join ITW or create something new/fork ITW. There are pros and cons for both sides.
>>>> 
>>>> I cannot really tell you right now how the native part and the Java part can be split. Our engineers are doing a two day workshop this week where they try to find good answers to these questions among others. Stay tuned! :)
>>> 
>>> Lets watch it:) ITW can become downstream of yours at the end...
>>>> 
>>>> Sorry for not being clear. When I wrote modules, I meant Maven modules. Our plan is to bundle OpenWebStart with a JRE in native installers, which would make us more flexibel in terms of which Java version we want to use. But the first version will probably run on Java 8 anyway.
>>> 
>>> ok. Multi jdk support is both advantage and pitfall of ITW.
>>> 
>>>> 
>>>> What is the policy of the ITW repo? Are you the only committer and people created pull requests? I guess this process will not work anymore unless you are willing to do nothing more but pull request reviews during the next couple of months. ;) How shall we setup the process?
>>> 
>>> There was about  20 commiters/reviwers on classapth servers I knew about, and aprox 100 I was not
>>> aware about. Unluckily all are inactive now. Anyway, we moved to new repo, so those are no longer
>>> valid, nor the workflow, nor the policies.
>>> 
>>> I definitely can not stay single commiter/reviwer. That would kill both me and ITW.  I'm definitely
>>> going to eyball all commits in  next few weeks, but I may be of for day or so, and I do not wont it
>>> to stay and wait. I can always speak my mind after merge, and you do not need to listen. Nor I can
>>> catch all, nor can I be the single stop show voice.
>>> 
>>> I guess all your fultimers on ITW should get commit review permissions right now, but all changes
>>> should go through PR, so other interested vocies can comment. Geerally untill there is anti voice,
>>> the PR should (SHOULD!) not be merged.
>>> 
>>> We are currently setting the process up. Lets it be square usable. My only note to it really is,
>>> that every change should go via PR, and your full timers shoudl get push/merge access.
>>> 
>>> TYVM!
>>> J.
>>> 
>>> 
>> 
> 
> 
> -- 
> Jiri Vanek
> Senior QE engineer, OpenJDK QE lead, Mgr.
> Red Hat Czech
> jvanek at redhat.com <mailto:jvanek at redhat.com>    M: +420775390109

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.java.net/pipermail/distro-pkg-dev/attachments/20190329/7c30a419/attachment.html>