Re: [iot-pmc] Request assistance for Eclipse Cyclone DDS

[Erik Boasson <erik.boasson@xxxxxxxxxxxxxx>]

Dear Kai and Jens,

I see your points, and cannot but agree with them.

There is, however, one detail that I would like to respond to, because I think that in trying to address a misunderstanding, I have created a new misconception …

On 18 Sep 2018, at 22:53, Jens Reimann <jreimann@xxxxxxxxxx> wrote:

I fully agree with Kai here. And I would like to add something.

The Eclipse project handbook says:

---

Each project is entitled to define how it evaluates "[people] who have the trust of the Project’s Committers … [through] contributing and showing discipline and good judgment". This definition needs to be a transparent and public document on the project’s website (the top-level project charter may provide this). It is extremely important to publish these criteria to avoid any issues around cliques or "the in-crowd" preventing others from joining a project.

---

Having personal conversations with a committer, not sharing work, doesn't really help as an argument in favor IMHO. The problem for me would be, how would other contributors be nominated in a similar fashion then? As Kai mentioned, there is no public record, which would make the decision transparent to others.

The case is not that he is not sharing work, the case is that he has not yet completed the work to the point that he feels it is ready for doing a pull request. This to me is a perfectly normal situation, it simply takes time to get it to that point. Given the as-yet very small commit record — the full commit history was squashed in the initial contribution — I thought it would be of value knowing that he (or anyone nominated for that matter) is actively working on contributions at this point in time, as that still shows a commitment. But it is equally fair to say that it then makes more sense to wait a little while longer until those contributions have been made.

I will be very happy merging pull requests from anyone in the larger community, provided the contribution has merit. Indeed, my point about the conflict of interest is exactly because I want to avoid situations where others might start wondering whether all contributions really are judged equally. In my opinion, reaching that state requires action (to have more than one active committer), and in the current circumstances there is precious little to rely on that is publicly visible. All that is left as input for the decision are the things we know through other means.

I also think something sounds a little odd here, if 3 project leads and 4 committers are inactive on the project, but still vote. I would of course understand if all (but one) committer are inactive and won't be involved in the project in the future. Then we might think about declaring the project dysfunctional (not in a technical sense of course) and reset the committer base. Things like this happen and there is a solution for that.

This raises a good point, but I subscribe to the view that the two issues should be considered separately from each other. A few inactive members are not a major impediment to progress, but not having the right people included in the set of committers is. But, as it happens, I myself am rather new to the process, and so my view may be affected by not having seen much yet …

Otherwise I think it maybe helpful help us understand the current situation of the project a bit better.

I would think that’s something to take up with the project leads. I’d be happy to contribute to the discussion, but to do it without involving them seems improper.

Best regards,

Erik

On Tue, Sep 18, 2018 at 4:39 PM, Kai Kreuzer <kai@xxxxxxxxxxx> wrote:

Dear Erik,

Thanks for this further information, it is much appreciated.

As I was the one that vetoed the election, let me respond to it:

First of all, it is very important to understand that this veto was not at all about Jeroen in person, but rather about the way the nomination was done. You are at any time free to do a new nomination of Jeroen.

The veto has been expressed because a nomination should show a track record of work of the nominee, which proves that he understands the project, its processes, that he did valuable contributions already and that it is likely that he will continue that way. All of this was missing in the description of the nomination.

has worked on what is now Eclipse Cyclone DDS internally before it was contributed to Eclipse

This is an important argument for the nomination. It should be possible to link to commits or code in the project that he has authored, so such links would be helpful.

 is currently undertaking relevant work, work that he simply hasn’t yet felt ready to share

This sounds as if it might simply be too early for a nomination. Let him contribute all those great things as a regular contributor and once it is accepted and merged, it should be a good time to make him a committer as well.

 that in itself does not eliminate the conflict of interest that is inherent in a committer approving his own contributions, which is basically what I have been forced to do, though it bothers me quite a bit. 

I think we all agree that no committer should be forced to approve his own contributions, so having more than one committer in the project is indeed essential. It leaves me quite puzzled though that https://projects.eclipse.org/projects/iot.cyclonedds/who lists 3 (!) project leads and 4 committers. So what exactly is wrong here that you so strongly require another committer?

Hope this clarifies the reasons for the veto and what is expected of a committer nomination.

Best regards,

Kai

On 18. Sep 2018, at 15:51, Erik Boasson <erik.boasson@xxxxxxxxxxxxxx> wrote:

Discussion regarding Election for Jeroen Koekkoek as Committer on Eclipse Cyclone DDS

https://projects.eclipse.org/projects/iot.cyclonedds/elections/election-jeroen-koekkoek-committer-eclipse-cyclone-dds

I believe there has been an unfortunate misunderstanding of the position of Jeroen that caused his nomination as a committer to have been rejected. If it is possible, I would like the decision to reject to be reconsidered. If I am wrong in sending to this list on this particular issue, then please forgive me and direct me to the appropriate procedure instead.

Two points were raised essentially. The first is that it carried a whiff of him having been hired for the express purpose of becoming a committer. That, however, is definitely not the case.

He has been with ADLINK for several years, has worked on what is now Eclipse Cyclone DDS internally before it was contributed to Eclipse, and is the only one besides myself who has actually made substantive contributions to the project: the work on continuous integration is his, and I know from personal conversations that he is a strong supporter of the project and is currently undertaking relevant work, work that he simply hasn’t yet felt ready to share. This includes:

• FreeRTOS support — quite valuable for embedded systems, which is one of the strong points of DDS
• various improvements to the testing, as we have some issues with intermittent tests on Windows due to a problematic library we depend on
• the beginnings of proper C++ support
• assembling a list of issues that need to be addressed, so that others in the community can actually see if there is anything they can do without first digging into all the details.

On the second point, regarding the existence of proper procedures for contributions from non-committers, I agree that these procedures are in place. However, that in itself does not eliminate the conflict of interest that is inherent in a committer approving his own contributions, which is basically what I have been forced to do, though it bothers me quite a bit. Having an active co-committer would solve that. In my personal opinion and having known Jeroen for years, he is someone who will take on that role in a meaningful manner.

I am aware that there are a number of other committers, but they have not taken that active role, as the record shows. Still, they have voted in favour. Finally, outside this group of people, the community really is but nascent, with only a few people having asked questions so far and only one truly external contribution.

Given all that, I feel my vote should carry some weight, and it is fully behind Jeroen becoming a committer.

Best regards,

Erik Boasson

_______________________________________________
iot-pmc mailing list
iot-pmc@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/iot-pmc

_______________________________________________
iot-pmc mailing list
iot-pmc@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/iot-pmc

--

Jens Reimann
Senior Software Engineer / EMEA ENG Middleware
Werner-von-Siemens-Ring 14
85630 Grasbrunn
Germany
phone: +49 89 2050 71286
_____________________________________________________________________________

Red Hat GmbH, www.de.redhat.com,
Registered seat: Grasbrunn, Commercial register: Amtsgericht Muenchen, HRB 153243,
Managing Directors: Paul Argiry, Charles Cachera, Michael Cunningham, Michael O'Neill

_______________________________________________
iot-pmc mailing list
iot-pmc@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/iot-pmc