Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [ee4j-community] [eclipse.org-membership-at-large] Proposed EE.next Working Group

While I had concerns initially as a representative of a small “vendor” so to speak.

 

I would encourage people to read and digest the FAQ https://www.eclipse.org/org/workinggroups/eclipse_ee_next_faq.php that goes along with the charter to place the working group committees in context vis a vis the actual projects and the role of committers on projects in setting technical direction.

 

Steve

 

From: ee4j-community-bounces@xxxxxxxxxxx [mailto:ee4j-community-bounces@xxxxxxxxxxx] On Behalf Of Mariano Amar
Sent: 07 February 2018 22:19
To: EE4J community discussions <ee4j-community@xxxxxxxxxxx>
Subject: Re: [ee4j-community] [eclipse.org-membership-at-large] Proposed EE.next Working Group

 

While a bit negative, I have to admit Markus' response hits the nail on the head. 

The charter, as is, is very heavily skewed towards the big industrial vendors (or equivalent).

 

Just from memory, we can easily accept that there will be at least a half dozen strategic members (probably more), and a dozen influencers.

 

With the current charter, this means:

6 Strategic Members, 2+ Influencers, 1+ Participants, and 1+ Commiters per Committee.

That means every single non-strategic member could vote on something non-major, and they would be irrelevant if the strategic members generally agree on the topic.

Meanwhile, depending on how you calculate 2/3rds of 10 votes, a major issue could still be decided on by just the strategic members. At best, they'd only need one extra vote.

 

It's even worse with the Enterprise Requirements Committee, where you'd get between one and two dozen members between strategics and influencers, with zero input from the rest of the community.

 

On the other hand, the closer we get to direct democracy on this issue, the more probable everything will be ground to a halt because of all the dissonant opinions.

 

I can also understand why there's no committer participation in the last committee. The enterprise requirements are the domain of enterprise actors.
That still doesn't explain why there's no participation from participant members, which is where most of the small companies and JUGs will be. 

I'll even posit that participant members are as important for that committee as strategic members, since they're the two sides of the enterprise coin: The vendors and the enterprise users. 

The enterprise requirements should not be left at the whim of what the vendors want to provide the rest of the world every year (just look at how Oracle pretty much dropped Java EE for Fn). The people that is going to use the tech at enterprise levels should have a say on what their requirements are. That's pretty much the point of this.

 

 

Now, leaving that specific committee aside, we're still left with a bad situation on the other three.

My knee-jerk reaction is to suggest/demand an equal representation for each member group. That would mean one of two solutions:

1) Reduce the amount of strategic members allowed into each committee, then turn those positions into elected ones too.

3 strategics, 3 influencers, 2 participants, and 2 committers, for example, would still give a bit of extra power to the people investing more resources, but it would not neuter the rest of the committee.

Forbid any member from occupying multiple seats, and have strategic members choose where they'll fall on each committee.

 

2) Increase the amount of elected seats until each group is relatively equal to strategic members. Maybe difference of one or two seats between groups if you consider that representation shouldn't be 100% equal.

6 strategics, 5 influencers, 4 participants, and 4 committers, for example (following the same weights as the other solution), would also keep more power in the strategics, but keep the rest of the committee relevant.

Raise or lower the elected seat amounts every year depending on the amount of strategic members and you're set.

 

Now, both solutions would require the two biggest member groups AND at last one seat from the other two in accord before anything major is passed, and pretty much two full groups agreeing over something non-major.

 

Now, I understand both solutions have big cons that make them unpalatable, so I won't push for any of them.

However, the general idea is still sound: Equalize power amongst the groups.

 

* Do we really need EVERY SINGLE strategic member to have influence on every single committee? Why?

I find it just as ridiculous as asking for direct democracy, where the committer group would pretty much stomp on every other group (and we'd probably end up with a bajillion forks of every project, one per committer).

 

* While I can see why the voting representation should not be equally weighted (where each seat represents the exact same amount of members), why is it that non-paying members are pretty much dismissed?

 

 

In the end, this group charter is pretty much the exact same working of the JCP, just without Oracle as the God-King over all.

We know it does work, but the point WAS to avoid the same pitfalls as before, and this is only removing a single issue (Oracle at the helm), while keeping all the others.

 

 

 

Now, to complete my review on this, since I've already written one of my "politely ranting text-walls":

I like how the committees are structured, how the power is divided, how the process is handled, etc.

My only gripes with the charter (outside of the seats) are:

 

1) Good Standing: There's a bit of a silly hole in this paragraph, where "[...] being in Good Standing [...]" is defined as (paraphrasing) "[...] having attended 3 of the last 4 meetings, if there have been at least four [...]". 

What happens for the first three meetings of the year? 

Is nobody considered to be in Good Standing (and thus no vote is possible until the fourth meeting)? 

Are all participants considered in Good Standing, even if they've missed two of the three meetings?

I understand it'll be that any participant of the first meeting can vote, but the next two meetings are up in the air. I'd suggest being a bit more explicit there.

 

2) Super Majority: Please, please, PLEASE define what the quorum will be. Even if it's just on relative values (Minimum majority, 2/3rds, 3/4ths, etc).

And, if at all possible, please avoid using minimum majority. In a 10 members committee, that would mean 6 seats present. At current numbers, that's around the amount of strategic seats.

This means we could just have a half dozen big companies attend a meeting by themselves, then just need 4 of them to agree on something before a major change is set in stone.

I'd even posit that, if there's a scheduled meeting, and only half the seats attend (after taking into account last minute representative changes, and proxy voters), then something is wrong with the committee, or the meeting was set at an unreasonable date.

I can understand provisoes for one or two seats missing the meeting in a dozen seats committee. It happens all the time.

I cannot agree to being able to affect major changes to policy if two to tree member groups out of four are missing from the meeting.

 

 

And that's it for my review of the charter.

There's some minor details missing, but I'm pretty sure they're actually covered by one of the other applicable documents (things like more exact definitions on "applicable" and "qualified" when dealing with elections, etc)

 


Mariano Amar

Senior Consultant

email/hangouts: mariano.amar@xxxxxxxxxx
skype: marianoamar

AVISO DE CONFIDENCIALIDAD DE CORREO ELECTRÓNICO

Esta comunicación contiene información que es confidencial y también puede contener información privilegiada. Es para uso exclusivo del destinatario. Si usted no es el destinatario tenga en cuenta que cualquier distribución, copia o uso de esta comunicación o la información que contiene está estrictamente prohibida. Si usted ha recibido esta comunicación por error por favor notifíquelo por correo electrónico(info@xxxxxxxxxx) o por teléfono (+54 11 3249 7503)

This communication contains information that is confidential and may also be privileged. It is for the exclusive use of the recipient. If you are not the intended note that any distribution, copying or use of this communication or the information it contains is strictly prohibited. If you have received this communication in error please notify us by email(info@xxxxxxxxxx) or phone (+54 11 3249 7503)

 

 

On Wed, Feb 7, 2018 at 5:57 PM, Markus KARG <markus@xxxxxxxxxxxxxxx> wrote:

This will not change the fact that committers have less votes than industry.
-Markus

-----Original Message-----
From: ee4j-community-bounces@xxxxxxxxxxx
[mailto:ee4j-community-bounces@xxxxxxxxxxx] On Behalf Of Mike Milinkovich
Sent: Dienstag, 6. Februar 2018 18:25
To: ee4j-community@xxxxxxxxxxx
Subject: Re: [ee4j-community] [eclipse.org-membership-at-large] Proposed
EE.next Working Group

On 2018-02-06 12:07 PM, Werner Keil wrote:
> https://www.eclipse.org/membership/exploreMembership.php#tab-associate
> shows, LJC, SouJava and other not-for-profit organizations like the
> Meruvian Foundation by the late Frans Thamura are Associate Members
> and even they (not to mention Individuals unless they pay at least
> 1500$ to become a Solutions Member) would be prohibited.

We will discuss the idea of including JUGs. I've seen first hand what a
force for good LJC and SouJava have been on the JCP.

Individuals can participate at no cost if they are committers. Note that we
anticipate that the individuals involved in working on a spec will all be
committers on that spec project. So if you're on the future equivalent of an
expert group, you will be able to participate in EE.next as a committer
member.

--
Mike Milinkovich
mike.milinkovich@xxxxxxxxxxxxxxxxxxxxxx
(m) +1.613.220.3223

_______________________________________________
ee4j-community mailing list
ee4j-community@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from
this list, visit
https://dev.eclipse.org/mailman/listinfo/ee4j-community

_______________________________________________
ee4j-community mailing list
ee4j-community@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/ee4j-community

 


Back to the top