Em janeiro 12, 2018 17:14 NicoHood escreveu:

Just to be sure: YES = Delete NO = Keep

Correct?

We had this discussion and more on the IRC channel. But you're correct on your assumption, YES mean TU removal. No, to keep them.

Regards, Giancarlo Razzolini

On 17.01.2018 19:51, Balló György via aur-general wrote:

The voting is over. Results:

Yes: 33 No: 5 Abstain: 5

This means that speps is no longer a Trusted User.

-- György Balló Trusted User

I'm no longer a TU so I can't see how active both speps and faidoc have been regarding participation in the votes. Yet the TU-Bylaws are pretty strict and given that Bluewind/Florian pointed out during the discussion period that both TUs had participated in proposal 99 that started on December 18 2017 which is in my TZ now exactly one month ago. Therefore the second requirement, to NOT do any special action on the AUR requiring TU privileges is not fulfilled, as participating in votes is exactly one of these TU privileges. Also the alternative option with inactivity due to non-participation in votes has been invalidated by this participation. Unless of course there have been six such votes before, again I can't see that nor if they themselves participated in the votes for proposal 100 and 101.

So I question whether both votes 100 and 101 have been valid and should have been created at all as the requirements for a removal vote defined in the TU-Bylaws have not been fulfilled.

The Bylaws don't exist without reason and these actions were initiated with references to them so I assume that we agree that they need to be respected as they currently are. If anyone sees the requirement to change the bylaws a new thread for a discussion about changing the bylaws can be started and depending on how this comes to an end it's possible to repeat such votes for faico and speps.

Cheers, Thorsten, who resigned September 2017

On 18.01.2018 01:43, Balló György via aur-general wrote:

On 18.01.2018 00.18, Thorsten Toepper via aur-general wrote:

...

Therefore the second requirement, to NOT do any special action on the AUR requiring TU privileges is not fulfilled, as participating in votes is exactly one of these TU privileges.

This is unclear in the bylaws. I assumed that AUR is the repository where the user-contributed packages are stored. Therefore the actions that require AUR privileges are: disown/merge/delete a package. These actions are tracked on the aur-requests mailing list only if a request was filed before the action, so we don't really know the last privileged action of a TU. I think it should be implemented within the AUR web interface.

Well, from my point of view being a TU is not solely about package management or taking care of what users posted about the packages stored in the AUR. It's also about what you did here with this action, keeping the group of users who have a higher influence on what happens with Arch Linux (be it what ends up in [community] or is dropped again, management of users in the AUR system) in a good shape.

Please don't misunderstand my mail from last night it's no "WHAT YOU DID WAS WRONG DON'T DO THAT EVER AGAIN!!!". I consider it great that you initiated these proposals, I just feel uneasy about the point that Florian pointing out for both that they still participate in the social part of being a TU. We wouldn't have this discussion if there had been a second person initiating this with you and the "normal" removal had been applied. Your arguments are fine, they are objective and there is no personal affection visible, so this should have sufficed.

As the special removal has been applied the votes need to be considered in both blocks, the one before the big OR and the one afterwards. My position is that in the first block the second point, has been fulfilled due to participation in the social part of being a TU and this also applied for the second block.

I can't remember how many years it's been since there was the last discussion that lead to this rule and I won't invest time (the lack of that made step down) searching the office, but what I [think to] remember pretty well that the problem point back then was people having the special privileges of being a TU didn't participate in votes which lead to failures due to quorums that couldn't be reached without their participation. And there were also discussions in case applications overlapped how in case the first applicant becomes a TU will be counted into the quorum of the second application. As vote participation was one of the central points back then I consider the votes to be an important factor if this "method" is chosen to be applied.

That's why I proposed at the end of last nights mail to consider a overhaul of the wording of the definition to make the definition more strict and reduce the room of interpretation. With a second vote afterwards there wouldn't be anything to question.

So again, this is no attack in any form, last night I just remembered old discussions in public and behind the curtains in the IRC about the problem with the quorums and that there was the need of a rule to strip TUs of their special status in case they ignored this social aspect of their task. And therefore I think that if that rule is chosen, the historic background should not be forgotten.

And again: It's good you looked at the status of all TUs and took action. :-)

Cheers, Thorsten

On 18.01.2018 02:11, Eli Schwartz via aur-general wrote:

On 01/17/2018 06:18 PM, Thorsten Toepper wrote:

...

I'm no longer a TU so I can't see how active both speps and faidoc have been regarding participation in the votes. Yet the TU-Bylaws are pretty strict and given that Bluewind/Florian pointed out during the discussion period that both TUs had participated in proposal 99 that started on December 18 2017 which is in my TZ now exactly one month ago. Therefore the second requirement, to NOT do any special action on the AUR requiring TU privileges is not fulfilled, as participating in votes is exactly one of these TU privileges. Also the alternative option with inactivity due to non-participation in votes has been invalidated by this participation. Unless of course there have been six such votes before, again I can't see that nor if they themselves participated in the votes for proposal 100 and 101.

"as participating in votes is exactly one of these TU privileges"

This does seem to be unclear in the Bylaws. As someone mentioned on IRC:

...

In my mind aurweb just happens to be a convenient place for the vote functionality to be located, but isn't actually part of the AUR.

Does voting, an action which doesn't seem to have a lot to do with *being* a TU, merely deciding who should be allowed to do so, and is more or less invisible to the community, constitute a "TU privilege" on the AUR?

Looking at the context of previous amendment discussions, it seems like the Special Removal was motivated by the desire to remove TUs who

EITHER block quorum by failing to vote,

OR fail to enhance the AUR as TUs are intended to do, by such qualifying actions as:

• elevating packages to [community]
• contributing to the AUR as a good example
• moderating the package list or users
• participated in general discussion about the AUR on this list

To that end, as bgyorgy said, the "Arch User Repository" would be "that which the regular users interact with to upload package recipes", and totally unrelated to an administrative voting interface which is only implemented in aurweb (not the AUR) insomuch as it would be bloat to host it separately.

First of all: Congratulations for becoming a TU, you should have applied sooner so I could've given you my "yes". :-)

As I just wrote in the other mail and as we all seem to agree the social management is also an important task a TU needs to do. As you say, the historic background was the problem that TUs being active in the formal part of being a TU by actively contributing to package management, but ignoring the social part could not be actively "motivated" so it was necessary to redefine the rules for reaching a valid quorum and getting rid of people not taking the social part serious.

Whether the votes happens in the AUR web interface or on a separate private mailing list is unimportant for the real process I agree, just at the moment it's the AUR webinterface and the second point is simply not too well formulated, in it's current form this also includes the votes and therefore in case someone participated in a vote both blocks are neglected. So the "either" "or" doesn't really work here.

Rephrasing the bylaw to something like

"performed any action that required TU privileges on the AUR (*excluding* participation in votes for there is below rule)"

would solve the problem, that TUs may only act in the absolute background of participation votes, yet not taking care of any binary packages or AUR package management any more.

As I wrote in the other mail I fully support Gjörgy's action to check all TUs and initiate votes, it's just that the rule that was chosen to be applied has this loophole with the participation in votes. And it should be considered to fix this.

Two TUs initiating the "normal" process based solely on arguments would've also been fine.

Cheers, Thorsten

PS: Sorry in case there are too many words missing in sentences, as I'm very tired I only proof-read once so I most likely missed several occasions.

On 18.01.2018 22:11, Eli Schwartz via aur-general wrote:

On 01/18/2018 04:05 PM, Thorsten Toepper wrote:

...

First of all: Congratulations for becoming a TU, you should have applied sooner so I could've given you my "yes". :-)

Thanks! :)

...

Rephrasing the bylaw to something like

"performed any action that required TU privileges on the AUR (*excluding* participation in votes for there is below rule)"

would solve the problem, that TUs may only act in the absolute background of participation votes, yet not taking care of any binary packages or AUR package management any more.

I would definitely agree that we need to clarify the bylaws on this point, and I'll probably propose something soon.

I'm not sure what that means we should do about the current vote though.

My idea last night was that the unquestionable move would be to declare the current results as something like "invalid"/"to be ignored" and after the bylaw has been adjusted to repeat them. I can't see the state of the quorum for the recent proposals, but as long as that wasn't in any danger I'd say there is no real urge as everyone knows (or should know) that the proposals and votes will be repeated afterwards in case the situation does not change and speps or faidoc become more active again.

Alternatives that currently come to my mind are to either start proposals based on the "normal" removal bylaw.

Or, the probably easier approach, that there is a public agreement of multiple TUs on this list that the voting results are ignored until an adjustment of the rule has been accepted and applied. In case they have not become more active until the change becoming accepted the results are acknowledged in retro perspective. This would give them themselves a chance to act by either becoming more active, explaining what happened or simply resigning and save all other TUs some time.

And now it's time for bed. Good night. :-)

On Thu, 18 Jan 2018 22:05:10 +0100 Thorsten Toepper atsutane@freethoughts.de wrote:

Whether the votes happens in the AUR web interface or on a separate private mailing list is unimportant for the real process I agree, just at the moment it's the AUR webinterface and the second point is simply not too well formulated, in it's current form this also includes the votes and therefore in case someone participated in a vote both blocks are neglected. So the "either" "or" doesn't really work here.

Yes, it is a bit ambiguous. The discussion in #archlinux-tu concluded that the voting being an the AUR was just happenstance and intent of the section was that voting not be included in point 2. With many/most of the most active TUs participating or present for that discussion, I would conclude that the general understanding of this section was followed in this case and the motions have passed.

On Sun, 21 Jan 2018 at 04:07:06, Xyne wrote:

Doug Newgard via aur-general wrote:

...

Yes, it is a bit ambiguous. The discussion in #archlinux-tu concluded that the voting being an the AUR was just happenstance and intent of the section was that voting not be included in point 2. With many/most of the most active TUs participating or present for that discussion, I would conclude that the general understanding of this section was followed in this case and the motions have passed.

I disagree. The intent of the first sectionm before the "OR", is to measure any sort of activity. Updating a package, voting or posting a comment shows that the TU is still logging in to the AUR and thus active in some sense. The point of the first section was to provide a way to remove TUs who had simply disappeared. This is as it should be. There is no mandated TU quota for package actions.

I find it ridiculous to call Trusted Users active ("in some sense") if all they do is vote. The actual job of a Trusted User is to maintain the AUR and the [community] repository. Imagine a world where all Trusted Users would do nothing but add/remove new Trusted Users; neither the AUR nor [community] are touched by anyone. Would you call such a group of Trusted Users active? I doubt so.

Also, as I already mentioned in another reply, the intent of the current statement in the bylaws is quite clear: voting should not be considered as some sort of activity in the first section before the "OR". If you count voting as activity, the condition "not active OR not voting" for special removal makes no sense: "voting" implies "active", so "not active" implies "not voting" and the statement "not active OR not voting" is equivalent to "not voting". This means that the whole section before the "OR" is unnecessary. I do not think we would have voted for an amendment of the bylaws adding unnecessary junk.

I really hope we will not start reading and writing our bylaws like pedantic lawyers, where every single formulation has to be chosen very carefully. Voting is just a tiny part of the things we do, our main focus should be on improving Arch Linux as a distribution.

Regards, Lukas