Abstentions and not voting

262 views
Skip to first unread message

Dracony

unread,
Nov 24, 2015, 12:21:14 PM11/24/15
to PHP Framework Interoperability Group
(Sorry for the rant, main point in the end)

Todays Icicle vote results and some previous votes make me really wonder. 

12 votes for, 1 against and 16 abstentions out of 43 members, 14 not voted.

Effectively only 1/4 of the members were in favor and that was enough for the vote to pass ( not that I wouldn't like the vote to go that way, since I voted +1 myself), but the number is a bit baffling.

Why request a membership, look for a sponsor, go through the process to get a permission to vote, if you are not going to vote at all?
I know it's now the job of the secretary to monitor who is actually voting or not, but it won't really change anything because of the +0 option.
If you press the non-voters you'll just get more +0.

A +0 vote is either an "I dont care and have no time to look into this" or "I don't support it, but don't want to offend anyone". If it's the former than the question remains of why bother requesting membership in the first place. If it's the latter than the entire voting system is kind of broken. 

It is now possible to pass a vote with a single +1 and a whole bunch of abstained votes, which is really weird to me. If only one person supports the topic and everybody else doesn't care it shouldn't pass really.

Not voting at all has basically become the new -1, since if enough people don't vote at all the quorun will not pass.


Main Point on fixing this

There are two ways essentially:

1. Drop the +0 vote. Don't let "I don't care" count towards a quorum.
2. Consider a +0 as if the person has not voted. The +0 still counts towards a quorum but you cannot use the +0 more than a set amount of times in a row.



Kinn Julião

unread,
Nov 24, 2015, 12:36:48 PM11/24/15
to php...@googlegroups.com
Just one point here: +0 isn't "I don't care" but "I support the team's decision"... that's when you work as a team, and as far as I know FIG is a team...
Dropping the +0 is taking out the option of support the team's decision and enforce the "agreement" or "denial"... this is not binary, neither a coin with 2 sides... this is a community and its voters must have the option to "I'm not sure, but I do support the team's decision".

The math shouldn't be made based on voting members, but on votes! so you count +1, -1 and +0.
And in reality, you're taking in consideration the vote for new members... but +0 is pretty useful on votes for implementations and new set of rules.

When you don't vote at all, you really don't care about the thread... when you vote +0, you do care about the thread but you don't have any opinion about it...

No need to be sorry... by your text I can't see if you're breaking the keyboard while ranting or laughing. 

--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/php-fig/1f22f30e-9612-406b-b5a9-4e52db211d81%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.



--
---------------------------------------------------
             Kinn Coelho Julião
          
               About.me/kinncj
---------------------------------------------------

Roman Tsjupa

unread,
Nov 24, 2015, 12:42:57 PM11/24/15
to FIG, PHP
I get it,  hence my option #2. The point of it is that if you are going to use a +0 too often you don't really need a vote at all, right?


--
You received this message because you are subscribed to a topic in the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/php-fig/Undf02UPb0U/unsubscribe.
To unsubscribe from this group and all its topics, send an email to php-fig+u...@googlegroups.com.

To post to this group, send email to php...@googlegroups.com.

guilher...@gmail.com

unread,
Nov 24, 2015, 12:43:52 PM11/24/15
to php...@googlegroups.com
For the last time, it's not up to anyone decide in binary for any single proposal or membership.
+0 can be "I don't care", but can also be "I support team decision", "No opinion formulated", "Couldn't get all core developers of my project to vote within a reasonable amount of time", etc

All I can say is... too much bureaucracy. That's what FIG is now.

[],


For more options, visit https://groups.google.com/d/optout.



--
Guilherme Blanco
MSN: guilher...@hotmail.com
GTalk: guilhermeblanco
Toronto - ON/Canada

will.g...@couchbase.com

unread,
Nov 24, 2015, 12:50:58 PM11/24/15
to PHP Framework Interoperability Group
Glad you brought this up as it did come across as a bit off.

My personal belief is that a minor change of requiring a quorum of '+1' instead of a quorum in general would work. There's actually currently a somewhat silly loophole at the moment where a '-1' vote can cause a vote to pass by causing it to exceed the quorum.

Larry Garfield

unread,
Nov 24, 2015, 1:46:09 PM11/24/15
to php...@googlegroups.com
I've seen similar patterns in other organizations; I know for a fact
that in some cases (and I suspect without hard data that in most cases)
+0s on membership is exactly "I want to vote no but not offend anyone or
be 'that guy'".

My gut feeling is that it's mainly an issue on membership votes; there's
less "offend people concern" on a bylaw or a spec, where what is being
voted on is a string of text, then on a membership vote when the person
really is what's being voted on.

So perhaps simply disallow "Present" votes for membership?

--Larry Garfield
> *Main Point on fixing this*
> *
> *
> There are two ways essentially:
>
> 1. Drop the +0 vote. Don't let "I don't care" count towards a quorum.
> 2. Consider a +0 as if the person has not voted. The +0 still
> counts towards a quorum but you cannot use the +0 more than a set
> amount of times in a row.
> *
> *
> *
> *
>
> --
> You received this message because you are subscribed to the Google
> Groups "PHP Framework Interoperability Group" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to php-fig+u...@googlegroups.com
> <mailto:php-fig+u...@googlegroups.com>.
> To post to this group, send email to php...@googlegroups.com
> <mailto:php...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/php-fig/fac70624-0f64-4530-870b-0692bc8563cf%40googlegroups.com
> <https://groups.google.com/d/msgid/php-fig/fac70624-0f64-4530-870b-0692bc8563cf%40googlegroups.com?utm_medium=email&utm_source=footer>.
> For more options, visit https://groups.google.com/d/optout.

--
--Larry Garfield

Michael Cullum

unread,
Nov 24, 2015, 1:48:45 PM11/24/15
to FIG, PHP
You raise an interesting constitutional point, made more interesting by taking it to the extreme that in theory, a vote can pass with, in effect, a single +1 vote, or potentially more controversially two +1 and one -1 vote. 

This came up on IRC recently and the only realistic way to fix this problem with our voting is to instead require that over half of all members vote +1 or complicate the matter with something along the lines of: To pass 1/3 of all members must vote +1, have a majority of those who do vote voting +1, and require a separate quorum including +1s of either 1/3 or of all members or 1/2 of all members.

If we look back historically this is a statistic that was on my voting sheet (When we have secretaries this will probably be brought up to date) https://docs.google.com/spreadsheets/d/1_6K1Qswau_5WqVUIz8ZVcYkdsGhglt10n15EvwALCJ8/edit?usp=sharing not all votes achieved a majority of the entire membership, but most of them did. And most that didn't get a majority of members voting +1 had a smaller number of people who voted. If it were to be changed in this way, it would become imperative that members vote regularly, much more so than the quorum of 1/3 than we have now.

Saying this, I think things are best as they are as it's simple and as Guilherme pointed out, it doesn't mean "I don't care", it can mean "I don't mind" or "I support the team decision", both of which have rather different meanings.

On your comment that non-voters are more likely to vote +/-0, I think that from what I've seen in the past past when we've prodded people to vote, they're more likely to vote +1 or -1 'with the crowd' than anything else.

--
Michael Cullum

--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/php-fig/fac70624-0f64-4530-870b-0692bc8563cf%40googlegroups.com.

guilher...@gmail.com

unread,
Nov 24, 2015, 2:02:11 PM11/24/15
to php...@googlegroups.com
Also keep in mind some doesn't even apply to the project at all.
For example PSR-7 would never apply for Doctrine, so we'll vote +0 (or not vote at all).

Abstaining or voting 0 doesn't mean anything to me tbh.

[]s,


For more options, visit https://groups.google.com/d/optout.

Matteo Beccati

unread,
Nov 24, 2015, 2:11:12 PM11/24/15
to php...@googlegroups.com
Hi,

On 24/11/2015 19:46, Larry Garfield wrote:
> I've seen similar patterns in other organizations; I know for a fact
> that in some cases (and I suspect without hard data that in most cases)
> +0s on membership is exactly "I want to vote no but not offend anyone or
> be 'that guy'".
>
> My gut feeling is that it's mainly an issue on membership votes; there's
> less "offend people concern" on a bylaw or a spec, where what is being
> voted on is a string of text, then on a membership vote when the person
> really is what's being voted on.
>
> So perhaps simply disallow "Present" votes for membership?

I agree that the problem comes from presence votes. Maybe we could allow
projects to explicitly abstain. Their non-vote would count as presence
token, but not increase the quorum.


Cheers
--
Matteo Beccati

Development & Consulting - http://www.beccati.com/

Korvin Szanto

unread,
Nov 24, 2015, 2:48:34 PM11/24/15
to php...@googlegroups.com
Should we make participation required? It seems like the FIG is a big enough group at this point that businesses are trying to join just so that they can say that they are a member of the great PHP-FIG. I think we might be able to afford to make membership require more involvement than just showing up for every 5th vote.
I don't think that it's unreasonable to say that if you're a member project, you need to participate in the discussions and not just the votes because we are a group that creates recommendations, we aren't a group assembled simply to vote.
I think if we made membership require more involvement, we could thin down the list of member projects to just the ones we see actively participating in the group. And we'll see the abstain votes less and less.

Thoughts?
  

Cheers
--
Matteo Beccati

Development & Consulting - http://www.beccati.com/

--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/php-fig/5654B63F.7060607%40beccati.com.

Larry Garfield

unread,
Nov 24, 2015, 2:59:37 PM11/24/15
to php...@googlegroups.com
That assumes a correlation between abstention and general inactivity. I
don't know if that's true.

Also, there are legitimately reasons why a given person or project rep
wouldn't have much to offer. There's very little that I can offer to
the event loop PSR, for instance, other than moral support. And if
someone is opposed to a given PSR in concept, there's little for them to
offer that wouldn't just be trolling. :-)

--
--Larry Garfield

Robert Hafner

unread,
Nov 24, 2015, 3:03:54 PM11/24/15
to php...@googlegroups.com
This is the real issue- we have a lot of members and very little in the way of member responsibility or filtering. I've only been +0 voting on new members because I think this group has grown entirely too unwieldy.

Rob 

Korvin Szanto

unread,
Nov 24, 2015, 3:51:38 PM11/24/15
to php...@googlegroups.com
I guess I see the real issue being the ratio of real votes to abstains and no-votes. It'd be better to say that we'll see people abstaining for no good reason less and less.
 
Also, there are legitimately reasons why a given person or project rep
wouldn't have much to offer.  There's very little that I can offer to
the event loop PSR, for instance, other than moral support.  And if
someone is opposed to a given PSR in concept, there's little for them to
offer that wouldn't just be trolling. :-)

That's fair.
With a secretary and better tracking of what's happening, we can better distinguish which projects are contributing and which aren't. And when the secretary calls out Drupal because you haven't been active for a year, you'll be available to make your case pretty quickly. I suspect that we have member projects who would not respond if called out in a thread due to not tracking on the mailing list at all.

Thanks for the response!
Korvin


--
--Larry Garfield


--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.

Dracony

unread,
Nov 24, 2015, 4:03:44 PM11/24/15
to PHP Framework Interoperability Group
And what then? Force them to vote ? They might as well just vote +1 on everything then.

Or expelled them ? But voting for that with the same system would definitely be victim to the "not offend anyone" votes.


The third option I can propose to the original ones are maybe anonymous membership votes ? We really need a voting system built. Then the secretary would also have way of an easier job

Korvin Szanto

unread,
Nov 24, 2015, 4:09:58 PM11/24/15
to php...@googlegroups.com
On Tue, Nov 24, 2015 at 1:03 PM Dracony <draco...@gmail.com> wrote:
And what then? Force them to vote ? They might as well just vote +1 on everything then.

No, I think then we ask them to start participating in discussion or resign from the group. Forcing them to vote is only going to give us more +0's, forcing them to be involved in discussion will make it much harder to not be involved in the group and I'd imagine will make it much easier to actually review what is being voted on.
 

Or expelled them ? But voting for that with the same system would definitely be victim to the "not offend anyone" votes.

 
If they don't start participating or maybe if they participate but fall off again, we call a vote to expell. Maybe this is just a yearly audit that we do, maybe a "spring cleaning" if you will.
 

The third option I can propose to the original ones are maybe anonymous membership votes ? We really need a voting system built. Then the secretary would also have way of an easier job
 
I think we need to keep representatives accountable with their communities, I agree that a voting platform would be awesome.
 
--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.

Robert Hafner

unread,
Nov 25, 2015, 4:33:01 AM11/25/15
to php...@googlegroups.com
I don’t think forcing people to email the list more as a gauge of participation is a good idea. We don’t want people writing emails that add nothing just for the sake of staying in the group. 

I’d like to step back and ask a question- what is it we are trying to solve? Is it that making quorum is too hard (or too easy)? That people are in the group for vanity/business reasons, rather than support of the group? Is the number of members itself a problem in that it slows down development?

Each of those problems will have very different solutions. We can change the quorum algorithm for the first, place restrictions and membership requirements in the second, and adopt different workflows for the third. However, we aren’t going to do any of those things until we decide on what we’re looking for in an outcome, which means defining what problem we’re trying to solve.

I think a big problem that is preventing this group from being as effective as it could be is that we simply do not define the problems we are trying to solve well enough (and I mean the meta issues of how the group functions perhaps more than I do the PSRs themselves), and this leads to a lot of circular discussions.

Rob


Jordi Boggiano

unread,
Nov 25, 2015, 5:49:13 AM11/25/15
to php...@googlegroups.com
The main reason I am not voting on most membership threads is that I
kinda lost interest in new memberships.. I don't really see the point,
we have enough people that no vote will pass just thanks to a little
clique of people colluding.

And we generally don't have very controversial votes anyway, either it's
a clear pass or a clear fail as there is quite a lot of discussion going
on the consensus is usually pretty high.

Having more people involved and reviewing and discussing PSRs? Cool.
Having a few membership threads a week where we re-hash the same
arguments and then everyone votes +0 because it doesn't really matter in
the end? Not so useful.

I think the main reason people want in is to be on The List (tm), so
perhaps we just need a more open list of projects that adhere to the PSR
ideals and try to follow them? So projects could just register
themselves there and we don't need a long debate about it.

Then only people with a strong and proven interest in participating can
request membership and we don't need to have big discussions anymore, we
can just summarily reject people that didn't participate yet because
they can still be on the list, just not have voting rights?

Cheers

--
Jordi Boggiano
@seldaek - http://seld.be

Larry Garfield

unread,
Nov 25, 2015, 11:47:44 AM11/25/15
to php...@googlegroups.com
A "known implementing projects" list is an interesting idea. The
challenge there is that it becomes less maintainable the more successful
we are. :-) There are tens of thousands of projects on Packagist that
depend on PSR-3, for instance. Obviously we can't list all of them.
PSR-2 is probably even more, if we allow for some fuzziness, and via
Composer I don't know that anyone that matters isn't using PSR-0/4 by
now, except WordPress[1]. So how would we manage this "supporting
projects" list that is distinct from a "membership list"?

[1] By which I mean, WordPress does matter but they're not using a
PSR-based autoloader.

--
--Larry Garfield

Lukas Smith

unread,
Nov 25, 2015, 2:33:24 PM11/25/15
to php...@googlegroups.com
Aloha,

honestly I have not read this entire thread but lets not try an optimize an edge case here. if people vote +0 where they want to vote -1, then obviously the concern was not too big. but creating yet another voting process for this seems overboard to me.

regards
Lukas

Evert Pot

unread,
Nov 26, 2015, 12:19:49 PM11/26/15
to php...@googlegroups.com
I agree with this too. I always felt that people who actively work on
PSR's and contribute should get a vote down the road. Don't give people
votes just because they do something with PHP.

I abstained from this vote because there was no clear consensus. There
is a big difference between a +0 and consciously abstaining. I didn't
want the vote to come through if there were not enough people who felt
that it should.

I think from this point forward I will -always- reply to member voting
threads with 'I abstain', unless there's been a precedent of
contributions for a fair amount of time. More votes slow us down and
will only increase the need for more bureaucracy.

Evert
Reply all
Reply to author
Forward
Message has been deleted
0 new messages